Sie sind auf Seite 1von 315

European Advertising Academy

George Christodoulides
Anastasia Stathopoulou
Martin Eisend Editors

Advances in
Advertising
Research (Vol. VII)
Bridging the Gap between Advertising
Academia and Practice
European Advertising Academy

Executive Board Members:


S. Okazaki, London, United Kingdom
P. De Pelsmacker, Antwerp, Belgium
H. Voorveld, Amsterdam, The Netherlands
M. Eisend, Frankfurt an der Oder, Germany
S. Diehl, Klagenfurt, Austria
T. Langner, Wuppertal, Germany
S. Rosengren, Stockholm, Sweden
S. Boerman, Amsterdam, The Netherlands
The objective of the European Advertising Academy (EAA) is to provide a profes­
sional association to academics and practitioners interested in advertising and its
applications that will promote, disseminate and stimulate high quality research in
the field.

Executive Board Members:


Prof. Shintaro Okazaki Prof. Tobias Langner
King’s College, United Kingdom Bergische Universität Wuppertal
Germany
Prof. Patrick De Pelsmacker
University of Antwerp, Belgium Prof. Sara Rosengren
Stockholm School of Economics
Prof. Hilde Voorveld Sweden
University of Amsterdam
The Netherlands Dr. Sophie Boerman
University of Amsterdam
Prof. Martin Eisend The Netherlands
Europa-Universität Viadrina
Frankfurt an der Oder, Germany

Prof. Sandra Diehl


Alpen-Adria-Universität Klagenfurt
Austria
George Christodoulides
Anastasia ­Stathopoulou · Martin Eisend
(Eds.)

Advances in
Advertising
Research (Vol. VII)
Bridging the Gap between Advertising
Academia and Practice
Editors
George Christodoulides Martin Eisend
London, UK Frankfurt (Oder), Germany

Anastasia Stathopoulou
London, UK

European Advertising Academy


ISBN 978-3-658-15219-2 ISBN 978-3-658-15220-8  (eBook)
DOI 10.1007/978-3-658-15220-8

Library of Congress Control Number: 2016948296

Springer Gabler
© Springer Fachmedien Wiesbaden 2017
This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part
of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations,
recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission
or information storage and retrieval, electronic adaptation, computer software, or by similar or
dissimilar methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this
publication does not imply, even in the absence of a specific statement, that such names are exempt
from the relevant protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this
book are believed to be true and accurate at the date of publication. Neither the publisher nor the
authors or the editors give a warranty, express or implied, with respect to the material contained
herein or for any errors or omissions that may have been made.

Printed on acid-free paper

This Springer Gabler imprint is published by Springer Nature


The registered company is Springer Fachmedien Wiesbaden GmbH
The registered company address is: Abraham-Lincoln-Strasse 46, 65189 Wiesbaden, Germany
Advances in Advertising Research: Bridging the Gap between
Advertising Academia and Practice

George Christodoulides, Anastasia Stathopoulou, and Martin Eisend

This collection of selected papers is associated with 14 th ICORIA conference


that took place at Birkbeck, University in London between July 2-4, 2015.
ICORIA is the annual meeting of the European Advertising Academy. All the
papers submitted to the conference were subject to a double blind reviewing
process. A total of 115 papers from a truly international pool of researchers
were accepted for presentation in London. From those the best papers were
invited to be developed further and be submitted to Advances in Advertising
Research Vol. VII: Bridging the Gap between Advertising Academia and
Practice.
This year’s theme is aligned with the theme of the conference and focuses on
the need to bridge the gap between advertising research and practice. It is for
this reason that all the authors were encouraged to discuss in detail the
managerial implications of their work.
We have grouped the papers in this collection in three main categories:
Online Advertising/Social Networks, Consumer Responses to Advertising and
Culture and Advertising.
Online Advertising/Social Networks is the largest section of the book
consisting of nine chapters. This reflects the huge interest from researchers in
digital, social media and mobile technologies which have undoubtedly
revolutionized the way organizations and brands communicate with consumers
(and vice versa). The first four chapters focus specifically on social networking
sites and their value in consumers’ information-seeking and content sharing
behaviors as well as collaborative consumption (e.g. with regard to New Product
Development). The next two chapters focus on corporate websites by examining
the interactivity and branding potential of websites. The subsequent two
chapters examine electronic word-of-mouth (eWOM) by investigating the
characteristics of eWOM diffusers and the effect of specific features of the
message on sales. The last chapter examines consumer’s privacy concerns in the
context of mobile apps.
Consumer Responses to Advertising
This section comprises eight chapters looking at various consumer responses
to various advertising formats and appeals. The first four chapters examine
emotional and rational appeals in marketing communications. The next three
chapters focus on consumer attitudes and behaviors instigated by various stimuli
in the advertising message or environment. The last chapter in this section deals
VI Preface

with the hot topic of engagement by putting forward a new framework to


conceptually enlighten consumer engagement with brands.
Culture and Advertising
The last section of the book consists of four chapters focusing on cultural
issues in advertising research. The first chapter provides an assessment of
Hofstede’s cultural framework in advertising research. The next chapter
examines appeals in a cross-national context. The penultimate chapter examines
‘adhocracy’ culture in the context of integrated marketing communications
whilst the last chapter profiles leading researchers in advertising by examining
their productivity in top international journals.

These three sections contain a total of 21 chapters representing the latest


cutting-edge research on advertising globally – research that covers a wide range
of topics in advertising with significant implications for managerial practice. The
chapters also provide new and innovative ideas for future research in this
exciting and fast-evolving field of study. We would like to take this opportunity
to thank the steering committee of the 14th ICORIA and everyone else at
Birkbeck, University of London and the EAA who has contributed to making
this conference a big success. We hope that you find the contents of this volume
interesting and thought-provoking in a way that will help you shape your ideas
for future research.
The objective of the association is to provide a professional association to
academics and practitioners interested in advertising and its applications that
will promote, disseminate and stimulate high quality research in the field.
The association particularly serves as a meeting and communication forum
for its members. It offers a network for the exchange of knowledge on an
international level and constitutes a framework allowing for a better
dissemination of information on research and teaching.
The association also aims at the development of relations with all other
professional and research-oriented associations which are active in the field, as
well as with European or international committees and authorities concerned
with political decision making, active in this field.
The EAA is closely related to the yearly International Conference on
Research in Advertising (ICORIA). The purpose of the conference is to create a
forum where people studying advertising in the academic world could exchange
ideas, and where they could meet with practitioners who have experience with
advertising in the commercial world.
Every natural person that is professionally concerned with or interested in
research or teaching in the field of advertising is, irrespective of nationality,
eligible to become a full member of the association.

For further information please visit our website: www.icoria.org


Table of Contents

Preface V

I. Online Advertising/Social Networks

Ofrit Kol, Shalom Levy, and Israel D. Nebenzahl


Consumer Values as Mediators in Social Network Information Search 3

Fabian A. Geise
Integration of Consumers into New Product Development by Social
Media-Based Crowdsourcing – Findings from the Consumer Goods
Industry in Germany 15

Vanessa Apaolaza, Patrick Hartmann, Jiaxun He, Jose M. Barrutia, and


Carmen Echebarria
The Relationship between Gratifications from Social Networking Site
Use and Adolescents’ Brand Interactions 29

Johanna Schwenk and Verena Hüttl-Maack


Promoting the Shareconomy: Effects of Beneficial Appeals and Personal
Characteristics on the Attractiveness of Renting and Reselling Platforms 43

Mototaka Sakashita
Communicating through Brand Websites to Create Unique Brands 57

Polyxeni (Jenny) Palla and Yorgos Zotos


Gaining Attention Online: Do the Levels of Product Involvement and
Website Interactivity Matter? An Eye-Tracking Approach 65

Anik St-Onge, Sylvain Senecal, Marc Fredette, and Jacques Nantel


Is Targeting Online Information Diffusers Based on Their Pesrsonality
Traits and Influencer Types Misleading? 79

Ewa Maslowska, Edward C. Malthouse, and Stefan F. Bernritter


The Effect of Online Customer Reviews’ Characteristics on Sales 87
X Table of Contents

Morikazu Hirose, Kei Mineo, and Keiya Tabe


The Influence of Personal Data Usage on Mobile Apps 101

II. Consumer Responses to Advertising

Stefan Thomas and Heribert Gierl


The Effect of Eroticism in Couple Depictions in Advertisements on Brand
Evaluations 117

Corinne Chevalier and Marie-Christine Lichtlé


Model’s Age and Target’s Age: Effects on Emotions towards and Beliefs
about an Ad 133

Gül Şener, Hasan Kemal Suher, and Ali Atıf Bir


Being Hooked by the Archetypal Characters in Drama TV Ads:
A Structural Equation Modeling Approach 151

Eeva-Liisa Oikarinen
The Moderating Role of Congruence between Humor and Fun
Climate of the Company on the Effects of Humor in Internet Job Ads 167

Tanja Schneider and Heribert Gierl


Put It on the Right Side: The Effect of Print Advertisement Location
on Product Evaluation 183

Franziska Oefele and Heribert Gierl


The Influence of Majority Agreements on Attitudes 199

Kang Li and Nora Rifon


The Effects of Message Framing and Reference Points of Public
Service Announcements on Bystander Intervention in College
Students’ Binge-Drinking 215

Ewa Maslowska, Edward C. Malthouse, and Tom Collinger


How Customers Engage with Brands: A New Framework 231
Table of Contents XI

III. Culture and Advertising

Salman Saleem and Jorma Larimo


Hofstede Cultural Framework and Advertising Research:
An Assessment of the Literature 247

Isabell Koinig, Sandra Diehl, and Barbara Mueller


The Effects of Different Ad Appeals in Non-Prescription
Drug Advertising: A Cross-Cultural Investigation 265

Lucia Porcu, Salvador del Barrio-García, Juan Miguel Alcántara-Pilar,


and Esmeralda Crespo-Almendros
Examining the Mediating Role of Integrated Marketing Communication
on the Relationship between Adhocracy Culture and Brand Advantage 281

Terri H. Chan and Caleb H. Tse


Profiling Lead Researchers in Advertising Research 297
Part I. Online Advertising/Social Networks
Consumer Values as Mediators in Social Network Information
Search

Ofrit Kol, Shalom Levy, and Israel D. Nebenzahl

1 Introduction

Social network sites (SNS) provide their users with new online information
sources. The information is actively created and distributed by consumers for
consumers, with the intention to inform, enrich and enlighten one another about
products, brands, services and more (Liu et. al., 2013; Deighton and Kornfeld,
2009). This information has been perceived to be trustworthy (Foux, 2006),
personal and with minimum cost, allowing the decision-making process to be
more effective (Chai et al., 2010) and improve the consumers' decision-making
process (Wang et Al., 2012; Constantinides et al., 2013).
Consumers, search for information on SNS, confront two types of information
sources: Non-commercial sources which are based on information created by
consumers on a personal profile or groups of interest and Commercial sources,
which are business originated and delivered through brand page or social
advertising tools. These diverse sources raise some important questions. What
influences consumer's choice? Why in some cases do consumers turn to non-
commercial sources, while in other cases the consumers turn to commercial
sources? This study attempts to answer these questions and enhance our
understanding of what motivates consumers' search behavior on SNS. The study
focuses on values the consumer receives from the information as motivating
factors in her / his information source selection.
The literature offers various studies on user information seeking behavior on
SNS in general (Lampe et al, 2012; Gray et al., 2013; Morris et al., 2010);
However, only few studies were conducted in the context of consumer’s search
behavior (Mikalef et al., 2013; Xiang and Gretzel, 2010). Furthermore, to the
best of our knowledge, there are no studies that integrate information values as
mediating factors in consumer’s selection of information sources on SNS. Thus,
there is a need for academic attention here and empirical evidence is needed to
enhance our understanding of the subject.
The current study goals are twofold. First, to apply and check the UTAUT
model (Venkatesh et al., 2003; Gefen et al., 2003; Shin and Kim, 2008) as a
theoretical base for consumer adoption of SNS as a tool for information search.
Second, to verify the suggestion that the values the consumer receives from the
information affect the consumer's search behavior and are mediating factors
influencing the consumer's choice of information sources on SNS. The suggested

© Springer Fachmedien Wiesbaden 2017


G. Christodoulides et al. (Hrsg.), Advances in Advertising Research (Vol. VII),
European Advertising Academy, DOI 10.1007/978-3-658-15220-8_1
4 Kol, Levy, and Nebenzahl

values are: Economic Value (EV), Psychological Value (PV) and Social Value
(SV). In the current study, we propose a conceptual framework, test it
empirically, and present related conclusions and managerial implications.

2 Literature Review

SNS is perceived by consumers to be an efficient source of information


(Lampe et al, 2012; Gray et al., 2013). Consumers frequently turn to various
types of SNS to search for information during their process of making purchase
decisions (Mangold and Faulds, 2009). The adoption of SNS as a legitimate
source of information is in the concern of the Technology Acceptance Model
(TAM). Davis (Davis, 1989; Davis et. al, 1989) introduced the TAM in order to
explain the acceptance of information technology (IT). The model was variously
extended by many researchers. Venkatesh et al. (2003) extended the two TAM
constructs and suggested the Unified Theory of Acceptance and Use of
Technology (UTAUT). Venkatesh et al. (2012) further extended the UTAUT to
predict the acceptance and use of technology in a consumer context. TAM and
UTAUT models have been studied in variety of technologies and their
effectiveness in predicting the adoption of technology in general and the Internet
in particular was supported (Venkatesh et al., 2003; Gefen et al., 2003; Shin and
Kim, 2008). In this study, we use the UTAUT model to explain consumer
adoption of SNS as a source of information search tool. Due to the handful
studies of UTAUT on SNS technologies (Constantinides et al., 2013; Willis,
2008), testing the UTAUT model’s prediction will be one of this study’s goals.
The UTAUT model suggests some antecedents (Performance Expectancy, Effort
Expectancy, Social Influence and Facilitating Conditions) leading towards
behavioral Intention which further leads to use behavior. In SNS consumer’s use
behavior is comprised of the engagement in consumer information sources.

2.1 Sources of Consumer Information on SNS

In SNS there are several different information sources, available to the


consumer who is looking for consumer related information. These channels can
be divided into two types: Non-commercial sources and Commercial sources.
Non-commercial information sources include information received from
personal profile (Boyd and Ellison, 2007; Muniz and O'Guinn, 2001) and groups
of shared interest. A SNS user can post a status message question asking
(SMQA) on these platforms in order to get consumer information (Oeldorf-
Hirsch et al., 2014, Morris et al, 2010). On the other hand, there are commercial
sources which include business or brand page (Lin and Lu, 2011; Kane et al.,
2009) and social advertising tools (Pate and Adams, 2013; Goyal, 2013).
Consumer Values as Mediators in Social Network Information Search 5

Consumers are exposed to ads and brand pages on SNS. By "Liking" a brand
page, the consumers can be exposed to information and special offers created by
a business company. In addition, a consumer can directly post SMQA and get a
personal response from the company.

2.2 SNS’s Information Search Motivators

What motivates consumer search behavior on SNS? The literature indicates


that by searching for information, consumers expect to receive some benefit or
value, which will help them make better purchasing decisions (Keeney, 1999).
Accordingly, this study suggests that the values consumers receive from their
information search effect their search behavior on SNS. The study suggests three
types of relevant consumer values: Economic Value (EV), Psychological Value
(PV) and Social Value (SV), and assumes they act as mediating factors in
consumer selection of information sources on SNS.
The economic value leans on the economics of information theory (Stigler,
1961; Nelson, 1974) and refers to the benefits of saving time, effort and money.
The interactive nature of the Internet in general and the SNS in particular
improves the access to information while reducing the cost of the search (Chai et
al, 2010; Lampe el al, 2012). Saving money comes from the special offers the
consumer receives from marketers, in the form of special offers, coupons and
discounts, which help the consumer, increase the economic benefit of his or her
decision. Following this perspective, we assume that in order to receive EV the
consumer will turn to both commercial and non-commercial sources of
information on SNS. Therefore, the hypothesis:

H1: To the extent that consumers are interested in increasing their EV they will
turn to both commercial information sources and non-commercial information
sources on SNS.

Psychological value refers to psychological benefits reached from the


reduced uncertainty involved in purchase decision. It is derived from the theory
of perceived risk, the two-dimensional construct, which includes the uncertainty
in a purchase decision and the results of a less than satisfactory decision
(Bettman, 1973; Cunningham, 1967; Mitra et al, 1999). The uncertainty can
result from one or more of six types of risks (monetary, performance, physical,
security, social and psychological (Jacoby and Kaplan, 1972) and time
(Roselius, 1971) which are involved in a purchase process and stem from a lack
of information. Information reduces uncertainty and thereby reduces the
perceived risk (Murray, 1991; Newman, 1977). Furthermore, according to a
study conducted by Nielsen (2013), people rely on information they get from
personal sources. According to this study, 92% rely on recommendations from
6 Kol, Levy, and Nebenzahl

people they know, and 70% rely on the opinions of persons writing on the
Internet, while only 33% rely on Internet ads. The non-commercial sources on
SNS provide reliable information on products and services (Mangold and Folds,
2009) as they are based on recommendations and opinions of friends and
acquaintances (Oeldorf-Hirsch et al, 2014; Morris et al, 2010). Following their
suggestions, consumers will look for various information sources to reduce
uncertainty and thereby reduce their perceived risk. Hereby, the hypothesis:

H2: To the extent that consumers are interested in increasing their PV they will
turn to both commercial information sources and non-commercial information
sources on SNS.

Social value refers to the social benefits users receive when connecting to
others via SNS (Deng et al., 2010).This value is located in the feeling of
belonging to a certain group (Deng et. al., 2010) and the need for a cognition
with those who share the same norms, values and interests (Giao et. al., 2015).
SNS offer SV by providing services that enable conversations and information
sharing, along with the possibility of gaining social approval, expressing
opinions, and influencing others (Gangadharbatla, 2008). Many researchers have
examined different aspects of the values a person gets from using Facebook,
including the desire to meet new people (Ellison et al., 2011), self-expression
(Lin and Lu, 2011; Hart et al., 2008), entertainment and even having fun
(Venkatesh et. al, 2012; Sledgianowski and Kulviwat, 2009). Accordingly, we
assume that consumers can reach SV via non-commercial information sources
and not via commercial sources on SNS. Therefore the following hypothesis:

H3: To the extent that consumers are interested in increasing their SV, they will
turn to non-commercial information sources and not to commercial information
sources on SNS.

3 Methodology

3.1 Sample

Data were collected through a web-based survey and a convenience sample.


The subjects were sent an invitation, including a short introduction and a request
to participate in a survey. The study was restricted to Facebook users (a
worldwide leading SNS). Overall, 214 usable responses were analyzed in this
study. The participants were highly experienced users of Facebook, with an
average of 653 friends. Participants were mostly females (70%), with an average
Consumer Values as Mediators in Social Network Information Search 7

age of 33. The education level of the majority of the participants was above high
school (94%), with an average or above-average income (68%).

3.2 Measurement

The survey instrument consisted of multiple items that were partly gathered
from prior studies and partly designed to measure the study's new constructs.
The items for the UTAUT model’s variables (including Performance
expectancy, Hedonic motivation, Social Influence, Facilitating conditions, Effort
Expectancy, Habit, and Behavioral intention) were adopted from Venkatesh et
al. (2012). Use behavior item scales were phrased to capture the commercial
(advertisement and Brand page) and non-commercial (personal profile and
groups) sources of consumer’s information search in SNS. Typical items in the
scales were “I am interesting in coupons and marketing suggestion, I receive
from brand pages in my personal profile” for commercial sources scale, and “I
am posting a question or asking for opinion from friends in my personal profile”
for non-commercial sources scale. For consumer’s values, the EV items were
gathered from Deng et al. (2010), Ailawadi et al. (2003) and Sweeney and
Soutar (2001). SV items were gathered from Deng et al. (2010) and Sweeney
and Soutar (2001). PV items were gathered from Ailawadi et al. (2003) and
Sweeney and Soutar (2001). All scales’ items were modified to suit the SNS
environment. Respondents were asked to indicate their level of agreement with
different statements. A seven-point Likert scale was used, ranging from 1 =
strongly disagree, to 7 = strongly agree. Demographic data were also collected.

4 Results

4.1 Validity and Reliability

The UTAUT model variables’ items were subjected to confirmatory factor


analysis (CFA) for constructs validity and reliability. The results confirm the
constructs (χ2 value (172) = 332.01, p < .05 (χ2/df, less than 2); Comparative Fit
Index (CFI) = .963; Normed Fit Index (NFI) = .927; and Root Mean Square
Error of Approximation (RMSEA) = .067). The CFA shows that all scale items
loaded satisfactorily (β’s > .5) on the relevant latent variables.
The values’ items and sources’ of information search items were subjected
first to exploratory factor analysis and next were subjected to CFA. The results
confirm the constructs (χ2 value (245) = 456.64, p < .05 (χ2/df, less than 2); CFI
= .953; NFI = .905; and RMSEA = .065). The CFA shows that scale items
loaded satisfactorily on the relevant latent variables. Convergent validity,
discriminant validity and internal consistency were examined using the following
8 Kol, Levy, and Nebenzahl

measurements: Average Variance Extracted (AVE), Composite Reliability (CR)


and Cronbach's alpha, displaying acceptable validity and reliability of the
measurements. Means were then calculated and examined for each factor. The
correlation pattern is provided in Table 1.

Table 1: Descriptive statistics and correlations

Variable Mean SD 2 3 4 5 6
1. BI 4.36 1.21 .535** .529** .563** .771** .616**
2. Commercial Sources 3.35 1.12 1.00 .556** .595** .638** .516**
3. Non Commercial Sources 4.21 1.20 -- 1.00 .606** .642** .543**
4. PV 3.76 1.62 -- -- 1.00 .768** .647**
5. EV 3.75 1.69 -- -- -- 1.00 .664**
6. SV 3.14 1.73 -- -- -- -- 1.00
Notes: N = 212; ** < .01

4.2 Model Testing

To examine the relationships among the constructs, two path analyses were
conducted using Structural Equation Modeling (SEM). First, the UTAUT model
was tested on SNS’ information search. The overall fit statistics (goodness of fit
measures) exhibit an acceptable level of fit (χ2 value (2) = 4.24, p = .12 (χ2/df,
less than 3); CFI = .998; NFI = .996; RMSEA = .073), indicating that the path
model is valid. This fit indicates an adequate application of UTAUT model to
SNS’ information search environment. Next, the suggested framework was
tested, while the two sources (Commercial and Non-commercial) were put
together as dependent variables. The overall fit statistics (goodness of fit
measures) exhibit an acceptable level of fit (χ2 value (33) = 49.01, p = .04
(χ2/df, less than 2); CFI = .991; NFI = .974; RMSEA = .048), indicating that the
path model is valid. The path model, regression standardized coefficients, and
their significance are illustrated in Figure 1.
The model depicts the direct and indirect paths toward the dependent
variables (The sources). As seen in Figure 1, Behavioral Intention (BI) has no
direct relationships, neither with the commercial sources nor with the non-
commercial sources. The relationships (β=.48 with Commercial sources and
β=.47 with Non-commercial sources) are indirect through the mediation of the
values. BI has a direct and positive effect on PV (β=.26), a direct effect on EV
(β=.58) and direct effect on SV (β=.59). Further, PV has direct and positive
effect on Commercial sources (β=.26) and on Non-commercial sources (β=.24).
EV has direct and positive effect on Commercial sources (β=.44) and on Non-
commercial sources (β=.33). However, SV has direct and positive effect only on
Non-commercial sources (β=.15), while the relationship with Commercial
Consumer Values as Mediators in Social Network Information Search 9

sources is indirect (β=.26) and through the relationships with PV (β=.49) and
EV (β=.31). Accordingly, hypotheses H1, H2 and H3 are accepted. Table 2
summarizes the relationships among variables.

Figure 1: Consumer’s values as mediators in consumer’s information search in SNS: A


path model a
a
Path parameters are standardized parameter estimates and only significant paths are
shown. Curved arrows indicate correlations. R2 are in the right corner. * p< .05; ** p<
.01.

Table 2: Study’s model’s relationships between variables: Direct and indirect


Relationships Standardized Effect Regression Weights (direct)
Total Direct Indirect Estimate C.R. p
BI  Comm.* .475 .000 .487
BI  Non Comm.** .468 .000 .468
BI  PV .550 .264 .286 .230 4.11 <.001
BI  EV .760 .580 .180 .525 11.14 <.001
BI  SV .587 .587 .000 .545 10.74 <.001
PV  Comm. .258 .258 .000 .250 3.15 <.01
PV  Non Comm. .236 .236 .000 .225 2.87 <.01
EV  Comm. .438 .438 .000 .410 5.36 <.001
EV  Non Comm. .328 .328 .000 .301 3.90 <.001
SV  Comm. .260 .000 .260
SV  Non Comm. .366 .151 .215 .134 2.18 <.05
SV  PV .487 .487 .000 .456 7.57 <.001
SV  EV .306 .306 .000 .298 5.88 <.001
* Comm. = Commercial Sources; ** Non Comm. = Non Commercial Sources.
10 Kol, Levy, and Nebenzahl

5. Discussion and Implications

The goals of this study were first, to empirically test the UTAUT model in the
context of consumer information search behavior on SNS. And second, to
suggest a theoretical framework integrating three types of values, EV, PV and
SV, as mediators towards information sources on SNS.
The results indicate that the UTAUT model (Venkatesh et al., 2003) is
applicable to the SNS information search context. Furthermore, the findings
show that the consumer’s expected values mediate information search behavior
in SNS and this mediation differs according to the information sources.
Commercial sources of information are mediated by EV and PV. This means
that the consumer applies to commercial sources, since he / she expects to
receive some deal or special marketing offer, or reduce uncertainty in buying
decision process. Nevertheless, non-commercial sources are mediated through
all values. This means the consumer applies to non-commercial sources for
different reasons, including SV such as philanthropy, self-expression and social
approval.

5.1 Theoretical and Managerial Implications

This study has several important theoretical and managerial implications.


From a theoretical perspective, the study supports the notion that consumer
values are mediators in information search behavior in SNS. This study suggests
three values, namely, EV, PV and SV. These consumer’s values affect the
consumer's search behavior and mediate the selection of information sources in
SNS. Second, the study suggests a conceptual framework for consumer’s
adoption of SNS as a tool for information search. It applies the UTAUT model
(Venkatesh et al., 2003; Venkatesh et al., 2012) in the context of SNS and
provides theoretical and empirical support for the conceptual framework.
The study has also significant implications for practitioners. Marketing
communication practitioners should note that consumer’s values and expected
benefits mediate the selection of information sources in SNS. The study
indicates that consumers turn to both channels, commercial and non-commercial,
while searching for information. However, the selection of the channels depends
on the potential benefits they pursue. First, in commercial channels, posting an
ad or a brand page on SNS must follow beneficial information, imparting
potential economic profit or psychological utility to the consumer. Messages
should contain rational appeal rather than emotional appeal and be more hard-
sell than soft-sell. Second, all values: social, psychological and economic, are
relevant in non-commercial sources of information. In order to provide social
value, it is important for marketing practitioners to develop a brand community
around consumers’ interests and common values. With this in mind, brands
Consumer Values as Mediators in Social Network Information Search 11

should provide added value supporting the interests and the community’s
common values, e.g. "Nike" developed an application that helps consumers keep
track of their sports activities and share their progress and activity with their
community. In addition, the brand should cultivate its loyal customers, turning
them into ambassadors who will share brand knowledge and experiences with
their community and warmly recommend the brand. This type of information
provides consumers psychological value, it is perceived to be more credible and
helps reduce their perceived risk. Furthermore, it gives economic value by
saving consumers time and money. Additionally, it will be highly beneficial to
integrate these values by distributing brand benefits via opinion leaders.

5.2 Limitations and Future Research

The current study has limitations that should be addressed in future research.
First, this study has limited the empirical testing to one SNS channel
(Facebook). Future research should further examine the current study’s
framework on varied digital SNS channels, such as Twitter and WhatsApp, to
enhance generalization possibilities. Second, though it is a common practice, the
convenience sample of the current study’s web-based survey could be another
limitation which needs to be addressed in future research. For increased
generalization, the study framework should be further tested under more
representative sample settings.

6. References

Ailawadi, K.L.; Gedenk, K. and Neslin, S. (2003), “Understanding Competition between


Retailers and Manufacturers: An Integrated Analysis of Store Brand and National
Brand Deal Usage”, Working Paper No. 03-16, Tuck Business School, Dartmouth
College.
Bettman, J.R. (1973), “Perceived risk and its components: a model and empirical test,”
in: Journal of Marketing Research, Vol. 10, 2, 184-190.
Boyd, D.M. and Ellison, N.B. (2007), “Social Network Sites: Definition, History and
Scholarship,” in: Journal of Computer-Mediated Communication, Vol. 13, 1, 210-
230.
Chai, L.T.; Ling, K.C. and Piew, T.H. (2010), ”The effects of shopping orientations,
online trust and prior online purchase experience toward customers' online purchase
intention,” in: International Business Research, Vol. 3, 3, 63-76.
Constantinides, E.; Lorenzo-Romero, C. and Alarcón-del-Amo, M.C. (2013), “Social
networking sites as business tool: a study of user behavior,” in: M. Glykas (ed.):
Business Process Management, Springer Berlin Heidelberg, 221-240.
Cunningham, S.M. (1967), “The major dimensions of perceived risk,” in: Cox, D.F.
(ed.): Risk Taking and Information Handling in Consumer Behavior, Boston
University Press, Boston, MA, 102-108.
12 Kol, Levy, and Nebenzahl

Davis, F.D. (1989), “Perceived usefulness, perceived ease of use, and user acceptance of
information technology,” in: MIS Quarterly, Vol. 13, 3, 319–340.
Davis, F.D.; Bagozzi, R.P. and Warshaw, P.R. (1989), “User acceptance of computer
technology: a comparison of two theoretical models,” Management science, Vol. 35,
8, 982-1003.
Deighton, J. and Kornfeld, L. (2009), “Interactivity's unanticipated consequences for
marketers and marketing,” in: Journal of Interactive Marketing, Vol. 23, 1, 4-10.
Deng, Z.; Lu, Y.; Wei, K.K. and Zhang, J. (2010), “Understanding customer satisfaction
and loyalty: An empirical study of mobile instant messages in China,” in:
International Journal of Information Management, Vol. 30, 4, 289-300.
Ellison, N. B.; Steinfield, C. and Lampe, C. (2011), “Connection Strategies: Social
Capital Implications of Facebook-enabled Communication Practices,” in: New Media
& Society, Vol. 13, 6, 873-892.
Foux, G. (2006), “Consumer-generated media: Get your customers involved,” in: Brand
Strategy, Vol. 8, May, 38-39.
Gangadharbatla, H. (2008), “Facebook me: Collective self-esteem, need to belong, and
internet self-efficacy as predictors of the iGeneration’s attitudes toward social
networking sites,” in: Journal of interactive advertising, Vol. 8, 2, 5-15.
Gefen, D.; Karahanna, E. and Straub, D.W. (2003), “Trust and TAM in online shopping:
An integrated Model,” MIS Quarterly, Vol. 27, 1, 51–90
Giao, Y.; Gao, J. and Yang, J. (2015), “Social Value and Content Value in Social Media:
Two Ways to Flow,” in: Journal of Advanced Management Science, Vol. 3, 4, 299-
306.
Goyal, S. (2013), “Advertising on social media,” in: Scientific Journal of Pure and
Applied Sciences, Vol. 2. 5, 220-223.
Gray, R.; Ellison, N.B.; Vitak, J. and Lampe, C. (2013), “Who wants to know?:
Question-asking and answering practices among Facebook users,” in: Proceedings of
the Conference on Computer supported cooperative work, ACM, 1213-1224.
Hart, J.; Ridley, C.; Taher, F.; Sas, C. and Dix, A. (2008), “Exploring the facebook
experience: a new approach to usability,” in: Proceedings of the 5th Nordic
conference on Human-computer interaction: building bridges, ACM, 471-474.
Jacoby, J. and Kaplan, L.B. (1972), “The components of perceived risk,” in:
M.Venkatesan (ed.): Third Annual Conference of the Association for Consumer
Research, Association for Consumer Research, College Park, MD, 382-93.
Kane, G.C.; Fichman, R.G.; Gallaugher, J. and Glaser, J. (2009), “Community relations
2.0,” in: Harvard Business Review, Vol. 87, 11, 45-50.
Keeney, R.L. (1999), “The value of Internet commerce to the customer,” in:
Management science, Vol. 45, 4, 533-542.
Lampe, C.; Vitak, J.; Gray, R. and Ellison, N. (2012), “Perceptions of Facebook's value
as an information source,” in: Proceedings of the SIGCHI Conference on Human
Factors in Computing Systems, ACM, 3195-3204.
Lin, K.Y. and Lu, H.P. (2011), “Intention to continue using Facebook fan pages from the
perspective of social capital theory,: in: Cyberpsychology, Behavior, and Social
Networking, Vol. 14, 10, 565-570.
Liu, I.L., Cheung, C.M. and Lee, M.K. (2013), “Customer Information Sharing Behavior
in Social Shopping Communities: A Social Capital Perspective,” in: Proceedings of
PACIS. Paper 227.
Consumer Values as Mediators in Social Network Information Search 13

Mangold, W.G. and Faulds, D.J. (2009), “Social media: The new hybrid element of the
promotion mix,” in: Business Horizons, Vol. 52, 4, 357-365.
Mikalef, P.; Giannakos, M. and Pateli, A. (2013), “Shopping and Word-of-mouth
intentions on social media,” in: Journal of theoretical and applied electronic
commerce research, Vol. 8, 1, 17-34.
Mitra, K.; Reiss, M.C. and Capella, L.M. (1999), “An examination of perceived risk,
information search and behavioral intentions in search, experience and credence
services,” in: Journal of Services Marketing, Vol. 13, 3, 208-228.
Morris, M.R.; Teevan, J. and Panovich, K. (2010), “What do people ask their social
networks, and why?: a survey study of status message Q&A behavior,” in:
Proceedings of the SIGCHI conference on Human factors in computing systems,
ACM, 1739-1748 .
Muniz, A.M. and O’Guinn, T.C. (2001), “Brand community,” in: Journal of Consumer
Research, 27(4), 412-432.
Murray, K.B. (1991), “A test of services marketing theory: consumer information
acquisition activities,” in: Journal of Marketing, Vol. 55, 1, 11-25.
Nelson, P. (1974), “Advertising as information,” in: The Journal of Political Economy,
Vol. 82, 4, 729-754.
Newman, J.W. (1977), “Consumer External Search: Amount and Determinants,” in: A.G.
Woodside, N.S Jagdish, P.D. Bennett (eds): Consumer and Industrial Buying
Behavior, New York: Elsevier North-Holland, 79-94.
Nielsen (2013), “Global-Trust-in-Advertising-Report,” Retrieved from
http://www.nielsen.com/us/en/insights/reports/2013/global-trust-in-advertising-and-
brand-messages.html
Oeldorf-Hirsch, A.; Hecht, B.; Morris, M. R.; Teevan, J. and Gergle, D. (2014), “To
search or to ask: the routing of information needs between traditional search engines
and social networks,” in: Proceedings of the 17th ACM conference on Computer
supported cooperative work & social computing, ACM, 16-27.
Pate, S.S. ans Adams, M. (2013), “The Influence of Social networking Sites on Buying
Behaviors of Millennials,” in: Atlantic Marketing Journal, Vol. 2, 1, 92-109.
Roselius, E. (1971), “Consumer rankings of risk reduction methods,” in: Journal of
Marketing, Vol. 35, 1, 56-61.
Shin, D.H. and Kim, W.Y. (2008), “Applying the technology acceptance model and flow
theory to cyworld user behavior: implication of the web2. 0 user acceptance,” in:
CyberPsychology & Behavior, Vol. 11, 3, 378-382.
Sledgianowski, D. and Kulviwat, S. (2009), “Using social network sites: the effects of
playfulness, critical mass and trust in a hedonic context,” in: Journal of Computer
Information Systems, Vol. 49, 4, 74-83.
Stigler, G.J. (1961), “The economics of information,” in: The Journal of Political
Economy, Vol. 69, 3, 213-225.
Sweeney, J.C. and Soutar, G. N. (2001), “Consumer perceived value: The development
of a multiple item scale,” in: Journal of retailing, Vol. 77, 2, 203-220.
Venkatesh, V.; Morris, M.G.; Davis, G.B. and Davis, F.D. (2003), “User Acceptance of
Information Technology: Toward a Unified View,” in: MIS Quarterly, Vol. 27, 3,
425-478
14 Kol, Levy, and Nebenzahl

Venkatesh V.; Thong J.Y.L. and Xin X. (2012), “Consumer acceptance and use of
information technology: extending the unified theory of acceptance and use of
technology,” in: MIS Quarterly, Vol. 36, 1, 157-225.
Wang, X.; Yu, C. and Wei, Y. (2012), “Social media peer communication and impacts
on purchase intentions: A consumer socialization framework,” in: Journal of
Interactive Marketing, Vol. 26, 4, 198-208.
Willis, T.J. (2008), “An evaluation of the technology acceptance model as a means of
understanding online social networking behavior,” Graduate School Theses and
Dissertations, University of South Florida. http://scholarcommons.usf.edu/etd/568
Xiang, Z. and Gretzel, U. (2010), “Role of social media in online travel information
search,” in: Tourism management, Vol. 31, 2, 179-188.
Integration of Consumers into New Product Development by
Social Media-Based Crowdsourcing – Findings from the
Consumer Goods Industry in Germany

Fabian A. Geise

1 Introduction
Successful new products are crucial for growth and the strengthening of a
company’s competitiveness. However, not every new product launch succeeds
on the market, i.e. the potential economic success is set against the risk of a new
product failure. The flop rates are up to 90 percent depending on the industry
(Gourville, 2006; Cooper, 2001; Crawford, 1987). The main reason for new
products failing to establish themselves is often that new offers do not fit the
needs of the customers (Reichwald and Piller, 2009, 128f.). This has been
proven in many empirical studies (Gruner and Homburg, 2000; Hanna et al.,
1995; von Hippel, 1986). These studies also show another relevant issue, that it
is necessary to integrate customers’ needs as early as possible into the process of
new product development (NPD), i.e. into the stages “search for new product
ideas” and “evaluation of ideas” (Kotler and Keller, 2012, 597; Bogers et al.,
2010). The question here is how customers can be deeply integrated into the
early stages of the development process of new products.
An effective strategy for integrating customers is the so-called open
innovation approach. The key assumption for open innovation is the fact that
innovation-related knowledge is omnipresent in the company’s environment, i.e.
this knowledge is held by various actors – in particular by suppliers and buyers
in the case of industrial goods and consumers in the case of consumer goods
(Spithoven et al., 2012; Gassmann et. al., 2010; Chesbrough, 2006; Prahalad and
Ramaswamy, 2004). Therefore, companies who work with an open innovation
strategy view customers as a valuable resource for new product ideas. Hence, the
challenging task is to integrate this knowledge systematically into the company’s
innovation management process.
The expansion of the internet to Web 2.0 offers companies the ideal
opportunity to realize open innovation strategies with customers on a new level
of collaboration (Chakravorti, 2010). This applies in particular for companies in
the consumer goods industry. A promising procedure is to use social media like
Facebook, blogs, brand communities, etc. On these virtual platforms you can
typically access many people outside the company cost-efficiently and quickly.
In doing so, innovation processes are outsourced to a crowd, thus to a plurality

© Springer Fachmedien Wiesbaden 2017


G. Christodoulides et al. (Hrsg.), Advances in Advertising Research (Vol. VII),
European Advertising Academy, DOI 10.1007/978-3-658-15220-8_2
16 Geise

of users. Crowdsourcing as a special open innovation strategy enables the


development of new products by direct integration of users into the early stages
of the innovation process. The special advantage of this strategy can be seen in
developed products which reflect the needs of the users and have for this reason
a greater likelihood of acceptance by the consumers. That is why more and more
companies are utilizing consumers as a collective source of knowledge
(“wisdom of the crowd”) for generating new product ideas (Fuchs and Schreier,
2011; Howe, 2008; Kleemann et al., 2008).
The results of an international cross-industry study by McKinsey underline
that by using social media platforms, the development of new products can be a
successful innovation strategy in various industries (McKinsey Global Institute,
2012). As figure 1 shows, about a quarter of surveyed companies reporting Web
2.0 technologies for internal purposes quote, among other things, a benefit for
new product development (see also Urban and Hauser, 2004).

Figure 1: Reported benefits from social media technologies


Source: McKinsey Global Institute, 2012, p. 28

Empirical studies regarding the integration of customers in the NPD are


mostly focused on industrial goods. Only relatively few studies show the
integration of users with respect to consumer goods (Bartl et al. 2012; Füller,
2006; Ogawa and Piller, 2006; Lüthje, 2004). The following remarks refer only
to the consumer goods industry in Germany.
Integration of Consumers into New Product Development 17

Crowdsourcing strategies can be carried out, for example on intermediary


innovation platforms (Innocentive, Atizo, Jovoto etc.) or on company-owned
platforms (tchibo-ideas, dellstorm, mystarbucksidea etc.). While intermediary
and company-owned platforms are focused on an ongoing generation of new
ideas or problem solving through the users, another development has appeared
over the last few years, especially in the consumer goods industry: so-called idea
contests. This kind of Web-based collaboration with consumers offers a further
option for co-creation to generate ideas for new product development (Ind and
Coates, 2013; Füller, 2010). Idea contests in the consumer goods industry are
temporary projects/campaigns which are carried out on a company-owned
platform including social media (especially social networks, in particular
Facebook) (Piller et al., 2012). It is typical for idea competitions that parti-
cipation itself is done selectively, i.e. only those users that feel attracted by the
innovation task contribute to the performance of tasks. For this reason, it is
important that the task and the incentives have to be communicated in a manner
that many users feel (intrinsically and/or extrinsically) motivated to participate.

2 Research Question
So far, there is still no empirical study for Germany, where virtual consumer
integration on social media platforms – particularly Facebook – has been
examined. The questions here are how widespread are social media-based idea
contests in the consumer goods industry in Germany, what are their common
characteristics and which success factors can be determined from idea contests
that have already taken place?

3 Method
A content analysis of idea contest websites, Facebook sites and press
information, etc., was conducted in order to discover such social media-based
idea contests realized by German companies. The companies that were
examined, with further information in parentheses including the year of the
contest, the new product to be developed and additionally further developing
tasks, are: Ritter Sport (2010, variety of chocolate and product packaging for it),
McDonald’s (2011, 2012, 2013 and 2014, hamburger), Griesson de Beukelaer
(2011, variety for biscuit brand Prinzenrolle), Vapiano (2011, pasta dishes),
Rügenwalder Mühle (2011, sausage), Bonprix (2011, designs for bedlinen),
Homann (2011 and 2012, varieties of potato salad), Edeka (2012 and 2013,
variety of ice cream, smoothie, biscuit and yoghurt), Beck’s (2013, variety of
mixed beer), Mondelez (2013, recipes for cakes with using the brand
Philadelphia) and Lidl (2013 and 2014, variety of yoghurt, smoothie and
doughnut). All idea contests took place in the period 2010 to 2015. As Facebook
18 Geise

has only been available in Germany since 2008 and required a certain time to
reach a large number of users to increase companies’ interest in this interaction
medium, it is understandable that the first Facebook-based contest was
conducted in 2010 (chocolate manufacturer Ritter Sport). With regard to
revealing success factors, open innovation literature (e.g. Howe, 2008; Franke et
al., 2013) and crowdsourcing blogs (e.g. socialnetworkstrategien.de,
crowdcourcingblog.de) were additionally analyzed.

4 Results

4.1 Typical Attributes of Social Media-based Idea Contests


The typical attributes of the analyzed idea contests can be described as
following:
- Types of new products
In all crowdsourcing campaigns, users were asked to generate relatively
simple, unproblematic goods like chocolate, hamburgers, cakes with cream
cheese, biscuits or pasta dishes, i.e. typical fast moving consumer goods.
- Scope of the task
In most contests, consumers were asked to generate new products and
evaluate them by Web-based voting procedures, i.e. crowdcreating and
crowdvoting were the main tasks for the users. The combination of
crowdcreating and crowdvoting uses the knowledge of consumers in two
ways and shows, in doing so, that the company takes consumers seriously
and considers them to be competent in not only generating new products
but also deciding which one is best. In five cases – Ritter Sport,
McDonald’s, Griesson de Beukelaer, Edeka and Lidl – the task included
idea creation, finding a name for the created product and Web-based
voting.
- Types of incentives for participation
Mostly, non-monetary incentives were promised as rewards for
participation (e.g. personal computer, product sample, shopping voucher,
participation in a commercial, invitation to the company’s headquarters,
etc.). In one case – Bonprix – a monetary incentive of 1.000 Euro plus
participation in sales of the winning product was offered as reward.
- “Mechanics” of the idea-generating process
The idea contests carried out so far indicate a generalized way of
functioning that includes four typical stages:
(1) Call for participation (on Facebook and the company’s website,
sometimes assisted by communication activities on TV, radio and/or print
media)
Integration of Consumers into New Product Development 19

(2) Registration (Users have to register on Facebook site or company’s


site. Here the users have to provide various kinds of personal information.
The analysis of this information can offer relevant insights on the
participants and the evaluation of the objectives of the idea contest, e.g. the
number of participants in total or specific groups of participants.)
(3) Idea generation (In some contests idea generation took place with the
assistance of a software tool, a so-called idea or product configurator (see
below).)
(4) Idea-evaluating procedure (Crowdvoting can be used for the prelim-
nary or the final selection of product ideas. Voting processes in the
network can also be executed in conjunction with an internal jury.)
- Results of idea contests (Direct and measurable results of a campaign show
up particularly in the number of participants, generated product ideas,
number of participants voting and the increase of followers on Facebook.)

4.2 Use of Product Configurator as a Toolkit for User Innovation


In the classical approach within the context of NPD, it is difficult to track
down need-related information via market research techniques. The problem is,
however, that consumers are often unable to accurately describe their ideas or
wishes for new products in written or oral form (Piller and Walcher, 2006; von
Hippel and Katz, 2002). So, need-related information remains often vague,
incomplete and ambiguous, i.e. it cannot be represented explicitly, or in other
words, the company has to do with so-called “sticky information” (von Hippel
and Katz, 2002; von Hippel, 1994). This makes it difficult to use consumers’
knowledge for the NPD process.
Acquisition, decoding and utilization of sticky information is often a time
consuming and cost-intensive iterative process between external knowledge
holders (crowd) and the organization. However, adaptation and utilization of
need-related information can fail due to a lack of “absorptive capacity” on part
of the company (Cohen and Levinthal, 1990; Zahra and George, 2002). This
primarily means the organisational and procedural interface between the
company and the crowd. Technical means to improve the knowledge transfer,
and therefore to improve the absorptive capacity, are toolkits for user innovation
(also called toolkits for user innovation and design or toolkits for idea
competition) (Piller and Walcher, 2006; Jeppesen, 2005). Such toolkits are
Web-based applications which are used at the user/company interface to gather
proposals for new products in a systematic manner (Prandelli et al., 2008; Prügl
and Schreier, 2006). Product configurators, as used in the social media-based
idea competitions, represent relatively simple structured toolkits. They make it
possible to collect a lot of product ideas from outside the company. Product
configurators are also known as a sales tool. They are used by many companies
20 Geise

in the context of mass customization to give the customer the opportunity to


have a product produced according to their individual preferences (e.g. when
configuring a car) (Franke and Piller, 2004 and 2003). For this, the customer
selects predetermined product characteristics/components from those that
characterize their specific product. However, production configurators can also
be used to generate new products within the scope of the innovation process
(Piller et al., 2011; Franke and Schreier, 2002).
In five of the analyzed contests (McDonald’s, Griesson de Beukelaer,
Homann, Edeka and Lidl) such configurators were used. It is typical for
configurators that the product development task is divided into several sub-tasks.
An easy and self-explanatory usability is another characteristic feature.
Particular product knowledge is not necessary for the configurator-guided
creation of new products.
Since the company defines the scope of possible product ideas (“solution
space”) through the deliberate setting of product components, it ensures that the
best ideas developed by the consumers can also be produced cost-efficiently
(von Hippel and Katz, 2002). Another advantage of such configurators can be
seen in that nonsensical, silly ideas are excluded from the outset. Figure 2 shows
the structure of configurators which have been used in the conducted contests.

Figure 2: Typical structure of product configurators used in social media-based idea


contests

After registration on the contest site, the user determines the product
components for the new product. Depending on the contest, there are a different
number of components to choose from. For example, in the first idea contest by
McDonald’s, 70 different components from the categories rolls, fish/meat etc.,
standard ingredients and extra ingredients could be selected. In this way, users
developed a total of 116,000 burger creations. In the campaign by Griesson de
Beukelaer, three varieties of biscuit (chocolate, multigrain and classic biscuit)
and a total of 140 components for the biscuit filling (from the categories
chocolate cream, spices, nuts/seeds, fruit and miscellaneous) were available. In
Integration of Consumers into New Product Development 21

this campaign, over 5,000 new sorts of biscuit were developed. In the five
configurators the naming of the new product was a further task before the
product proposal was submitted. After submitting the proposals they are usually
listed in a virtual product gallery, and users have the opportunity to vote for the
best products in their opinion. The product proposals receiving the most votes
are finally evaluated by an internal jury to determine the winning product(s).

4.3 Success Factors of Social Media-based Idea Contests


As success factors, three main factors (each with two or four underlying
single success factors) were discovered: interaction competence, user
orientation and task content (see figure 3). The main success factors represent
the three components as well as the analytical levels of idea contests: company,
user (consumer) and task.

Figure 3: Main success factors of social media-based idea contests


22 Geise

- Main success factor “Interaction competence”:


What is meant by the success factor organisational structures and
strategies to support interaction is that the company-internal NPD tasks,
responsibilities and communication systems at the interface between
company and crowd (Web site, Facebook fan page, blog, Twitter etc.) must
be aligned with the expected interaction. In addition, sufficient personnel
resources are necessary to ensure smooth procedures within the idea
contest campaign (e.g. quick responses to participants’ e-mails, evaluation
of product ideas, etc.). Finally, interaction competence includes a coherent
social media strategy as an integral part of the overarching online
marketing and innovation strategy.
A company shows credibility if it performs an idea contest as previously
announced, e.g. if the results from crowdvoting are fully accepted and the
winning product is actually produced.
The factor transparency refers to the openness by which the company
communicates the rules and procedures of the crowdsourcing campaign.
Transparency particularly plays a role in evaluating ideas through crowd-
voting. If a jury is established, their members and the criteria for the
selection of ideas should be announced in a timely manner.
Finally, fairness is closely related to the incentives a company offers as
rewards for participating in an idea contest. Incentives should match the
cognitive performance of the participants when generating new product
ideas (it makes a difference if a consumer composes a new hamburger by
using a software configurator or if he or she creates a new design for bed
linen). In addition, fair incentives should also consider that consumers
usually assign the usage rights for a product idea to the company.

- Main success factor “User orientation”:


User orientation is closely connected to two special success factors:
selection of the right crowd and incentives to encourage participation. To
select the right crowd means addressing such users in social media in a way
that corresponds with the target group of the brand. These are primarily
users of Facebook. In addition, users of Twitter, blogs, brand or company-
related communities can be addressed. Because Facebook represents the
leading medium for idea contests for consumer goods, the company should
have a large fan base if it starts with a crowdsourcing campaign to generate
product ideas.
To ensure that enough users participate in a crowdsourcing campaign, the
right incentives must be offered to address the intrinsic and extrinsic
motives of the users (Füller, 2006). As the analysis of idea contests shows,
different incentives were offered as rewards for participation. They should
Integration of Consumers into New Product Development 23

correspond to the task, i.e. the relatively simple one of generating new
product ideas with the assistance of a product configurator (e.g. in the case
of McDonald’s or Griesson de Beukelaer) call for other incentives than the
creation of new package or product designs (see the cases of Ritter Sport
and Bonprix).

- Main success factor “Task content”:


Two success factors with regard to task content can be identified:
relevance of the new product development task and simplicity of the task.
The first success factor points out that the crowd should see an idea contest
as an interesting task. Here it is necessary to determine whether the
campaign offers a task to users with a highly stimulative nature (see e.g. the
McDonald’s campaign “Build your own burger”, where consumers were
able to create new hamburgers in a “playful” way by using a hamburger
configurator and giving the new burger its own name). Furthermore, the
relevance of the task means that the call for participation should be
communicated credibly, i.e. the company should stress that it is very
interested in new product ideas or, more generally, the needs of the users
are a vital source of information for the company’s innovation strategy.
Simplicity of the task refers to the conception of the content, i.e. a task
should be not too complex and difficult. This also infers that a toolkit that
assists the generation of ideas should be easy to use. If a task consists of
several subtasks (e.g. including idea creating, designing the package and
voting for the best idea, as in the Ritter Sport campaign), this should be
carried out in reasonable time intervals. Furthermore, this success factor
includes simple and transparent voting procedures and communication of
jury decisions as well as quick and transparent feedback to the participants
(see success factor “transparency”).

5 Managerial Implications and Outlook


Although there are a variety of studies on open innovation and co-creation
with customers, studies about social media-based (particularly Facebook-based)
crowdsourcing in the form of consumer-oriented idea contests with focus on
consumer goods are not available. About a dozen idea contests in the field of
consumer goods in Germany have been analyzed. In addition to determining the
characteristic features of such contests, specific success factors have also been
worked out. The three main success factors as well as the underlying single
success factors show how to implement a successful social media-based
crowdsourcing campaign with the aim to generate ideas for new products in the
field of consumer goods.
24 Geise

From the main success factor interaction competence some managerial


recommendations can be derived: Before realizing a social media-based idea
competition the company should ensure that this kind of new product
development task is an integral part of the pursued social media strategy and an
integral part of the overarching online marketing and innovation strategy. Hence,
this (normally new) strategic orientation requires a commitment from all
departments concerned (e.g. marketing/communication/social media/CRM,
R&D/product development, etc.). It should also be ensured that enough
experiences with social media are available (in particular in the marketing
department) and that the decision makers in the departments affected by the new
task of idea generating have a positive attitude towards the integration of users
into the NPD process. Finally, the company should make sure that in the context
of an idea contest, the rules of credibility, transparency and fairness are strictly
respected.
With regard to the main success factor user orientation, it should be ensured
that in particular the follower base on Facebook is big enough. However, what a
big enough follower base means cannot be generalized. It depends on the type of
product development task; to develop a new packaging design requires fewer
followers for acquiring enough participants than to create a new product, for
example a new sort of chocolate. As a rule of thumb it can be recommend that a
company should have a minimum of several thousand followers when an idea
contest is to be carried out. Howe (2008) for example suggests the minimum
number as 5,000 followers. If the number of Facebook fans is not large enough,
the company should recruit new followers by continuously practiced social
media marketing before it carries out such a contest. Moreover, depending on
the NPD task and presumed motivation of the participants, appropriate
incentives should be awarded to encourage the greatest possible participation.
Intellectually demanding NPD tasks should be rewarded more significantly than
a “simple” product development task by using a product configurator.
In connection with the main success factor task content, some more
recommendations for practice can be derived: The company should ensure that
the NPD task – from the perspective of the users – appears to be interesting. The
call for participation in an idea contest should clearly communicate that users
play a very important role in the innovation policy of the company. In addition,
it must be ensured that the task does not overwhelm the user. If a toolkit is used,
it must be designed in a user-friendly manner and with many selectable product
components. Voting procedures should also be made easy and transparent.
Social media-based idea contests in the consumer goods industry are a new
kind of consumer integration into the NPD process. This specific crowdsourcing
practice underlines in particular the following advantages for the company
(Reichwald and Piller, 2009; Enkel et al., 2005): time advantage when
Integration of Consumers into New Product Development 25

developing new products (reduction of time-to-market) and revised market


acceptance of a launched new product (enhancement of fit-to-market).
Furthermore, this kind of Web-based innovation strategy can also be seen as an
efficient instrument for gathering deep customer insights (i.e. information about
relevant consumer needs or product-related expectations) and for strengthening
customer loyalty. Finally, the image of the brand or the company can hereby be
improved.
However, with social media-based idea contests some disadvantages or
problems may also be associated. A disadvantage can be seen in the costs of
such an action (this applies particularly to idea contests in which one or more
external service providers, e.g. social media or crowdsourcing companies, are
involved in implementing such a campaign). Such contests are primarily suitable
for the development of relatively “simple” consumer products of frequent
demand. For the development of “complex” consumer goods (e.g. technically
oriented goods) intermediate development platforms with an appropriate
specialization should be used. Finally, massive negative feedback from the
crowd in form of a controversy can result, particularly if the voting procedure
during the campaign is changed by the company. Another loss of control may
result if new product ideas are to be generated without or with too generalized
regulations so that nonsensical or silly product ideas are provided by the users.

6 References

Bartl, M., Füller, J., Mühlbacher, H. and Ernst, H. (2012), “A manager’s perspective on
virtual customer integration for new product development,” in: Journal of Product
Innovation Management, Vol. 29, 6, 1031-1046.
Bogers, M., Afuah, A. and Bastian, B. (2010), “Users as innovators: A review, critique,
and future research directions,” in: Journal of Management, Vol. 36, 4, 857-875.
Chakravorti, B. (2010), “Stakeholder marketing 2.0,” in: Journal of Public Policy &
Marketing, Vol. 29, 1, 97-102.
Chesbrough, H.W. (2006), Open innovation. The new imperative for creating and
profiting from technology, Boston, Mass.: Harvard Business School.
Cohen, W.M. and Levinthal, D.A. (1990), “Absorptive capacity: A new perspective on
learning and innovation,” in: Administrative Science Quarterly, Vol. 35, 1, 128-152.
Cooper, R.G. (2001), Winning at new products: Accelerating the process from idea to
launch, 3. ed., Cambridge, Mass.: Perseus Books.
Crawford, C.M. (1987), “New product failure rates: A reprise,” in: Research
Management, Vol. 30, 4, 20-24.
Enkel, E., Perez-Freije, J. and Gassmann, O. (2005), ”Minimizing market risks through
customer integration in new product development: Learning from bad practice,” in:
Creativity and Innovation Management, Vol. 14, 4, 425-437.
26 Geise

Franke, N., Keinz, P. and Klausberger, K. (2013), “‘Does this sound like a fair deal?’
Antecedens and consequences of fairness expectations in the individual’s decision to
participate in firm innovation,” in: Organizational Science , Vol. 24, 5, 1495-1516.
Franke, N. and Piller, F. (2004), “Value creation by toolkits for user innovation and
design: The case of the watch market,” in: Journal of Product Innovation
Management, Vol. 21, 6, 401-415.
Franke, N. and Piller, F. (2003), “Key research issues in user interaction with
configuration toolkits in a mass customization system,” in: The International Journal
of Technology Management, Vol. 26, May-June, 578-599.
Franke, N. and Schreier, M. (2002), ”Entrepreneurial opportunities with toolkits for user
innovation and design,” in: The International Journal on New Media Management,
Vol. 4,4, 225-234.
Fuchs, C. and Schreier, M. (2011), “Customer empowerment in new product
development,” in: Journal of Product Innovation Management, Vol. 28, 1, 17-32.
Füller, J. (2010), “Refining virtual co-creation from a consumer perspective,” in:
California Management Review, Vol. 52, 2, 98-122.
Füller, J. (2006), “Why consumers engage in virtual new product developments initiated
by producers,” in: Advances in Consumer Research, Vol. 33, 639-646.
Gassmann, O., Enkel, E. and Chesbrough, H. (2010), “The future of open innovation,”
in: R&D Management, Vol. 40, 3, 213-221.
Gourville, J. (2006), “Eager sellers & stony buyers. Understanding the psychology of
new product adoption,” in: Harvard Business Review, Vol. 84, 6, 99-106.
Gruner, K. and Homburg, C. (2000), “Does customer interaction enhance new product
success?,” in: Journal of Business Research, Vol. 49, 1, 1-14.
Hanna, N., Ayers, D.J., Ridnour, R.E. and Gordon, G.L. (1995), “New product
development practices in consumer versus business products organizations,” in:
Journal of Product and Brand Management, Vol. 4, 1, 33-45.
Howe, J. (2008), Crowdsourcing. Why the power of the crowd is driving the future of
business, New York: Three Rivers Press.
Ind, N. and Coates, N. (2013), “The meanings of co-creation,” in: European Business
Review, Vol. 25, 1, 86-95.
Jeppesen, L.B. (2005), “User toolkits for innovation: Consumers support each other,” in:
Journal of Product Innovation Management, Vol. 22, 4, 347-362.
Kleemann, F., Voß, G.G. and Rieder, K. (2008), ”Un(der)paid innovators: The
commercial utilization of consumer work through crowdsourcing,” in: Science,
Technology & Innovation Studies, Vol. 4, 1, 5-26.
Kotler, P. and Keller, K.L. (2012), Marketing management, 14. ed., Boston: Pearson.
Lüthje, C. (2004), “Characteristics of innovating users in a consumer goods field. An
empirical study of sport-related product consumers,” in: Technovations, Vol. 24, 9,
683-695.
McKinsey Global Institute (2012), The social economy. Unlocking value and
productivity through social technologies, http://www.mckinsey.com/insights/
high_tech_telecoms_ internet/ the_ social_economy, access 16.9.2015.
Ogawa, S. and Piller, F. (2006), “Reducing the risks of new product development,” in:
MIT Sloan Management Review, Vol. 47, 2, 65-72.
Piller, F., Vossen, A. and Ihl, C. (2012), “From social media to social product
development: The impact of social media on co-creation of innovation,” in: Die
Unternehmung, Vol. 66, 1, 7-27.
Integration of Consumers into New Product Development 27

Piller, F., Ihl, C. and Vossen, A. (2011),”Customer co-creation: Open innovation with
customers,” in: Wittke, V. and Hanekop, H. (Eds.), New forms of collaborative
innovation and production on the Internet, Göttingen: Universitätsverlag, 31-61.
Piller, F. and Walcher, D. (2006), “Toolkits for idea competitions: a novel method to
integrate users in new product development,” in: R&D Management, Vol. 36, 3, 307-
318.
Prahalad, C.K. and Ramaswamy, V. (2004), The future of competition: Co-Creating
unique value with customers, Boston, Mass.: Harvard Business School.
Prandelli, E., Sawhney, M. and Verona, G. (2008), Collabrating with customers to
innovate, Cheltenham: Edward Elgar.
Prügl, R. and Schreier, M. (2006), “Learning from leading-edge customer at The Sims:
opening up the innovation process using toolkits,” in: R&D Management, Vol. 36, 3,
237-250.
Reichwald, F. and Piller, F. (2009), Interaktive Wertschöpfung, 2. ed., Wiesbaden:
Gabler.
Spithoven, A., Teirlinck, P. and Frantzen, D. (2012), Managing open innovation.
Connecting the firm to external knowledge, Cheltenham: Edward Elgar Publishing.
Urban, G.L. and Hauser, J.R. (2004), “’Listening in’ to find and explore new
combinations of customer needs,” in: Journal of Marketing, Vol. 68, April, 72-87.
von Hippel, E. and Katz, R. (2002), “Shifting innovation to users via toolkits,” in:
Management Science, Vol. 478, 7, 821-833.
von Hippel, E. (1986), Lead users: A source of novel product concepts, in: Management
Science, 32 (3), 791-805.
von Hippel, E. (1994), “’Sticky information’ and the locus of problem solving:
implications for innovation,” in: Journal of Product Innovation Management, Vol.
18, 4, 247-257.
Zahra, S.A. and George, G. (2002), “Absorptive capacity: A review, reconceptualization,
and extension,” in: Academy of Management Review, Vol. 27, 2, 185-203.
The Relationship between Gratifications from Social
Networking Site Use and Adolescents’ Brand Interactions

Vanessa Apaolaza, Patrick Hartmann, Jiaxun He, Jose M. Barrutia, and


Carmen Echebarria

1 Introduction
Online social networking sites (SNS) such as Facebook or Twitter provide
advertiser with new channels to contact and interact with potential target groups.
Brand profiles enable companies to interact with consumers in a direct
personalized way (Kelly et al., 2010; Rowley, 2009). The interactivity between
consumers and the brand is based on playful experiences, entertainment,
information and active participation. Consumers can dialogue with the brand,
share their opinions, and receive information on new products or promotion
campaigns (Othman-Yousif, 2012). Young consumers and even adolescents are
often targeted with SNS brand profiles. However, despite considerable efforts
being made by some brands, the degree to which adolescents interact with
brands through social networks is still quite low (Cooper, 2011; Hadija et al.,
2012), and brand messages pass by often undetected amongst the rest of the
information (Kelly et al., 2010). SNS users, and particularly adolescents, focus
their attention on other SNS contents of greater personal relevance, such as
friends’ profiles, videos, photos, etc. (Zeng et al., 2009). Overall, adolescent
audiences are only reached if they are themselves interested and search for
contact with a specific brand (Hadija et al., 2012).
Knowing what motivates target groups to interact with brands on SNSs is
therefore essential for advertisers aiming to develop effective communication
strategies on SNS. However, research on the motives for brand interactions on
SNSs is still scarce. One of the few studies we identified was conducted by Tsai
and Men (2013), who analyze users’ motivations for using SNS brand pages.
They found that respondents typically used a Facebook brand page to search for
news on discounts or sales promotion, and to exchange information with other
SNS users. On the other hand, Kwon et al. (2014) examined what factors
affected attitudes towards brand communications on Twitter, based on consumer
socialization patterns. Their findings suggest consumers follow brands on
Twitter because of four primary motivations: incentive seeking, social-
interaction seeking, brand usage/likeability, and information seeking. The latter
three were found to be significant predictors of the consumer–brand relationship
variables. Also, individuals who communicate frequently with peers about

© Springer Fachmedien Wiesbaden 2017


G. Christodoulides et al. (Hrsg.), Advances in Advertising Research (Vol. VII),
European Advertising Academy, DOI 10.1007/978-3-658-15220-8_3
30 Apaolaza, Hartmann, He, Barrutia, and Echebarria

brands and generally hold more favorable attitudes towards advertising were
more likely to have positive attitudes towards brand communications via
Twitter.
This paper proposes a conceptual model addressing the motives for the
interaction of SNS users with brands on the SNS. Drawing on Uses and
Gratifications (U&G) Theory three main gratifications are proposed, which are
potentially related to the motivation to interact with brands on SNSs:
gratifications derived from socializing, information seeking and entertainment.
The aim of our study is to analyze the relationship of these gratifications with the
degree to which users engage with brands on the SNS. We also focus on a
specific target group, adolescents, since young SNS users have been less
attended by previous research, but may show a particular pattern of online
networking behavior, which may affect brand interactions. The specific
conceptual framework proposed in this research has not been addressed in prior
research.
Most previous SNS research has concentrated on western and mostly
American SNSs such as Facebook Myspace and Twitter. A perspective from
other online social networks from different cultural environments is still lacking.
Despite the fact that a very significant share of global SNS users are Chinese,
specific Chinese SNSs have been mostly neglected by research so far, with few
exceptions (Chu and Choi, 2011; Jackson and Wang, 2013; Wang et al., 2012).
The Chinese SNS Qzone, for instance, in spite of the popularity of this SNS
among Chinese adolescents, has been hardly studied previously. While
government censure in China blocks the entry of most western SNSs into the
country, highly developed local social networks such as RenRen, Weibo, and
Qzone have been very successful. Little known outside of China, these networks
have millions of users (Sansone et al., 2012). Qzone is the most popular SNS
among Chinese youth aged between 15 and 20. The functioning of Qzone is
similar to other SNSs such as Facebook including personal pages, comments,
pictures and videos. In 2013, Qzone had 600 million registered users, placing it
internationally in second position, with Facebook in first place (Millward,
2013a, 2013b). Commercial brands can be present in Qzone through their
personal profiles or by creating events that Qzone users are invited to take part
in. Many brands have created official brand pages in this SNS in recent years.
Since there are significant cultural differences between Western and Eastern
cultures, it is important not to concentrate SNS research only on the USA and
Facebook but to study other online social networks from different cultural
environments as well. Thus, a secondary aim of this research is contributing to a
broader perspective of user behavior in SNSs with a study conducted among
Qzone users.
The Relationship between Gratifications from Social Networking Site Use 31

2 Conceptual Framework
Uses and gratifications (U&G) theory (Katz and Foulkes, 1962) proposes
that consumers use media because they actively seek to satisfy needs and to
obtain gratifications that can be hedonic (excitement of watching a movie) or
utilitarian (receive relevant information). U&G theory studies media content
(including advertising) analyzing the extent to which it meets consumers’ needs
in terms of information, and emotional stimulation, such as entertainment,
escapism, and fun (McQuail, 1983; Sansone et al., 2012). Previous research has
shown the relevance of U&G theory as a specific framework to predict SNS
usage behavior (Lee and Ma, 2012). According to the application of the U&G
approach to SNS use, principal motivations of SNS users are socializing,
information search, and entertainment (Chua et al., 2012; Diddi and LaRose,
2006; Dunne et al., 2010; Ko et al., 2005). Previous research has confirmed this
gratification pattern for the specific case of Chinese SNSs too (Apaolaza et al.,
2014; Chu and Choi, 2011; Jackson and Wang, 2013; Ku et al., 2013; Wang et
al., 2012).

Deriving personal gratifications from socializing is one of the principal


motives for SNS use, particularly for adolescents (Chen and Marcus, 2012;
Howard and Corkindale, 2008; Kim, Sohn and Choi, 2011; LaRose and Eastin,
2004; Lee and Ma, 2012). For college students, for instance, socializing is one
of the main gratifications from participation in Facebook groups (Park et al.,
2009). Consumers frequent SNSs to engage in immediate, selective, and
efficient interpersonal communication with others. Such communication
provides a mean to achieve acceptance and endorsement from others at an
interpersonal level (Urista et al., 2008). Socializing on SNSs can boost self-
esteem, too. Facebook use, for instance, can assist low self-esteem individuals to
overcome inhibitions in their off-line relationships. SNSs provide users with
gratifications from bridging and bonding in their social relationships, as well as
from preserving social capital (Ellison et al., 2007; Steinfield et al., 2008).
Indeed, particularly for adolescents, maintaining relationships is a central motive
for SNS use, since young individuals have a strong need for group belonging
and self-expression. Facebook is being used by college students to locate old
friends and ex-school friends, contact them, and catch up on their life (Dunne et
al., 2010).
Based on the U&G approach to SNS use, we propose a positive relationship
between the degree to which an adolescent SNS user perceives gratifications
from online socializing and the degree of interacting with online brand
personalities on the SNS. The greater the perceived socializing gratifications, the
higher should be the motivation to interact with other SNS members. Since
online brand personalities appear in a very similar way to people’s SNS
32 Apaolaza, Hartmann, He, Barrutia, and Echebarria

identities, this effect should also be present for brand personalities on the SNS.
Thus, perception of socializing gratifications should relate positively to
adolescent user’s online brand interactions.

H1: Gratifications from socializing on a SNS have a positive relationship with


adolescents’ brand interactions on the SNS.

According to U&G theory, media users also derive significant gratifications


from information obtained through media usage. Information seeking may thus
constitute a significant motive for SNS use for the majority of users, including
adolescents. Information seeking on SNSs can offer gratifications to users since
SNSs can constitute a useful and up to date source of information on fashion,
events, news, etc. Information can also be interactively shared among peers
(George et al., 2013; Lee and Ma, 2012; Luo, 2002; Papacharissi and Rubin,
2000). Thus, gratifications from information search on a SNS are not only
delivered by receiving information, but also by sharing it with other users (Ko et
al., 2005; LaRose and Eastin, 2004; Leung, 2007). We propose that the
perception of SNS-information seeking gratifications relate positively to the
motivation to interact with brands online.

H2: Gratifications from information seeking on an SNS have a positive


relationship with adolescents’ brand interactions on the SNS.

A further significant motive of media usage, indicated by U&G theory, and


which applies also to SNS use, is the search for gratifications from
entertainment. Apart from the entertaining value of social interaction and
consumption of information, many SNSs, including Qzone, offer specific
entertainment such as the possibility to engage in online games. Entertainment
through games and leisure activities provides emotional gratifications, based on
positive affective stimulation and providing remedy against boredom.
Enjoyment and pleasure have been identified as significant motives for the usage
of pleasure-oriented or hedonic information systems including online games
(Van der Heijden, 2004). A number of researchers have considered SNSs as
recreational information systems of this type (Kang and Lee, 2010; Lee et al.,
2010; Lin and Lu, 2011; McQuail, 2005; Nov et al., 2010; Sledgianowski and
Kulviwat, 2009). We propose that gratifications from entertainment may
constitute significant a motive for engaging in SNS brand interactions (see
conceptual framework depicted in Figure 1).

H3: Gratifications from entertainment on an SNS have a positive relationship


with adolescents’ brand interactions on the SNS.
The Relationship between Gratifications from Social Networking Site Use 33

3 Method

3.1 Sample and Procedure


To test the hypothesized relationships, we conducted a study with self-
administered surveys on a sample of Chinese adolescent users of the SNS
Qzone. The sample consisted of Shanghai resident adolescents (N=220) aged 14
to 19 years (M=16.71), 58.6% female and 41.4% male, who possessed Qzone
personal profiles. Participants were recruited from seven Shanghai high-school
centers, selected by convenience in locations adjacent to the university campus
where the study was based. Participation was voluntary and the participants were
informed that their responses would remain anonymous. Prior to the study we
obtained ethical clearance as well as permissions from the high-school centers
and expressed parental consent.

3.2 Measurement
A self-administered questionnaire was designed to measure consecutively
SNS brand interaction, gratifications from socializing, information seeking and
SNS entertainment gratifications.
Brand interactions were measured on 4-point Likert-type agreement scales
with response categories ranging from not at all = 1 to very much = 4 by rating
the following items: “I usually contact brands within Qzone”, “I like browsing
brands on Qzone”, “I usually interact with brands on Qzone”.
The items for the measurement of SNS gratifications were based on accepted
measures from previous research on Uses and Gratifications in SNSs (Lee and
Ma, 2012; Lee et al., 2010; Park et al., 2009). Indicators for all three constructs
were rated on identical agreement scales as the brand interaction measure.
Indicator sentences and constructs are presented in Table 1.
Scale validity was addressed by confirmatory factor analysis using the SPSS
AMOS 20 Maximum Likelihood algorithm. Goodness-of-fit index (GFI) = 0.94,
adjusted goodness-of-fit index (AGFI) = .91 (Jöreskog and Sörbom, 1984;
Bollen, 1989), normed fit index (NFI) = .93, comparative fit index (CFI) = .97
(Bentler, 1990), root mean square error of approximation (RMSEA) = .05
(Steiger and Lind, 1980) and root mean square residual (RMR) = .03 (Byrne,
2001; Steiger, 1990) confirmed a satisfactory adjustment of the measurement
model to the underlying data. With variance extracted ranging .56 to .67 and
composite reliability ranging .79 to .86 as shown in Table 1 (Bagozzi and Yi,
1994; Hair, Anderson, Tatham and Black, 1998), convergent validity can be
considered adequate (Bagozzi and Yi, 1994; Hair, Anderson, Tatham and Black,
1998). Discriminant validity was also confirmed with the square of variable
34 Apaolaza, Hartmann, He, Barrutia, and Echebarria

correlations (Table 2) not exceeding explained variance in any case (Anderson


and Gerbing, 1988; Fornell and Larcker, 1981).

Table 1: Measurement indicators and constructs


Factor Variance
Factor Reliability
loadings explained
Socializing gratification
I can interact with my friends when
.81 .63 .84
sharing news.
I feel I can talk more with my friends. .88
I can stay in contact with my friends
.69
and get closer to them.
Information seeking
I like using Qzone for keeping up to
.74 .56 .79
date on the latest news and events.
It helps me to find helpful information. .75
It is easy to retrieve information when
.75
I need to.
Entertainment gratification
I find entertainment on Qzone
.75 .67 .86
pleasant.
I find entertainment on Qzone
.92
enjoyable.
I find entertainment on Qzone exciting. .77
Brand interaction
I like browsing brands on Qzone. .76 .65 .84
I usually contact brands within Qzone. .87
I usually interact with brands on
.78
Qzone.

4 Results
Table 2 shows the correlation of variables. The conceptual model was
addressed by structural equation analysis with Amos 20. Model fit was
appropriate with all fit indicators close to the perfect fit criterion of 1.0 (GFI =
.94, AGFI = .91, NFI = .93, CFI = .97), RMR = .03, and RMSEA = .05 (Bollen,
1989). Also the chi-square/df value = 1.75 was below the 3.0 cut-off criterion
(Kline, 2010).
The Relationship between Gratifications from Social Networking Site Use 35

Table 2: Construct correlations


Information
Socializing Entertainment
seeking
Information seeking .39
Entertainment .53 .43
Brand interaction .17 .18 .27

As shown in Figure 2, findings did not support the relationship of brand


interaction with gratifications from socializing, proposed in hypothesis one
(SRC=.02, p=.84). In addition, also the postulated relationship with information-
seeking gratifications (H2) was disconfirmed (SRC=.07, p=.49). Results did
confirm the proposed influence of gratifications derived from entertainment on
interaction with brands within the SNS (SRC=.23, p=.01). The third hypothesis
was thus supported by the empirical study. Overall, the search for gratifications
from entertainment seems to constitute the main and possibly only motive for
engaging with brands on SNSs, at least with regard to the studied Shanghai
resident teenagers.

Socializing Information Entertainment


gratification seeking gratification

.07
.02
.23*

SNS
Brand interaction

Notes: * p<.05; standardized regression coefficients; model fit: GFI = .94;


AGFI = .91; NFI = .93; CFI = .97; RMR = .03; RMSEA = .05; Chi-
square = 84.21; Chi-square/df = 1.75

Figure 1: Effects of SNS user gratifications on brand interactions.


36 Apaolaza, Hartmann, He, Barrutia, and Echebarria

5 Discussion and Implications


The literature still lacks a comprehensive theoretical approach to the
advertising effects of the interaction with brand personalities on SNSs. The aim
of this study is to address this gap by studying the motives that induce
consumers to engage with brand on SNSs. The specific focus of this research is
on adolescent SNS users. We propose a theoretical framework based on Uses
and Gratifications theory (Katz and Foukles, 1962) comprising three categories
of gratifications potentially derived from SNS usage which may positively relate
to the motivation to interact with brands on SNSs: gratifications from socializing
on SNSs, information seeking and entertainment.
Overall, this study supports the Uses and Gratifications approach to SNS
usage behavior. The principal contribution of this research is to show that of the
three gratification classes proposed by our theoretical model and derived from
Uses and Gratification theory, only entertainment gratifications seem to have a
positive relationship with SNS brand interaction. Findings did not provide
support for a significant influence of either socializing gratifications or
gratifications derived from information search. Findings therefore contrast to
recent results for the specific case of Facebook user highlighting the importance
of relationship-oriented factors in consumer’s motivations to engage with
Facebook brand pages (Tsai and Men, 2013). Adolescent users on Qzone seem
to be primarily driven by entertainment gratifications but not by the need for
socializing or the search for specific information. Findings are in line with a
number of researchers recommending brand communications on SNSs that focus
on delivering contents which entertain SNS users, instead of information about
product and brand features (Fosdick, 2012; Kelly et al., 2010; Stafford, 2008;
Williams, 2010). Particularly for adolescents, it has been suggested that brand
page contents on SNSs should be centered on emotional experiences offering
hedonic value rather than on product benefits (Goldsmith and Lafferty, 2002).
As a secondary contribution, this research offers a perspective of user
behavior on a Chinese SNS, while most existing research has focused on western
SNSs such as Facebook and Twitter. Specifically Qzone, popular among young
Chinese and with a user base which rivals that of popular and better known
western SNSs, has not been studied previously in terms of user behavior.

Our findings have significant implications for advertisers on SNSs. To


successfully engage with SNS users, advertisers need to understand the motives
for user’s interaction with brands on the SNS. Our results show that users who
perceive more entertainment value in SNS usage also engage more with brand
pages on the SNS. This seems to be particularly true for young SNSs users.
Advertisers who deliver entertainment through their SNS brand pages will likely
be more successful in engaging consumers to interact with the brand. Brand
The Relationship between Gratifications from Social Networking Site Use 37

pages delivering mostly information will likely be less engaging. Apparently,


consumers also do not visit brand pages looking for social contact. Brand
content on SNSs should therefore principally offer entertainment and fun. SNSs
offer many possibilities to entertain users through videos and online games.
Specific advergames have been shown to be very effective in achieving
advertising effects (Terlutter and Capella, 2013; Waiguny, Nelson and Terlutter,
2012). Advertisers should use the opportunity to engage SNS users in such
brand related gaming activities.

6 Limitations and Future Research


A number of limitations of this research are the consequence of its specific
scope focusing on Chinese adolescent users of the Chinese SNS Qzone. The
generalizability of results to SNS use in general may be limited by the type of
data collected. Particularly, users of Western SNSs such as Facebook and users
from different cultural backgrounds may show different usage patterns and
underlying motives. Still, given the international and metropolitan character of
Shanghai, consumer behavior may be not much different from that of residents
of other international metropolis. On the other hand, the generalizability for
Chinese adolescents may be somewhat limited as well, since Shanghai residents
may have a consumer culture which differs from that of average Chinese
mainland inhabitants. Future research should therefore address the proposed
conceptual model in different SNSs and cultural environments, for instance, for
Western users of a SNS such as Facebook.
The further development of the explanatory framework of motivations for
interacting with brands on SNSs could benefit from considering a broader
approach than the three SNS usage gratifications proposed in this research based
on Uses and Gratifications theory. Thus, future research should extend the
analysis to specific hedonic characteristics of SNS brand pages and the
gratifications provided to users by those features. The role of additional
variables such as brand involvement and trust, as well as user perception of
brand communications, should also be considered. The SNS brand interaction
construct still lacks a comprehensive measurement scale. The need for such a
measurement instrument provides further avenues for future research which
should address this construct in detail and analyze its different aspects.

7 Acknowledgments
This study received financial support from National Natural Science
Foundation of China (Grants No. 71372177 and 71072152), as well as research
grants GIU 11/17, STK S-PE 10UN29, EHU 10/13 and FESIDE Foundation.
38 Apaolaza, Hartmann, He, Barrutia, and Echebarria

8 References
Anderson, J. C. and Gerbing, D. W. (1988), “Structural equation modeling in practice: A
review and recommended two-step approach,” in: Psychological Bulletin, Vol. 103,
411-423.
Apaolaza, V.; He, J. and Hartmann, P. (2014), “The effect of gratifications derived from
use of the social networking site Qzone on Chinese adolescents’ positive mood,” in:
Computers in Human Behavior, Vol. 41, 203-211.
Bagozzi, R. P. and Yi, Y. (1994), “Advanced Topics in Structural Equation Models” in:
Bagozzi, R. P., ed. Advanced Methods of Marketing Research. Oxford: Blackwell, 1-
51.
Bentler, P. M. (1990), “Comparative fit indexes in structural models,” in: Psychological
Bulletin, Vol. 107, 238-246.
Bollen, K. A. (1989), Structural equations with latent variables. New York, NY: John
Wiley & Sons, Inc.
Byrne, B. M. (2001), Structural Equation Modeling with AMOS. Mahwah, New Jersey:
Lawrence Erlbaum.
Chen, B. and Marcus, J. (2012), “Students’ self-presentation on Facebook: An
examination of personality and self-construal factors,” in: Computers in Human
Behavior, Vol. 28, 2091-2099.
Chu, S. C. and Choi, S. M. (2011), “Electronic Word-of-Mouth in social networking
sites: A cross-cultural study of the United States and China,” in: Journal of Global
Marketing, Vol. 24, 3, 263-281.
Chua, A. Y. K.; Goh, D. H. and Lee, C. S. (2012), “Mobile content contribution and
retrieval: An exploratory study using the uses and gratifications paradigm,” in:
Information Processing & Management, Vol. 48, 1, 13–22.
Cooper, L. (2011), “Blend in with the scene to get maximum ad stand-out (the
effectiveness of advertising based on social media and networks),” in: Strategic
Direction, Vol. 27, 8, 24-26.
Diddi, A. and LaRose, R. (2006), “Getting hooked on news: Uses and gratifications and
the formation of news habits among college students in an Internet environment,” in:
Journal of Broadcasting & Electronic Media, Vol. 50, 2, 193–210.
Dunne, A.; Lawlor, M. and Rowley, J. (2010), “Young people's use of online social
networking sites – a uses and gratifications perspective,” in: Journal of Research in
Interactive Marketing, Vol. 4, 1, 46-58.
Ellison, N. B.; Steinfield, C. and Lampe, C. (2007), “The benefits of Facebook “friends:”
social capital and college students’ use of online social network sites,” in: Journal of
Computer-Mediated Communication, Vol. 12, 4, 1143-1168.
Fornell, C. and Larcker, D. F. (1981), “Evaluating Structural Equation Models with
unobservable Variables and Measurement Error,” in: Journal of Marketing Research,
Vol. 18, 39-50.
Fosdick, M. (2012), “The globalization of social media: Consumer relationships with
brands evolve in the digital space,” in: Strategic Direction, Vol. 28, 6, 564-570.
George, D. R.; Dellasega, C.; Whitehead, M. M. and Bordon, A. (2013), “Facebook-
based stress management resources for first-year medical students: A multi-method
evaluation,” in: Computers in Human Behavior, Vol. 29, 559-562.
Goldsmith, R. E. and Lafferty, B. A. (2002), “Consumers’ response to websites and their
influence on advertising effectiveness,” in: Internet Research, Vol. 12, 4, 318-328.
The Relationship between Gratifications from Social Networking Site Use 39

Hadija, Z.; Barnes, S. B. and Hair, N. (2012), “Why we ignore social networking
advertising,” in: Qualitative Market Research: An International Journal, Vol. 15, 1,
19-32.
Hair, J. F.; Anderson, R. E.; Tatham, R. L. and Black, W. C. (1998), Multivariate Data
Analysis. NJ: Prentice Hall.
Howard, Y. and Corkindale, D. (2008), “Towards an understanding of the behavioral
intention to use online news services,” in: Internet Research, Vol. 18, 3, 286–312.
Jackson, L. A. and Wang, J. L. (2013), “Cultural differences in social networking site
use: A comparative study of China and the United States,” in: Computers in Human
Behavior, Vol. 29, 910-921.
Jöreskog, K. G. and Sörbom, D. (1984), LISREL VI user’s guide. Mooresville, IN:
Scientific Software.
Kang, Y. S. and Lee, H. (2010), “Understanding the role of an IT artifact in online
service continuance: An extended perspective of user satisfaction,” in: Computers in
Human Behavior, Vol. 26, 353-364.
Katz, E. and Foukles, D. (1962), “On the use of the mass media as escape: Clarification
of a concept,” in: The Public Opinion Quarterly, Vol. 26, 3, 377-388.
Kelly, L.; Kerr, G. and Drennan, J. (2010), “Avoidance of Advertising in Social
Networking Sites: The Teenage Perspective,” in: Journal of Interactive Advertising,
Vol. 10, 2, 16–27.
Kim, Y.; Sohn, D. and Choi, S. M. (2011), “Cultural difference in motivations for using
social network sites: A comparative study of American and Korean college students,”
in: Computers in Human Behavior, Vol. 27, 1, 365-372.
Kline, R. B. (2010), Principles and practice of structural equation modeling. New York,
NY: The Guilford Press.
Ko, H.; Cho, C. H. and Roberts, M. S. (2005), “Internet uses and gratifications: A
structural equation model of interactive advertising,” in: Journal of Advertising, Vol.
34, 2, 57–70.
Ku, Y. C.; Chu, T. H. and Tseng, C. H. (2013), “Gratifications for using CMC
technologies: A comparison among SNS, IM, and e-mail,” in: Computers in Human
Behavior, Vol. 29, 226-234.
Kwon, E.S.; Kim, E.; Sung, Y. and Yoo, C.Y. (2014), “Brand followers. Consumer
motivation and attitude towards brand communications on Twitter,” in: International
Journal of Advertising, Vol. 33, 4, 657-680.
LaRose, R. and Eastin, M. S. (2004), “A social cognitive theory of Internet uses and
gratifications: Toward a new model of media attendance,” in: Journal of
Broadcasting & Electronic Media, Vol. 48, 3, 358–377.
Lee, C. S.; Goh, D. H.; Chua, A. Y. K. and Ang, R. P. (2010), “Indagator: Investigating
perceived gratifications of an application that blends mobile content sharing with
gameplay,” in: Journal of the American Society for Information Science and
Technology, Vol. 61, 6, 1244–1257.
Lee, C. S. and Ma, L. (2012), “News sharing in social media: The effect of gratifications
and prior experience,” in: Computers in Human Behavior, Vol. 28, 331-339.
Leung, L. (2007), “Stressful Life Events, Motives for Internet Use, and Social Support
among Digital Kids,” in: CyberPsychology & Behavior, Vol. 10, 2, 204-214.
Lin, K. Y. and Lu, H. P. (2011), “Why people use social networking sites: An empirical
study integrating network externalities and motivation theory,” in: Computers in
Human Behavior, Vol. 27, 3, 1152-1161.
40 Apaolaza, Hartmann, He, Barrutia, and Echebarria

Luo, X. (2002), “Uses and gratifications theory and e-consumer behaviors: A structural
equation modeling study,” in: Journal of Interactive Advertising, Vol. 2, 2, 44–54.
McQuail, D. (1983), Mass Communication Theory. London: Sage.
McQuail, D. (2005), McQuail’s mass communication theory. Singapore: Sage
Publications Ltd.
Millward, S. (2013a), “Tencent: WeChat now has 271.9 million monthly active users
around the world.” Techinasia 10/06. http://www.techinasia.com/tencent-wechat-272-
million-activer-users-q3-2013.
Millward, S. (2013b), “Check Out the Numbers on China’s Top 10 Social Media Sites
(Infographic).” Techinasia 10/06. http://www.techinasia.com/2013-china-top-10-
social-sites-infographic.
Nov, O.; Naaman, M. and Ye, C. (2010), “Analysis of participation in an online photo
sharing community: A multidimensional perspective,” in: Journal of the American
Society for Information Science and Technology, Vol. 61, 3, 555–566.
Othman-Yousif, R. (2012), “The Extent of Facebook Users’ Interest in the Advertising
Messages,” in: International Journal of Marketing Studies, Vol. 4, 3, 122-133.
Papacharissi, Z. and Rubin, A. M. (2000), “Predictors of Internet use,” in: Journal of
Broadcasting & Electronic Media, Vol. 44, 2, 175–196.
Park, N.; Kee, K. F. and Valenzuela, S. (2009), “Being immersed in social networking
environment: Facebook groups, uses and gratifications, and social outcomes,” in:
CyberPsychology & Behavior, Vol. 12, 6, 729-733.
Rowley, J. (2009), “Online branding strategies of UK fashion retailers,” in: Internet
Research, Vol. 19, 3, 348-369.
Sansone, M.; Moretta, A. and Bruni, R. (2012), “How do companies achieve their
marketing goals with social networks?,” in: Chinese Business Review, Vol. 11, 11,
970-980.
Sledgianowski, D. and Kulviwat, S. (2009), “Using social network sites: The effects of
playfulness, critical mass and trust in a hedonic context,” in: Journal of Computer
Information Systems, Vol. 49, 74-83.
Stafford, T. F. (2008), “Social and usage-process motivation for customer internet
access,” in: Journal of Organizational & End User Computing, Vol. 20, 3, 1-21.
Steiger, J. H. (1990), “Structural model evaluation and modification: An interval
estimation approach,” in: Multivariate Behavioral Research, Vol. 25, 2, 173-180.
Steiger, J. H. and Lind, J. C. (1980), Statistically-based tests for the number of common
factors. Annual Meeting of the Psychometric Society, Iowa City, IO.
Steinfield, C.; Ellison, N. and Lampe, C. (2008), “Social Capital, Self-esteem, and Use of
Online Social Network Sites: A longitudinal analysis,” in: Journal of Applied
Developmental Psychology, Vol. 29, 434−445.
Terlutter, R. and Capella, M. L. (2013), “The gamification of advertising: analysis and
research directions of in-game advertising, advergames, and advertising in social
network games,” in: Journal of Advertising, Vol. 42, 2, 95-112.
Tsai, W. S. and Men, L. R. (2013), “Motivations and Antecedents of Consumer
Engagement with Brand Pages on Social Networking Sites,” in: Journal of Interactive
Advertising, Vol. 13, 2, 76-87.
Urista, M. A.; Dong, Q. and Day, K. D. (2008), “Explaining why young adults use
MySpace and Facebook through uses and gratifications theory,” in: Human
Communication, Vol. 12, 2, 215-229.
Van der Heijden, H. (2004), “User acceptance of hedonic information systems,” in: MIS
Quarterly, Vol. 28, 4, 695-704.
The Relationship between Gratifications from Social Networking Site Use 41

Waiguny, M. K.; Nelson, M. R. and Terlutter, R. (2012), “Entertainment matters! The


relationship between challenge and persuasiveness of an advergame for children,” in:
Journal of Marketing Communications, Vol. 18, 1, 69-89.
Wang, J. L.; Jackson, L. A.; Zhang, D. J. and Su, Z. Q. (2012), “The relationships among
the Big Five Personality factors, self-esteem, narcissism, and sensation-seeking to
Chinese University students' uses of social networking sites (SNSs),” in: Computers
in Human Behavior, Vol. 28, 2313-2319.
Williams, E. (2010), La nueva publicidad. Las mejores campañas. Barcelona: Gustavo
Gili, SL.
Zeng, F.; Huang, L. and Dou, W. (2009), “Social Factors in User Perceptions and
Responses to Advertising in Online Social Networking Communities,” in: Journal of
Interactive Advertising, Vol. 10, 1, 1-13.
Promoting the Shareconomy: Effects of Beneficial Appeals and
Personal Characteristics on the Attractiveness of Renting and
Reselling Platforms

Johanna Schwenk and Verena Hüttl-Maack

1 Introduction
Recently, many reports about the “Shareconomy” and collaborative
consumption (CC) could be found in the media. The Time Magazine even
denoted sharing as one of the “10 ideas that will change the world” and asserted
that the ownership society of the 20th century is more and more replaced by a
sharing society (Walsh, 2011). However, academic literature in this field is
scarce and it seems that academics have just started to investigate this
phenomenon more thoroughly. In this paper, we study the effectiveness of
different advertising messages for transactions that are characterized by the
shared usage of goods. The first transaction we look at is peer-to-peer-renting,
characterized by a shared usage while ownership remains at one party (Durgee
and O’Connor, 1995). The second transaction is reselling, i.e., the sale of a used
good, where the shared usage takes place consecutively and the ownership
changes with the transaction. We put our study in an online context by
investigating ad appeals for online platforms since the internet is an important
driver of the shared usage and shared ownership trend.
In contrast to the common purchase of new goods from a retailer, several
types of benefits can be obtained by forms of CC and are most frequently used in
ads for respective online platforms. Besides economic benefits (i.e., saving
money), the consumer can become part of a community of like-minded
individuals (Hamari, Sjöklint, and Ukkonen, 2015). Moreover, consumers can
save natural resources. With the climate change in mind, the possibility to
behave more environmentally conscious becomes more and more important for
many consumers. In the following paragraphs, we refer to the benefits described
above as financial, community, and environmental benefits. We propose in our
theorizing that the transaction of reselling more strongly corresponds to financial
benefits whereas the transaction of renting is more strongly associated with
community and environmental benefits. Therefore, we expect financial benefit
appeals to be more effective for a reselling platform and community as well as
environmental benefit appeals for a renting platform.

© Springer Fachmedien Wiesbaden 2017


G. Christodoulides et al. (Hrsg.), Advances in Advertising Research (Vol. VII),
European Advertising Academy, DOI 10.1007/978-3-658-15220-8_4
44 Schwenk and Hüttl-Maack

Moreover, we consider three personal characteristics which are of specific


relevance when it comes to CC. The first is the individual’s need to belong. This
need can be one motivational factor for consuming collaboratively (Felson and
Spaeth, 1978). We assume that there is a relationship between the need to belong
and the susceptibility to different beneficial appeals. The second characteristic
we examine is a person’s materialism. There is a wide range of research
describing how materialism affects consumption behavior in general (e.g.,
Richins, McKeage, and Najjar, 1992; Rindfleisch, Burroughs, and Denton,
1997; Tatzel, 2002). We expect this trait to affect the willingness to engage in
renting versus reselling. The third characteristic is environmental awareness. We
assume that green consumer values positively affect CC independent of the
platform type, since both considered transactions (renting and buying used
products) help conserve natural resources.

2 Effects of Beneficial Ad Appeals for Different Platform Types


To the best of our knowledge, there is no research dealing with advertising
messages in the context of CC. To gain insight into how the beneficial appeals
described above may influence consumers, we consult literature on human
motivations for sharing in different network contexts. Referring to self-
determination theory a distinction between intrinsic and extrinsic motivations
can be made that either directly or indirectly determine behavioral intentions
(e.g., Hamari et al., 2015; Lin and Lu, 2011; Nov, Naaman, and Ye, 2009).
Intrinsic motivation reflects the propensity of “doing something because it is
inherently interesting or enjoyable” and extrinsic motivation refers to doing
something because it leads to a certain outcome (Ryan and Deci, 2000, p. 55).
The cost-benefit rationale (financial advantages due to sharing) determined by
Lamberton and Rose (2012) as a main motivational factor constitutes an
extrinsic motivation (Kasser and Ryan, 1996). In line with this, prior research
was able to show that economic drivers are important motivational factors for
CC (e.g., car2go, Airbnb) (Moehlmann, 2015) as well as for buying used
products and renting in general (Durgee and O’Connor, 1995; Roux and Guiot,
2008). Besides the cost-benefit rationale, community aspects play an important
role in sharing contexts. Several community related variables (e.g., social
identity, affiliation, sense of community, connectedness) can influence
consumption behaviors (Bock et al., 2005; Kwon and Wen, 2010; Rosen,
Lafontaine, and Hendrickson, 2011). These community aspects refer to a sense
of belonging, i.e., belonging to a specific social group, the wish to have
relationships with others, and the desire for social contact (Atkinson, Heyns, and
Veroff, 1954; French and Chadwick, 1956; Hill, 1987; Hogg and Terry, 2000;
Kwon and Wen, 2010). These aspects can be considered as key facets of the
concept of belongingness as motivation (Lee and Robbins, 1995). According to
Promoting the Shareconomy 45

belongingness theory, humans have an innate need to belong (Baumeister and


Leary, 1995) which can be classified as intrinsic motivation that serves to satisfy
basic and inherent psychological needs (Kasser and Ryan, 1996). The need to
belong does not refer to a particular relationship but can be satisfied by any
social interaction (Baumeister and Leary, 1995). Online selling or renting
platforms, which offer the possibility to be part of a community, provide an
opportunity to seek out interpersonal contacts to satisfy a need to belong.
Moreover, the motive of behaving environmentally conscious is increasingly
relevant in the context of CC (Briceno and Stagl, 2006). Environmental concern
can be defined as “a general attitude toward preserving the environment”
(Minton and Rose, 1997) with indirect effects on behaviors through behavioral
intentions (Gill, Crosby, and Taylor 1986). As far as ideology and norms instead
of reputation are the consumer’s motivation, environmental concerns can be
classified as intrinsic (Hamari et al., 2015). Previous research on social networks
showed that extrinsic as well as intrinsic factors affect community participation
(Nov et al., 2009) and the use of network sites (Lin and Lu, 2011). Moreover, it
has been shown that extrinsic motivations (e.g., economic benefits) directly
determine behavioral intentions and intrinsic motivations (e.g., sustainability)
indirectly affect intentions through attitude toward CC (Hamari et al., 2015).
If we compare the purchase of a used good with the transaction of renting,
the latter one – especially peer-to-peer – is more strongly characterized by an
interaction between the person who temporarily gives his/her possession away
and the person who rents it (Durgee and O’Connor, 1995). Reliance plays a
central role since the lender of the good has to trust that the renter returns the
good in an immaculate condition and the renter relies on the behavior of the
lender in terms of the condition of the good and timely provision (Botsman and
Rogers, 2010). Furthermore, a connection between two individuals is built
insofar that one person uses a good that the other one owns. Thus, we posit that
community benefits possess a higher fit to associations that are activated in a
renting context. In case of a green appeal, the concept of environmental
protection is activated. Since renting a product is most efficient for saving
resources in terms of maximizing use and extending the life span of items, we
expect that environmental benefits can be pursued to a higher extend by the
transaction of renting.
In contrast to that, when purchasing a used product the peer-to-peer
interaction is typically rather low or implicit. The relation between people only
consists of the usage of a good the other person has used before. The connection
between these two individuals (the new owner and the pre-owner) is rather
subtle in nature. Thus, we argue that social benefits like being part of a
community are not activated to a considerable extent. The transaction is mainly
characterized by an exchange of product and money at one time. Hence, we
46 Schwenk and Hüttl-Maack

argue that from the perspective of the buyer, the motive of saving money
compared to the other motives we consider dominates when a purchase of used
goods is made. Thus, we expect that in this case, advertising the extrinsic benefit
of saving money fits better to the transaction type than using community or
environmental benefits in ads. Therefore, we hypothesize that communicating
financial benefits is more effective for the transaction type of reselling. Although
buying used products is associated with saving resources it can be assumed that
this is not the driving motivation (Botsman and Rogers, 2010). Figure 1
summarizes our theoretical considerations and empirical findings that lead us to
the following hypotheses:

H1: Appeals addressing intrinsic (community, environmental) in contrast to


extrinsic motivations cause a higher attitude toward the platform in the case of a
renting platform.

H2: Appeals addressing extrinsic (financial) in contrast to intrinsic motivations


cause a higher behavioral intention in the case of a reselling platform.

Figure 1: Illustration of theoretical considerations on benefit appeals and platform types

3 Effects of Personal Characteristics


We assume that the need to belong, materialism, and environmental
awareness are personal characteristics of specific relevance when it comes to
CC. According to Baumeister and Leary (1995) the need to belong provokes
Promoting the Shareconomy 47

goal-directed activity expressed by a tendency to form social bonds until a


satiation is achieved. Therefore, we assume that individuals with a high need to
belong express higher behavioral intentions towards a community appeal since
the opportunity for interpersonal contacts is emphasized. When the need to
belong is already well satisfied (i.e., need to belong is low) a community appeal
loses its persuasive influence. Literature shows that belongingness - often
manipulated by using social exclusion (e.g., Baumeister et al., 2002; Gardner,
Pickett, and Brewer, 2000; Twenge et al., 2001, 2007) - has an effect on
consumption behavior (Mead et al., 2011) and the readiness to spend money
(Carter-Sowell, Chen, and Williams, 2008). It is interesting with respect to our
study that Janssen, van Beest, and Mead (2014) could ascertain that social
exclusion enhances the susceptibility to certain persuasion strategies. The
authors found that a social proof tactic influences socially excluded participants
more strongly than socially accepted ones since they seek social connection. Our
considerations lead us to the following hypothesis:

H3: Using a community benefit appeal as an advertising message leads to higher


behavioral intention for individuals with a high need to belong compared to a
low need to belong.

Moreover, we consider the devotion to material needs and desires.


Materialistic individuals place greater emphasis on possession and have a
greater interest in getting and spending (Belk, 1984; Rassuli and Hollander,
1986; Richins and Dawson, 1992). We argue that materialistic consumers have
higher behavioral intentions to purchase a used good than consumers that score
low on the materialism scale. Note that we do not expect materialism to affect
renting intentions since no possession can be obtained. We hypothesize:

H4: In case of a reselling platform, individuals with a high materialistic attitude


show a higher behavioral intention in comparison to individuals with a low
materialistic attitude.

Further, we consider environmental awareness in terms of green consumers


whose purchase behavior is influenced by environmental concerns (Haws,
Winterich, and Naylor, 2014). Renting products as well as buying used products
offer the possibility to save resources. Thus, in the context of CC a positive
effect of environmental awareness can be expected. However, academic
literature mentions a gap between positive attitudes toward sustainability and
actual consumption behavior (Prothero et al., 2011). Minton and Rose (1997),
for instance, could ascertain that while environmentally concerned attitudes
positively affect intentions to act in environmentally concerned ways, action
48 Schwenk and Hüttl-Maack

(e.g., environmentally friendly product choices) is not affected. Considering


environmental concern as an attitude separate from its subsequent intentions and
behaviors, we only expect a positive effect on the attitude toward a CC platform.

H5: Environmental awareness has a positive influence on the attitude toward a


CC platform.

4 Experimental Study
Experimental design: The study was based on a 3 (benefit appeal: financial,
community, environmental) × 2 (platform type: reselling, renting) between-
subjects design. In the experimental scenarios, we used two existing platforms
from Germany where the study was conducted. The reselling platform was
“rebuy.de”. The renting platform we used was “leihdirwas.de” which enables
registered users to rent and borrow a wide range of used goods. As a test
product, we chose a stereo shelf system of high quality. To induce a need for the
test product, we created scenarios.
Pretest: One purpose of our pretest was to ensure that the chosen platforms
do not differ substantially in terms of familiarity or trustworthiness. We
interviewed 33 students of a German university and got similar scores of
awareness, prior usage and perceived trustworthiness. Moreover, in order to
identify reasonable prices for the main study scenarios, the pretest respondents
were given a short scenario in which they were instructed to imagine that they
needed a powerful stereo system since they were planning a big party; thus, they
were searching for a stereo system online. Then, the stereo system with its main
features was presented and the respondents were asked which price they were
willing to pay for renting or repurchasing it. We used the median prices (60 € for
renting the product and 300 € for buying it) in our main study.
Test stimuli: We created fictitious front pages of the platform websites. The
pages were created by modifying the design existing in reality. Besides the front
pages, we created a follow-up page with the concrete product offer. The
financial benefit conditions consist of a persuasive message with focus on
financial advantages a customer is able to obtain, e.g., customers can save up to
80% by renting products from another platform user instead of purchasing them
from a retailer. The community benefit conditions (see Fig. 2 for an example)
concentrated on community aspects with the promise to become part of a
community and get related to people. The environmental benefit pointed out that
environmental pollution caused by unused products is avoidable through
sharing. We added message-supporting pictorial elements such as money for the
financial benefit or a group of people for the community benefit. We also
created two product displays that showed the offer for renting and reselling
including a brand name, prices, and general features of the product. To enhance
Promoting the Shareconomy 49

the authenticity, we arranged a positive evaluation of the person that provides


the product and a website certification seal.
Sample, procedure, and measures: We conducted an online survey with 273
students from a German university (Mage= 25 years; 52.4% females). According
to the experimental design the participants were randomly assigned to one of the
six groups what resulted in a minimum cell size of 27. At first, respondents were
given the short scenario described in the pretest section to create the need for a
stereo system. Subsequently, the fictitious front page of the platform with a
financial, community or environmental benefit appeal was presented. Then, the
product offer of the stereo system followed. Next, the respondents reported their
attitude towards the platform and behavioral intention. Attitude was assessed by
using two items (“The platform is unattractive/attractive”; “I do not like it/I like
it very much”; r=.84). In the case of the reselling platform, the behavioral
intention was the intention to purchase the stereo system. The respondents
reported their agreement to the statement “If needed I would buy the product”.
Analogously, we assessed the intention to rent the product in the renting
platform conditions. The need to belong was assessed with five items adopted
from Leary et al. (2012) and materialism with three items from the scale
proposed by Richins (1987) (see Tab. 1). Environmental awareness was
measured by the green consumer value scale developed by Haws et al. (2014).
All scales were seven-point-scales. Finally, we measured demographics.

Translation: Speech bubble left: reBuy.de is not only a great place for used stuff, but
also to get to know new people. Speech bubble right: Join us and become part of our
group! Just click through and benefit from the constantly growing community.
Header: Your place for used stuff. Text on the bottom: Join the community as well
and take a look at reBuy.de!

Figure 2: Example of stimuli: Fictitious starting page with community benefit appeal
50 Schwenk and Hüttl-Maack

Table 1: Scales used to assess the personal characteristics


CONSTRUCTS/SCALES
Need to belong (adapted from Leary et al. 2012, p. 624; Cronbach’s α=.682)
“If other people don't seem to accept me, I don't let it bother me”, reversed;
“I do not like being alone”
“I have a strong need to belong”
“It bothers me a great deal when I am not included in other people's plans”
“My feelings are easily hurt when I feel that others do not accept me”
Materialism (adapted from Richins 1987, p. 354; Cronbach’s α=.59)
“It is important to me to have really nice things”
“I`d be happier if I could afford to buy more things”
“It sometimes bothers me quite a bit that I can’t afford to buy all things I would like”
Environmental awareness (adapted from Haws et al. (2014); Cronbach’s α=.90)
“It is important to me that the products I use do not harm the environment.”
“I consider the potential environmental impact of my actions when making many of my
decisions.”
“My purchase habits are affected by my concern for our environment.”
“I am concerned about wasting the resources of our planet.”
“I would describe myself as environmentally responsible.”
“I am willing to be inconvenienced in order to take actions that are more
environmentally friendly.”

Results: To test H1 and H2, we compared the effects of the benefit appeals
addressing the extrinsic motive with the benefit appeals addressing the intrinsic
motives (i.e. environmental and community appeals collapsed) by using
univariate ANOVAs with planned comparisons for both types of online
platforms. The results are presented in Table 2 and reveal that when intrinsic
motivations are addressed, a higher attitude towards the renting platform results
(t(145)=2.175; p=.016) what is in line with H1. As anticipated, no effect on the
renting intention occurs (t(145)=.582; p=.282). For the reselling platform, we
found a higher purchase intention when extrinsic motivations are activated
(t(122)=2.061; p=.021) (see Tab. 3). Thus, H2 is supported. As expected the
attitude towards the reselling platform was not affected by the ad appeals
(t(121)= -.078, p=.469).
Promoting the Shareconomy 51

Table 2: Effects of the beneficial appeals in ads for a renting platform (H1)

Renting Platform (H1)


Intrinsic motive
Extrinsic motive Environmental Community Test
(financial benefit) benefit benefit values

3.80 4.08 F(2,145) =2.463; p=.09


Aplatform
3.28 3.94 t145=2.175; p=.016

4.25 4.39 F(2,145)=.17; n.s.


Behavioral
intention
4.13 4.32 t145= .582; n.s.

Table 3: Effects of the beneficial appeals in ads for a reselling platform (H2)

Reselling Platform (H2)


Intrinsic motive
Extrinsic motive Environmental Community Test
(financial benefit) benefit benefit values

4.59 4.35 F(2,121)=.30; n.s.


Aplatform
4.44 4.47 t121= -.078; n.s.

4.48 4.43 F(2,122)=2.162; p=.12


Behavioral
intention
5.26 4.46 t122=2.061; p=.021

Due to the fact that the need to belong, materialism, and environmental
awareness was not manipulated but measured, we use moderated regression
models and simple slope analyses as proposed by Fitzsimons (2008) and Irwin
and McClelland (2001) to test H3, H4, and H5. The results are reported in Table
4, 5, and 6. We regressed behavioral intention on the reported need to belong,
the type of benefit appeal coded as dummy variable and the interaction of these
two independent variables. In order to test the influence of the need to belong
on the susceptibility to community benefit appeals, we examine the slope of the
need to belong at the level of the dummy variable where the community benefit
appeal (instead of the financial appeal) is present. Since neither the interaction
term nor the slope is significant (β=.194, t=.953, p=.342), H3 has to be rejected.
52 Schwenk and Hüttl-Maack

Table 4: Moderated regression results of the effect of need to belong and benefit appeal
on behavioral intention (H3)

β SE t p
Intercept 3.571 .865 4.127 .000
Need to belong (NTB)a .194 .204 .953 .342
Benefit appeal (BA)b 1.045 1.249 .837 .404
NTB x BA -.187 .297 -.632 .529
Notes: Coefficients reported in the table are unstandardized.
a Measured on scale ranging from 1=low to 7=high.
b Coded as dummy variable with 0= community benefit and 1=financial benefit.

The test of H4 was performed analogously. In H4, we argued that in case of a


reselling platform, individuals with a high materialistic attitude show a higher
behavioral intention in comparison to individuals with a low materialistic
attitude. We regressed behavioral intentions on reported materialism, the portal
type and their interaction. The results (see Tab. 5) reveal that the slope of
materialism is significant and positive (β=.246, t=2.037, p=.043) when the portal
type variable is set to zero which indicates the presence of the reselling platform.
This result provides strong support for H4.

Table 5: Moderated regression results of the effect of materialism and platform type on
behavioral intention (H4)

β SE t p
Intercept 3.592 .533 6.743 .000
Materialism (M)a .246 .121 2.037 .043
Portal type (P)b .834 .731 1.140 .255
MxP -.288 .169 -1.706 .089
Notes: Coefficients reported in the table are unstandardized.
a Measured on scale ranging from 1=low to 7=high.
b Coded as dummy variable with 0=reselling portal and 1=renting portal.

In H5, we argued that environmental awareness positively affects the attitude


toward a sharing platform independent of the transaction type. The results (see
Tab. 6) reveal that the portal type and its interaction with environmental
awareness have no effects on attitude toward the platform as expected. However,
environmental awareness has a positive influence (β=.230, t=2.264, p=.024).
Thus, H5 can be supported. A regression on behavioral intention showed no
significant effects as it was anticipated.
Promoting the Shareconomy 53

Table 6: Moderated regression results of the effect of environmental awareness and


platform type on attitude toward the platform (H5)

β SE t p
Intercept 3.548 .440 8.059 .000
Env. awareness (E)a .230 .102 2.264 .024
Portal type (P)b -.576 .579 -.993 .321
ExP -.039 .136 -.288 .774
Notes: Coefficients reported in the table are unstandardized;
a Measured on scale from 1=low to 7=high;
b Dummy variable with 0=reselling portal and 1=renting portal.

5 Conclusion
Our findings lead to the recommendation to emphasize community and
environmental benefits in ads for renting platforms and financial benefits for
reselling platforms. We also show that depending on the type of motivations that
are addressed (intrinsic vs. extrinsic), different dependent variables are affected
(attitudes vs. behavioral intentions). The community and environmental benefit
only affected the attitude towards the renting platform but not the behavioral
intention. Contrarily, in the case of the ad containing the financial benefit for the
reselling platform, only the behavioral intention was affected. This result
corresponds to the theoretical notion that intrinsic motivations affect behavior in
a rather indirect way via attitudes while the extrinsic financial motivation works
more directly on behavioral intent. Our results regarding materialism suggest
that providers of reselling platforms could also address the desire of possession
in their advertising messages. Moreover, environmental awareness is revealed as
an important personal characteristic that positively affects the attitude toward
CC independently of the transaction or platform type that is involved.
A limitation of this study is the lack of systematic manipulation of variables
such as the need to belong as it is often proposed in literature. This reduces the
meaningfulness of our rather preliminary analysis regarding the effect of the
need to belong since our procedure must have resulted in data with
comparatively low variance in this variable. This limitation may at least partly
explain the weak results on H3. Since the materialism and environmental
awareness construct is a more stable trait, measuring it seemed reasonable.
54 Schwenk and Hüttl-Maack

6 References
Atkinson, J. W., Heyns, R. W., and Veroff, J. (1954), „The Effect of Experimental
Arousal of the Affiliation Motive on Thematic Apperception,” in: The Journal of
Abnormal and Social Psychology, Vol. 49, 3, 405-410.
Baumeister, R. F. and Leary, M. R. (1995), “The Need to Belong: Desire for
Interpersonal Attachments as a Fundamental Human Motivation,” in: Psychological
Bulletin, Vol. 117, 3, 497-529.
Baumeister, R. F., Twenge, J. M. and Nuss, C. K. (2002), “Effects of Social Exclusion
on Cognitive Processes: Anticipated Aloneness Reduces Intelligent Thought,” in:
Journal of Personality and Social Psychology, Vol. 83, 4, 817-827.
Belk, R. W. (1984), “Three Scales to Measure Constructs Related to Materialism:
Reliability, Validity, and Relationships to Measures of Happiness,” in: Kinnear, T.
C., ed. Advances in Consumer Research, Vol. 11, 1. Provo, UT, Associations for
Consumer Research, 291-297.
Bock, G.-W., Zmud, R. W., Kim, Y.-G. and Lee, J.-N. (2005), “Behavioral Intention
Formation in Knowledge Sharing: Examining the Roles of Extrinsic Motivators,
Social-psychological Forces, and Organizational Climate,” in: MIS Quarterly, Vol.
29, 1, 87-111.
Botsman, R. and Rogers, R. (2010), “What’s Mine Is Yours: Rhe Rise of Collaborative
Consumption. New York: HarperCollins.
Briceno, T. and Stagl, S. (2006), “The Role of Social Processes for Sustainable
Consumption,” in: Journal of Cleaner Production, Vol. 14, 17, 1541-1551.
Carter‐Sowell, A. R., Chen, Z. and Williams, K. D. (2008), “Ostracism Increases Social
Susceptibility,” in: Social Influence, Vol. 3, 3, 143-153.
Durgee, J. F. and O'Connor, G. C. (1995), “An Exploration into Renting as Consumption
Behavior,” in: Psychology and Marketing, Vol. 12, 2, 89-104.
Fitzsimons, G. J. (2008), “Death to Dichotomizing,” in: Journal of Consumer Research,
Vol. 35, 1, 5-8.
Felson, M. and Spaeth, J. L. (1978), “Community Structure and Collaborative
Consumption,” in: American Behavioral Scientist, Vol. 21, 4, 614-624.
French, E. G. and Chadwick, I. (1956), “Some Characteristics in Affiliation Motivation,”
in: The Journal of Abnormal and Social Psychology, Vol. 52, 3, 296-300.
Gardner, W. L., Pickett, C. L. and Brewer, M. B. (2000), “Social Exclusion and Selective
Memory: How the Need to Belong Influences Memory for Social Events,” in:
Personality and Social Psychology Bulletin, Vol. 26, 4, 486-496.
Gill, J. D., Crosby, L. A. and Taylor, J. R. (1986), “Ecological Concern, Attitudes, and
Social Norms in Voting Behavior,” in: Public Opinion Quarterly, Vol. 50, 4, 537-
554.
Hamari, J., Sjöklint, M. and Ukkonen, A. (2015, in press), “The Sharing Economy: Why
People Participate in Collaborative Consumption,” in: Journal of the Association for
Information Science and Technology. Available at SSRN:
http://papers.ssrn.com/sol3/Papers.cfm? abstract_id=2271971 (online retrieval 11.
November 2015).
Haws, K. L., Winterich, K. P. and Naylor, R. W. (2014), “Seeing the World Through
Green-tinted Glasses: Green Consumption Values and Responses to Environmentally
Friendly Products,” in: Journal of Consumer Psychology, Vol. 24, 3, 336-354.
Promoting the Shareconomy 55

Hill, C. A. (1987), “Affiliation Motivation: People Who Need People…But in Different


Ways,” in: Journal of Personality and Social Psychology, Vol. 52, 5, 1008-1018.
Hogg, M. A. and Terry, D. I. (2000), “Social Identity and Self-categorization Processes
in Organizational Contexts,” in: Academy of Management Review, Vol. 25, 1, 121-
140.
Irwin, J. R. and McClelland, G. H. (2001), “Misleading Heuristics and Moderated
Multiple Regression Models,” in: Journal of Marketing Research, Vol. 38, 1, 100-
109.
Janssen, L., van Beest, I. and Mead, N. (2014), “When the Need for Affiliation
Heightens Consumers’ Susceptibility to Persuasion Tactics,” in: Bigné, E.,
ed.Proceedings of the 43rd European Marketing Academy Conference (EMAC).
Valencia, Grupo Diario Imprementa, 132.
Kasser, T. and Ryan, R. M. (1996), “Further Examining the American Dream:
Differential Correlates of Intrinsic and Extrinsic Goals,” in: Personality and Social
Psychology Bulletin, Vol. 22, 3, 280-287.
Kwon, O. and Wen, Y. (2010), “An Empirical Study of the Factors Affecting Social
Network Service Use,” in: Computers in Human Behavior, Vol. 26, 2, 254-263.
Lamberton, C. P. and Rose, R. L. (2012), “When is Ours Better than Mine? A
Framework for Understanding and Altering Participation in Commercial Sharing
Systems,” in: Journal of Marketing, Vol. 76, 4, 109-125.
Leary, M. R., Kelly, K. M., Cottrell, C. A. and Schreindorfer, L. S. (2012), “Construct
Validity of the Need to Belong Scale: Mapping the Nomological Network,” in:
Journal of Personality Assessment, Vol. 95, 6, 610-624.
Lee, R. M. and Robbins, S. B. (1995), “Measuring Belongingness: The Social
Connectedness and the Social Assurance Scales,” in: Journal of Counseling
Psychology, Vol. 42, 2, 232-242.
Lin, K. Y. and Lu, H. P. (2011), “Why People Use Social Networking Sites: An
Empirical Study Integrating Network Externalities and Motivation Theory,” in:
Computers in Human Behavior, Vol. 27, 3, 1152-1161.
Mead, N. L., Baumeister, R. F., Stillman, T. F., Rawn, C. D. and Vohs, K. D. (2011),
“Social Exclusion Causes People to Spend and Consume Strategically in the Service
of Affiliation,” in: Journal of Consumer Research, Vol. 37, 5, 902-919.
Minton, A.P. and Rose, R.L. (1997), “The Effects of Environmental Concern on
Environmentally Friendly Consumer Behavior,” in: Journal of Business Research,
Vol. 40, 1, 37-48.
Moehlmann, M. (2015), “Collaborative Consumption: Determinants of Satisfaction on
the Likelihood of Using Sharing Economy Options Again,” in: Journal of Consumer
Behavior, Vol. 14, 193-207.
Nov, O., Naaman, M. and Ye, C. (2009), “Analysis of Participation in an Online
Photo‐sharing Community: A Multidimensional Perspective,” in: Journal of the
American Society for Information Science and Technology, Vol. 61, 3, 555-566.
Prothero, A., Dobscha, S., Freund, J., Kilbourne W.E., Luchs, M.G., Ozanna, L.K. and
Thøgersen, J. (2011), “Sustainable Consumption: Opportunities for Consumer
Research and Public Policy,” in: Journal of Public Policy and Marketing, Vol. 30, 1,
31-38.
56 Schwenk and Hüttl-Maack

Rassuli, K. M. and Hollander, S. C. (1986), „Desire-induced, Innate, Insatiable?” in:


Journal of Macromarketing, Vol. 6, 2, 4-24.
Richins, M. L. (1987), “Media, Materialism, and Human Happiness,” in: Wallendorf, M.
and Anderson, P., eds.Advances in Consumer Research, Vol. 14, 1. Provo, UT,
Associations for Consumer Research, 352-356.
Richins, M. L. and Dawson, S. (1992), “A Consumer Values Orientation for Materialism
and its Measurement: Scale Development and Validation,” in: Journal of Consumer
Research, Vol. 19, 3, 303-316.
Richins, M. L., McKeage, K. K. R. and Najjar, D. (1992), “An Exploration of
Materialism and Consumption-related Affect,” in: Sherry, J. F. and Sternthal, B., eds.
Advances in Consumer Research, Vol. 19. Provo, UT, Associations for Consumer
Research, 229-236.
Rindfleisch, A., Burroughs, J. E. and Denton, F. (1997), “Family Structure, Materialism,
and Compulsive Consumption,” in: Journal of Consumer Research, Vol. 23, 4, 312-
325.
Rosen, D., Lafontaine, P. R. and Hendrickson, B. (2011), “CouchSurfing: Belonging and
Trust in a Globally Cooperative Online Social Network,” in: New Media and Society,
Vol. 13, 6, 981-998.
Roux, D. and Guiot, D. (2008), “Measuring Second-hand Shopping Motives,
Antecedents and Consequences,” in: Recherche et Applications en Marketing, Vol.
23, 4, 63-91.
Ryan, R. M. and Deci, E. L. (2000), “Intrinsic and Extrinsic Motivations: Classic
Definitions and New Directions,” in: Contemporary Educational Psychology, Vol.
25, 1, 54-67.
Tatzel, M. (2002), ““Money worlds” and Well-being: An Integration of Money
Dispositions, Materialism and Price-related Behavior,” in: Journal of Economic
Psychology, Vol. 23, 103-126.
Twenge, J. M., Baumeister, R. F., DeWall, C. N., Ciarocco, N. J. and Bartels, J. M.
(2007), “Social Exclusion Decreases Prosocial Behavior,” in: Journal of Personality
and Social Psychology, Vol. 92, 1, 56-66.
Twenge, J. M., Baumeister, R. F., Tice, D. M. and Stucke, T. S. (2001), „If You Can't
Join Them, Beat Them: Effects of Social Exclusion on Aggressive Behavior,” in:
Journal of Personality and Social Psychology, Vol. 81, 6, 1058-1069.
Walsh, B. (2011), “Today’s Smart Choice: Don’t Own. Share,” Time Magazine 15/03.
http://content.time.com/time/specials/packages/article/0,28804,2059521_2059717_2
0597 10,00.html.
Communicating through Brand Websites to Create Unique
Brands

Mototaka Sakashita

1 Purpose of the Study


Differentiating own brands is the key strategy to establish a strong brand with
unique associations (Keller, 1998). However, it is getting extremely difficult to
do so because of the flood of new products launched every year. For example,
Japanese market experiences an enormous number of new products introduced
within one year; 213 canned coffee products, 765 snacks, and 1928 chocolate-
related products all in the year of 2005 (Tanaka and Hosoda, 2006). Consumers
are now facing lots of products in each category, thus differences among brands
are hardly noticeable. As a result, all the different brands look very similar,
which results in consumers’ frequent switching activities.
This perceived similarity among different brands is very important in
capturing the whole market picture, and is conceptualized by Muncy (1996) as
brand parity. According to Muncy, brand parity refers to an overall perception
held by the consumer that the differences between the major brand alternatives
in a product category are small. For a marketer, it is extremely important to
lower the level of brand parity, so that the customers perceive the brand unique,
thus giving a plausible reason to purchase it repeatedly.
Previous researches have revealed the outcomes of brand parity, all
indicating that high level of brand parity creates various kinds of undesirable
outcomes. For example, Muncy (1996) found that high level of brand parity
results in higher level of price sensitivity. Other studies discovered that high
brand parity leads to more use of price cue (Obermiller and Wheatley, 1984),
lower value perception of market information (Muncy, 1996), and lower level of
loyalty toward a particular brand (Jacoby, 1871). Also, brand parity weakens the
relationship between brand loyalty and satisfaction/perceived quality (Iyer and
Muncy, 2005). However, despite of its importance, little have been known about
the antecedents of brand parity.
At the same time, technological advancement has changed the way
consumers learn about a brand. Marketing communications including
advertisements on mass media, commercial information from salesperson at
retailers, or promotional messages from brand-related events were more or less
effective in convincing customers of the positive aspects of their brands.
However, consumers are now searching necessary information to make purchase

© Springer Fachmedien Wiesbaden 2017


G. Christodoulides et al. (Hrsg.), Advances in Advertising Research (Vol. VII),
European Advertising Academy, DOI 10.1007/978-3-658-15220-8_5
58 Sakashita

decisions at different websites on internet. This sheds light on the importance of


brand websites, because marketers can educate their customers effectively by
carefully designing these new media. Therefore, the purpose of this study is to
understand the characteristics of consumers’ browsing activity in a brand
website, and to investigate its impact on their level of perceived brand parity.

2 Method
In order to gather empirical evidence, it conducted two studies. The first
study implemented an online survey to capture the importance of the internet
usage in purchase decisions in Japan. To better understand the role of a brand
website in creating a more unique brand, the second study was conducted, where
two different sets of data, a browsing data from an existing brand website in
Japan, and an attitudinal data from online survey, were combined.

2.1 Study 1: Online Survey


The purpose of study 1 is to understand the importance of internet in
consumers’ purchase decisions. An online survey (N=1036) was conducted by a
top research institute with the biggest consumer panels (over 1.1 million people)
covering all over Japan, so that the age and the residence population of the
participants could be designed to be similar to that of Japan. The male
participants consisted of 48.2% of the samples while 51.8% were female. 36.9%
were working full-time, 25.3% were domestic engineers, 12.5% were working
part-time, 7.2% were students, and 18.0% were not working. Participants had to
complete all the questions in the survey in order to be rewarded with special
points that were to be converted to cash, so there were no void responses. All the
questionnaires were sent out via emails, where they were to click on the link that
would direct them to the survey pages.
The online questionnaire consists of three parts. In the first part, the
participants were asked about a recently purchased product that they
remembered well, as well as the price, purchasing place, and overall evaluation
of it. Also, brand parity of the product category that each participant had
purchased was measured using the five items by Muncy(1996): “I can’t think of
any differences between the major brands of this product category”, “To me,
there are big differences between the various brands of this product category
(reverse)”, “The only difference between the major brands of this product
category is price”, “About this product category, most brands are basically the
same”, “All major brands of this product category are the same.” For each
question, they were asked to rate on a five point scale from “1=I do not agree at
all” to “5=I agree very much.” In the second part, they were asked how much
they searched information in the following media to make the purchase decision:
Communicating through Brand Websites to Create Unique Brands 59

word of mouth from friends and family members (WOM), magazines, TV, radio,
newspaper, internet, retail stores, company campaign events, and product
sampling. They were asked to rate from “1=I did not search at all” to “6=I
searched very much” for each media. Also they were asked how much they
searched online in the following websites in the same manner: search engines
(such as google or yahoo), comparison sites, brand websites, WOM forums, and
social network websites. In the third part, the demographic characteristics of the
participants were measured.

2.2 Result of Study 1


Participants bought various kinds of products such as automobiles (104), PC and
related products (101), clothing (97), foods (94), cosmetics (78), household
durables (64), daily supplies (63), and so on. Among these different media,
participants search for internet most intensely (average=3.55), followed by retail
stores (3.34) and WOM (2.49). The means are displayed in Figure 1. There were
strong correlations between radio and newspaper (r=.813, p<.01), TV and radio
(r=.688, p<.01), and TV and newspaper (.670, p<.01); however, the correlations
between internet and other media were all low (correlation coefficients ranged
from .010 to .231). Therefore, it seems that participants search for online
information relatively more, but they do not simultaneously search for the
information in the other media.

Figure 1: Media search


60 Sakashita

Figure 2: Online search

In online search (Figure 2), brand websites are visited most (2.96),
followed by search engines (2.87) and comparison sites (2.74). Correlation
matrix revealed that there were strong correlations between search engines and
comparison sites (r=.776, p<.01), comparison sites and WOM forums (.687,
p<.01), and search engines and brand websites (r=.624, p<.01). We can make a
reasonable assumption that participants directly search for brand websites, or via
search engines, to get necessary information to make purchase decisions. In
order to see the relationship between search activities at brand websites and
perceived brand parity, first the average of all the items were calculated after
reversing the second item, to get a single BP variable (Cronbach’s alpha=.673).
The results showed that BP was negatively correlated with search activities at
brand websites (r=-.129, p<.01), while none of the other online websites were
significantly correlated with BP. It seems that the more consumers search for
information at brand websites, the lower the level of brand parity becomes;
however, the relationship seems weak. In order to further investigate the
relationship between brand website search and brand parity in details, the second
study was conducted.

2.3 Study 2: Field Experiment


In study 2, two different sets of data, a browsing data from an existing top
brand website in Japan, and an attitudinal data from online survey, were
combined. For the browsing data collection, all the browsing activities on an
existing leading brand website in a sports nutrition product category of all the
Communicating through Brand Websites to Create Unique Brands 61

pre-registered customers (N=20000) were recorded during three-month period.


The website was selected because sports nutrition products are generally
purchased with high involvement, therefore it would make it easier to track
consumers’ online browsing. Each page views (PV) were aggregated per
customer, forming accumulated PV during the period per customer. Email-
magazines were then sent out to all of them, to encourage them to participate in
an online survey, where brand parity among all the brands in the sports nutrition
product category was measured using the same five item scale from Muncy
(1996). 2711 valid samples were collected, then merged to the browsing data
using unique customer ID. About 65 % of them had actually visited the website
during the period, to form the final valid 1770 observations. Among those 1770
samples, 59.9% were male and 40.1% were female. 59.4% were working full-
time, 19.5% were domestic engineers, 4.6% were working part-time, 5.3% were
students, and 11.2% were not working. The average of total visits to the website
during the period was 3.15 per customer, that of total PV was 9.33, and that of
total duration in seconds was 354.57.

2.4 Result of Study 2


Using the PV data, cluster analysis was applied to detect four browsing
clusters; non-searchers (NS), campaign hunters (CH), nutrition information
hunters (NIH), and extreme searchers (ES). Further ANOVA revealed
significant differences among them (p<.01, see Table 1). Precisely, NS (consists
of 72.5% of the whole samples) hardly searched for any pages (average of total
PV in three-month period was 4.5), hardly visited the website (average of total
visit frequency was 2.2), and rarely spent any time (average duration of each
visit was 2.5 minutes). CH (12.9%) typically searched only for the campaign
related information (8.6 PV), visited the website occasionally (4.7 visits), and
spent little time (3 minutes). NIH (11.3%) browsed moderate amount of
information typically in nutrition information of the products (24.7 PV),
occasionally visited the website (5.3 visits), and spent moderate amount of time
(18.5 minutes). ES (3.3%) searched for large amount of information (66.1 PV),
with highest website visit frequency (11.4 visits) and longest duration (45
minutes).
62 Sakashita

Table 1: Cluster characteristics

ANOVA was conducted to the average score of BP (Cronbach’s


alpha=.754), using those four clusters as factors, to find that there was a
significant differences among them (p<.01). NIH and ES showed lower BP
scores than NS and CH (means=2.53(.73) and 2.59(.73) < 2.77(.72) and
2.77(.69), respectively). The means are displayed in Figure 3. NIH and ES
browse more pages in the brand website and seemingly learn more about the
brand, which enables them to realize the unique characteristics of the brand, thus
lowering the level of brand parity. Therefore, the more they browse, the more
likely they become to perceive that all the brands are unique from each other.

Figure 3: Brand parity average

3 Conclusion
This study sheds light on the importance of brand websites in consumers’
purchase decisions, to create unique brand associations through browsing
Communicating through Brand Websites to Create Unique Brands 63

behavior. As study 1 shows, consumers’ search activities tend to rely on the


internet sources, and especially brand websites are important in constructing
unique brand associations. However, as study 2 shows, even though the
customers visit the brand website, very few of them actually browse the various
kinds of contents related to the brand. In fact, NIH and ES segments together
consist only 14.6 % of the whole visitors. Though small in numbers, however,
these segments with relatively high page views, understand the unique
characteristics of the brand better, indicating the low level of brand parity.
This study offers some implications to both academic and managerial
fields. Theoretical implication is that it identifies the antecedent of brand parity
in online context. Especially it highlights the importance of brand websites that
create unique brand associations, through obtaining and analyzing the actual
data from an existing brand website in Japan. As for managerial implication,
marketers can use these results to implement a better segmentation strategy
based on the online browsing behavior. It provides information which helps to
construct a better brand website that educates its customers effectively.
Despite of its contributions, there are limitations to be noted. Study 2 offers
only one product category with relatively high level of involvement, so future
research should be conducted using different products with different level of
involvement. The samples in study 2 were the members of the brand website,
therefore their favorability toward the brand was relatively high, which may
affect the results. It is highly important to use the real data from the natural field
study, however, more heterogeneous samples under controlled settings are
equally important for the validity of the findings. This study focused on the
antecedent of the brand parity, but the direction of the causality is also
questionable. The measurement of brand parity took place after recording the
clickstream data, however, this by itself cannot rule out the possibility of the
opposite direction. More controlled laboratory experiments are expected to solve
this problem. Finally, the outcomes of brand parity in online context need to be
studied as well in future studies. Brand parity serves as a key concept to
understand the whole market picture, thus more researches are needed.

4 References
Iyer, R., Muncy, J. A. (2005), "The Role of Brand Parity in Developing Loyal
Customers," in: Journal of Advertising Research, 45(2), 222-228.
Jacoby, J. (1971), "A Model of Multi-Brand Loyalty," in: Journal of Advertising
Research, 11(2), 25-31.
Keller, K. Lane. (1998), Strategic Brand Management, Prentice-Hall.
Muncy, J. A. (1990), "Involvement and Perceived Brand Similarities/Differences: The
Need for Process Oriented Models," in: Advances in Consumer Research, 17, 144-
148.
64 Sakashita

Muncy, J. A. (1996), "Measuring Perceived Brand Parity," in: Advances in Consumer


Research, 23, 411-417.
Obermiller, C., Wheatley, J. J. (1984), "Price Effects on Choice and Perception under
Varying Conditions of Experience, Information, and Belief in Quality Differences,"
in: Advances in Consumer Research, 11, 453-58.
Tanaka, S., Hosoda, T. (2006), "Products last only for three weeks," in: Nikkei Business,
May 29, 26-29.
Gaining Attention Online: Do the Levels of Product
Involvement and Website Interactivity Matter? An Eye-
Tracking Approach

Polyxeni (Jenny) Palla and Yorgos Zotos

1 Introduction

Differences between digital and traditional media may reflect differences in


consumers’ attention (Liu and Shrum, 2009). Attention capture is the starting
point of any further cognitive process according to hierarchy of effect
advertising models (eg. AIDA proposed by Strong, 1925). The level of product
involvement and the level of interactivity in a website are examined as two
variables that may have an effect on individuals’ attention. Understanding what
attracts individuals’ attention in a website is of main importance for both
academics and practitioners. Kahneman (1973) (cited by Mackenzie, 1986)
defines attention as the amount of the mental effort or cognitive capacity
devoted to a specific stimuli. Individuals sometimes pay more or sometimes less
attention to different stimulus due to several reasons.
First, people differ in their level of involvement. There is empirical
evidence supporting that the product involvement is a variable that has a direct
effect on attention as well as on the process of the provided information in
traditional forms of advertising (Petty and Cacioppo, 1979; 1986; Rucket et al.,
2007). Highly involvement products motivate consumers to focus more
processing resources to a stimulus leading to higher levels of attention (Petty
and Cacioppo 1986). The question proposed here is whether the product
involvement has any effect on attention in the online environment. Previous
studies investigating the website effectiveness have not examined any potential
effect of product involvement on attention in an online context (e.g. Coyle and
Thorson, 2001; Bezjian-Avery et al., 1998). Thus, our study could contribute to
current research and shed light on the possible moderation effects of this factor
with the use of eye-tracking system.
Second, in the online environment, individuals tend to be more goal-oriented
and tend to pay attention to the provided information (Stanaland and Tan,
2010). However, attention is limited and selective and only a piece of
information on a web site gains users’ attention (Molosavljevic and Cerf, 2008).
Increased number of interactivity elements may distract the consumers’
attention, especially when it is visually complicate (Bezjian-Avery et al., 1998).
Recently, Voorveld et al (2011) point out six website elements that contribute

© Springer Fachmedien Wiesbaden 2017


G. Christodoulides et al. (Hrsg.), Advances in Advertising Research (Vol. VII),
European Advertising Academy, DOI 10.1007/978-3-658-15220-8_6
66 Palla and Zotos

positively to perceived interactivity [feedback form, product registration,


product customization, information customization, tell a friend and dropdown
menu]. In the study reported here, we employ an experimental design with the
use of eye-tracking system to assess whether the interactivity elements proposed
by Voorveld et al. (2011) gain users’ attention in a website that presents a high
and a low involvement product. The present study seeks to examine both;
medium and high interactive versions of both websites [presenting high and low
involvement products].
The purpose of this study is twofold. First, it makes an attempt to
investigate the moderating role of product involvement on attention with the
use of eye-tracking system. Second, it attempts to assess whether the six
interactive elements proposed by Voorveld et al. (2011) employed in a high and
in a medium interactive websites grasp users’ attention.

2 Theoretical Background

Product involvement is considered as a variable that reflects the level of


personal relevance with an issue (Zaichkowsky, 1985). Personal relevance is
defined as the extent to which consumers perceive the object to be self-related
or in some ways crucial to achieve their personal goals and values (Celsi and
Olson, 1988). When the product under consideration is of high personal
relevance then individuals allocate a greater degree of attention (Petty and
Cacioppo, 1986). On the contrary, an advertising message of a low-involvement
product gains less attention. The number and not the quality, the source expert
and other cues gain consumers’ attention and may lead to persuasion (Zotos et
al., 1992). The moderating role of product involvement on attention and thought
elicitation has also replicated in the web environment (Liu and Shrum, 2009). It
is hypothesized that:

H1: Individuals exposed to the website [a) whole homepage, b) text and c)
images provided by the homepage] of a high involvement product will pay more
attention as compared to those exposed to the website that presents a low
involvement product.

Authors agree that there is a great incongruence between the level of actual
and the level of perceived interactivity (Voorveld et al., 2011). Three
interactivity dimensions are specified to the literature; two-way communication,
synchronicity and perceived control (e.g. McMillan and Hwang, 2002; Song
and Zinkhan, 2008; Liu and Shrum, 2002). Recently, Voorveld, Neijens and
Smit (2011) make an invaluable contribution by providing six unique
characteristics that make the brand websites truly interactive. The three
interactive elements, which belong to the two-way communication dimension,
Gaining Attention Online 67

are (1) the option to recommend the site/product to a friend, (2) feedback form,
(3) product registration online. The other three interactive elements, within the
active control dimension, are (1) product customization (2) information
customization (3) dropdown menu. The above findings are based on
respondents’ answers. Thought they are very helpful they cannot answer to
which elements individuals pay attention.
It should be added that the level of attention paid to a stimulus is expected to
vary based on the level of involvement (Rossiter, Percy and Donovan, 1991).
Under high involvement conditions individuals allocate a greater degree of
attention (Petty and Cacioppo, 1986). We use eye-tracking data to examine
whether the elements that are considered to increase the perceived interactivity
of a website that presents a high involvement product are seen and used:

H2: Participants will pay attention to the interactive elements in a website of a


high involvement product.

Under low involvement conditions consumers’ attention is gained mainly by


easy-to-identify peripheral cues (Petty and Cacioppo, 1986). Low involvement
consumers tend to perceive the presence of interactivity as a positive peripheral
cue (Liu and Shrum, 2009). Increased number of interactive elements in a high-
interactivity web-site seems to persuade low-involvement individuals, whether
those elements are relevant or beneficial to them or not (Liu and Shrum, 2009).
However, an individual could become highly involved when a new, and
maybe low-risk, brand needs to be assessed (Rossiter et al. 1991). An individual
who is unfamiliar with a new product may acquire more communication effects
and eventually pay attention to the interactive elements provided in the website.
We use eye-tracking data to realize if the interactive elements available in a
website that presents a product of low involvement are seen and used.

H3: Participants will pay attention to the interactive elements in a website of a


low involvement product.

3 Method
The Personal Involvement Inventory proposed by Zaichkowsky (1985)
employed to conclude that students accept the laptop as a high involvement
product and the refreshment as a low (sample: 32 students)1. Two fictitious
brands, one for a laptop and one for a website were designed in order to avoid
pre-attitudinal effects. Two websites were constructed for each brand. Two

1 For ease of reporting and consistency reasons we use the terms high and low to identify the

level of product involvement. Actually the differences are relative and the terms higher and lower
are the most appropriate.
68 Palla and Zotos

interactive versions (high and medium)2 of each website were designed. The
informational text and the images remained constant in both versions of each
website. The interactivity elements employed in every level followed the
recommendations of the relevant literature (Table 1). Two experienced web-
designers served as judges to verify the actual level of interactivity on each
website. 17 items were used taken from Liu (2003), McMillan and Hwang
(2002) and Song and Zinkhan (2008) to measure the level of perceived
interactivity in an additional online pre-test (sample: 40 students). The results
confirm that each webpage provided the expected level of interactivity.

Table 1: Interactivity elements employed in every interactivity level


Dimensions of Interactivity elements Interactivity level
Interactivity
High Medium
Two way option to recommend the site to a  
communication friend
Liu and Shrum 2002; capability to register the product 
McMillan and online
Hwang 2002; feedback form  
Voolverd et al 2011 telephone number  
e-mail  
online service 
e-shop  
Time or click to call – we call you back 
synchronicity now
Liu and Shrum online service 
2002; McMillan and
number of clicks required to reach  
Hwang 2002;
certain information
Johnson et al. 2006;
Song and Zinkhan response time 
2008; Voolverd et al
2011
Control an option to customize products 
Liu and Shrum 2002;
McMillan and capability to customize 
Hwang 2002; Song information
and Zinkhan 2008; dropdown menu  
Voolverd et al 2011

2
For ease of reporting and consistency reasons we use the terms high and medium to describe
the level of perceived website interactivity. The actually differences should be described as higher
and lower. However, we avoid use the term low/lower interactivity level because previous studies
(e.g. Sohn et al., 2007) identify the low interactive website as one-page document containing
nothing but the text information and photo images of the product.
Gaining Attention Online 69

Seventy two [72] students participated in a controlled experiment for extra


credit in a marketing module. Half of the participants [36] were exposed to the
high involvement website from which eighteen [18] were exposed to the high
and eighteen [18] to the medium interactive webpage. The other half [36] were
exposed to the website that presents the low involvement product, eighteen [18]
to the low and eighteen [18] to the medium version [Table 2].

Table 2: Webpages – participants

Interactivity High Medium


Product Involvement
High (Laptop) 18 participants 18 participants

Low (Refreshment) 18 participants 18 participants

One student at the time participated in the experimental process. One by one
participant sat in a quiet room in front of a typical PC and asked to hold his/her
head relatively steady. Participants’ eyes were calibrated to the screen using
Tobii Studio software to ensure accurate recording of participants’ fixations.
Each student was instructed to navigate a website for a new brand of a laptop/
refreshment which is not marked in his country. He was asked to relax and
navigate to the website as he did at home at his own time and pace. Each
participant was informed that they were free to proceed with virtual e-shopping
if he wished, in order to assess his actual behavior. Once the navigation process
was completed, the participant was instructed to fill in the questionnaire
regarding perceived interactivity, product involvement (for manipulation check)
and demographics. Manipulation check indicated the appropriate level of
interactivity in each website. The laptop was recognized as a high and the
refreshment as a low involvement product by the participants.
For collecting data on eye movement, a Tobii X2-30 Eye tracker (Sweden)
was used. Eye movements were collected with 30-Hz frequency. Following Lee
and Ahn (2012) attention was measured with total fixation duration [fd] and
relative fixation frequency [rff]. Relative fixation frequency [rff] was
calculating by taking the number of fixation on the element (Area of Interest)
divided by the total number of eye fixations landing on the screen (Kuisma et al.
2010). Fixation is a stable gaze focused on a specific area (Balatsoukas and
Ruthven, 2012).
70 Palla and Zotos

4 Findings

Fixations were identified by using the Tobii Studio software. The Tobii
Studio software provides the opportunity to define Areas of Interest (AoI),
specified areas which may be boxes or rectangular of content, to find how
many times (number of fixations) and for how long (fixation duration)
participants fixated on the specific area (AoI) of the webpage. We define an
AoI for each interactivity element as well as for the informational text and the
images. Tobii Studio provides the ability to export heatmaps for every website.
In order to answer to the H1 we analyzed and compared the data released
from the participants exposed to the website that presents the high and the low
involvement product for every interactivity level. Attention refers to sensations.
It has been demonstrated that shifts in visual attention and eye movements are
related (Kuisma et al., 2010). Attention was measured with the total fixation
duration to the screen, the text and the images as well as with relative fixation
frequency to the text and the images. Three AoI were defined responding to the
whole homepage, the main text and the images. Relative fixation frequency is
calculated as the number of fixation to the element [area of interest – text or
images] divided by the number of fixations landing in the screen. The relative
fixation frequency for the images for the individuals exposed to the website of
high involvement product is higher (mean: .0850) as compared to those
exposed to the website of low (mean: .1162) and this difference is statistically
important [F(1.104) =6.270; Sig.: .014]. The relative fixation frequency
regarding the text is higher for those exposed to the website of the low
involvement product [mean: high= .4542; low= .5069] but this difference is
insignificant [F(1.105) =1.663; Sig.: .200]. These findings are depicted in
Tables 3.

Table 3: Anova RFF images- text


Relative fixation frequency Sum of df Mean F Sig.
Squares Square
images Between .026 1 .026 6.270 .014
Groups
Within .427 104 .004
Groups
Total .453 105
text Between .074 1 .074 1.663 .200
Groups
Within 4.673 105 .045
Groups
Total 4.747 106
Gaining Attention Online 71

It is also examined the effect of product involvement on fixation duration


towards the screen, the text and the images. Individuals exposed to the website
of the high involvement product indicated higher fixation duration towards the
screen [mean: 45.3107] as compared to those exposed to the low [31.7769] and
this difference is statistically important [F (1,106) = 6.476, Sig. = 0.44].
Individuals’ fixation duration towards the text is higher for those exposed to the
website of the high involvement product but this difference is insignificant
[mean: high= 20.0091, low= 16.9885] [F(1,106) = 6.476, Sig. = .205].
Participants’ fixation duration towards the images is higher for those exposed
to the website of the low involvement product [mean: high= 3.3087, low =
3.7487] but this difference is also insignificant [F(1.104) = .158, Sig. = .692].
Findings are exposed to table 4. Therefore, H1 is accepted only if attention is
assessed as in terms of fixation duration towards the screen.

Table 4: Anova fixation duration


Fixation Duration Sum of Squares df Mean F Sig.
Square
Screen Between 4945.486 1 4945.48 6.476 .012
Groups
Within 80950.395 106 763.68
Groups
Total 85895.881 107
Text Between 244.036 1 244.03 1.623 .205
Groups
Within 15785.765 105 150.34
Groups
Total 16029.801 106
Images Between 5.130 1 5.130 .158 .692
Groups
Within 3378.382 104 32.48
Groups
Total 3383.512 105

In order to answer the H2 eighteen (18) participants were exposed to the


high and eighteen (18) to the medium interactive version of the website that
presents the high involvement product [RQ 1]. In the high interactive webpage
ten [10] AoI were specified. According to the Table 5, individuals pay more
attention to the texts and information provided by the website Participants saw
and used all the interactive elements that were available to the high interactive
website. The “dropdown menu” was used at least four times by each participant.
“Product customization” is the interactivity element that was seen by all
72 Palla and Zotos

participants [rff mean: 9.91%; fd: 87.61 sec.]. The ability for “product
registration” was seen by 13 out of 18 participants [rff mean: 6.25%; fd 41.4
sec.]. The ability to “order now” was seen by 17/18 [rff mean: 4.70%; fd: 37.29
sec.], to “tell a friend” by 15/18 [rff: 2.78%; fd: 28.61 sec.]. “Feedback form”
was visited by only one participant.

Table 5: Interactive elements that are seen and used


Website of high involvement product – high interactive version
Areal of Relative Fixation Other (click
Interest fixation duration (sec.) mouse – visit the
frequency participants site)
(mean)
Text – 25.8% 232.73 total 1 click / 2
information [17/18] persons
Info 19.42% 128.51 total 8 click /5
customization [18/18] persons
Product 9.91% 87.61 2 clicks / 2
customization [18/18] persons
Images 9.72% 68.85 total 8 clicks / 6
[17/18] persons
Product 6.25% 41.4 total 1 click/ 1
registration [13/18] person
Order now 4.70% 37.29 total
[17/18]
Tell a friend 2.78% 28.61 1 click/ 1
[15/18] person
Feedback 1visit – 1
form person
Dropdown at least 4
menu times each
participant

Regarding the medium interactive website of a high involvement product, all


participants saw the interactive element that provides “product customization”
[mean rff: 10.44%; fd: 92.43 sec.], 18 out of 18 saw the “information
customization” [mean rff:18.45% ; fd: 160.77 sec.] and 17 out of 18 the “e-
order” [mean rff: 3.64%; fd: 28.22 sec.]. The “feedback form” was visited by 1
person and the “dropdown menu” was used at least 3 times by each participant
(Table 6). H2 is accepted since all the elements of website interactivity were
seen apart from the feedback form.
Gaining Attention Online 73

Table 6: Interactive elements that are seen and used


Website of high involvement product – medium interactive version
Relative Fixation Other (click
Areas of fixation duration (sec.) mouse – visit the
Interest frequency participants site)
(mean)

Text – 29.79% 260.53


information 18/18
Info 18.45% 160.77 18 click/ 10
customization 18/18 person
Product 10.44% 92.43 2 click – 2
customization 18/18 person
Images 7.87% 70.35
18/18
Order now 3.64% 28.22 2 click
17/18
Feedback One person
form visit the site
Dropdown At least 3
menu times by each
participant.

In order to answer the H3 we analyze the data released from 36 participants.


Eighteen (18) participants were exposed to the high interactive version of the
website that presents the low involvement product. Nine (9) Areas of Interest
(AoI) were defined, seven (7) correspond to the interactive elements and the rest
correspond to the text and the images. Analysis with Tobii Studio indicated that
participants saw and used all the interactive elements that were available to the
website. “Info customization” was seen by all participants [mean rff: 17.8%; fd:
43.74 sec.], “product registration” by 17/18 [mean rff: 6%; fd: 26.09 sec.],
“product customization” by 10/18 [mean rff: 10%; fd: 9.37 sec.], “order now”
9/18 [mean rff:2.5% ; fd: 13.11 sec.] and “tell a friend” by 14/18 [mean rff: 1%;
fd:14.46 sec.]. The “dropdown menu” was used at least 4 times by each
participant [Table 7].
74 Palla and Zotos

Table 7: Interactive elements that are seen and used


Website of low involvement product – high interactive version
Areal of Relative Fixation Other
Interest fixation duration (sec.) (click mouse
frequency participants – visit the
(mean) site)
Text – 24% 421.95 total 1 click /
information [18/18] 2 persons
Info 17.8% 43.74 1 click /6
customization [18/18] persons
Images 11.5% -
124.23 total
[18/18]
Product 9.37 -
customization 10% [10/18]
Product 6% 26.09 total 1 click/ 7
registration [17/18] persons
Order now 2.5% 13.11 total 1 click /3
[9/18] persons
Tell a friend 1% 1 click/ 1
14.46 person
[14/18]
Feedback - -
form
Dropdown at least 4
menu times each
participant

Eighteen (18) participants were exposed to the medium interactive version


of the website that presents the low involvement product. Seven (7) Areas of
Interest were defined to the medium interactive website. Participants see and
use most of the available interactive elements. The “dropdown menu” was used
at least 3 times by each participant. “Info customization” was seen by 17 out of
18 [mean rff: 7.25%; fd: 32.49 sec.] and clicked by 3 out of 18 participants.
“Order now” was seen by 12 out of 18 [mean rff: 1.84%; fd: 9.22 sec.] and
clicked by 2 out of 18 participants. Feedback form was seen by 7 out of 18
participants [mean rff: 4.07%; fd: 20.6 sec.] [Table 8]. H3 is accepted for since
all the elements of website interactivity were seen apart from the feedback form.
Gaining Attention Online 75

Table 8: Interactive elements that are seen and used


Website of low involvement product – medium interactive version
Relative Fixation Other
fixation duration (click
Areas of Interest frequency (sec.) mouse –
(mean) participants visit the
site)
Text – information 54.2% 349.77
18/18
Info customization 7.25% 32.49 3 click –
17/18 3 persons
Images 8.71% 46.42
18/18

Order now 1.84% 9.22 2 click –


12/18 2 person
Feedback form 4.07% 20.6
7/18
Dropdown menu - - At least
3 times each
participant

5 Discussion and Results


Eye tracking proved to be a useful method for understanding which
elements gain users’ attention in a website. We were able to answer two
questions. The first regards with the level of product involvement and the
second to the availability of interactive elements.
Findings of the H1 underlined the moderating role of product involvement
on the attention paid in the whole homepage in terms of fixation duration.
However, this difference seems to be unrelated with the attention paid on the
text and images provided in the homepage. A plausible explanation may be the
fact that other elements, such as the option to customize the product, to suggest
the website to a friend etc., available in the website of the high involvement
product gain individuals’ attention.
Another interesting finding relates to the insignificance difference on the
attention paid on the text under both conditions; high and low involvement. The
text gain most attention, under every interactivity and involvement condition,
with both metrics [rff and fd]. This may be explained by the fact that when an
individual visits for first time a website for a new brand, usually searches for
information regarding the product, regardless his level of involvement. It is,
therefore, reinforced the notion that attention is not totally responsive;
76 Palla and Zotos

individuals voluntarily direct their attention in line with their goals in the online
environment (Lee and Ahn, 2012).
Product involvement is directly related to the risk experienced by the
consumers. However, it should be underlined in our study the fact that both
brands were totally new. An individual could become highly involved when a
new, and maybe low-risk, brand needs to be assessed (Rossiter et al. 1991). It
seems obvious that a consumer who has never heard of the brand may attain or
elaborate more informational elements in a website and eventually become more
involved as compared to a familiar brand.
Data analysis indicated that the majority of individuals pay attention (see
and use) to all the interactive elements that are available to the high and medium
interactive version of the website that presents a high involvement product,
except from the “feedback form”. Similarly, low involvement individuals see
and use the available interactive features, apart from the feedback form,
provided in the website that presented the refreshment.
It seems that the five out of the six interactive elements proposed by
Voorveld et al. (2011) [information customization, product customization,
dropdown menu, tell a friend, product registration] make a website truly
interactive, regardless the conditions of product involvement.
Under low involvement conditions, no one user visited the “feedback form”.
Moreover, under high involvement the feedback form was visited by only one
in each interactivity level (high and medium). A plausible explanation may be
the fact that it is a service provided by the website in stages after the awareness/
information of the product. When an individual visits for first time a website for
a new brand, usually searches for information regarding the product.
The current research suggests that when designing a website for a new
brand, online marketers should consider that the first objective of the visitors is
to search for information regarding the product. The results of this research can
help business to employ the appropriate web features in order to design truly
interactive web pages and therefore, increase their marketing edge.
This study supports Liu and Shrum (2002) suggestion that the rush to
employ interactive elements into the marketing context should be mediated or
tempered by fully understanding both; what interactivity can do well and most
importantly what it cannot do. Before adopting the latest technological advances
firms need to take into consideration first their advantages and limitations.
Future eye-tracking research may focus on other forms of online advertising
messages. Additional research that examines experiential/utilitarian products
would provide useful insight to online marketers. The assessment of the facial
expresses could also help academics and practitioners to answer unresolved
questions regarding users’ emotions. Finally further investigation in the
cognitive process of the interactive elements would provide insights into the
cognitive psychology in the online environment.
Gaining Attention Online 77

Several limitations of the study encompassing the nature of the sample, data
collection procedures, the conditions of the lab experiment that may decrease
external validity, should be taken into consideration when interpreting the
study’s results and developing future research to extend and expand its scope.

6 References

Balatsoukas, P. and Ruthven I. (2012), “An eye-tracking approach to the analysis of


relevance judgments on the web: the case of Google search engine,” in: Journal of the
American Society for Information Science and Technology, Vol. 63, 9, 1728-1746.
Bezjian-Avery, A.; Calder, B. and Iacobucci, D. (1998), “Interactive advertising vs.
traditional advertising,” in: Journal of Advertising Research, Vol. 38, 4, 23-32.
Celsi, R.L. and Olson, J. C. (1988) “The role of involvement in attention and
comprehension processes,” in: Journal of Consumer Research, 210-224.
Coyle, J.R. and Thorson, E. (2001), “The Effects of Progressive Levels of Interactivity
and Vividness in Web Marketing Sites,” in: Journal of Advertising, Vol. 30, Fall, 65-
77.
Kuisma, J.; Simola, J.; Uusitalo, L. and Öörni, A. (2010), “The effects of animation and
format on the perception and memory of online advertising,” in: Journal of
Interactive Marketing, Vol. 24, 269-282.
Lee, J. W. and Ahn, J-H. (2012), “Attention to banner ads and their effectiveness: an
eye-tracking approach” in: International Journal of Electronic Commerce, Vol.17, 1,
119-137.
Liu, Y. and Shrum, L. J. (2002), “What is interactivity and is it always such a good
thing? implications of definitions, person, and situation for the influence of
interactivity on advertising effectiveness,” in: Journal of Advertising, Vol. 31, 4, 53-
64.
Liu, Y. and Shrum L. J. (2009), “A Dual-Process Model of Interactivity Effects” in:
Journal of Advertising, Vol. 38 (Summer), 53-68.
Mackenzie, S.B. (1986) “The role of attention in mediating the effect of advertising on
attribute importance,” in: Journal of Consumer Research, Vol. 13 (September), 174-
195.
McMillan, S. and Hwang, J. S. (2002), “Measures of perceived interactivity: An
exploration of the role of direction of communication, user control, and time in
shaping perceptions of interactivity” in Journal of Advertising, Vol. 31, 3, 29-42.
Molosavljevic, M. and Cerf, J.D. (2008), “First attention then intention: Insights from
computational neuroscience of vision,” in: International Journal of Advertising, Vol.
27, 3, 381-398.
Petty, R.E. and Cacioppo J.T. (1979), “Issue Involvement Can Increase or Decrease
Persuasion by Enhancing Message-Relevant Cognitive Responses,” in: Journal of
Personality and Social Psychology, Vol. 37, 10, 1915-1926.
Petty, R. E. and Cacioppo, J. T. (1986), “The elaboration likelihood model of persuasion
(In L. Berkowitz (Ed.)” in: Advances in experimental social psychology, New York:
Academic Press ed. vol.19, 123-205.
78 Palla and Zotos

Percy, L., and Donovan, R. J. (1991), “A better advertising planning grid” in: Journal of
Advertising Research, Vol. 31, 11-21.
Rucket D.D.; Petty R.E. and Priester, J.R. (2007), “Understanding advertising
effectiveness from a psychological perspective: The importance of attitudes and
attitude strength,” in Tellis G.J. and Ambler T. (Eds.) The handbook of advertising,
(73-88). Los Angeles, CA: Sage
Rayner, K. (2009), “Eye movements and attention in reading, scene perception and
visual search” in: The Quarterly Journal of Experimental Psychology, Vol. 62, 1457-
1506.
Sohn, D.; Ci, C. and Lee, B.K. (2007), “The moderating effects of expectation on the
patterns of the interactivity-attitude relationship” Journal of Advertising, Vol. 36, 3,
109-119.
Song, J. H. and Zinkhan, G. M. (2008), “Determinants of perceived web site
interactivity,” in: Journal of Marketing, Vol. 72, 99-113.
Stanaland A.J.S. and Tan J. (2010), “The impact of surfer/seeker mode on the
effectiveness of website characteristics,” in: International Journal of Advertising,
Vol. 29, 4, 569-595.
Strong, E.K. (1925), “Theories of Selling,” in Journal of Applied Psychology, Vol. 9, 75-
86.
Tellis, G.J. (2004), “Effective Advertising, Understanding when, how, and why
advertising works. Sage Publications, printed in United States of America.
Thorson, K. S. and Rodgers, S. (2006), “Relationships between blogs as eWOM and
interactivity, perceived interactivity, and parasocial interaction,” in: Journal of
Interactive Advertising, Vol: 6, 2, 39-50.
Voorveld, H.A.M.; Neijens, P.C. and Smit, E.G. (2011), “The relation between actual
and perceived interactivity” in: Journal of Advertising, Vol: 40, 2, 77-92.
Zaichkowsky, J. L. (1985), “Measuring the involvement construct” in: Journal of
Consumer Research, Vol: 12, 3, 341-352.
Zotos, Y. and Lysonski, S. (1993), “An exploration of the quantity surcharge concept in
Greece,” in: European Journal of Marketing, Vol: 27, 10, 5-18.
Information regarding the eye-tracking Tobii X2-30 : http://www.tobii.com/de/eye-
tracking-research/global/products/hardware/tobii-x2-30-eye-tracker/
Is Targeting Online Information Diffusers Based on Their
Personality Traits and Influencer Types Misleading?
Anik St-Onge, Sylvain Senecal, Marc Fredette, and Jacques Nantel

1 Introduction
In 2014, 250 millions of dollars has been spent in US email advertising and
according to eMarketer (2014), this amount will grow by approximately 7 % for
the next few years. This amount can seem small, but email adverting is the cheap
pest way to reach clients and is the digital advertising tool with the better return
on investment (eMarketer, 2014).
The objective of this research was to investigate the relationship between
consumers’ characteristics (e.g., opinion leadership, traits) and their intention to
engage in word-of-mouth (WOM) and these characteristics and actual behavior
in engaging in WOM. In the context of electronic WOM, prior research has
mostly used WOM intention (e.g., Chi et al., 2007), not actual behavior, to
identify information diffusers. Since intention and behavior are not always
strongly correlated, the proposed study aimed at addressing this gap in the
literature.

2 Literature Review and Hypotheses

The literature regarding traditional WOM indicates that certain persons have
a greater reach with their messages due to their numerous contacts (Goldenberg
et al., 2010, Hinz et al., 2011). With their status, knowledge and / or social
networks, these persons are more credible and inclined to diffuse information to
other consumers. The success of WOM is due in great part to these persons,
generally named influencers (Godes and Mayzlin, 2009) or hubs (Goldenberg et
al., 2010). Influencers have the ability to influence the choice of products for
others (Weimann, 1991), we know them under the name of opinion leaders
(Rogers and Cartano, 1962) and mavens (Feick and Price, 1987). For their part,
hubs possess a great number of relationships in their social network (Goldenberg
et al., 2010). Other studies reveal that brand loyalty (Iyengar, Van den Bulte and
Valente, 2011) and personality traits could explain certain attitudes and
behaviors of consumers (Black et al., 2010) such as WOM.
The literature concerning the characteristics associated with persons
diffusing information via traditional WOM is scarce and practically inexistent
for electronic WOM. The only study identified is the one by Chiu et al., (2007).
However, this study only determines the persons’ characteristics according to the

© Springer Fachmedien Wiesbaden 2017


G. Christodoulides et al. (Hrsg.), Advances in Advertising Research (Vol. VII),
European Advertising Academy, DOI 10.1007/978-3-658-15220-8_7
80 St-Onge, Senecal, Fredette, and Nantel

intent of diffusing an email, not according to the actual diffusion of an email.


Although a person’s behavior is determined by its behavioral intention (Fishbein
and Ajzen, 1975), it remains that certain intentions are never translated into
behavior (Sheppard, Hartwickrt Warshaw, 1988). Moreover, certain researchers
question the relationship between the use of influencers and the success of viral
marketing campaigns (Watt and Dodds, 2007). This study also allows testing the
“tipping point” principle popularized by Gladwell (2000) about influencers.

Figure 1: Conceptual framework

3 Method
The data collection for the study was performed with the collaboration of a
radio station, which regularly sends out a newsletter to its listeners advertising
several promotions from various companies. This radio station newsletter
database contained 186 479 email addresses. With the collaboration of one of
the frequent newsletter advertisers, a restaurant chain, we created five different
email advertisements, each communicating a distinct promotion. This study was
performed in two phases.
Phase I: WOM intentions. First, the radio station newsletter recipients
received an invitation to fill-out a questionnaire. They were randomly assigned
to one of the five restaurant promotions. The questionnaire used the following
measurement scales, which were derived from the literature: personality traits
(extraversion, conscientiousness, and openness) (Goldberg, 1990), opinion
leader (King and Summers, 1970), maven (Feick and Price, 1987), intention to
diffuse the advertisement to their friends (Harrison-Walker , attitude toward the
ad, attitude toward the advertised brand (i.e., restaurant chain) attitude toward
the radio station (Zhang and Gelb, 1996), and loyalty toward the advertised
product (e.g., pizza) (Jacoby and Kyner, 1973). A total of 3550 usable surveys
Is Targeting Online Information Diffusers Based on Their Personality 81

were completed. The sample was composed of a majority of women (68 %) and
the average age was 35.
Phase II: WOM behavior. The second phase consisted in broadcasting the
advertisement used in the Phase 1 questionnaire and tracking its diffusion. For
the purpose of this study, a virtual platform was constructed. In each sent
advertisement it was clearly stated that in order to be eligible for the promotion,
the recipient had to forward the promotional email to his/her friends by using the
virtual platform. This platform allowed us to identify diffusers and retrace their
characteristics which they provided in Phase I.
Two months after completing the questionnaire (Phase I), the 3550 Phase I
respondents received an email promotion from the same radio station containing
the same promotion as in Phase I. This time, they were not asked their intention
to forward it, but they were asked to actually forward it using the virtual
platform. Of these 3550 respondents, 1019 forwarded the email to friends using
the virtual platform. However, most of these diffusers were second and third
wave diffusers (i.e., did not directly receive the promotion from the 3550
newsletter subscribers, but from one of their contacts). A total of 168 of these
information diffusers, were part of the 3550 newsletter subscribers who
completed the Phase 1 questionnaire.

4 Results
Characteristics of information diffusers based on Intention to diffuse
information. Using Phase I data, a T-test between the participants having a high
and a low intention of diffusing information showed that participants with a
greater intention were significantly more opinion leaders and mavens. With
regards to personality traits, consumers with greater intention to diffuse were
generally more extroverted and open, corroborating the study of Chiu et al.
(2007). They were also more loyal to the restaurant chain represented in the
email ads.
82 St-Onge, Senecal, Fredette, and Nantel

Table 1: Characteristics of information diffusers based on intention to diffuse


information
Hypothesis N = 3 555 Group (1) Group(3) F P-
Average Low intent High intent value
and to to
(standard distribute distribute
deviation) advertising advertising
email email
n=654 n=676
H2a. Leader 3.57 (.87) 3.30 (.95) 3.87 (.83) 134.1 .000
H3a. Maven 5.00 (1.27) 4.58 (1.37) 5.54 (1.13) 194.9 .000
H4a. Nb of email
addresses X2(4)=
0 and 29 3.18 (1.25) 37.3 % 32.5 % 5.938.
30 and 59 3.45 (1.35) 24.0 % 28.6 % p>0.0
60 and 89 3.82 (1.02) 13.5 % 12.4 % 5
90 and 119 4.18 (.87) 7.6 % 7.0 % p-
More than 120 4.30 (1.10) 17.6 % 19.5 % value
Average 3.47 (1.16) =
0.204
H5a. Extraversion 3.54 (.77) 3.38 (.80) 3.80 (0.77) 95.5 .000
trait
H6a. Openness 3.81 (.88) 3.66 (.90) 4.00 (0.86) 49.3 .000
trait
H7a.Conscientious- 4.15 (.85) 4.03 (.89) 4.37 (0.77) 53.3 .000
ness trait
H8a. Loyalty X2(4)=
Never 3.18 (1.25) 37.8 % 20.6 % 132,
One or twice in the 3.45 (1.13) 54.4 % 48.5 % p<.05
last 3 months
One or twice per 3.82 (1.02) 7.6 % 25.9 %
month
Once or two per 4.18 (.87) .2 % 4.6 %
week
Almost daily 4.30 (1.09) 0% 0.4 %
Average 3.47 (1.16)

From intention to diffuse information. Phase 2 results showed that the


behavioral intention to diffuse an email ad is an important variable in its actual
diffusion (χ² (1)= 8,618, p<0,05). Moreover, results of a Poisson regression
demonstrated that when the intention to diffuse increased by one on the Likert-
type scale (ranging from 1 to 5), actual diffusion raised by 36%.
Is Targeting Online Information Diffusers Based on Their Personality 83

Table 2: From intention to diffuse information


Coefficients Hypothesis Exp (B)
B Std. Error Wald Chi-square
Intercepter -3.245** .4145
Intention to .308 .1049 8.618* 1.36
diffuse
(χ² (1)= 8,618, p<0,05) * p<0,05, **p<0,01

Characteristics of information diffusers based on actual behavior. Using


Phase II data, results suggested that only the loyalty differentiated diffusers from
non-diffusers. It has to be noted that the probability of diffusion of an email
promotion was higher when the participant was moderately patronizing the
restaurant chain, compared to participants who never visited the restaurant. This
supports the literature on the satisfaction and the intention to diffuse
information: the more a person is satisfied from a product, the more this person
is susceptible to refer it to others (Kassim and Abdullah, 2010).

Table 3: Characteristics of information diffusers based on actual behavior


Hypothesis Non- Diffusion T P-
diffusion n=168 value
n=3382
H2a. Leader 3.57 (.87) 3.30 (.95) 3.87 (.83) .000
H3a. Maven 5.00 (1.27) 4.58 (1.37) 5.54(1.13) .000
H4a. Nb. of email addresses
0 and 29 3.18 (1.25) 37.3 % 32.5 % X2(4)=
30 and 59 3.45 (1.35) 24.0 % 28.6 % 5.938.
60 and 89 3.82 (1.02) 13.5 % 12.4 % p>0.0
90 and 119 4.18 (0.87) 7.6 % 7.0 % 5
More than 120 4.30 (1.10) 17.6 % 19.5 %
Average 3.47 (1.16)
H5a. Extraversion trait 3.54 (0.77) 3.38 (.80) 3.80 (.77) .000
H6a. Openness trait 3.81 (0.88) 3.66 (0.90) 4.00 (.86) .000
H7a.Conscientiousness trait 4.15 (0.85) 4.03 (0.89) 4.37 (.77) .000
H8a. Loyalty X2(4)=
Never 3.18 (1.25) 37.8 % 20.6 % 132,
One or twice in the last 3 3.45 (1.13) 54.4 % 48.5 % p<.05
months
One or twice per month 3.82 (1.02) 7.6 % 25.9 %
Once or two per week 4.18 (0.87) 4.6 %
Almost daily 4.30 (1.09) 0.2 % 0.4 %
Average 3.47 (1.16) 0%
84 St-Onge, Senecal, Fredette, and Nantel

5 Discussion, Limitations and Managerial Implications


Globally, results suggest that even if a large number of internet users have
the intention of diffusing a promotion, very few actually engaged in WOM.
Furthermore, results suggested that this relationship is not linear (See Graph 1).
Thus, a large number of moderators may interfere between the intention and the
behavior. Results support prior research asserting that there are no precise
characteristics profiling information diffusers.
This study has important implication for theory. Results suggest that
WOM intention, at least in an electronic context, is not a good basis to
discriminate between information diffusers and non-diffusers. Hence, research
should turn its attention to the moderators between WOM intention and actual
WOM behavior in order to gain a better understanding of the phenomenon. For
managers, results suggest that loyalty is a better predictor of electronic WOM
than consumers’ characteristics and traits.
This study faces a number of limits. Even when starting with a large
database it is quite possible that certain email addresses were no longer valid or
simply that recipients decided to not open their emails. Unfortunately, we could
not obtain the number of emails that were opened but only the number that were
opened and clicked on. This information would have allowed us to evaluate the
incidence between the promotional message in the email title and the actual
opening of the email.
One of the most limiting factors in this study was the use of a virtual
platform to track the spread of the email advertisement. It is quite conceivable
that some recipients did open the message but decided to just forward it to
contacts instead of enlisting them on the platform. The virtual platform used for
this study would not have been able to track them in this case, perhaps
contributing to the identification of only 168 persons from the 1019 diffusers.
Some recipients were perhaps not inclined to give out the addresses of their
friends to the radio station. Still more recipients perhaps did not know or have
the email addresses of their friends as a great number of internet users now use
the social site Facebook to communicate with their friends (eMarketer, 2014).
For many younger persons, the use of email addresses is outdated.

6 References
Black, Iain R., George C. Organ and Peta Morton (2010), “The effect of personality on
response to sexual appeals”, in: European Journal of Marketing, Vol. 44, 9/10,
1453-1477.
Chiu, Hung-Chang, Yi-Ching Hsieh, Ya-Hui Kao ND Monle Lee. (2007), “The
Determinants of Email Receivers' Disseminating Behaviors on the Internet”, in
Journal of Advertising Research, vol. 47, 4, 524 - 534.
Is Targeting Online Information Diffusers Based on Their Personality 85

Bagozzi, Richard P, et Paul R. Warshaw (1990), “Trying to consume”, in: Journal of


Consumer Research, Vol.17, 2, 127-140.
Bagozzi, Richard P. (1992), “The Self-Regulation of Attitudes, Intentions, and
Behavior”, in: Social Psychology Quarterly, Vol. 55, 2, 178-204.
Feick, Lawrence F., AND Linda L. Price (1987), “The Market Maven: A Diffuser of
Marketplace Information”, in: Journal of Marketing, Vol.51, 1, 83-97.
Fishbein, Martin and Icek Ajzen (1975), Belief, Attitude, Intention and Behavior:
An Introduction to Theory and Research, Boston, Addison-Wesley.
Gladwell, Malcom (2000), The Tipping Point : How Little Things Can Make a Big
Difference. New York, Little, Brown and Company.
Goldberg, Lewis R. (1990)), “An Alternative "Description of Personality": the Big-Five
Factor Structure”, in: Journal of Personality and Social Psychology, Vol, 59, 6,
1216-1229.
Goldenberg, Jacob, Barak Libai, Eitan Muller and Stefan Stremersch (2010), “The
Evolving Social Network of Marketing Scholars”, in: Marketing Science, vol. 29, 3,
561-567.
Godes, David and Dina Mayzlin (2009), “Firm-Created Word-of-Mouth
Communication: Evidence From a Field Test”, in: Marketing Science, vol. 28, 4,
721-739.
Harrison-Walker, L. Jean. (2001), “The Measurement of Word-of-Mouth
Communication and an Investigation of Service Quality and Customer Commitment
as Potential Antecedents,” in: Journal of Service Research, Vol. 4, 1, 60-75.
Hinz, Olivier, Bernd Skiera, Christian Barrot and Jan U. Becker (2011),”Seeding
Strategies for Viral Marketing: an Empirical Comparison”, in: Journal of Marketing,
Vol. 75, 6, 55-71.
Iyengar, Raghuram, Christophe Van den Bulte and Thomas W. Valente (2011), “Opinion
Leadership and Social Contagion in New Product Diffusion”, in: Marketing Science,
Vol.30, 2, 195-212.
Jacoby, Jacob and David B. Kyner (1973), “Brand Loyalty vs. Repeat Purchasing
Behavior”, in: Journal of Marketing Research, Vol. 10, 1, 1-9.
Kassim, Norizan, et Nor Asiah Abdullah (2010), “The Effect of Perceived Service
Quality Dimensions on Customer Satisfaction, Trust, and Loyalty in E-Commerce
Settings”, in Asia Pacific Journal of Marketing and Logistics, Vol. 22, 3, 351-371.
King, Charles W., et John Summers (1970), “Overlap of Opinion Leadership Across
Consumer Product Categories, in: Journal of Marketing Research, Vol. 7, 1, 43-50.
Rogers, Everett M. and David G. Cartano (1962), “Methods of Measuring Opinion
Leadership”, in: Public Opinion Quarterly, Vol.26, 3, 435-441.
Sheppard, Blair H., Jon Hartwick and Paul R. Warshaw (1988), “The Theory of
Reasoned Action: a Meta-Analysis of Past Research with Recommendations for
Modifications and Future Research”,in: Journal of Consumer Research, Vol. 15, 3,
325-343.
Smith, Ted, James R. Coyle, Elizabeth Lightfoot and Amy Scott (2007), “Reconsidering
Models of Influence: The Relationship Between Consumer Social Networks and
Word-of-Mouth Effectiveness”, in: Journal of Advertising Research, vol. 47, 4,
387-397.
86 St-Onge, Senecal, Fredette, and Nantel

Watts, Duncan and Peter Dodds (2007). “Influentials, Networks, and Public Opinion
Formation, in: Journal of Consumer Research, Vol. 34, 4, 441-458.
Weimann, Gabriel (1991), “The influentials: back to the concept of opinion leaders?”, in:
Public Opinion Quarterly, Vol. 55, 2, 267-279.
The Effect of Online Customer Reviews’ Characteristics on
Sales

Ewa Maslowska, Edward C. Malthouse, and Stefan F. Bernritter

1 Introduction

Online customer reviews help consumers make decisions, such as


purchasing products, watching movies, or joining a sports club. Online reviews
have become a major driving force in marketing (Cui et al., 2012) and are a
common feature on many websites. Information from other consumers, such as
online reviews, is thought to be more persuasive because it is allegedly written
by other consumers rather than brands, and is therefore perceived as being more
credible and trustworthy (Willemsen et al., 2012). In addition, online reviews are
available wherever and whenever a consumer needs them. Due to their
effectiveness and prevalence, online reviews attracted substantial attention from
both researchers and practitioners. While previous research has looked into
different review features, such as valence, length, and volume, the interaction
between different features has been mostly neglected. Also, a large body of
research on online reviews is based on experimental or survey based work and
assumes that effects of online reviews are linear. We aim to address these
shortcomings by relating actual sales in an online shop to various features of
online reviews, namely: the number of reviews (review volume), the average
number of stars (review valence), and length.
The literature shows ambiguous results about the effects of all of the three
review features that are covered in this paper. With regard to the valence of
online reviews, it is the popular belief that it positively influences attitudes
towards the brand and purchase behavior. However, the results of past studies
are equivocal (see King et al., 2014 for a review). Some studies have shown that
there is a disproportionally high number of positive online reviews, which may
cause customers to discount positive reviews as not reliable (Chevalier and
Mayzlin, 2006). Furthermore, it has been demonstrated that consumers hold a
negativity bias in perception of reviews (Ba and Pavlou, 2002), whereas other
studies have demonstrated consumers’ positivity (Li and Hitt, 2008) or
confirmation biases.
Concerning the number of reviews, Bazaar (2015) showed that the count is
generally positively correlated with sales volume. However, previous research
has shown that for attribute-focused reviews, a large number of reviews creates
information overload for consumers (Park and Lee, 2008), which might result in

© Springer Fachmedien Wiesbaden 2017


G. Christodoulides et al. (Hrsg.), Advances in Advertising Research (Vol. VII),
European Advertising Academy, DOI 10.1007/978-3-658-15220-8_8
88 Maslowska, Malthouse, and Bernritter

negative effects on consumers’ purchase behavior. Finally, the length of a review


may influence consumers’ behavior, but existing results are inconsistent. For
example, Chevalier and Mayzlin (2006) found contradicting effects of review
length between reviews on Amazon.com and barnesandnoble.com and
concluded that longer reviews do not necessarily enhance sales.
These inconsistencies emphasize that there is still much to be known about
the workings and limits of consumers’ online reviews (De Bruyn and Lilien,
2008; Kimmel and Kitchen, 2014). We think that the reasons for these
inconsistencies rely on several factors. First, the majority of studies has focused
on the relative strength of the effect of review valence and tested positivity
versus negativity biases. Second, past studies have treated the effect of review
valence on dependent variables as linear and have not looked at non-linear
effects. Third, length has often been treated as a control variable (e.g., Kim and
Gupta, 2012; Xu, 2014) rather than a factor. Fourth, previous studies have
mostly dealt with information and entertainment products, such as books,
movies, and television shows (Cui et al., 2012), which may be perceived as high-
involvement, incidentally purchased products. In order to address these gaps, the
aim of this study is to investigate the effects of three online customer reviews’
features, that is, valence, volume, and length, on sales of a common use product–
a light bulb.

2 Review Valence

Star rating is an important feature of product reviews, as product reviews


have been shown to influence readers’ heuristic processing (Forman et al.,
2008), during which readers use peripheral cues to form opinions. The average
rating of a product is in many online shopping environments summarized by the
mean number of stars. Usually, these stars are rated on a 5-point scale. Simple
cues in shopping environments, such as these stars might thus be perfectly suited
as source of information for consumers who are processing product information
based on heuristics. The advantage of these kinds of summaries is that they
cannot be misinterpreted in contrast to information provided in the text of a
review. Findings of previous studies are ambiguous regarding the effects of
ratings (King et al., 2014). Some studies show that the effect of ratings on sales
is positive (e.g., Chevalier and Mayzlin, 2006). Others look at the strength of the
ratings’ effect. Chiou and Cheng (2003) explain that negative information should
weigh more because consumers expect life events to be moderately positive.
Hence they perceive negative information as far from normative expectancy,
which makes it more salient. Others suggest that ratings do not matter. In their
study of the online movie reviews on Yahoo.com, for instance, Duan et al.
(2008) found that the star ratings have no significant effect on box office sales.
The Effect of Online Customer Reviews’ Characteristics on Sales 89

Similarly, Clemons, and Gao (2008) found that star ratings do not predict sales
accurately. Still, the majority of studies demonstrate effects of review valence.
The effects vary though.
Chevalier and Mayzlin (2006) showed that 1-star reviews can negatively
affect sales on Amazon.com. Clemons et al. (2006) demonstrated that while high
ratings can predict sales, bad ratings do not predict poor sales. However, there
are exceptions to this rule showing that negative evaluations can enhance sales,
as they have the potential to increase product awareness (Berger et al., 2010).
Bosman et al. (2013) showed that star rating significantly affects review
credibility and that for every additional star, the credibility decreases on average
by 2.39 per cent (if all other factors remain unchanged). This suggests that a
review with a poor rating is perceived as more trustworthy. In line with this
notion, O’Reilly and Marx (2011) showed that readers are skeptical of
reviews that are too positive. In other words, readers who see only 5-star
reviews become suspicious (Dholakiya, 2014). Consumers also know that firms
can alternate or remove reviews, or stimulate positive reviews with financial
rewards to create high ratings (Li and Hitt, 2008).
The assumption that too positive reviews can have negative effects is
supported by research of Mudambi et al. (2010), who suggest that when
consumers rate a product on a 5-point scale, a mean rating of three may indicate
indifference of or variance in reviewers. The authors furthermore explain that
the positive effect of moderate reviews relies on the fact that two-sided messages
can enhance source credibility and brand evaluation. To disentangle extant
findings, we build on Mudambi et al. (2010) who showed that extreme reviews
will be less helpful than moderate ones, producing an inverted U-shape effect,
and propose that the relationship between the star ratings and sales is non-linear.
We expect that more positive reviews will lead to more positive effects (i.e.,
higher sale probability), but extremely good reviews (i.e., 5-star reviews), will
lead to less positive outcomes, because they imply a lack of variation and induce
suspicion. Hence, we hypothesize that:

H1: The relationship between an average review rating and sale probability is
non-linear, where higher average rating leads to higher sale probability, unless
the average rating is 5.

3 Review Volume and Length

The number of reviews is believed to have a strong effect on sales, since it


signals the popularity of a product and triggers consumers’ awareness (Duan et
al., 2008; Cui et al., 2012; Liu, 2006). The majority of studies confirms this
belief (see King et al., 2014 for a review), although there are also studies
90 Maslowska, Malthouse, and Bernritter

suggesting that volume on its own is not enough (Chintagunta et al., 2010). The
idea behind the alleged positive effect of the number of reviews is that more
information can reduce consumers’ uncertainty and strengthen their confidence
in a product, leading to a greater willingness to pay for it (Brynjolfsson and
Smith, 2000). Therefore, increasing the number of messages should bring
positive change in attitudes regardless the mode of information processing. If a
reader is in a deep processing mode, message repetition will lead to deeper
thinking about the message and the arguments, leading consumers to generate
more favorable attitudes in response to issue-relevant arguments (Chiou and
Cheng, 2003). If a reader is in a shallow processing mode, an increased number
of messages can also trigger a more positive attitude (Chiou and Cheng, 2003).
According to the elaboration likelihood model (Petty and Cacioppo, 1986),
individuals use a shortcut and follow the peripheral route of processing, which
means they focus on heuristics when they are not motivated to process a
message. Hence, they look for cues that could signal the value of a message,
such as the number of reviews that can indicate the popularity of a product, but
also, as noticed by Chiou and Cheng (2003), that consumers care about the
brand or product, because they take their time to write a review about it. Indeed
Chiou and Cheng (2003) found that participants exposed to 12 messages (vs.
six) evaluated the product more positively, had a higher overall attitude, and
showed stronger overall liking. Park et al. (2007) noticed that the number of
online reviews of a product may indicate the product's popularity, as consumers
can assume that the number of reviews is related to the number of consumers
who have bought the product, and hence, purchase intention increases with the
number of reviews. Also, Bazaar (2015) and Matfield (2011) stated that the
number of reviews had a positive impact on revenues regardless of their
valence.
However, we suggest that the effect of the number of reviews is conditional
upon another characteristic of online reviews, namely their length. Length of a
message may be perceived as a proxy of the number of arguments that it entails
and may increase information diagnosticity, which would mean that consumers
do not need to spend any more resources on searching (Mudambi and Schuff,
2010). The number of arguments can be a relevant cue in consumers’ processing
of reviews and longer reviews can be perceived as more diagnostic, which might
thus strengthen consumers’ confidence in their decisions (Mudambi and Schuff,
2010). A longer message can suggest that the reviewer thought that the product
was important and hence spent more time and effort on writing the review. In
addition, longer reviews may signal that they were written by more
knowledgeable consumers, and as such can be more persuasive (Bansal and
Voyer, 2000). If consumers scrutinize information, the longer messages can
The Effect of Online Customer Reviews’ Characteristics on Sales 91

offer additional explanation and as such be perceived as more helpful and lead
to more positive attitudes (Mudambi and Schuff, 2010).
Indeed, previous research has confirmed that the length of a review is an
important characteristic. Bosman et al. (2013) showed that text length is a
relevant predictor of review credibility. In their study, each additional word
increased review credibility on average by 0.04 per cent (when all other factors
remain constant). In a similar vein, Mudambi and Schuff (2010) found that
review depth (i.e., word count) had a positive effect on the credibility of the
review and predicted its helpfulness.
Both the number and the length of reviews reflect the cognitive effort that
the sender and the receiver need to make to, respectively, write and read the
review (Ghose and Ipeirotis, 2008). However, while the number of reviews
reflects the opinions and effort of a variety of consumers, the length of reviews
relates to the effort of individual consumers. In line with theories on social proof
(see Cialdini, 2009 for an overview), we suggest that the number of reviews is a
more vital sign of trustworthiness (i.e., many made the effort to review this
product) than the length of the reviews (i.e., this consumer made particularly
much effort). Consumers are cognitive misers with limited cognitive resources
and as such they often use heuristics instead of elaboration. Therefore, we expect
that consumers would rather rely on cues in the shopping environment that are
easier to process. Hence, when they are exposed to a large number of reviews,
they may experience cognitive overload (cf. Park and Lee, 2008), which might
turn out negative if this large number of reviews is also lengthy. On the other
hand, when there are only a few reviews, consumers have enough cognitive
resources to take another cue into account and focus on their length, which
indicates the number of arguments and hence the increased availability of
reasons for a decision, which would lead to a bigger confidence in the decision
(Tversky and Kahneman, 1974). We thus hypothesize that:

H2: The effect of the number of reviews is stronger for short reviews than for
long reviews.

4 Method

To test our hypotheses, we use data from an e-commerce store that sells
different types of light bulbs. Since it does not have physical stores, all
purchases can be recorded and linked to the reviews that were shown at the time
of ordering. We are thus able to link online reviews to real purchase behavior.
We have six weeks of sales data from August 31 through October 11, 2014. We
focus on the six largest brands and the five largest categories of bulbs, which
gives 919 unique stock keeping units (SKUs). Over the six weeks these 919
92 Maslowska, Malthouse, and Bernritter

SKUs were displayed 275,289 times resulting in 4546 purchases. Our focus is
on whether or not an item was purchased rather than predicting the quantity or
purchase amount. Table 1 gives summary statistics for the categories and
brands. For each exposure we know the following about the reviews to which
the customer was exposed: (1) the number of reviews for the SKU (num), (2)
the average number of stars across the reviews for the SKU (stars), and (3) the
average length of the reviews, measured in the number of characters (long). Our
interest is in an interaction between length and the number of reviews, and we
divide length at the mean so that there are long or short reviews; the dummy
variable long equals 1 for long reviews and 0 for short ones. We estimate the
following logistic regression using generalized additive models (Hastie and
Tibshirani, 1990):

where π is the probably of purchase, s is a univariate smoothing spline with


3 degrees of freedom modeling the effect of the average number of stars on log
odds of purchase, and the last term is the interaction effect between the number
of reviews and dichotomized length. The number of reviews is logged because
it is right skewed with outliers. As some previous studies have shown that the
effect of reviews may differ for different brand familiarity and price (e.g., Ba
and Pavlou, 2002; Berger et al., 2010; Chiou and Cheng, 2003), we included the
brand and the product category in our model to control for such effects.

Table 1: Share or orders, items and sales by brand and category

Number
Brand % Orders % Items % Sales
Orders
Brand 1 610 6.43 5.98 3.79
Brand 2 3182 33.5 36.5 42.2
Brand 3 369 3.89 3.18 4.23
Brand 4 383 4.04 1.99 6.14
Brand 5 340 2.88 3.96 2.94
Brand 6 2568 27.06 28.6 18.0
Category
Incandescent 2415 25.33 37.4 18.8
Halogen 2155 22.61 18.4 18.2
Linear 1621 17.00 21.05 18.6
LED 835 8.76 3.95 16.03
CFL 1238 12.99 8.83 14.60
The Effect of Online Customer Reviews’ Characteristics on Sales 93

5 Results

Table 2 provides the results of the analysis of deviance for the different
terms in the model, which gives the change in deviance when each term is
dropped from the model, and the value of the likelihood ratio test (LRT)
statistic. It shows for example that when the four dummies for category are
dropped from the model, the deviance increases by 396.54, which has a χ2(4)
distribution (p<.0001). There are also significant differences across brands
(χ2(5)=61.64, p<.0001). Table 3 gives parameter estimates.

Table 2: Analysis of deviance table

Term DF Deviance LRT p


Category 4 46,752 396.54 <.0001
Brand 5 46,417 61.64 <.0001
S(Avg Stars) 3 46,383 26.86 <.0001
Logreview*Long 1 46,360 4.36 .03701
Residual 46,356

Left panel of Figure 1 presents marginal plots for the effect of the number
of stars on the log odd ratio, where the dotted lines indicate 95% confidence
intervals for the mean prediction. The results are presented on a logit scale,
where a logit of -1 gives a probability of 1/(1+exp(- -1)) = 27%, a logit of -2
gives a probability of 12%, a logit of -3 gives a probability of 4.7%, and a logit
of -4 gives a probability of 1.8%. A logit of 0 corresponds to a purchase
probability of 50%=1/(1+exp(-0)). The right panel of Figure 1 shows the
interaction between length and log number of reviews.
94 Maslowska, Malthouse, and Bernritter

Table 3: Parameter estimates

Parameter Estimate SE Z p
Intercept -4.422 0.135 -32.72 <.0001
Category
Halogen 0.543 0.085 6.39 <.0001
LED -0.401 0.087 -4.63 <.0001
Linear 0.098 0.083 1.18 0.238
CFL
Brand
Brand 1 -0.179 0.066 -2.72 .007
Brand 2 0.048 0.038 1.26 0.209
Brand 3 0.499 0.093 5.37 <.0001
Brand 4 -0.107 0.085 -1.26 0.208
Brand 5 0.321 0.082 3.94 <.0001
Brand 6
Logreview 0.136 0.031 4.38 <.0001
Long -0.059 0.044 -1.34 0.179
Logreview*
-0.082 0.039 -2.09 0.037
Long

Concerning the first hypothesis, we can see in Table 2 that the nonlinear
effect of review valence (i.e., average stars) is significant (χ2(3)=26.86,
p<.0001), indicating that the spline explains significantly more variance than a
linear term. The left panel of Figure 1 shows that the number of stars has no
effect between 1 and about 3, but the function increases after 3 stars, indicating
that, for instance, products with 4-star reviews are more likely to be purchased
than those with 3 stars. The function achieves a maximum around 4.2, and
decreases somewhat thereafter. This suggests that a SKU with an average of 4.2
stars is more likely to be purchased than one with 5 stars, which confirms our
first hypothesis.
With regard to the second hypothesis, Table 2 shows a significant
interaction effect (Est.=-.08, SE=.04, p=.037), indicating that the effect of the
number of reviews depends on their length. The right panel of Figure 1 shows
the interaction effect between review length and the amount of reviews. The
effect of the log number of reviews is stronger when the average length of the
review is short (solid line, slope=0.136) than when the reviews are long (dashed
line, slope=0.136–0.082=0.054), which supports our second hypothesis.
The Effect of Online Customer Reviews’ Characteristics on Sales 95

1 2 4 8 16
0.2

0.2
Short
Long
0.1

0.1
36
.1

Effect on Log Odds


e=0
lop
s(Avg Stars)

0.0

0.0
S
-0.1

-0.1
-0.2

-0.2
4
e= 0.05
Slop
-0.3

-0.3
1 2 3 4 5 0.0 0.5 1.0 1.5 2.0 2.5 3.0

Average Number of Stars log(Number Reviews)

Figure 1: Marginal plots for (left) average number of stars and (right) the interaction
between length and log number of reviews

6 Summary, Managerial Implications, Limitations

Despite the substantial interest in the workings and limits of online reviews,
there are still lacunae and inconsistencies in the literature. Moreover, neither
non-linear effects of review features nor interaction effects between them have
been sufficiently studied, especially in the context of actual sales. The aim of
this study was to address some of these gaps by investigating the effect of star
rating of reviews and the interaction between the number of reviews and their
length on purchase probability.
Although extant research is not consistent when it comes to the effects of
valence with some studies showing it has a positive effect (e.g., Chevalier and
Mayzlin, 2006; Chintagunta, Gopinath, and Venkataraman, 2010; Zhang and
Dellarocas, 2006) and others failing to find any effect (e.g., Amblee and Bui,
2011; Chen, Wu, and Yoon, 2004; Liu, 2006), a higher average star rating is
generally assumed to be a good thing. We found that this effect is not linear and
products with an average star rating of about 4.2 are most likely to be purchased.
In a similar vein, there is no consensus regarding the effect of volume of
reviews. Some past studies have proved that volume affects market outcomes
(e.g., Cui, Lui, and Guo, 2012; Duan, Gu, and Whinston, 2008; Gu, Park, and
Konana, 2012; Liu, 2006), but there is also literature suggesting that volume on
its own does not suffice (e.g., Chintagunta, Gopinath, and Venkataraman, 2010;
Gopinath, Thomas and Krishnamurthi, 2014). We suggest that the effect of
96 Maslowska, Malthouse, and Bernritter

review volume depends on their length. Indeed, the results show that consumers
appreciate short reviews that get to the point, and that quantity matters more for
short reviews than long ones.
There are several explanations for our results, which may provide
directions for future research. Concerning the effect of review valence, it may be
that consumers perceive extremely positive ratings (i.e., 5-star ratings) as being
too good to be true. Consumers may think that such reviews were possibly
created by representatives from the company. Consumers are aware that firms
can edit or remove negative reviews, or stimulate positive reviews with financial
rewards to create high ratings (Li and Hitt, 2008). Another explanation may be
that positive product characteristics may be perceived as not very indicative of
product quality (Herr, Kardes, and Kim, 1991; Mizerski, 1982), because of
consumers’ naive theories about the sources of positive information, suggesting
that people sometimes share positive information to signal competence or for
self-presentation reasons (Chen and Lurie, 2014). These notions should be
investigated in future studies, which could take into account the content of the
reviews. Bazaar’s report (2015) showed that reviews with different star ratings
contain different features, with 1-3 star reviews containing product shortcomings
and 3-4 star reviews suggesting product improvements. Hence, the content of the
reviews may explain the non-linear effect of review valence.

6.1 Limitations
Our findings contribute to our understanding of the role of reviews, but our
study has some limitations as well. One of them may be an omitted variable
bias—whether or not the consumer read the reviews. We were not able not
compare consumers who rely on the heuristic (i.e., valence and volume) with
those who actually read the content of the reviews. It is plausible that only
customers who rely on heuristics are affected in a non-linear way by review
valence. Similarly, the interaction effect of review volume and length may
depend on review readership. It is possible that only those who read reviews
carefully prefer shorter reviews when there are many of them due to the
experienced cognitive overload.
It is also important to notice that even though we find an effect for reviews,
the detected effect is almost flat. That would suggest that the number of reviews
does not have such a strong effect, which would be more in line with previous
studies finding no effects of volume (e.g. Chintagunta, Gopinath, and
Venkataraman 2010).
We used real-life data. Hence, our sample is not a random sample
representative of population. Consumers differ in their attention to and reliance
on reviews, and their previous experience. Also, some consumers are more likely
to write online reviews (Zhu and Zhang 2010). However, we did not have a
The Effect of Online Customer Reviews’ Characteristics on Sales 97

chance to control for such variables. Consumers also have pre-existing product
knowledge, attitudes towards brands and stores. As Chiou and Cheng (2003)
argue, the effect of valence may depend on pre-existing brand image. Although
we did control for the effect of brand, we did not know consumers’ brand
preferences. Therefore, these personal factors should be included in the future.
Correspondingly, past research has also shown that the effect of brand
diminishes when reviews are included (Kostyra et al. 2015), but the relationship
between brands and online reviews has received little attention (Lovett, Peres,
and Shachar 2013). For example, Berger, Sorensen, and Rasmussen (2010)
suggest that negative reviews can increase sales of unestablished brands by
increasing their awareness, Chiou and Cheng (2003) show that the effect of
reviews differs for low- vs. high-image brands, and Ho-Dac, Carson, and Moore
(2013) show that online reviews matter less for strong brands. Therefore, future
research should not only control for the effect of brand, but should examine how
different brands are affected by online reviews, and marketers should measure
what the effects of reviews for their brand are.
In our study we looked at one product category, which is used offline and
is a rather commonly purchased good. The results may differ for products from
other categories, for example, for goods that are used online (Zhu and Zhang
2010) or depend on product lifetime (Chen, Wang, and Xie 2011). Future
studies should look at the effects of reviews across different categories of
products.

6.2 Managerial Implications


Our results also bring significant practical implication. First, the finding
showing that a higher average star rating does not always mean better sales of
the product imply that companies should not moderate or remove negative
reviews. A small amount of less positive reviews may increase credibility and
trust, improving sales, while extremely positive reviews can undermine sales.
Hence, we agree with Zhang, Craciun, and Shin (2010) and Zhang, Li, and Chen
(2012) that companies should not censor negative reviews. In the case of
extraordinary well evaluated products (i.e., > 4.5 stars), it might even be
beneficial to explicitly motivate consumers to suggest points for improvement of
the product in their reviews in order to slightly lower the overall valence of the
reviews.
Second, our results concerning the interaction effect of review volume and
length suggest that when a new product is introduced that does not have many
reviews, brands should stimulate consumers to write more elaborate reviews. On
the other hand, when it comes to an established product with many reviews,
brands may want to change their review solicitation procedures and ask for
short, concise reviews.
98 Maslowska, Malthouse, and Bernritter

7 References

Amblee, N., and Bui. T. (2011), “Harnessing the Influence of Social Proof in Online
Shopping: The Effect of Electronic Word of Mouth on Sales of Digital
Microproducts,” in: International Journal of Electronic Commerce, 16(2): 91–114.
Ba, S. and Pavlou, P. A. (2002), “Evidence of the Effect of Trust Building Technology in
Electronic Markets: Price Premiums and Buyer Behavior,” in: MIS quarterly, 243-
268.
Bansal, H. S. and Voyer, P. A. (2000), “Word-of-Mouth Processes Within a Services
Purchase Decision Context,” in: Journal of service research, 3(2), 166-177.
Bazaar (2015). The Conversation Index, Vol. 8.
Berger, J. Sorensen, A. T. and Rasmussen, S.J. (2010), “Positive Effects of Negative
Publicity: When Negative Reviews Increase Sales,” in: Marketing Science 29(5),
815-827.
Bosman, D. J., Boshoff, C. and van Rooyen, G. J. (2013), “The Review Credibility of
Electronic Word-of-Mouth Communication on E-commerce Platforms,” in:
Management Dynamics,22(3), 29-44.
Brynjolfsson, E. and Smith, M. D. (2000), “Frictionless Commerce? A Comparison of
Internet and Conventional Retailers,” Management Science, 46(4), 563-585.
Chen, P.-Y., Wu, S., and Yoon, J. (2004), “The Impact of Online Recommendation and
Consumer Feedback on Sales,” in: Proceeding of the International Conference on
Information Systems: 711–724.
Chen, Y., Wang, Q. and Xie, J. (2011), “Online Social Interactions: A Natural
Experiment on Word of Mouth Versus Observational Learning,” in: Journal of
Marketing Research, Vol. 48, No. 2, pp. 238-254.
Chen, Z., and Lurie. N. H. (2013), “Temporal Contiguity and Negativity Bias in the
Impact of Online Word of Mouth,” in: Journal of Marketing Research, 463–476.
Chevalier, J. A. and Mayzlin, D. (2006), “The Effect of Word of Mouth on Sales: Online
Book Reviews,” in: Journal of Marketing Research, 43(3), 345-354.
Chintagunta, P. K., Gopinath, S. and Venkataraman, S. (2010), “The Effects of Online
User Reviews on Movie Box Office Performance: Accounting for Sequential Rollout
and Aggregation across Local Markets,” in: Marketing Science 29(5), 944-957.
Chiou, J. S. and Cheng, C. (2003), “Should a Company Have Message Boards on Its
Web Sites?” in: Journal of Interactive Marketing, 17(3), 50-61.
Cialdini, R. B. (2009), Influence: Science and Practice (5th ed.). Boston: Pearson
Education.
Clemons, E. K. and Gao, G. (2008), “Consumer Informedness and Diverse Consumer
Purchasing Behaviors: Traditional Mass-Market, Trading Down, and Trading Out
into the Long Tail,” E-Commerce Research and Applications, 7(1), 3 - 17.
Clemons, E. K., Gao, G. and Hitt, L. M. (2006), “When Online Reviews Meet
Hyperdifferentiation: A Study of the Craft Beer Industry,” in: Journal of
Management Information Systems, 23(2), 149-171.
Cui, G. Lui, H.K. and Guo, X. (2012), “The Effect of Online Consumer Reviews on New
Product Sales,” in: International Journal of Electronic Commerce, 17(1), 39-58.
The Effect of Online Customer Reviews’ Characteristics on Sales 99

De Bruyn, A. and Lilien, G.L. (2008), “A Multi-Stage Model of Word-of-Mouth


Influence Through Viral Marketing,” International Journal of Research in Marketing
25, 151–163.
Dholakiya, P. (2014), “Are Excessively Positive Customer Reviews Hurting Your
Conversion Rates?” http://unbounce.com/conversion-rate-optimization/customer-
reviews-conversion-rates/.
Duan, W., Gub, B. and Whinston, A. B. (2008), “The Dynamics of Online Word-of-
Mouth and Product Sales—An Empirical Investigation of the Movie Industry,” in:
Journal of Retailing 84(2), 233–242.
Forman, C., Ghose, A. and Wiesenfeld, B. (2008), “Examining the Relationship Between
Reviews and Sales: The Role of Reviewer Identity Disclosure in Electronic
Markets,” in: Information Systems Research, 19(3), 291-313.
Ghose, A. and Ipeirotis, P. G. (2008), “Estimating the Socio-Economic Impact of
Product Reviews: Mining Text and Reviewer Characteristics,” NYU Stern
Research Working Paper 1–26.
Gopinath, S., Thomas, J. S. and Krishnamurthi. L. ( 2014), “Investigating the
Relationship Between the Content of Online Word of Mouth, Advertising, and
Brand Performance,” in: Marketing Science, 33(2): 241–258.
Gu, B., Park, J. and Konana, P. (2012), “The Impact of External Word-of-Mouth Sources
on Retailer Sales of High-Involvement Products,” in: Information Systems
Research, 23(1): 182–196.
Hastie, T. J. and Tibshirani, R. J. (1990), Generalized Additive Models, Vol. 43. CRC
Press.
Herr, P. M., Kardes, F. R and Kim. J. (1991), “Effects of Word-of-Mouth and Product-
Attribute Information on Persuasion: An Accessibility-Diagnosticity Perspective,”
in: Journal of Consumer Research 17(4): 454–462.
Ho-Dac, N. N., Carson, S. J. and Moore. W. L. (2013), “The Effects of Positive and
Negative Online Customer Reviews : Do Brand Strength and Category Maturity
Matter?” in: Journal of Marketing 77(November): 37–53.
Kim, J. and Gupta, P. (2012), “Emotional Expressions in Online User Reviews: How
They Influence Consumers' Product Evaluations,” in: Journal of Business Research,
65, 985–992.
Kimmel, A.J. and Kitchen, P.J. (2014), “WOM and Social Media: Presaging Future
Directions for Research and Practice,” in: Journal of Marketing Communications,
20(1)-2, 5-20.
King, R.A., Racherla, P. and Bush, V.D. (2014), “What We Know and Don't Know
About Online Word-of-Mouth: A Review and Synthesis of the Literature,” in:
Journal of Interactive Marketing, 28, 167–183.
Kostyra, D. S., Reiner, J., Natter, M. and Klapper, D. (2015), “Decomposing the Effects
of Online Customer Reviews on Brand, Price, and Product Attributes,” in:
International Journal of Research in Marketing.
Li, X. and Hitt, L. M. (2008), “Self-Selection and Information Role of Online Product
Reviews,” in: Information Systems Research, 19(4), 456-474.
Liu, Y. (2006), “Word of Mouth for Movies: Its Dynamics and Impact on Box Office
Revenue,” in: Journal of Marketing, 70(3): 74–89.
100 Maslowska, Malthouse, and Bernritter

Lovett, M. J., Peres, R. and Shachar. R. (2013), “On Brands and Word of Mouth,” in:
Journal of Marketing Research 50(August): 427–444.
Matfield (2011), “The Impact of Customer Reviews and Ratings on Conversion Rates,”
http://www.smartinsights.com/conversion-optimisation/product-page-
optimisation/reviews-conversion-rate-impact/.
Mizerski, R. W. (1982), “An Attribution Explanation of the Disproportionate Influence
of Unfavorable Information,” in: Journal of Consumer Research, 9(3), 301.
Mudambi, S. M. and Schuff, D. (2010), “What Makes a Helpful Review? A Study of
Customer Reviews on Amazon. com.,” in: MIS quarterly, 34(1), 185-200.
O'Reilly, K. and Marx, S. (2011), “How Young, Technical Consumers Assess Online
WOM Credibility,” in: Qualitative Market Research: An International Journal,
14(4), 330-359.
Park, D. H. and Lee, J. (2008), “eWOM Overload and Its Effect on Consumer
Behavioral Intention Depending on Consumer Involvement, in: Electronic
Commerce Research and Applications, 7, 386–398.
Park, D. H., Lee, J. and Han, I. (2007), “The Effect of On-Line Consumer Reviews on
Consumer Purchasing Intention: The Moderating Role of Involvement Source,” in:
International Journal of Electronic Commerce,11(4), 125-148.
Petty, R. E. and Cacioppo, J. T. (1986), “The Elaboration Likelihood Model of
Persuasion,” in: Advances in Experimental Social Psychology, 19, 123-205.
Tversky, A. and Kahneman, D. (1974), Judgment Under Uncertainty: Heuristics and
Biases,” in: Science, 185(4157), 1124-1131.
Willemsen, L. M., Neijens, P.C. and Bronner, F. (2012), "The Ironic Effect of Source
Identification on the Perceived Credibility of Online Product Reviewers," in: Journal
of Computer-Mediated Communication, 18(1), 16–31.
Xu, Q. (2014), “Should I Trust Him? The Effects of Reviewer Profile Characteristics on
eWOM Credibility,” in: Computers in Human Behavior, 33, 136–144.
Zhang, J. Q., Craciun, G. and Shin, D. (2010), “When Does Electronic Word-of-Mouth
Matter? A Study of Consumer Product Reviews,” Journal of Business Research, 63,
1336–1341.
Zhang, Z., Li, X. and Chen, Y. (2012), “Deciphering Word-of-Mouth in Social Media,”
in: ACM Transactions on Management Information System,s 3(1): 1–23.
Zhu, F., and Zhang. X. (Michael) (2010), “Impact of Online Consumer Reviews on
Sales:TheModerating Role of Product and Consumer Characteristics,” in: Journal
of Marketing 74(March): 133–148.
The Influence of Personal Data Usage on Mobile Apps

Morikazu Hirose, Kei Mineo, and Keiya Tabe

1 Introduction
Smartphones and mobile devices have been used as powerful and popular
marketing tools. In fact, the mobile advertising market has seen swift growth.
Traditionally, the mobile advertising has been recognized as mobile website
advertising. However, as smartphones penetrate markets worldwide, a new type
of form advertising, called mobile in-apps advertising, has become more
popular. Recent surveys have indicated that smartphone users spend much more
time on their apps while spending less time on mobile Internet access through
web browsers (Perez, 2014). According to Statista (2015), worldwide mobile in-
apps advertising expenditure in 2013 was $3.5 billion (USD). They also
estimated that this figure would increase to $16.9 billion (USD) by 2018.
In-apps advertising has several advantages over mobile website advertising.
First, in-apps advertising is less clutter than websites advertising because
smartphone users access the Internet directly through apps. Once users download
an app and like it, they tend to use it frequently. Second, advertisers can easily
select their advertising media. Because most apps have a specific purpose,
advertisers can develop in-apps advertising that fits the context. Third,
advertisers can develop highly personalized advertising. Compared to mobile
website advertising, in-apps advertising can be related to personal information
collected with global positioning system (GPS).
However, serious concerns associated with user privacy are growing among
mobile marketing and advertising. Smartphone users can enjoy personalized
information while they provide their privacy to service providers. Location-
based or transaction information are highly personalized information. Even if
they are anonymized, users can be identified with integrated personal datasets.
Therefore, it is important that advertisers and apps providers understand how
to balance these issues. Few studies have focus on these issues in the advertising
research context. To fill this gap, this study examines how users’ perceptions of
the personalization of mobile apps and their privacy concerns influence their
acceptance of mobile apps.

© Springer Fachmedien Wiesbaden 2017


G. Christodoulides et al. (Hrsg.), Advances in Advertising Research (Vol. VII),
European Advertising Academy, DOI 10.1007/978-3-658-15220-8_9
102 Hirose, Mineo, and Tabe

2 Literature Review and Hypotheses

2.1 The Influences of Personalization and Privacy Concerns in Mobile App


Usage
Mobile apps are an emerging advertising platform. Advertisers can deliver their
message while users can search for what they want through the app. Providers
design their apps independently. The graphical user interfaces or functions are
not always built in the same manner. Therefore, users need to understand and
evaluate how the mobile app works, then decide if they should use (accept) it or
not.

Technology Acceptance Model


To explain mobile apps usage, we applied the technology acceptance model
(TAM) as the core model. Because mobile apps are an emerging area and prior
studies are limited, an interdisciplinary approach was appropriate. The TAM
explains the process of an individual’s adoption of new information technology,
especially computer usage behaviors (Davis et al., 1989). Moreover, in the field
of mobile advertising, TAM is the most frequently used model (Okazaki and
Barwise 2011). Therefore, we decided to apply this framework to our research.
According to TAM, evaluations of apps lead to the intention to apps, and as
variables of evaluation for mobile apps, perceived ease of use and perceived
usefulness simultaneously work as antecedents for attitudes toward apps.
Formally:

H1: Perceived ease of use positively influences perceived usefulness.

H2: Perceived ease of use positively influences attitudes toward mobile apps
usage.

H3: Perceived usefulness positively influences attitudes toward mobile apps


usage.

H4: Attitudes toward apps positively influence intentions to use mobile apps.

Personalization
Consumer perception toward personalization is an important factor in online
services. Developments of Internet technology have enabled service providers to
deliver and users to receive valuable information. Various online marketing
contexts have well discussed the concept of personalization. Personalization is
defined as at which “the firm decides, usually based on previously collected
The Influence of Personal Data Usage on Mobile Apps 103

customer data, what marketing mix is suitable for the individual” (Arora et al.
2008, p. 306). Wolfinbarger and Gilly (2003) recognized personalization as an
antecedent of online service quality. Additionally, Ha and Stoel (2009) implied
that online service quality positively influences on TAM variables. Thus;

H5: Perceived personalization has a positive effect on (a) perceived ease of use,
(b) perceived usefulness, and (c) attitudes toward mobile apps usage.

Information Privacy Concerns


Privacy concerns is another important factor in online communications.
Research has discussed this issue often with information personalization (e.g.,
Awad and Krishnan, 2006). In various online contexts, users’ concerns for
privacy issues have also been examined (Milberg et al., 2000; Rose, 2006;
Stewart and Segars, 2002). The concern for privacy is regarded as one risk
perception within the online context (Wu and Wang 2005). Especially in mobile
advertising, this risk perception negatively effects user acceptance of
technologies (Okazaki et al., 2009). Therefore, users’ information privacy
concerns should be negatively related to perceptions toward mobile apps. Given
previous research, we propose the following hypotheses:

H6: Information privacy concerns negatively affects (a) perceived ease of use,
(b) perceived usefulness, and (c) attitudes toward mobile apps usage.

2.2 Situational Differences


To gain further insights, we considered situational differences among mobile
apps use. Prior research has suggested that personalization and privacy have a
contextual nature (Sheehan and Hoy, 2000; Smith, Dinev, and Xu, 2011). Thus,
investigating the situational influences seems warranted.

Data Collection and Privacy Concerns


Mobile apps can take a different approach to personalization. The amount of
information gathered differs depending on the mobile app services. The amount
of information collected is related to privacy concerns as such. Concerns for
online information has specific dimensions that are distinct from traditional or
offline direct marketing.
Privacy concerns for online information have specific dimensions that are
distinct from traditional or offline direct marketing. The concept of privacy
concerns consists of three antecedents; collection, control, and awareness
(Malhotra et al., 2004). Collection refers to the degree to which a person is
concerned about the amount of individual-specific data possessed by others
104 Hirose, Mineo, and Tabe

relative to the value of benefits received. Sheehan and Hoy (2000) suggested
that collection is the most important factor among other dimensions. Arora et al.
(2008) pointed out the need to consider the degree of personalization because it
generates psychological factors. From this perspective, it is easily assumed that
the amount of private data collected should affect mobile apps usage.
To provide highly personalized information, mobile apps collect a lot of
private information. Therefore, it is assumed that users access these services
outside or in urgent situations. Thus, the degree of personalization means a lot to
the evaluation of mobile apps. On the other hand, the apps, which collect private
information, put individuals’ privacy at risk. Therefore, in the situation where
users have to give much more private information, privacy concerns would work
strongly and negatively on perceptions to mobile apps. In brief, the amount of
personal data collected will moderate the relationship between personalization
and privacy concerns and TAM variables, thereby suggesting that the
relationships of these factors in rich data collection apps situation will be
different to those in poor data collection. Thus:

H7: TAM variables, (a) perceived ease of use, (b) perceived usefulness, and (c)
attitudes toward mobile apps usage are highly evaluated for poor data
collection apps use compared to rich data collection apps.

H8: The influences of personalization on (a) perceived ease of use, (b)


perceived usefulness, and (c) attitudes toward mobile apps usage are
positively stronger for rich data collection apps use compared to poor data
collection apps.

H9: The influences of information privacy concerns on (a) perceived ease of


use, (b) perceived usefulness, and (c) attitudes toward mobile apps usage
are negatively stronger for rich data collection apps use compared to poor
data collection apps.

Figure 1 illustrates the research model proposed in this study.


The Influence of Personal Data Usage on Mobile Apps 105

The Amount of The Amount of


Data Collection Data Collection

H8 H9 H7

H5c

Personalization H5b Usefulness

H3
H5a
H4 Intention
H1 Attitude
To Use
H6b
H2

Privacy H6a Ease of Use


Concerns

H6c
Figure 1: Proposed research model

3 Analysis and Results

3.1 Procedure
The research survey was conducted in Japan to test our hypotheses. A
professional research firm conducted the online surveys. All scales were adapted
from prior research and were measured using 7-point Likert scales. In TAM,
perceived usefulness was adapted from Nysveen, Pedersen, and Thorjoensen
(2005), perceived ease of use from Davis (1989), attitude toward using the
mobile apps service from Dickinger and Kleijnen (2008), intention to use mobile
apps from Dabholkar and Bagozzi (2002). The scales of perceived
personalization were adapted from Xu, Luo, Carroll, and Rosson (2011) and
information privacy concerns from Awad and Krishnan (2006).
These scale items were translated into Japanese and back-translated to ensure
consistency and accuracy of meaning. A scenario method was employed. Since
smartphone users read an explanation of the apps at the down load sites, a
scenario method seems to be reasonable in this study.
106 Hirose, Mineo, and Tabe

Two types of situational scenarios were prepared, and respondents were


asked to imagine a fictious mobile apps service for short trips (see Appendix).
According to the MIAC (2015), the major online shopping categories included
tourism related (19.5%), followed by foods (15.9%) and fashion (12.5%) in
Japan. Therefore, adapting mobile apps for short trips was reasonable for the
study in Japan.

Table 1: Respondents’ profiles


Poor data collection appRich data collection app Total sample
Demographics ( n = 327 ) ( n = 337 ) ( N = 664 )
Gender
Male 51.99 52.23 52.11
Felmale 48.01 47.77 47.89
Age
20-29 37.61 41.25 39.46
30-39 34.86 28.49 31.63
40-49 27.52 30.27 28.92
Occupations
Executive/managerial 1.53 2.37 1.96
Administrative/clerical 48.32 49.55 48.95
Self-employed 3.67 3.26 3.46
Part-time workers 12.54 10.68 11.60
Public employee 5.20 3.86 4.52
Freelance 2.14 1.48 1.81
Homemakers 14.07 14.54 14.31
Students 6.42 8.61 7.53
Retired 5.50 4.45 4.97
Unemployed 0.61 1.19 0.90

One scenario was for poor data collection (PDC) condition and the other was
for rich data collection (RDC) condition. In advance, we conducted interviews
of experts who worked for an advertising agency. Based on these interviews, we
selected consumers’ interest and location information as means to personalize
the information provided by suppliers. The online panel produced a total of
1,000 responses. Then, only smartphone users were selected. Finally, 664
samples were adapted (see Table 1).

3.2 Results
To analyze our model, we followed the 2-step approach presented by
Anderson and Gerbing (1988). Firstly, we conducted confirmatory factor
analysis (CFA) with the maximum likelihood method (Byrne, 2009). We used
multiple indices to assess the fit of the proposed model (Bagozzi and Yi, 1989).
The results of CFA indicated that except the χ2 value, the other indices that were
within an acceptable range (Hair et al., 2009). The measurement model was
deemed statistically acceptable (see Table 2). Next, the CR and AVE were
The Influence of Personal Data Usage on Mobile Apps 107

calculated to determine the level of internal consistency. All scores highly


exceeded the recommended minimum level (Hair, Black, Babin, & Anderson,
2009). Then, we conducted discriminative validity analysis. We compared the
correlations among constructs to the square root of AVEs. The results indicated
that discriminant validity is acceptable in our model (Fornell and Larker, 1981)
(see Table 3).

Table 2: Fit indices


Confirmatory factor analysis
χ2 df CFI TLI RMSEA
Total sample 508.824 188 0.9 0.98 0.051
PDC 441.012 188 8 0.9 0.96 0.064
RDC 401.664 188 7 0.9 0.97 0.058
8

Structural equation modeling


?² df CFI TL RMSEA
Total sample 627.894 193 0.98 I 0.9 0.058
PDC 542.98 193 0.96 7 0.9 0.075
RDC 442.357 193 0.97 5 0.9 0.062
7
To test the hypothesized relationships, the structural paths between the
proposed constructs were examined. The fit indices of our model were
acceptable, except for the χ2 statistic (see also Table 2). All paths of the pooled
sample in TAM were significant. Therefore, H1 through H4 are all supported.
The paths from personalization were also statistically significant. Therefore, H5
was also fully supported.
On the other hand, the paths for privacy were positively significant except
for the path between privacy concerns and ease of use. Thus, H6 was rejected.
Figure 2 shows the model estimation results.

Table 3: Reliability and convergent validity, discriminant validity

Total
Variable Name ? CR AVE 1 2 3 4 5 6
1. Perceived ease of use 0,94 0.94 0.83 0.91
2. Perceived usefulness 0.93 0.94 0.79 0.66 0.89
3. Attitude towards mobile apps 0.96 0.96 0.88 0.68 0.81 0.94
4. Intention to use mobile apps 0.97 0.97 0.87 0.56 0.67 0.78 0.93
5. Perceived personalization 0.96 0.96 0.88 0.68 0.79 0.85 0.80 0.94
6. Information privacy concerns 0.82 0.84 0.59 0.24 0.30 0.24 0.19 0.20 0.77
108 Hirose, Mineo, and Tabe

PDC
Variable Name α CR AVE 1 2 3 4 5 6
1. Perceived ease of use 0.94 0.94 0.83 0.91
2. Perceived usefulness 0.92 0.93 0.76 0.66 0.87
3. Attitude towards mobile apps 0.96 0.96 0.89 0.66 0.83 0.94
4. Intention to use mobile apps 0.97 0.97 0.85 0.48 0.69 0.75 0.92
5. Perceived personalization 0.95 0.95 0.87 0.63 0.77 0.84 0.81 0.93
6. Information privacy concerns 0.81 0.83 0.57 0.33 0.47 0.41 0.30 0.31 0.75

RDC
Variable Name α CR AVE 1 2 3 4 5 6
1. Perceived ease of use 0.94 0.94 0.84 0.91
2. Perceived usefulness 0.95 0.95 0.83 0.65 0.91
3. Attitude towards mobile apps 0.96 0.96 0.88 0.68 0.79 0.94
4. Intention to use mobile apps 0.98 0.97 0.88 0.62 0.65 0.79 0.94
5. Perceived personalization 0.96 0.96 0.90 0.72 0.80 0.86 0.80 0.95
6. Information privacy concerns 0.83 0.85 0.61 0.16 0.17 0.09 0.11 0.10 0.78

NOTE: Diagonal elements are the square root of AVE and highlighted in bold. Off-
diagonal elements are simple bivariate correlations between the constructs.

0.556***

0.648***
Personalization Usefulness

0.298 ***
0.671***
Attitude
Towards 0.794 *** Intention
0.196 *** To Use
Mobile Apps
0.137 *** 0.108 ***

Privacy
Ease of Use
Concerns 0.114***

n.s.

Figure 2: Model results (H1 to H6)


NOTE: *p < 0.05, **p < 0.01, ***p < 0.001.

To compare two groups, we conducted one-way ANOVA and multi-group


analysis. Firstly, we tested differences of the latent mean variables. One-way
ANOVA showed that the mean values of TAM variables are higher for PDC
situation compared to RDC situation. However, statistical differences were not
found for perceived ease of use and perceived usefulness. Hypotheses H7c to
The Influence of Personal Data Usage on Mobile Apps 109

H7d were statistically supported. However, H7a and H7b were not supported in
our data. Therefore H7 was partially supported (see Table 4).

Table 4: One-way ANOVA for TAM variables (H7)

Atttitude towad Intention to use


Degree of Ease of Use Usefulness
mobile apps usage mobile apps
RDC 5.00 5.26 5.03 4.38
PDC 4.92 5.15 4.82 4.14
Pooled sample 4.96 5.20 4.93 4.26
ANOVA F (1, 662) = 1.60 F (1, 662) = 0.98 F (1, 662) = 5.46 F (1, 662) = 4.97
N.S. N.S. p<.05 p<.05

Secondly, we tested for metric invariance and confirmed that it was


supported (see Table 5). The paths from personalization to all variables for
TAM were significantly and positively stronger for RDC apps compared to the
PDC apps (see Table 6). H8 was supported fully and statistically.
The path from privacy concerns to variables for TAM was negatively
stronger for RDC apps compared to the PDC apps. Additionally, differences in
the path to perceived usefulness and attitudes toward the apps were statistically
significant; however, a significant difference in the path to ease of use was not
found. H9b and H9c were supported, while H9a was rejected. Therefore H9 was
partially supported (see also Table 6).

Table 5: Goodness-of-fit statistics for tests of multi-group invariance


Statistical
χ² df Δχ² Δdf
significance
Unconstrained Model 985.343 386 ― ―
Mesurement Weights 1004.372 402 19.029 16 p = 0.27 (N.S.)

Table 6: Individual group estimations (H8 to H9)

NOTE: *p < 0.05, **p < 0.01, ***p < 0.001.


110 Hirose, Mineo, and Tabe

4 Discussion, Limitations and Managerial Implications


The current research investigated the causal relationships between mobile
apps usage, information privacy concerns, and perceived personalization and the
moderating effects of the amount of data collection. The results of this study
provide some theoretical and practical implications.
First, the results were consistent with previous research, which indicate that
the TAM is applicable in the mobile advertising context. Our data also suggests
that attitudes toward the mobile apps usage work as strong motivators.
Second, personalization indicated positive relationships to TAM variables. In
particular, personalization was relatively strong antecedent of ease of use and
usefulness. This relationship was consistent with the mobile marketing research,
which suggests that personalization is a key element in mobile apps usage.
Third, information privacy concerns suggested very weak but positive effects
on ease of use and usefulness. In addition, no significant relationships existed
between information privacy concerns and attitudes toward mobile apps usage.
These results are not consistent with prior research possibly because mobile
users may have recognized that personal information adopted in our research
hardly invade their privacy resulting in the positive relationships. It was also
suggested that privacy concerns might be associated with favorable evaluations
toward low risk apps.
Finally, the amount of information provided to users had moderating effects.
TAM variables in PDC situation were highly evaluated to RDC situation.
Attitude towards mobile apps and intention to use mobile apps, in particular, are
statistically different. In addition, the relationships between perceived
personalization and TAM variables were strong when mobile apps users needed
to provide personal information. On the other hand, the path between
information privacy concerns and usefulness and attitudes toward the apps were
weak in the same situation. The overall trend in our data was consistent with
prior studies, although no significant difference existed from information privacy
concerns to ease of use.
In practice, our results make some valuable contributions. Mobile apps
become a major advertising platform. To develop effective mobile apps and in-
apps advertising, advertisers have to know how consumers perceive marketing
initiatives.
First, the results indicate that personalization has a strong influence on
mobile apps usage. Therefore, users are willing to use apps if they can get the
relevant information just in time. To provide the relevant information, it is
necessary for providers to include well-designed personalization.
Second, personalization has a strong influence when users provide much of
their information to service providers. On the contrary, information privacy
concerns have a weak influence in the same condition. Our results imply that
The Influence of Personal Data Usage on Mobile Apps 111

mobile users are motivated to use apps if they realize that the apps use certain
amounts of personal information to provide them with valuable information.
Service providers and advertisers should consider the amount of information
collected from users to enhance their service usage.
Finally, the perception of privacy may have changed as time goes by.
Personal interests and location-based information are highly personalized
information. However, mobile users may give their personal information if they
can get valuable information. Service providers and advertisers need to know
what type of personal information is appropriate for personalization. To develop
effective mobile advertising, advertisers have to consider how much information
should be collected.
Also, there remains room for further study. This research model has been
tested within very limited situations. Further, the influence of privacy concerns
and the amount of data collection are not explained enough in this study. To
better understand mobile apps usage as an advertising media behavior, further
research should be conducted to test this model in different situations, such as
within other industries or for different occasions.

5 Acknowledgement

This work was supported by the Japan Academy of Advertising, grant for the
research project 2014.

6 References
Anderson, J. C., Gerbing, D. W. (1988). Structural equation modeling in practice: A
review and recommended 2-step approach, Psychological Bulletin, 103(3), 411-423.
Arora, N., Dreze, X., Ghose, A., Hess, J. D., Iyengar, R., Jing, B., et al. (2008). Putting
one-to-one marketing to work: Personalization, customization, and choice, Marketing
Letters, 19(3-4), 305-321.
Awad, N. F., Krishnan, M. S. (2006). The personalization privacy paradox: An empirical
evaluation of information transparency and the willingness to be profiled online for
personalization, Mis Quarterly, 30(1), 13-28.
Bagozzi, R. P., Yi, Y. J. (1989). On the use of structural equation models in
experimental-designs, Journal of Marketing Research, 26(3), 271-284.
Byrne, B. M. (2010). Structural equation modeling with AMOS: Basic concepts,
applications, and programming (2nd ed.), New York: Routledge Academic.
Ministry of Internal Affairs and Communications (2015). Survey of household economy,
Retrieved from http://www.stat.go.jp/info/guide/public/joukyou/pdf/net0306.pdf
Dabholkar, P. A., Bagozzi, R. P. (2002). An attitudinal model of technology-based self-
service: Moderating effects of consumer traits and situational factors, Journal of the
Academy of Marketing Science, 30(3), 184-201.
112 Hirose, Mineo, and Tabe

Davis, F. D., Bagozzi, R. P., Warshaw, P. R. (1989). User acceptance of computer-


technology: A comparison of 2 theoretical-models, Management Science, 35(8), 982-
1003.
Dickinger, A., Kleijnen, M. (2008). Coupons going wireless: Determinants of consumer
intentions to redeem mobile coupons, Journal of Interactive Marketing, 22(3), 23-39.
Fornell, C., Larcker, D. F. 1981. Evaluating structural equation models with
unobservable variables and measurement error. Journal of Marketing Research 18,
no.1, 39-50.
Ha, S., Stoel, L. (2009). Consumer e-shopping acceptance: Antecedents in a technology
acceptance model, Journal of Business Research, 62(5), 565-571.
Hair, J. F., Black, W. C., Babin, B. J., Anderson, R. E. (2009). Multivariate data analysis
(7th ed.), New Jersey: Prentice Hall.
Malhotra, N. K., Kim, S. S., Agarwal, J. (2004). Internet users' information privacy
concerns (IUIPC): The construct, the scale, and a causal model, Information Systems
Research, 15(4), 336-355.
Milberg, S. J., Smith, H. J., Burke, S. J. (2000). Information privacy: Corporate
management and national regulation, Organization Science, 11(1), 35-57.
Nysveen, H., Pedersen, P. E., Thorbjornsen, H. (2005). Intentions to use mobile services:
Antecedents and cross-service comparisons, Journal of the Academy of Marketing
Science, 33(3), 330-346.
Okazaki, S., Barwise, P. (2011). Has the time finally come for the medium of the future?:
Research on mobile advertising, Journal of Advertising Research, 51(1), 59-71.
Okazaki, S., Li, H. R., Hirose, M. (2009). Consumer privacy concerns and preference for
degree of regulatory control: A study of mobile advertising in Japan, Journal of
Advertising, 38(4), 63-77.
Perez, S. (2014). Mobile app usage increases in 2014, As Mobile Web Surfing Declines,
Retrieved from http://techcrunch.com/2014/04/01/mobile-app-usage-increases-in-
2014-as-mobile-web-surfing-declines/
Rose, E. A. (2006). An examination of the concern for information privacy in the New
Zealand regulatory context, Information & Management, 43(3), 322-335.
Sheehan, K. B., Hoy, M. G. 2000. Dimensions of privacy concern among online
consumers. Journal of Public Policy & Marketing 19, no.1, 62-73.
Smith, H. J., Dinev, T., Xu, H. (2011). Information privacy research: an interdisciplinary
review. Mis Quarterly 35, no.4, 989-1015.
Statista. (2015). Mobile advertising spending worldwide from 2010 to 2017 (in billion
U.S. dollars), Retrieved from http://www.statista.com/statistics/280640/mobile-
advertising-spending-worldwide/
Stewart, K. A., Segars, A. H. (2002). An empirical examination of the concern for
information privacy instrument, Information Systems Research, 13(1), 36-49.
Wolfinbarger, M., Gilly, M. C. (2003). eTailQ: Dimensionalizing, measuring and
predicting etail quality, Journal of Retailing, 79(3), 183-198.
Wu, J. H., Wang, S. C. (2005). What drives mobile commerce?: An empirical evaluation
of the revised technology acceptance model, Information & Management, 42(5), 719-
729.
Xu, H., Luo, X., Carroll, J. M., Rosson, M. B. (2011). The personalization privacy
paradox: An exploratory study of decision making process for location-aware
marketing, Decision Support Systems, 51(1), 42-52.
The Influence of Personal Data Usage on Mobile Apps 113

7 Appendix
Scenario

Scenario for poor data collection (PDC) apps


This mobile apps service is for short trips. Users can see the recommended
information about shopping and new topical spots. The information for short
trips includes an amusement park, a movie theater, and eating place such as
restaurants and bars.

Scenario for rich data collection (RDC) apps


This mobile apps service is for short trips. Users can see the recommended
information about shopping and new topical spots. The information for short
trips includes an amusement park, a movie theater, and eating place such as
restaurants and bars, which are selected based on your interests.
Furthermore, users can search for the information on nearest spots with GPS.
Part II. Consumer Responses to Advertising
The Effect of Eroticism in Couple Depictions in
Advertisements on Brand Evaluations

Stefan Thomas and Heribert Gierl

1 Introduction

Reichert and Carpenter (2004, p. 828) analyzed the content of mainstream


magazine ads in the U.S. and found for 1983 that 3.7% of all ads contained
motifs of couples in intimate or very intimate contact. In 2003, this portion was
7.5% and thus has doubled over the course of two decades. Obviously, in
practice, this type of advertising is widely used.

Figure 1: Sample of images of erotic couples in ads of the Calvin Klein brand

© Springer Fachmedien Wiesbaden 2017


G. Christodoulides et al. (Hrsg.), Advances in Advertising Research (Vol. VII),
European Advertising Academy, DOI 10.1007/978-3-658-15220-8_10
118 Thomas and Gierl

Especially, marketers of numerous apparel brands rely on this kind of market


communication (Reichert et al., 2012, p. 9). But also numerous companies
offering perfume, cosmetics, jewelry, and alcoholic beverages utilize this
strategy. As an example, we provide some ad versions for the Calvin Klein
brand containing erotic couple motifs in Figure 1. Contrarily to the use in
advertising practice, academic investigations of the effectiveness of this type of
ads for improving brand or product evaluations are very rare. We only found
four studies that tested the effect of depictions of different degrees of couple
eroticism depending on the recipients’ gender, and the results of these
investigations are contradictory. Thus, we contribute to research by providing
results about the effectiveness of ads containing motifs of couples in different
intensities of eroticism.

2 Theoretical Considerations

Sexual economics theory: Economic exchange theories are based on the


assumption that people give money to get goods and services that are at least as
valuable to them as the given money (i.e., the price paid). This economic
principle has been adopted for explaining all kinds of interpersonal relations:
Social exchange theories state that people give resources such as time,
information, friendship, and social recognition, if they receive adequate
resources from the exchange partner. A special variant of these theories, the
“sexual economics theory,” focuses on the exchange of sex-related resources
(Baumeister et al., 2001; Baumeister and Vohs, 2004). Its key proposition is that
“sex is the resource that women have and men want” (Vohs and Lasleta, 2008, p.
787) and, thus, sex is of greater worth for females compared to males. The
authors who advanced this view provided two arguments in favor of this
proposition. Vohs et al. (2014, p. 278) state that “sex is costlier for women
(biologically, physically, and socially).” For instance, women have higher
parental investment costs to raise children compared to men what makes
“women choosier.” As a second reason, the authors argue that the sex drive in
men is generally higher (Oliver and Hyde, 1993). As a consequence, these
authors (e.g., Vohs et al., 2014, p. 278) hypothesize that, “for a woman, sex
should take place when it is worth the risk, meaning that there ought to be
something that she gets of it (more than the sex itself). (…) Sex should be
accompanied by gains to the women.” Baumeister and Vohs (2004, p. 343)
count affection and respectful attention, the promise to share wealth and
earnings with the woman exclusively, a long series of compliments, material
gifts, and fancy dinners to the rewards men can offer as gains to the female
partner. Dahl et al. (2009) provided evidence to the sexual economics theory in
The Effect of Eroticism in Couple Depictions in Advertisements 119

the field of advertising. These researchers analyzed the effectiveness of an image


of an erotic couple on the attitudes toward an ad that promoted wristwatches for
females depending on whether the wristwatches were or were not advertised as a
gift from men to women. I.e., the erotic ad either contained or did not contain
the text “a gift from a man to the special woman in his life” and showed a red
ribbon around the watch. Females reported more favorable attitudes toward the
ad when the erotic couple motif was accompanied by these relationship cues
indicating gains for women.
Undoubtedly, using this kind of rational, economic-like, and cognition-based
perspective is insufficient for explaining interpersonal and even intimate
relationships between females and males that are primarily affect-based and,
thus, one may find this perspective embarrassing. However, it sheds light on the
key resources which are valued differently by female and male individuals who
are engaged in intimate relations in general: Women tend to appreciate mild
forms of intimate relations (warmth, tenderness, gentle embrace) from a
committed partner whereas men are prone to favor more intense sexual contacts
that do not necessarily involve a committed relationship.
Feelings in response to erotic couple depictions: In line with these
arguments, Vohs et al. (2014, p. 278) posit that erotic couple images shown in
advertisements are liked more if sex is depicted in a manner consistent with the
sexual values held by females and males. We presume that low-erotic or
moderately erotic depictions of couples, compared to highly erotic motifs, are
liked more by females because they evoke feelings about warmth, tenderness,
and affection. Moreover, we posit that highly erotic motifs of couples, compared
to low and moderately erotic motifs, cause pleasant feelings and desires in male
persons.
Spillover of feelings onto product evaluations: There are numerous theories
in literature that are used to explain the transfer of affect (which is, for instance,
associated with a peripheral cue in ads such as an erotic motif) to the target
object (e.g., the promoted product). First, a fundamental principle suggests that
people tend to reduce even pleasant emotions to achieve a state of homeostasis.
If individuals cannot respond to the cue that elicited the emotion (e.g., the erotic
cue itself), an easy possibility to regulate pleasant feelings is to respond
favorably to the product or brand that is associated with the cue. Second,
Reichert (2002, p. 258) posits that feelings act as “energizers.” Positive
energizers put individuals into the mental state of approaching to any stimulus
connected to the positive feelings (e.g., products and brands that are promoted
by the images). Third, from the perspective of neuroscience, the reward circuits
of the brain are activated due to appropriate erotic cues while the product is
presented. Thus, the product itself is liked to a higher extent. Fourth, individuals
could mistakenly consult their feelings for evaluating the product that in fact
120 Thomas and Gierl

resulted from the erotic cues (Baron, 1982, p. 429). In this case, more favorable
product evaluations result from a cognitive misinterpretation of the source of
the feelings. Fifth, it could be argued that individuals tend to preserve pleasant
affective mental states. When, for instance, an erotic motif has put the consumer
into a pleasant affective state s/he is expected to prevent herself/himself from an
intense processing of information about the advertised product which otherwise
could cause an end to this feeling (e.g., Isen et al., 1982, p. 246). Fredrickson
(2011) provides a similar argument. She posits that pleasant emotions motivate
people to “approach or continue consuming whatever stimulus is biologically
useful for them at the moment.” Thereby, individuals are less likely to generate
counterarguments against the product (Patzer, 1980, p. 363). We test:

H1: Compared to low and moderately erotic motifs of couples shown in


advertisements, brand attitudes of female consumers are less favorable when the
brand is associated with highly erotic motifs.
H2: Compared to low and moderately erotic motifs of couples shown in
advertisements, brand attitudes of male consumers are more favorable when the
brand is associated with highly erotic motifs.

3 Prior Research

We found four studies in prior research that investigated the effect of


different levels of eroticism of couples depicted in advertisements on the
attitudes of female and male consumers toward the promoted brands (Black et
al., 2010) or toward the advertisement (Belch et al., 1982; LaTour and
Henthorne, 1994; Pope et al., 2004), respectively.
The Effect of Eroticism in Couple Depictions in Advertisements 121

Table 1: Effects of couple eroticism on attitudes toward the brand and the advertisement
depending on consumer gender reported in prior research

Detailed information about these studies is contained in Table 1. As can be


seen in this table, the findings are mixed. Thus, we looked at the study details to
explore the reasons for the heterogeneous findings. Unfortunately, with the
exception of Black et al. (2010), the authors did only provide brief verbal
descriptions of the used test material what made it impossible to find reasons for
the mixed results. Thus, we focus on the findings of Black et al. (2010). These
authors illustrated the images used in their experiment; according to our
interpretation, they compared a low to a moderately erotic level (Figure 2). They
found that the moderate level (half-naked couple) deteriorated brand evaluations
compared to the low level of eroticism (fully dressed couple) for both consumer
genders and for a wide range of products. However, the negative effect might
have resulted from the specific mode they used to present the test stimuli: Each
test person had contact to four ad versions (an ad for body oil, perfume, USB-
stick, and muesli) that showed the same couple for each product. Watching the
same half-naked couple several times for different products might have caused
higher aversion against the experimental setting than viewing the same low-
erotic depictions repeatedly.
122 Thomas and Gierl

Figure 2: Test stimuli used for promoting massage oil in the study of Black et al. (2010)

Moreover, there are studies that compared different levels of eroticism


without providing data separately for male and female consumers (e.g., Reichert
et al., 2001; Huang, 2004; Severn et al., 1990) or that compared the
effectiveness of only one level of couple eroticism to different types of ad cues
(e.g., Mittal and Lassar, 2000; Sengupta and Dahl, 2008; Vohs et al. 2014). We
refrain from discussing these results. In summation, we found no results in
academic research that are suitable for answering the question about which level
of couple eroticism is most advantageous for influencing the evaluations of the
promoted product depending on the consumer gender.

4 New Experiments

4.1 Experiment 1

Pretest: We bought motifs from picture agencies showing the same couples
in different degrees of nudity or erotic poses, respectively. The pretest’s
experimental design was a 3 (degree of eroticism: low, moderate, or high) × 2
(source of eroticism: nudity or pose) × 2 (consumer gender) factorial between-
subjects design. A sample of 420 students (50% female, Mage = 22.96 years, SD
= 1.962) were assigned to the resulting twelve conditions. Each test person had
contact to one of the motifs and assessed the degree of eroticism by agreeing or
disagreeing with “This motif is strongly sexually-related” and “This motif is very
erotic” on a seven-point scale (α = .845). The images and the results are shown
in Figure 3. When the data were combined across the source of eroticism and the
consumers’ gender, the perceptions of eroticism are Mlow = 2.48, Mmoderate = 4.24,
and Mhigh = 5.69 (F2;417 = 291.445, p < .001) which indicates a successful
manipulation of eroticism. Similar findings resulted for the disaggregated data.
The Effect of Eroticism in Couple Depictions in Advertisements 123

Experimental design of the main study: The study was based on a 3 (degree
of eroticism: low, moderate, or high) × 2 (source of eroticism: nudity or pose) ×
2 (consumer gender) × 2 (products: deodorant and wristwatch) factorial
between-subjects design.
Test stimuli: We used the motifs selected in the pretest to create print
advertisements. The ads showed the couple on a sailing yacht and the promoted
product. For the case of the deodorant, we used an unknown brand name
(Malizia) and developed ad versions targeting female consumers (“Maliza for
her”) and ad versions targeting male consumers (“Malizia for him”). For the
wristwatch, we chose a fictitious brand name (Vade) and included a “female”
and a “male” watch into the same ad. We illustrate the test stimuli for the
wristwatch in Figure 4.
Sample, procedure, and measures: Data were collected in face-to-face
interviews at universities located in Germany. Overall, 756 students (51.6%
female, Mage = 22.53 years, SD = 2.458) participated in the main experiment.
They were asked to watch one ad version as long as they wanted. Then, they
indicated the degree to which they evaluated the advertised product as
“appealing,” “attractive,” likeable,” and “good” on a seven-point scale (α =
.855).
Results: In Table 2, we present the findings for the brand attitudes depending
on the level of eroticism, source of eroticism, product, and consumer gender. For
male consumers, brand attitudes increased with the level of eroticism
independently of whether the couple’s nudity or the couple’s pose was used to
manipulate eroticism and independently of the promoted product. For female
consumers, inverted-U-shaped relations were observed. Thus, we collapsed the
data across these conditions. For female consumers, we observed the following
brand attitudes depending on the level of eroticism: Mlow = 3.43, Mmoderate = 4.15,
Mhigh = 3.13 (F2;387 = 24.139, p < .001). In H1, we postulated that brand
evaluations are lower in the high-eroticism condition compared to the low-
eroticism condition (3.13 < 3.43, t258 = -1.937, p < .05) and compared to the
moderate-eroticism condition (3.13 < 4.15, t258 = -6.762, p < .001). Our findings
support H1. For male consumers, the respective mean values are Mlow = 3.14,
Mmoderate = 3.54, and Mhigh = 3.89 (F2;363 = 9.961, p < .001). In H2, we presumed
that the brand attitudes are higher in the high-eroticism condition compared to
the low-eroticism condition (3.89 > 3.14, t244 = 4.340, p < .001) and compared
to the moderate-eroticism condition (3.89 > 3.54, t239 = 1.989, p < .05) which
was confirmed.
124 Thomas and Gierl

Figure 3: Perceptions of eroticism associated with the motifs used in Experiment 1


The Effect of Eroticism in Couple Depictions in Advertisements 125

Figure 4: Test stimuli for the wristwatch used in Experiment 1


126 Thomas and Gierl

Table 2: Brand attitudes depending on the level and source of eroticism, consumer
gender, and product (Experiment 1)

4.2 Experiment 2
In Experiment 2, we additionally considered product categories (cookies and
pain pills) where erotic advertising is less common and used real brands (Armani
perfume, Bahlsen cookies, and Stada pain pills). We focused on moderate and
high levels of eroticism expressed by the couple’s pose.
Pretest: The pretest’s experimental design was a 2 (degree of eroticism:
moderate or high) × 2 (consumer gender) factorial between-subjects design. A
sample of 124 students (48.4% female, Mage = 23.44 years, SD = 1.375) were
assigned to the two levels of eroticism and assessed the degree of eroticism. The
items were adopted from Experiment 1 (α = .949).

Figure 5: Perceptions of eroticism associated with the motifs used in Experiment 2


The Effect of Eroticism in Couple Depictions in Advertisements 127

The motifs and the results are contained in Figure 5. Combined across the
consumer gender, the perceptions of eroticism are Mmoderate = 4.02 and Mhigh =
6.14 (F1;122 = 97.739, p < .001). Overall, the pretest shows that we successfully
manipulated the level of eroticism.
Experimental design of the main study: The experiment was based on a 2
(degree of eroticism: moderate or high) × 2 (consumer gender) × 3 (products:
perfume, cookies, and pain pills) factorial between-subjects design.
Test stimuli: Together with depictions of the promoted products, we included
the motifs of the couples into print advertisements. To reduce irritations due to
the use of products that are less frequently combined with erotic motifs, we
additionally included product claims (for the perfume: “a new direction of sense:
Armani mania”, for the cookies: “scrumptious”, and for the pain pills: “for a
clear head”). In the case of the perfume, the ad versions targeting female
consumers showed a perfume “for women,” and the versions targeting male
consumers contained the information “for men.” Figure 6 shows how the test
stimuli looked like.

Figure 6: Test stimuli used in Experiment 2

Sample, procedure, and measures: In total, 537 students (50.3% female,


Mage = 23.61 years, SD = 2.850) took part in the main experiment. The
procedure and the measures for assessing brand attitudes (α = .886) were
adopted from Experiment 1.
128 Thomas and Gierl

Results: In Table 3, we summarize the findings for the brand attitudes


depending on the level of eroticism and the product. On the aggregate level, we
found for female consumers, that the brand attitudes were lower in the high- than
in the moderate-eroticism condition (Mmoderate = 4.98 and Mhigh = 4.54, t268 = -
3.401, p < .001) which is in line with H1. For male consumers, brand attitudes
were higher in the high- than in the moderate-eroticism condition (Mmoderate =
4.32 and Mhigh = 4.98, t265 = 4.529, p < .001) which supports H2. Figure 7
illustrates that Experiment 2 replicated the findings of Experiment 1.

Figure 7: Summary of the results of Experiment 1 and Experiment 2


The Effect of Eroticism in Couple Depictions in Advertisements 129

Table 3: Brand attitudes depending on eroticism, gender, and product (Experiment 2)

4.3 Experiment 3

We replicated Experiment 2 but aimed at limiting attention to the erotic


advertisement. We included the test stimulus (an ad for the om/one Bluetooth
speaker set) in the middle of a sequence of additional seven advertisements (e.g.,
for Sixt car rental, Pedigree dog food, and Sensodyne toothpaste) that were
constant across both experimental conditions. Each of the eight ads could be
viewed for seven seconds within a power-point presentation. After the
presentation of all ads, all products including the test product had to be
evaluated (“makes me curious,” “appealing,” “interesting,” and “would buy”) on
a seven-point scale (α = .929). Then, the test ad was shown again and the test
persons (60 female and 60 male students) had to rate the ad’s level of eroticism;
the measures were adopted from Experiment 1. The manipulation check of the
ad’s eroticism proved to be successful (t118 = 7.101, p < .001). Although
distracted by other advertisements, the level of eroticism of the test ad improved
product evaluations among male consumers (t58 = 2.083, p < .05) and
deteriorated evaluations among female consumers (t58 = 1.266, p = .105). Thus,
we conclude that couple eroticism also affects product evaluations under more
realistic conditions (see Table 4).
130 Thomas and Gierl

Table 4: Brand attitudes depending on eroticism and consumer gender (Experiment 3)

5 Implications for Practice

Our results indicate that high levels of eroticism in depictions of couples are
disadvantageous if female consumers are targeted. When this finding is
compared to the advertising practice of some famous brands (e.g., Calvin Klein,
Abercrombie & Fitch, Gucci, and Dolce and Gabbana), we conclude that these
marketers did not choose the optimum level of eroticism. However, we do not
recommend to reposition these well-known brands because repositioning (e.g.,
reducing the current level of eroticism) would confuse consumers. Instead, we
recommend avoiding the use of highly erotic couple depictions when brands
enter a market consisting of female consumers. For male consumers, we found
that high levels of eroticism are advantageous.
We found these effects independently of whether products with a high fit to
eroticism (deodorant, wristwatch, and perfume) or a low fit to eroticism
(cookies, pain pills, and speaker set) were promoted. Thus, couple eroticism
could also be used for promoting products which are normally not associated
with eroticism. However, marketers should pay attention to the importance of
using cues that are unique for their brands. Thus, simply relying on the
effectiveness of eroticism goes along with the danger that the brand’s
positioning is unclear and that the brand becomes interchangeable. Moreover,
practitioners should consider the aspect that highly erotic advertising may be
criticized in media reports and in Internet communities. Thus, this kind of
response may have detrimental effects for the promoted brand.
The Effect of Eroticism in Couple Depictions in Advertisements 131

6 References

Baron, R. S. (1982), “Sexual Content and Advertising Effectiveness: Comments on


Belch et al. (1981) and Caccavale et al. (1981),” in: A. Mitchell (ed.), Advances in
Consumer Research (Vol. 9), 428-430. Ann Arbor, MI: Association for Consumer
Research.
Baumeister, R. F. and K. D. Vohs (2004), “Sexual Economics Theory: Sex as a Female
Resource for Social Exchange in Heterosexual Interactions,” in: Personality and
Social Psychology Review, 8 (4), 339-363.
Baumeister, R. F.; Catanese, K. R. and K. D. Vohs (2001), “Is There a Gender
Difference in Strength of Sex Drive? Theoretical Views, Conceptual Distinctions,
and a Review of Relevant Evidence,” in: Personality and Social Psychology Review,
5 (3), 242-273.
Belch, M. A.; Holgerson, B. E.; Belch, G. E. and J. Koppman (1982),
“Psychophysiological and Cognitive Responses to Sex in Advertising,” in: A.
Mitchell (ed.), Advances in Consumer Research (Vol. 9), Ann Arbor, MI:
Association for Consumer Research, 424-427.
Black, I. R.; Organ, G. C. and P. Morton (2010), “The Effect of Personality on Response
to Sexual Appeals,” in: European Journal of Marketing, 44 (9), 1453-1477.
Dahl, D.; Sengupta, J. and K. D. Vohs (2009), “Sex in Advertising: Gender Differences
and the Role of Relationship Commitment,” in: Journal of Consumer Research, 36
(2), 215-231.
Fredrickson, B. L. (2011), “The Broaden-and-Build Theory of Positive Emotions,” in:
American Psychologist, 53 (3), 218-226.
Huang, M.-H. (2004), “Romantic Love and Sex: Their Relationship and Impacts on Ad
Attitudes,” in: Psychology & Marketing, 21 (1), 53-73.
Isen, A. M.; Means, B.; Patrick, R. and G. P. Nowicki (1982), “Some Factors Influencing
Decision-Making Strategy and Risk Taking,” in: M. S. Clark and S. T. Fiske (eds.),
Affect and Cognition: The 17th Annual Carnegie Symposium on Cognition,
Hillsdale, NJ: Erlbaum, 243-261.
LaTour, M. S. and T. L. Henthorne (1994), “Ethical Judgments of Sexual Appeals in
Print Advertising,” in: Journal of Advertising, Vol. 23 (3), 81-90.
Mittal, B. and W. M. Lassar (2000), “Sexual Liberalism as a Determinant of Consumer
Response to Sex in Advertising,” in: Journal of Business and Psychology, 15 (1),
111-127.
Oliver, N. B. and J. S. Hyde (1993), “Gender Differences in Sexuality: A Meta-
Analysis,” in: Psychological Bulletin, 114 (1), 29-51.
Patzer, G. L. (1980), “A Comparison of Advertisement Effects: Sexy Female
Communicator vs. Non-sexy Female Communicator,” in: J. C. Olson (ed.), Advances
in Consumer Research (Vol. 7), Ann Arbor, MI: Association for Consumer
Research, 359–364.
Pope, N. K. Ll.; K. E. Voges, K. E. and M. R. Brown (2004), “The Effect of Provocation
in the Form of Mild Erotica on Attitude to the Ad and Corporate Image: Differences
between Cause-Related and Product-Based Advertising,” in: Journal of Advertising,
33 (1), 69-82.
132 Thomas and Gierl

Reichert, T. (2002), “Sex in Advertising Research: A Review of Content, Effects, and


Functions of Sexual Information in Consumer Advertising,” in: Annual Review of
Sex Research (Vol. 13), 241-273.
Reichert, T. and C. Carpenter (2004), “An Update on Sex in Magazine Advertising: 1983
to 2003,” in: Journalism & Mass Communication Quarterly, 81 (4), 823-837.
Reichert, T.; Carpenter, C. and L. N. Reid (2012), “How Sex in Advertising Varies by
Product Category: An Analysis of Three Decades of Visual Sexual Imagery in
Magazine Advertising,” in: Journal of Current Issues and Research in Advertising,
33 (1), 1-19.
Reichert, T.; Heckler, S. E. and S. Jackson (2001), “The Effects of Sexual Social
Marketing Appeals on Cognitive Processing and Persuasion,” in: Journal of
Advertising, 30 (1), 13-27.
Sengupta, J. and D. W. Dahl (2008), “Gender-Related Reactions to Gratuitous Sex
Appeals in Advertising,” in: Journal of Consumer Psychology, 18 (1), 62-78.
Severn, J.; Belch, G. E. and M. A. Belch (1990), “The Effects of Sexual and Non-Sexual
Advertising Appeals and Information Level on Cognitive Processing and
Communication Effectiveness,” in: Journal of Advertising, 19 (1), 14-22.
Vohs, K. D. and J. Lasaleta (2008), “Heterosexual Sexual Behavior Is Governed by
Social Exchange and Basic Economic Principles: Sexual Economics Theory,” in:
Minnesota Journal of Law, Science, and Technology, 9 (2), 785-802.
Vohs, K. D.; Sengupta, J. and D. W. Dahl (2014), “The Price Had Better Be Right:
Women’s Reactions to Sexual Stimuli Vary with Market Factors,” in: Psychological
Science, 25 (1), 278-283.
Model’s Age and Target’s Age: Effects on Emotions towards
and Beliefs about an Ad
Corinne Chevalier and Marie-Christine Lichtlé

1 Introduction

Most ads use models, whether famous or unknown, who are an attractive
feature playing the part of consumers or in some instances offering expert advice
(Friedman and Friedman, 1979). While the message put across by an ad is
important, features such as the physical characteristics of the models also have a
non-negligible influence on the process of persuasion, and research in this area
is expanding (Bisseil and Rask, 2010). Advertisers often choose young models,
but there are examples to show that older models are preferable, even if the
target audience is made up in part of young consumers.
This paper looks at how targets react to models of various ages in ads. In
particular, the effect of the model’s perceived age on emotions towards and
beliefs about an ad is ascertained for young people and seniors. In terms of
theory, the paper contributes to research into the process by which an ad design
variable - the model’s perceived age - influences targets. Age is held to be one of
the most relevant criteria to allow for when evaluating a model (Milliman and
Erffmeyer, 1989-1990), but little research has taken this variable into account. In
managerial terms, the point of the research is to identify the conditions in which
emotions toward and beliefs about the ad will be enhanced, depending on the
perceived age of the model.

2 The Conceptual Framework: The Model’s Perceived Age and the


Target’s Age

2.1 The Effect of the Perceived Age of the Model on the Variables of the
Advertising Persuasion Process

If the use of models in advertising has been studied for many years (Grier
and Deshpandé, 2001; Mastro and Stern, 2003), little research has been
conducted into the specific effects of the perceived age of models as part of the
persuasion process. A few dependent variables have been examined:
- perceived attractiveness: generally young models are found more attractive
regardless of the respondent’s age (Vermeir and Sompel, 2014);
- model credibility: young targets find young models more credible; the
opposite is true for seniors (Milliman and Erffmeyer, 1989-1990);

© Springer Fachmedien Wiesbaden 2017


G. Christodoulides et al. (Hrsg.), Advances in Advertising Research (Vol. VII),
European Advertising Academy, DOI 10.1007/978-3-658-15220-8_11
134 Chevalier and Lichtlé

- beliefs about the product: these are more positive when the model’s age is
consistent with the target’s age (Rotfeld et al., 1982);
- attitude towards the brand: for seniors, a mature model prompts a more
favorable attitude than a young or elderly model (Bristol, 1996);
- intention to purchase: the perceived age of the model comes into play when
the product targets a specific age group. In this event, the intention to buy of
seniors is greater when the model is elderly (Nelson and Smith, 1988; Konzar,
2010). However, no effect of the model’s perceived age is observed when the
product is intended for the population as a whole (Nelson and Smith, 1988;
Greco et al., 1997).

2.2 The Role of Congruency

For Mandler (1982), congruency expresses the fact that two entities go well
together. In the case of ‘personal congruency’, one of those entities is the
consumer, who compares the entity presented with her own schemas or personal
requirements, her emotional state, or with her self-image, or the image she would
like to project. Perceived congruency with the model may enhance the ad’s
persuasive power (Petty and Cacioppo, 1986). In the case of models used in ads,
such congruency may be achieved by presenting people of the same ethnic origin
(Whittler and Spira, 2002) or the same sex (Petroschius and Crocker, 1989) as
the target.
More especially, with the perceived age of the model, it is important to take
the target’s age into account. When the product is intended for a specific age
group, young people and seniors think that the model’s age must be congruent
with the product (Rotfeld et al., 1982; Chang, 2008). Conversely, when the
product is intended for all age groups, young adults and seniors do not consider
it essential to adjust the age of the model to the target (Greco et al., 1997).
Looking more specifically at seniors as targets, they are predisposed to judge
their degree of congruency with the model (Nelson and Smith, 1988; Kozar and
Damhorst, 2008). It is important to select the model with which the elderly will
identify most (Kozar and Damhorst, 2008; Kozar, 2010). The use of older
models may make the message more credible (Milliman and Erffmeyer, 1989-
1990) and more persuasive (Nelson and Smith, 1988; Bristol, 1996; Kozar and
Damhorst, 2008; Kozar, 2010). Conversely, showing an atypical or stereotyped
old person may make the ad wholly unbelievable and drive people away from
the product or service advertised (Festervand and Lumpkin, 1985).

2.3 The Target’s Subjective Age

In studying behavioral models for seniors, some investigators prefer to use


cognitive or subjective age instead of the chronological age of respondents
Model’s Age and Target’s Age 135

(Barak and Schiffman, 1981; Markides and Boldt, 1983). This concept of
subjective age, also called ‘personal age’ (Kastenbaum et al., 1972) or
‘perceived age’ (Linn and Hunter, 1979), has been used in marketing by Barak
and Schiffman (1981) to characterize the age people think they are in terms of
‘sentiments’, physical appearance, activities and centres of interest. Seniors
commonly see themselves as younger than their chronological age (by 10 or 15
years on average) and are therefore predisposed to identify their feelings and
their actions with a younger age group than their true age group (Barak and
Stern, 1985; Leventhal, 1997).
In practice, two separate categories of seniors can be identified in terms of
their subjective age (Underhill and Cadwell, 1983; Barak and Stern, 1985):
subjectively young seniors (who consider themselves 10 to 15 years younger
than they actually are) and objective seniors (whose subjective age is the same as
their chronological age). Subjective age influences seniors’ purchasing behavior:
subjectively young seniors tend to consume many products in agreement with
their subjective age. This suggests that people’s attitudes and behaviors may
depend as much on their perceived age as on their chronological age (Smith and
Moschis, 1984; Moschis and Mathur, 2006).
Which are the results in advertising? Smith and Moschis (1984) argue that
attitudes to ads are closely related to seniors’ subjective ages. Subjectively
young seniors are an attractive target for advertisers, above all for age-related
products (Barak and Gould, 1985).
The impact of the age of the model in an ad therefore varies with the target’s
chronological and subjective age (Nelson and Smith, 1988). Seniors who
consider themselves younger than they actually are have a predisposition to
perceive congruency with younger models, that is, models who are
representative of their subjective age. Conversely, seniors who consider they are
their chronological age will perceive greater similarities with an elderly model
(Nelson and Smith, 1988). If respondents think they perceive congruency
between the model and themselves, they will pay more attention to the use of the
model in the ad, the product used, the arguments about the product, and they will
have more favorable attitudes towards the ad.

3 Research Hypotheses

In this research, we look into the effect of the model’s perceived age on the
emotions elicited by the ad and on beliefs about it.
There are a variety of definitions of emotions (Batra and Ray, 1986;
Chamberlain and Broderick, 2007). According to Dubé and Menon (2000),
emotions can be defined as a complex set of interactions among subjective and
objective factors giving rise to affective experiences. In order to measure each of
these processes, different measures exist. According to Mehrabian and Russell
136 Chevalier and Lichtlé

(1974), any emotional response can been represented in a space with three
independent and bipolar dimensions: pleasure which refers to a positive state
and reflects the person’s degree of well-being and satisfaction; arousal, the
physiological dimension characterizing an organism’s physical activity or mental
alertness, which has two extremes of sleep and overexcitement; and finally
dominance, which refers to the sensation of power, control or influence as
opposed to an inability to control or influence a situation. Like other studies
(Olney et al., 1991; Lichtlé, 2007), we omit the third dimension (dominance) as
it does not seem to influence attitudes toward advertising (Olney et al., 1991).
Beliefs about an ad are taken to be ‘recipients’ perceptions of the ad itself
(Lutz et al., 1983). Such beliefs are broken down in accordance with the
advertising feature to which they relate: beliefs about the advertising message
and beliefs about advertising design.
As seen before, the literature has shown that the model’s perceived age has
an effect on perceived attractiveness, credibility, attitude toward the brand,
beliefs about the brand and intention to purchase. The present research seeks to
ascertain the effect of the model’s age on emotions toward and beliefs about the
ad.
Looking first at subjectively young seniors, they tend to identify with a
younger age group (Underhill and Cadwell, 1983; Barak and Stern, 1985). It
might therefore be thought that they are not convinced by ads using elderly
models. Chevalier (2010) reported a positive effect of congruency of
subjectively young seniors with a model on beliefs about the product, attitudes
toward the ad and the brand, and on intention to purchase. We think that the
effect on emotions toward and beliefs about the ad is similar: these should
therefore be more favorable when the model is young or mature (Hypotheses
H1). The following hypotheses can therefore be inferred.

H1: For subjectively young seniors, ads with young or mature models are more
effective.

H1a: Subjectively young seniors get more pleasure from ads with young or
mature models.
H1b: Subjectively young seniors are more aroused by ads with young or mature
models.
H1c: Subjectively young seniors have greater belief in the message of ads
including young or mature models.
H1d: Subjectively young seniors have greater belief in the design of ads with
young or mature models.

Subjectively young seniors apart, we give precedence to the hypothesis of


congruency and think that emotions felt and beliefs toward ads are more positive
Model’s Age and Target’s Age 137

if the model’s age corresponds to the target’s age (Hypotheses H2). Several
scholars have shown the importance of such congruency, especially for the
credibility of the model (Milliman and Erffmeyer, 1989-1990), and beliefs about
the product (Rotfeld et al., 1982). The following hypotheses can therefore be
framed.

H2: For young people and objective seniors, emotions toward and beliefs about
ads are more positive when there is congruency between the model’s age and the
target’s age.

H2a: Young people have more positive emotions toward and beliefs about ads
when the models are young:
H2.a1: they get more pleasure from ads with young models.
H2.a2: they are more aroused by ads with young models.
H2.a3: they have more positive beliefs about the message of ads with young
models.
H2.a4: they have more positive beliefs about the design of ads with young
models.

H2b: Objective seniors have more positive emotions and beliefs about ads when
the models are elderly:
H2.b1: they get more pleasure from ads with elderly models.
H2.b2: they are more aroused by ads with elderly models.
H2.b3: they have more positive beliefs about the message of ads with elderly
models.
H2.b4: they have more positive beliefs about the design of ads with elderly
models.

4 Method

Two products that could be purchased by individuals of various ages were


chosen: mineral water and coffee. Three print versions of each ad were
produced, the only changes were the different aged models: a young model, a
mature model and an elderly model (Appendix 1 and 2).
So as to have sufficient variance in their individual characteristics, two
groups of respondents were selected: young adults (20-35 age range) and seniors
(60-75 age range). The sample comprised 194 men and 155 women, 146 young
people and 203 seniors. The seniors were subdivided into two categories by
subjective age:
- subjectively young seniors are those for whom the difference between their
true age and their subjective age is 10 years or more. These seniors had a
subjective age of 50-65 years;
138 Chevalier and Lichtlé

- objective seniors are those for whom the difference between their true age
and their subjective age is less than 10 years. The subjective age of these seniors
is much the same as their true age (60-75 years).
In order to measure the subjective age of seniors, we used Barak and
Schiffman’s scale (1981). This scale is constructed from the average of four
components: feel-age, the age people feel they are; look-age, the age people look
to be; do-age, the age reflected by what people do; and interest-age, the age
corresponding to people’s centres of interest.
To measure emotions elicited by the ad, we used a verbal scale: the PAD
(Pleasure, Arousal, Dominance) scale of Mehrabian and Russell (1974).
Factorial analysis allowed two dimensions to be selected. The first, pleasure,
includes the items (7-point differential semantic scale): happy/unhappy,
content/discontented, dissatisfied/satisfied, joyful/sad (α = 0.95). The second,
arousal, also comprised four items: stimulated/relaxed, irritated/calm,
energetic/lifeless, not awake/awake (α = 0.89).
The scale developed by Derbaix (1995) and inspired by Miniard et al.’s
research (1990) was used to measure beliefs about the ad: the arguments of the
message are measured against bipolar scales and relate to their power of
persuasion, their clarity, etc. and the design elements are rated using scales in the
same format with respect to model, colours, atmosphere, etc. Factorial analysis
allowed two dimensions to be selected: beliefs about the message (α = 0.92) and
beliefs about design (α = 0.86).

5 Results

The set of hypotheses H1 relate to subjectively young seniors. To test the


effect of the age of the model, as we used experimentation, analyses of variance
were made (Table 1).

Table 1: Analysis of variance for H1


Pleasure Arousal Beliefs Beliefs
about the about the
message design
F (p-value)
Main effects
Model 10.50 9.96 9.99 33.00
(.000) (.000) (.000) (.000)
Numbers in bold indicate the variable has a significant effect.

For pleasure in the ad, the results show a significant main effect of the model
used in the ad (F = 10.50, p = .000): the results confirm that subjectively young
Model’s Age and Target’s Age 139

seniors get more pleasure from a mature model. The young model comes in
second position (Figures 1). Hypothesis H1a is therefore confirmed.
The effects of the same variables on arousal at the ads were then ascertained.
Once again, as Table 1 shows, the model has a significant influence (F = 9.96, p
= .000): the mature model is more arousing (Figures 1). Hypothesis H1b is
therefore confirmed.
The influence of the same explanatory variables about beliefs (about the
message and then the design) was then ascertained. With respect to beliefs about
the message, the results show a preference for the mature model (F = 9.99, p =
.000): hypothesis H1c, about greater belief in the message when the model is
young or mature, is therefore confirmed (Figures 1). Lastly, with respect to
design, the results reveal a main effect of the model (F = 33.00, p = .000).
Hypothesis H1d is confirmed: the mature model is preferred (Figures 1).
140 Chevalier and Lichtlé
Model’s Age and Target’s Age 141

Figures 1: Effects of model on emotions toward and beliefs about the ad for subjectively
young seniors

Analyses of variance including the model and target age were made for the
sample of young people and objective seniors (Hypotheses H2) (Table 2).
142 Chevalier and Lichtlé

Table 2: Analysis of variance for H2


Pleasure Arousal Beliefs Beliefs
about the about the
message design
F (p-value)
Main effects
Model 2.26 3.03 6.49 5.85
(.105) (.049) (.002) (.003)
Respondent 3.49 .57 11.37 2.88
age (.062) (.449) (.001) (.090)
Two-way interaction effects
Model x 6.54 10.54 2.37 3.06
Respondent age (.002) (.000) (.095) (.048)
Numbers in bold indicate the variable has a significant effect.

For pleasure, the results show an interaction effect between the model’s age
and the target’s age (F = 6.54, p = .002): the elderly model elicits more pleasure
for elderly respondents; conversely, young people get more pleasure from an ad
with a young model (Figures 2). Hypotheses H2a1 and H2b1 are therefore
confirmed.
For arousal, the results shown in Table 2 confirm the interaction between the
model’s age and the target’s age (F = 10.54, p = .000): young people are aroused
more by ads with young models, while seniors are aroused more by ads with an
elderly model (Figures 2). Hypotheses H2a2 and H2b2 are therefore confirmed.
Lastly, the influence of the same variables on beliefs (about the message and
design) was tested. For beliefs about the message, the interaction between the
model’s age and the respondent’s age is not significant: hypotheses H2a3 and
H2b3 are therefore not confirmed. For beliefs about design, the results confirm
an interaction effect between the target’s age and the model’s age (F = 3.06, p =
.048): young people’s beliefs about design are greater when the model is young,
while an elderly model elicits more favorable beliefs about design among
objective seniors (Figures 2). Hypotheses H2a4 and H2b4 are therefore
validated.
Model’s Age and Target’s Age 143
144 Chevalier and Lichtlé

Figures 2: Effects of model on emotions toward and beliefs about the ad for young
people and objective seniors

6 Discussion, Limitations and Managerial Implications

The results show an effect of the model used in an ad on the emotions


elicited by and beliefs about the ad. As expected, a different effect is observed
for young people, subjectively young seniors and objective seniors.
For subjectively young seniors, all the ads with seniors have a negative effect
on their emotions and beliefs. Their pleasure and arousal are greater with young
or mature models. Likewise, beliefs about the ad, whether the message or the
design, will be more favorable if the model used is young or mature. In any
event an elderly model should be avoided. This category of seniors refuses to
identify with an elderly model, considering themselves to be younger. It may be
their fear of ageing that leads these respondents to reject the elderly model. The
literature had already shown congruency of subjectively younger seniors with a
mature model when examining the effect of the model’s age on beliefs about the
product, attitude towards the ad and the brand, and intention to purchase
(Chevalier, 2010). Those results are confirmed here for emotions toward the ad
and beliefs about it.
For young people and objective seniors, the results show, for the most part,
the need to ensure congruency between the model’s age and the target’s age:
Model’s Age and Target’s Age 145

young people feel more emotion and have greater belief about the design when
the model is young; and for objective seniors, emotions and beliefs about the
design are greater with an elderly model. The importance of the hypothesis of
congruency shown up in particular by Milliman and Erffmeyer (1989–1990) or
Greco et al. (1997) is therefore confirmed. However, such congruency is not
essential for improving beliefs about the message. The power of persuasion
(belief) may correspond to a greater extent with the contents of the message of
the ad, which did not vary between the groups.
There are multiple theoretical contributions from this research. First, the
effect of the age of a model used in an ad on the effectiveness of the ad is a
variable that has been little studied to date. This research shows that its influence
is considerable and differs with the target’s age. In fact, a young model is better
for a young target; when the target is old, it is recommended to avoid an elderly
model. Moreover, the present results do bring out the crucial role played by the
subjective age of seniors. Consideration of both true age and subjective age
provide greater insight into the reactions of elderly consumers toward ads.
At methodological level, a distinction was made between beliefs about the
message and about design, which has not always been the case in previous
research. In this work, care was taken that the sample of young adults was not
made up essentially of students but also of people in employment. This enhances
its external validity and means the results can be more readily generalized.
Managerial implications arise from these results on various levels. The
results will make it possible to improve and guide the creative process of
advertising through better knowledge of the reactions of young people and
seniors toward ads with models of different ages. The results have shown the
effect of the perceived age of the model: the emotions elicited by the ad and the
beliefs about it. Consequently, advertisers must consider this variable in order to
produce positive emotions toward and beliefs about their ads. Moreover, they
will also have to consider the subjective age of seniors in order to reach them
effectively. They will have to use models of different ages in their ads if they
wish to communicate with subjectively young and objective seniors: for
example, the model chosen has to be congruent with the subjective age of the
seniors. Thus, our results lead to the conclusion, in the case of products
consumed by all age brackets, that advertisers must at all costs avoid using an
elderly model when addressing subjectively young seniors. A mature or young
model is preferable in this case. Conversely, congruency between the model’s
age and the target’s age must be given precedence for young and objective
senior targets.
This research has limits that point to directions for future research. First, only
the emotions elicited by the ad and beliefs about it have been taken into account.
It would be worth checking the influence of the model on other variables such as
attitude towards the ad, attitude toward the brand and beliefs about it. Secondly,
146 Chevalier and Lichtlé

only young people and seniors have been considered here. It would be
interesting to ascertain the age of models for the middle aged or for subjectively
elderly seniors. Lastly, it would be interesting to reiterate this research with a
larger number of ads and above all ads with more varied characteristics. For
example, an ad with emotional content and a more informative one might be
tested with the assumption that the former would elicit more emotion from
targets and the second greater belief.
Given the scarcity of investigations into the age of models in ads, many
avenues of enquiry are opened up by which to further our understanding of the
part this variable plays in persuasive advertising.

7 References

Barak, B., Gould, S. (1985). Alternative age measures: a research agenda, Advances in
Consumer Research, 12, 53-58.
Barak, B., Schiffman, L.G. (1981). Cognitive age: a nonchronological age variable,
Advances in Consumer Research, 8, 602-606.
Barak, B., Stern, B. (1985). Fantastic at forty! The new young woman consumer, Journal
of Consumer Marketing, 2, 41-54.
Batra, R., Ray, M.L. (1986). Affective responses mediating acceptance of advertising,
Journal of Consumer Research, 13, 234-249.
Bisseil, K., Rask, A. (2010). Real women on real beauty: self-discrepancy, internalization
of the thin ideal, and perceptions of attractiveness and thinness in Dove's
Campaign for Real Beauty, International Journal of Advertising, 29 (4), 643-668.
Bristol, T. (1996). Persuading senior adults: the influence of endorser age on brand
attitudes, Journal of Current Issues and Research in Advertising, 18(2), 59-67.
Chamberlain, L., Broderick, A.J. (2007). The application of physiological observation
methods to emotion research, Qualitative Research: an International Journal, 10 (2),
199-216.
Chang, C. (2008), Chronological age versus cognitive age for younger consumers,
Journal of Advertising, 37 (3), 19-32.
Chevalier, C. (2010). L’influence de l’âge perçu du personnage inclus dans une annonce
sur les variables d’efficacité publicitaire, Conference of the French Marketing
Association, Le Mans.
Derbaix, C.M. (1995). The impact of affective reactions on attitudes toward the
advertisement and the brand: a step toward ecological validity, Journal of Marketing
Research, 32, 470-79.
Dubé, L., Menon K. (2000). Multiple roles of consumption emotions in post-purchase
satisfaction with extended service transactions, International Journal of Service
Industry Management, 11 (3), 287-304.
Festervand, T.A., Lumpkin, J.R. (1985). Response of elderly consumers to their portrayal
by advertisers, Current Issues and Research in Advertising, 1, 203-226.
Friedman, H.H., Friedman, L. (1979). Endorser effectiveness by product type, Journal of
Advertising Research, 19 (5), 63-71.
Model’s Age and Target’s Age 147

Greco, A.J., Swayne, L.E., Johnson, E.B. (1997). Will older models turn off shoppers?,
International Journal of Advertising, 16 (1), 27-36.
Grier, S.A., Deshpandé, R. (2001). Social dimensions of consumer distinctiveness: the
influence of social status on group identity and advertising persuasion, Journal of
Marketing Research, 38 (2), 216-224.
Kastenbaum, R., Derbin, V., Sabatini, P., Artt, S. (1972). The ages of me: toward
personal and interpersonal definitions of functional aging, Aging and Human
Development, 3 (2), 197-211.
Kozar, J.M. (2010). Women's responses to fashion media images: a study of female
consumers aged 30–59, International Journal of Consumer Studies, 34 (3), 272-278.
Kozar, J.M., Damhorst, M.L. (2008). Older women's responses to current fashion
models, Journal of Fashion Marketing and Management: An International Journal,
12 (3), 338- 350.
Leventhal, R.C. (1997). Aging consumers and their effects on the marketplace, Journal
of Consumer Marketing, 14 (4), 276-281.
Linn, M.W., Hunter, K. (1979). Perception of age in the elderly, Journal of Gerontology,
34(1), 46-52.
Lichtlé, M.C. (2007). The effect of an advertisement’s color on emotions evoked by an
ad and attitude toward the ad: the moderating role of the optimal stimulation level,
International Journal of Advertising, 26 (1), 32-62.
Lutz, R.J., MacKenzie, S.B., Belch, G.E. (1983). Attitude toward the ad as a mediator of
advertising effectiveness: determinants and consequences, Advances in
ConsumerResearch, 10, 532-539.
Mandler, G. (1982). The structure of value: accounting for taste, in affect and cognition:
The 17th Annual Carnegie Symposium, Eds. Margaret S. Clark and Susan T. Fiske,
Hillsdale, NJ: Lawrence Erlbaum Associates, 3-36.
Markides, K.S., Boldt, J.S. (1983). Change in subjective age among the elderly: a
longitudinal analysis, The Gerontologist, 23 (4), 422-427.
Mastro, D.E., Stern, S.R. (2003). Representations of race in television commercials: a
content analysis of prime-time advertising, Journal of Broadcasting & Electronic
Media, 47 (4), 638-647.
Mehrabian, A., Russell, J.A. (1974). An approach to environmental psychology, Ed. MIT
Press.
Milliman, R.E., Erffmeyer, R.C. (1989-1990). Improving advertising aimed at seniors,
Journal of Advertising Research, 29 (6), 31-36.
Miniard, P.W., Bhatla, S., Rose, R.L. (1990). On the formation and relationship of ad
and brand attitudes: an experimental and causal analysis, Journal of Marketing
Research, 27, 290-303.
Moschis, G.P., Mathur, A. (2006). Older consumer responses to marketing stimuli: the
power of subjective age, Journal of Advertising Research, 46 (3), 339-346.
Nelson, S.L., Smith, R.B. (1988). The influence of model age or older consumers
reactions to print advertising, Current Issues and Research in Advertising, 11 (1),
189-212.
Olney, T.J., Holbrook, M.B., Batra, R. (1991). Consumer responses to advertising: the
effects of ad content, emotions and attitude toward the ad on viewing time, Journal
of Consumer Research, 17 (4), 440-453.
148 Chevalier and Lichtlé

Petty, R.E., Cacioppo, J.T. (1986). Message elaboration versus peripheral cues, In
Communication and persuasion central and peripheral routes to attitude change, 141-
172.
Petroshius, S.M., Crocker, K.E. (1989). An empirical analysis of spokesperson
characteristics on advertisement and product evaluations, Journal of the Academy of
Marketing Science, 17 (3), 217-225.
Rotfeld, H.J., Reid, L.N., Wilcox, G.B. (1982). Effect of age of models in print ads on
evaluation of product and sponsor, Journalism Quarterly, 59 (3), 374-381.
Smith, R.B., Moschis, G.P. (1984). Consumer socialization of the elderly: an exploratory
study, Advances in Consumer Research, 11, 548-552.
Underhill, L., Cadwell, F. (1983). « What age do you feel. » age perception study, The
Journal of Consumer Marketing, 1 (1), 18-27.
Vermeir, I., Sompel, D. (2014). Assessing the what is beautiful is good stereotype and
the influence of moderately attractive and less attractive advertising models on self-
perception, ad attitudes, and purchase intentions of 8-13-year-old children, Journal
of Consumer Policy, 37 (2), 205-233.
Whittler, T.E., Spira, J.S. (2002). Model’s race: a peripheral cue in advertising
messages?, Journal of Consumer Psychology, 12 (4), 291-301.
Model’s Age and Target’s Age 149

APPENDIX 1: THE MINERAL WATER ADS


150 Chevalier and Lichtlé

APPENDIX 2: THE COFFEE ADS


Being Hooked by the Archetypal Characters in Drama TV
Ads: A Structural Equation Modeling Approach

Gül Şener, Hasan Kemal Suher, and Ali Atıf Bir

1 Introduction
In today’s advertising where the management of brand meaning becomes
more and more vital, archetypes provide marketers the means to cut through the
competition (Mark and Pearson, 2001; Batey, 2008; Hartwell and Chen, 2012).
The concept of archetype has been originally defined by Jung (1968) as human
figures that can be arranged around similar traits and typical mode of conducts.
In their neo-archetypal theory, Faber and Mayer (2009) described archetypes as
mental models of others represented through generic story characters that create
emotion laden automatic response in the viewer. By manifesting particular
modes of thinking and acting, archetypes become the agents of a brand’s
worldview. They have the power to shape consumers’ behavior by providing
role models with distinct approaches to life in general and to the advertised
brand in specific. From this perspective, archetypal characters may serve as
means of self-relevance for consumers from whom they learn how the brand is
instrumental in attaining their desired goals (Boller and Olson, 1991).
A decade ago, effective advertising had been reformulated as “personally
meaningful, culturally relevant and creating a subjective feeling of warmth and
positive affect” (Gordon, 2006, p.4). Due to their emotional potency and the
collective and individual meanings they embody, archetypes can pave the way
for effective advertising. However, existing research on the subject is still
limited. The current study tries to expand the literature in question by shedding
light on the mechanisms through which archetypal characters in drama TV ads
affect consumers’ attitude toward the ad (Aad). This relationship is examined
through a new research model involving a viewer’s immersive experience of
being hooked by the ad and the affective reactions (sympathy and empathy)
induced by the resonance to archetypal characters.

2 Literature Review and Hypotheses


Past body of work provides that archetypal images are one of the influencing
factors in consumer-ad interaction. Gröppel-Klein et. al. (2006) found that TV
spots with typical fairytale archetypes produced more arousal reactions in the
viewers than informational TV spots without an archetype. Maloney (1999)
empirically concluded that archetypal themes (i.e. quest, attachment, conflict)
shape adults’ affective responses to certain figures (i.e. mother, hero). As

© Springer Fachmedien Wiesbaden 2017


G. Christodoulides et al. (Hrsg.), Advances in Advertising Research (Vol. VII),
European Advertising Academy, DOI 10.1007/978-3-658-15220-8_12
152 Şener, Suher, and Bir

theorized by Faber and Mayer (2009), frequent use of archetypes in TV, cinema
and music is related to the affective charge they contain. Aylesworth et. al.
(1999) confirmed that feelings are more responsive to archetypally suggestive
embeds in advertisements than cognitive judgments. Another stream of research
examined such emotional relationship in terms of archetypal brand personalities
(Roberts, 2010; Mark and Pearson, 2001) and yielded parallel results.

2.1 Relating with the Archetypal Characters: Resonance


Faber and Mayer (2009) identified resonance as “the striking experience of
encountering a psychologically meaningful character” (p.314). To measure
viewers’ emotional connection to archetypes (Table 1) in rich cultural media
they developed a unitary resonance construct consisting of interest, familiarity,
liking and disliking dimensions. Familiarity was excluded from the final
construct due to its low correlation with other dimensions.
Table 1: Archetype definitions (Faber, 2009)
A devoted, sacrificing, and nurturing person: Compassionate, generous,
Caregiver protective, parental. One who is benevolent, friendly, helping and trusting.
An innovative and artistic person; perhaps a dreamer, looking for novelty and a standart of
Creator aesthetic beauty. Emphasizes quality (over quantity), highly internally driven and inventive.
An independent, free-willed adventurer. One who seeks self-discovery: Solitary but
Explorer strong-willed, an observer of the environment. Constantly moving, a ‘wanderer’.
A courageous warrior or crusader undertaking arduous tasks to prove themselves
Hero and become an inspiration. A symbol of redemption and human strenght.
A pure, faithful, childlike character. Humble and tranquil; longing for happiness
Innocent and simplicity. A traditionalist; perhaps naive but symbolizing renewal.
Living for fun and amusement, a playful and mischievous comedian. One who is
Jester sometimes irresponsible, a prankster. Enjoys most a good time and diversion from care.
An intimate, sensual and passionate person. Seeking mainly pleasure, delightful but
Lover often jelous and impulsive. A warm, erotic and enthusiastic partner.
A rebellious rule-breaker, misfit or iconoclast. May be angry and vengeful. Can be wild,
Outlaw destructive and provoking from a long time spent surviving, struggling or injured.
Everyman/ The common person, the underdog, the working class. One who is persevering, ordered,
Everywoman wholesome, and candid. Self-depreciating, perhaps cynical but realistic.
A highly influential leader, boss or judge. Possessing a strong sense of power and control.
Ruler Stubborn, even tyrannical. Highly dominant, as an administrator or a manager of others.
One who values enlightenment, truth and understanding. A wise guide and counselor:
Sage Scholarly, philosophical, intelligent; perhaps a bit pretentious.

The current study proposes that the original resonance scale can be extended
with the inclusion of perceived similarity as a new dimension and re-inclusion of
familiarity. Such a scale can be instrumental in determining a viewer’s varying
Being Hooked by the Archetypal Characters in Drama TV Ads 153

resonance levels with respect to different archetypal characters in TV ads.


Consequently, suggested hypotheses were as follows:

H1: The relationship between the viewer and the archetypal character in a TV ad
can be operationalized and measured by the unitary archetype resonance
construct.
H2: Viewers resonate with some archetypal characters more than the others.

Based on the previous literature regarding the subject, the current research
model proposed that the degree of resonance with the archetypal character in an
ad produces a deeper experience of being hooked by this ad and that resonance
generates affective reactions. Accordingly, following hypotheses were
formulated:

H3a: Archetype resonance is positively related to being hooked.


H3b: Archetype resonance is positively related to sympathy.
H3c: Archetype resonance is positively related to sympathy with the mediation
of being hooked.

2.2 Engaging with the Ad Narrative: Being Hooked


While watching a drama ad (Wells, 1988) where the brand message is
communicated through a plot and characters, viewers become cognitively and
emotionally invested in the narrative and are absorbed by it (Slater and Rouner,
2002). Such a state of experiential involvement with the ad narrative is identified
as transportation (Green and Brock, 2000; Green, 2004) or being hooked
(Escalas et. al., 2004). The persuasiveness of an ad depends heavily on the
extent of being hooked by the narrative (Slater and Rouner, 2002). An ad can
hook the viewers by making them experience the feelings of the characters and
immersing them in the characters’ world. In such a case, viewers are less
inclined to counterargue the ad message (Dal Cin et. al., 2004). Also,
identification with the characters endorses a modeling behavior (Slater, 2002)
and a change in the attitudes and beliefs (Morgan et. al., 2009; Jones, 2014;
Chen, 2015). In parallel with the literature, subsequent hypotheses were
proposed:

H4a: Being hooked is positively related to sympathy.


H4b: Being hooked is positively related to empathy.
H4c: Being hooked is positively related to Aad.
154 Şener, Suher, and Bir

2.3 Emotional Responses to the Characters: Sympathy and Empathy


Cognitive and emotional processing of the drama ads is related to the
viewers’ empathetic relationship with the characters (Boller and Olson, 1991). In
the sense that the viewer vicariously merges with the character’s feelings
(Escalas and Stern, 2003), empathy is one of the key components of
identification process (Oatley, 1994). Identification occurs by adopting the
character’s point of view, experiencing the story from that position (Cohen,
2001) and it reinforces story-consistent attitudes (De Graaf et. al., 2012). Also,
the viewer may feel sympathy for the character (Tan, 1996). Sympathy implies a
spectatorship position where the viewer watches the character from an emotional
distance as the story unfolds (Oatley, 1999). For viewers, experiencing
sympathetic characters can be informational and impactful (Green, 2008). Past
research has shown that sympathy has an impact on the attitude toward the ad
(Aad) and this impact is partially mediated by the feelings of empathy (Escalas
and Stern, 2003). Following previous findings, final hypotheses suggested by the
research model were:

H5a: Sympathy and empathy mediate the relation between being hooked and
Aad.
H5b: Sympathy is positively related with empathy.
H5c: Empathy mediates the relation between sympathy and Aad.
H5d: Empathy is positively related to Aad.

2.4 Attitude Toward the Ad (Aad) as an Outcome Variable


Aad is “a predisposition to respond in a favorable or unfavorable manner to a
particular advertising stimulus during a particular exposure” (MacKenzie et. al.,
1986, p.130-131). Advertising research has already shown that affective
reactions help to predict Aad (Edell and Burke, 1987). When the viewer is swept
away by the narrative (or being hooked) this situation generates positive feelings
(Deighton et. al., 1989) which in turn causes a positive Aad (Escalas and Stern,
2003; Escalas et. al., 2004).

3 Methodology
3.1 Procedure

3.1.1 Study 1
To test the extended version of Archetype Resonance Scale (perceived
similarity and familiarity items were added to the original construct), 115
graduate student subjects (41% male, 59% female) were recruited by
convenience sampling to participate in the study. The mean age of subjects was
Being Hooked by the Archetypal Characters in Drama TV Ads 155

21, with a range of 17-27. Subjects have watched 22 TV ads that contained 11
different archetypal characters (each represented by 2 different TV ads) as their
protagonist. After watching each TV ad subjects were asked to administer
a survey questionnaire that consisted of a five-item Archetype Resonance Scale.
The responses were subjected to statistical analyses including correlation and
factor analysis with SPSS.

3.1.2 Study 2
To test the structural model (Figure 1) involving Archetype Resonance,
Being Hooked, Sympathy and Empathy and Aad constructs; 168 graduate and
postgraduate student subjects (91% graduate, 9% postgraduate level and 37,5%
male, 62,5% female) were recruited by convenience sampling to participate in
the study. The mean age of subjects was 21, with a range of 17-43. Subjects
have watched 11 TV ads that contained 11 different archetypal characters as
their protagonist. After each TV ad subjects were asked to administer a survey
questionnaire that consisted of a five-item Archetype Resonance Scale, eight-
item Being Hooked Scale, five-item Sympathy Scale, five-item Empathy Scale,
and four-item Aad Scale.
Based on the two-step approach proposed by Anderson and Gerbing (1988),
first the reliability and validity of the measurement model were assessed and
then the structural model was tested by using LISREL 8.51 program.
The measurement and structural models were tested for each archetypal
character category separately. Before continuing with Confirmatory Factor
Analysis (CFA) and Structural Equation Modeling (SEM), item parceling was
applied to all scales in the study (Matsunaga, 2008; Little et. al., 2002).

SYMP1 SYMP2 EMP1 EMP2


H5a,H5b,
H5c
Sympathy Empathy

H3b
H5

a b
d

RES1 H4 H4 Aad1
H3a,H3c Being H4c
Resonance Aad
Hooked
RES2 Aad2

HOOK1 HOOK2 HOOK3

Figure 1: Structural equation model of the relationship between archetype resonance,


being hooked, affective responses and Aad
156 Şener, Suher, and Bir

3.2 Measures

3.2.1 Archetype Resonance

The new archetype resonance scale developed within the confines of this
research contained five-item (perceived similarity, familiarity, interest, like,
dislike) 5-points Likert scale (1=Not at all, 5=Extremely).

3.2.2 Being Hooked


Being hooked measures were adopted from Escalas et. al. (2004) and
original seven-point Likert scale was modified into a five-point Likert scale
(1=Strongly disagree, 5=Strongly agree). The items of being hooked scale are
as follows: (1) “This commercial did not really hold my attention” (2) “This ad
did not draw me in” (3) “This ad really intrigued me” (4) “If I had seen this ad at
home, I’d have watched the whole thing” (5) “I could not relate to this
commercial” (6) “This commercial reminded me of experiences and feelings
I’ve had in my own life” (7) “I felt as though I was right there in the commercial
experiencing the same thing” (8) “I would like to have an experience like the
one shown in the commercial”

3.2.3 Sympathy and Empathy

Affective reactions towards the characters in TV ads were assessed using


Escalas and Stern’s (2003) Ad Response Sympathy and Empathy Scales (ARS
and ARE). Original seven-point Likert scale was modified into a five-point
Likert scale (1=Strongly disagree, 5=Strongly agree). The items for both scales
are as follows:
• ARS: (1) “Based on what was happening in the commercial, I understood what
the characters were feeling” (2) “Based on what was happening in the
commercial, I understood what was bothering the characters” (3) “While
watching the ad, I tried to understand the events as they occurred” (4) “While
watching the ad, I understand the characters’ motivation” (5) “I was able to
recognize the problems that the characters in the ad had”
• ARE: (1) “While watching the ad, I experienced feeling as if the events were
really happening to me” (2) While watching the ad, I felt as though I were one of
the characters” (3) “While watching the ad, I felt as though the events in the ad
were happening to me” (4) “While watching the commercial, I experienced
many of the same feelings that the characters portrayed” (5) “While watching the
commercial, I felt as if the characters’ feelings were my own”.
Being Hooked by the Archetypal Characters in Drama TV Ads 157

3.2.4 Attitude Toward the Ad (Aad)


Aad was assessed using four five-point bipolar scales adapted from Mitchell
and Olson (1981) and was anchored as “good”/”bad”, “like”/”dislike”,
“irritating”/”not irritating” and “interesting”/”not interesting”.

4 Analysis and Results


4.1 Study 1

The reliability test of the “archetype resonance” as a unitary construct with


five participant items (familiarity, perceived similarity, interest, liking and
disliking) proved that the scale is reliable (α=0,805).
To explore the interrelationships among items and to test whether all items
represent one single dimension of “archetype resonance” as predicted, an
exploratory factor analysis was conducted. Inspection of the correlation matrix
revealed the presence of significant inter-item correlations, the Pearson
correlation coefficients ranging from r(115)=-0,157 to r(115)=0,925, p<0,05.
Majority of the coefficients exceeds the recommended value of r=0,30 (Pallant,
2013). Familiarity, perceived similarity, interest and liking are positively
correlated whereas disliking is negatively correlated with the other participant
items of the scale. Kaiser-Meyer Olkin value was 0,705, exceeding the
recommended value of 0,6 and Bartlett’s Test of Sphericity was significant
(X2=323,386 sd=10 p<0,05) (Tabachnick and Fidell, 2014), supporting the
factorability of the correlation matrix. The result of the principal component
analysis revealed one component with eigenvalue over 1, explaining %58,79 of
the variance. As a result, the findings of the factor analysis have confirmed that
the extended version of the original “Archetype Resonance Scale” by Faber and
Mayer (2009) is a reliable scale and a unitary construct. Archetype Resonance
can be operationalized as the totality of a viewer’s familiarity, perceived
similarity, interest, liking and disliking evaluations of the archetypal character.
Therefore, H1 is supported.
Descriptive statistics of the archetype resonance scores for each archetypal
character under investigation revealed that characters reflecting Caregiver, Jester
and Ruler archetypes had the highest resonance scores respectively 17,69;
16,59; 16,23). On the other hand, characters exemplifying
Everyman/woman and Outlaw archetypes had the lowest scores respectively
( 13,18; 13,70). Therefore, H2 is supported.
158 Şener, Suher, and Bir

4.2 Study 2

4.2.1 Reliability and Validity of the Measurement Model

All the composite reliability values for 11 different measurement models with
5 constructs ranged from 0,645 to 0,985 and can be considered above acceptable
threshold values of 0.6 or 0.7 (Bagozzi and Yi, 1988; Hair et. al, 2010). The
factor loadings of the items were significant and higher than 0.5 for 11 different
measurement models, at the same time Average Variance Extracted (AVE) values
were 0.5 and higher indicating acceptable convergent validity for 11
measurement models (Fornell and Larcker, 1981; Anderson and Gerbing, 1988;
Hair et. al., 2010) (Table 2).

Table 2: Composite reliability and average variance extracted values


Archetype Being
Resonance Hooked Sympathy Empathy Aad
AVE CR AVE CR AVE CR AVE CR AVE CR
Caregiver 0,867 0,928 0,823 0,933 0,819 0,901 0,951 0,974 0,875 0,928
Creator 0,832 0,907 0,763 0,907 0,941 0,972 0,941 0,972 0,913 0,955
Explorer 0,913 0,955 0,758 0,904 0,961 0,977 0,941 0,972 0,970 0,985
Hero 0,495 0,645 0,822 0,932 0,866 0,930 0,865 0,928 0,970 0,982
Innocent 0,867 0,930 0,871 0,951 0,903 0,953 0,951 0,972 0,951 0,977
Jester 0,811 0,898 0,764 0,909 0,766 0,872 0,922 0,961 0,932 0,966
Lover 0,837 0,913 0,792 0,918 0,839 0,910 0,856 0,919 0,931 0,964
Outlaw 0,703 0,823 0,706 0,878 0,726 0,843 0,856 0,919 0,884 0,936
Everyman/ 0,847 0,916 0,823 0,933 0,931 0,961 0,913 0,953 0,932 0,969
Everywoman
Ruler 0,629 0,771 0,755 0,900 0,835 0,907 0,904 0,950 0,849 0,919
Sage 0,821 0,901 0,823 0,934 0,812 0,895 0,913 0,955 0,951 0,977
AVE: Average Variance Extracted CR: Composite Reliability

Discriminant validity could be assessed in a rigorous test by examining


whether the square root of the AVE of each construct was larger than the
correlation between constructs (Fornell and Larcker, 1981; Chin, 1998; Hair et.
al., 2010).
In eight measurement models the smallest value of the square root of the
AVE of the constructs was larger than the correlation between constructs. In the
measurement model of Lover, Hero and Explorer, certain correlation values
were larger than the smallest value of the square root of the AVE of the
Being Hooked by the Archetypal Characters in Drama TV Ads 159

constructs but this condition was only limited with Archetype Resonance and
Being Hooked constructs (Table 3).
In conclusion 11 measurement models proposed in the current study showed
adequate reliability and convergent validity in all the measurement models; and
acceptable discriminant validity in the majority of the measurement models.

Table 3: Square root of AVE and correlation between constructs


The Smallest Value The Largest Value
of the Square Root of the Correlation
of the AVE between Constructs
Caregiver 0,905 0,830
Creator 0,873 0,800
Explorer 0,871 0,910
Hero 0,704 0,910
Innocent 0,931 0,870
Jester 0,874 0,760
Lover 0,890 0,900
Outlaw 0,838 0,670
Everyman/ 0,907 0,760
Everywoman
Ruler 0,793 0,790
Sage 0,901 0,820

4.2.2 Performance of the Measurement Model

For each archetype category (11 in total), the measurement model


was estimated with LISREL 8.51. Chi-square fit index (X2) / degrees of
freedom (df), root mean square error of approximation index (RMSEA),
comparative fit index (CFI), goodness of fit index (GFI) and normed fit
index (NFI) statistics provide evidence of an acceptable model fit. The
results of the study (Table 4) show that all the indices are in the
acceptable range (Hair et. al., 2010) for each archetype category.
160 Şener, Suher, and Bir

Table 4: Fit indices of the measurement model based on the archetype category

4.2.3 Performance of the Structural Model

After estimating the structural model and testing the proposed hypotheses, it
was found that 7 archetype models out of 11 yielded fit indices that were above
the recommended values (Table 5). Hence, the results for the majority of the
archetype categories showed that the proposed structural model fitted well with
the data. Fit indices of the Caregiver, Creator, Explorer and Lover archetype
models revealed that these models were not eligible for further investigation.

Table 5: Fit indices of the structural model based on the archetype category
X2/df RMSEA CFI NFI GFI
Caregiver Not acceptable
Creator Not acceptable
Explorer Not acceptable
Hero 1,73 0,066 0,98 0,96 0,93
Innocent 1,60 0,060 0,99 0,97 0,94
Jester 1,60 0,060 0,99 0,97 0,94
Lover Not acceptable
Outlaw 1,19 0,034 0,99 0,97 0,95
Everyman/ 1,77 0,068 0,99 0,97 0,93
Everywoman
Ruler 2,57 0,097 0,96 0,94 0,91
Sage 1,96 0,074 0,98 0,96 0,93
Being Hooked by the Archetypal Characters in Drama TV Ads 161

The results of the standardized path estimates of the five constructs under
investigation (Table 6) significantly supported that;
 In all eligible archetype models: Archetype resonance is positively related to
being hooked. Being hooked is positively related to empathy and Aad.
 In the majority of the eligible archetype models: Archetype resonance and
being hooked are positively related to sympathy.
The standardized path estimates did not provide significant results regarding
hypothesized relationships between sympathy and empathy; also between being
hooked, sympathy, empathy and Aad. Accepted and rejected hypotheses were
presented in Table 7.

Table 6: Estimated standardized path coefficients based on the archetype category

Archetype Resonance Resonance Being Hooked Being Hooked Being Hooked Sympathy Empathy
Model → Being Hooked → Sympathy → Sympathy → Empathy → Aad → Empathy → Aad
Everyman/woman
Standardized
Estimate Positive/0,74 Positive/0,16 Positive/0,54 Positive/0,65 Positive/0,71 Positive/0,03 Positive/0,09
Significance Significant Not significant Significant Significant Significant Not significant Not significant
Hero
Standardized
Estimate Positive/0,93 Positive/0,37 Positive/0,30 Positive/0,68 Positive/0,90 Positive/0,14 Negative/-0,09
Significance Significant Not significant Not significant Significant Significant Not significant Not significant
Innocent
Standardized
Estimate Positive/0,75 Positive/0,27 Positive/0,45 Positive/0,75 Positive/0,85 Positive/0,06 Positive/0,03
Significance Significant Significant Significant Significant Significant Not significant Not significant
Jester
Standardized
Estimate Positive/0,73 Positive/0,39 Positive/0,30 Positive/0,69 Positive/0,83 Positive/0,14 Negative/-0,09
Significance Significant Significant Significant Significant Significant Not significant Not significant
Outlaw
Standardized
Estimate Positive/0,69 Positive/0,08 Positive/0,38 Positive/0,44 Positive/0,82 Positive/0,32 Positive/0,02
Significance Significant Not significant Significant Significant Significant Significant Not significant
Ruler
Standardized
Estimate Positive/0,83 Positive/0,91 Negative/-0,32 Positive/0,66 Positive/0,82 Positive/0,12 Negative/-0,18
Significance Significant Significant Not significant Significant Significant Not significant Significant
Sage
Standardized
Estimate Positive/0,81 Positive/0,40 Positive/0,36 Positive/0,70 Positive/0,97 Positive/0,03 Negative/-0,2
Significance Significant Significant Significant Significant Significant Not significant Significant
162 Şener, Suher, and Bir

Table 7: Results of the hypothesis testing

Definition Results

H3a Archetype resonance is positively related to being hooked. Accepted (All models)
H3b Archetype resonance is positively related to sympathy. Accepted (Model majority)
H3c Achetype resonance is positively related to sympathy Accepted (Model majority)
with the mediation of being hooked.

H4a Being hooked is positively related to sympathy. Accepted (Model majority)


H4b Being hooked is positively related to empathy. Accepted (All models)
H4c Being hooked is positively related to Aad. Accepted (All models)

H5a Sympathy and empathy mediate the relation Rejected


between being hooked and Aad.
H5b Sympathy is positively related with empathy. Rejected
H5c Empathy mediates the relation between sympathy and Aad. Rejected
H5d Empathy is positively related to Aad. Rejected

5 Discussion, Managerial Implications and Limitations


The present study investigated the relationship between resonance to
archetypal characters in drama ads and viewers’ experience of being hooked,
feelings of sympathy and empathy, and Aad.
Within the confines of this study, an extended Archetype Resonance Scale
was developed and was proved as valid and reliable. Participants in this study
resonated the most with characters that represented Caregiver, Ruler and Jester
archetypes. The results of the structural model had shown that degree to which a
viewer resonates with an archetypal character affects positively the experience of
being hooked which in turn affects viewers’ attitude towards the ad. Moreover,
the experience of being hooked by the ad as an indication of the viewers’
immersion in the narrative affects positively the feelings of sympathy and
empathy. However, the relation between sympathy and empathy and their effect
on Aad were rejected despite previous evidence on such relationships (Escalas
and Stern, 2003; Taute et. al., 2011). These findings demonstrate that archetypal
characters create narrative impact and their contribution to advertising
effectiveness is not through viewers’ emotions but narrative transportation.
The results indicate that practitioners should pay special attention to
character development while creating drama ads. Archetypes are impactful tools
in the sense that they reflect specific personalities. The more viewers resonate
with such clearly defined characters the more they become part of the world
evoked by the narrative. In addition, being hooked is proven to be a gateway to
Being Hooked by the Archetypal Characters in Drama TV Ads 163

empathy. This finding implies that once the viewers are swept away by the
narrative they can easily identify with the characters. Practitioners can use this
formula as a way to encourage empathetic feelings in the viewer towards the
characters.
There are several limitations to this study. First, data was collected by using
an all-university student convenience sample. Future research should include
viewers from different age groups to increase the generalizability of the findings.
Further, the selection of the drama TV ads was based solely on the archetypal
characters. However, the plotline is also one of the key elements of a story and
contribute to the narrative impact (Banerjee and Greene, 2012; Van Laer et. al.,
2014). In that sense, characters are figures that bring the plotline into action. In
the future, plotline can be integrated into the model as a variable to develop a
broader understanding of the narrative impact and its effect on the Aad.

6 References
Anderson, J. C. and Gerbing, D. W. (1988), “Structural Equation Modeling in Practice:
A Review and Recommended Two-Step Approach” in: Psychological Bulletin,
Vol. 103, No: 2, 411-423.
Aylesworth, A.; Goodstein, R.C. and Kalra, A. (1999), “Effect of Archetypal Embeds on
Feelings: An Indirect Route to Affecting Attitudes” in: Journal of Advertising, Vol.
28, No: 3, 73-81.
Bagozzi, R. and Yi, Y. (1988), “On the Evolution of Structural Equation Models” in:
Journal of the Academic of Marketing Science,Vol.16, No.1, 74-94.
Banerjee, S. C. and Greene, K. (2012), “I Quit vs I’m Sorry I Used: A Preliminary
Investigation of Variations in Narrative Ending and Transportation” in: Psychology
and Health, Vol.27, No.11, 1308-1322.
Batey, M. (2008), Brand Meaning. Routledge, New York, London.
Boller, G. W. and Olson, J. C. (1991), “Experiencing Ad Meanings: Crucial aspects of
Narrative/Drama Processing” in: Holman, R. H. and Solomon, M. R., Eds.
Advances in Consumer Research, Association for Consumer Research, Provo, Vol.
18, 164-171.
Chen, T. (2015), “The Persuasive Effectiveness of Mini-films: Narrative Transportation
and Fantasy Proneness” in: Journal of Consumer Research, Vol. 14, 21-27.
Chin, W. W. (1998), “The Partial Least Squares Approach to Structural Equation
Modeling” in: Marcoulides, G. A. Ed. Modern Methods for Business Research,
Lawrence Erlbaum, Mahwah, NJ, 295-336.
Cohen, J. (2001), “Defining Identification: A Theoretical Look at the Identification of
Audiences with Media Characters” in: Mass Communication and Society, Vol.4,
245-264.
Dal Cin, S., Zanna, M. P. and Fong, G. T. (2004), “Narrative Persuasion and
Overcoming Resistance” in: Knowles, E. S. and Linn, J. A. Eds. Resistance and
Persuasion, Lawrence Erlbaum, Mahwah, NJ, 175-192.
164 Şener, Suher, and Bir

De Graaf, A., Hoeken, H., Sanders, J. and Beentjes, J. W. J. (2012), “Identification as a


Mechanism of Narrative Persuasion” in: Communication Research, Vol. 39, 802-
823.
Deighton, J., Romer, D. and McQueen, J. (1989), “Using Drama to Persuade” in:
Journal of Consumer Research, Vol. 16, No: 4, 335-343.
Edell, J. A. and Burke, M. C. (1987), “The Power of Feelings in Understanding
Advertising Effects” in: Journal of Consumer Research, Vol. 14, 421-433.
Escalas, J. E. and Stern, B. B. (2003), “Sympathy and Empathy: Emotional Responses to
Advertising Dramas” in: Journal of Consumer Research, Vol. 29, No: 4, 566-578.
Escalas, J. E., Moore, B. C. and Britton, J. E. (2004), “Fishing for feelings? Hooking
viewers help!” in: Journal of Consumer Psychology, Vol.14, No: 1-2, 105-114.
Faber, M. A. (2009), “Understanding Personality through Preferences in Popular Mass
Media: An Archetypal Approach”, Unpublished Doctoral Dissertation, University
of New Hampshire.
Faber, M. A. and Mayer, J. D. (2009), “Resonance to Archetypes in Media: There’s
Some Accounting for Taste” in: Journal of Research in Personality, Vol. 43,
No: 3, 307-322.
Fornell, C. and Larcker, D. F. (1981), “Evaluating Structural Equation Models with
Unobservables and Measurement Error” in: Journal of Marketing Research,
Vol.18, No.1, 39-50.
Gordon, W. (2006), “What Do Consumer Do Emotionally with Advertising?” in: Journal
of Advertising Research, Vol. 46, No:1, 2-10.
Green, M. C. and Brock, T. C. (2000), “The Role of Transportation in the persuasiveness
of public narratives” in: Journal of Personality and Social Psychology, Vol. 79,
701-721.
Green, M. C. (2004), “Transportation into Narrative Worlds: The Role of Prior
Knowledge and Perceived Realism” in: Discourse Processes, Vol.38, No: 2, 247-
266.
Green, M. C. (2008), “Research Challenges in Narrative Persuasion” in: Information
Design Journal, Vol. 16, No: 1, 47-52.
Gröppel-Klein A.; Domke, A. and Bartmann, B. (2006), “Pretty Woman or Erin
Brockovich? Unconscious and Conscious Reactions to Commercials and Movies
Shaped by Fairytale Archetypes – Results from Two Experimental Studies” in:
Pechmann, C. and Price, L., Eds. Advances in Consumer Research, Association for
Consumer Research, Duluth, MN, Vol.33, 163-174.
Hair, J. F.; Black, W. C.; Babin, B. J. and Anderson, R. E. (2010), Multivariate Data
Analysis (7th Edition), Pearson Prentice Hall, NJ.
Hartwell, M. and Chen, J. C. (2012), Archetypes in Branding: A Toolkit for Creatives
and Strategists. How Books, Blue Ash.
Jones, M. D. (2014), “Communicating Climate Change: Are Stories Better Than ‘Just the
Facts’?” in: The Policy Studies Journal, Vol.42, No: 4, 644-673.
Jung, C. G. (1968), Archetypes and the Collective Unconscious. Princeton University
Press, Princeton, NJ.
Princeton University Press.
Being Hooked by the Archetypal Characters in Drama TV Ads 165

Little, T. D., Cunningham, W. A., Shahar, G. and Widaman, K. F. (2002), “To Parcel or
Not to Parcel: Exploring the Question, Weighing the Merits” in: Structural
Equation Modeling, Vol. 9, No: 2, 151-173.
MacKenzie, S. B., Lutz, R. J. and Belch, G. E. (1986), “The Role of Attitude Toward the
Ad as a Mediator of Advertising Effectiveness: A Test of Competing Explanations”
in: Journal of Marketing Research, Vol.23, No:2, 130-143.
Maloney, A. (1999), “Preference Ratings of Images Representing Archetypal Themes:
An Empirical Study of the Concept of Archetypes,” in: Journal of Analytical
Psychology, Vol. 44, No:1, 101-116.
Mark, M. and Pearson, C. S. (2001), The Hero and The Outlaw: Building Extraordinary
Brands Through the Power of Archetype. McGraw-Hill, New York.
Matsunaga, M. (2008), “Item Parceling in Structural Equation Modeling: A Primer”
in: Communication Methods and Measures, Vol. 2, No: 4, 260-293.
Mitchell, A. A. and Olson, J. C. (1981), “Are Product Attribute Beliefs the Only
Mediator of Advertising Effects on Brand Attitude?” in: Journal of Marketing
Research, Vol.18, No:3, 318-332.
Morgan, S. E., Movius, L. and Cody, M. J. (2009), “The Power of Narratives: The Effect
of Entertainment Television Organ Donation Storylines on the Attitudes,
Knowledge and Behaviors of Donors and Non-Donors” in: Journal of
Communication, Vol.59, 135-151.
Oatley, K. (1994), “A Taxonomy of the Emotions in Literary Response and a Theory
of Identification in Fictional Narrative” in: Poetics, Vol. 23, 53-74.
Oatley, K. (1999), “Meetings of Minds: Dialogue, Sympathy and Identification in
Reading Fiction” in: Poetics, Vol. 26, 439-454.
Pallant, J. (2013), SPSS Survival Manual, Open University Press, Berkshire.
Roberts, C. (2010), Exploring Brand Personality Through Archetypes (Unpublished
Doctoral Dissertation), East Tennessee State University.
Slater, M. D. and Rouner, D. (2002), “Entertainment-Education and Elaboration
Likelihood: Understanding the Processing of Narrative Persuasion”, in:
Communication Theory, Vol. 12, No: 2, 173-191.
Slater, M. D. (2002), “Entertainment Education and the Persuasive Impact of Narratives”
in: Green, M. C., Strange, J. J. and Brock, T. C. Eds. Narrative Impact: Social and
Cognitive Foundations, Lawrence Erlbaum, Mahwah, NJ, 157-181.
Tabachnick, B. G. and Fidell, L. S. (2014), Using Multivariate Statistics (6th Edition),
Pearson, New York.
Tan, E. S. (1996) Emotion and Structure of the Film: Film as an Emotion Machine,
Lawrence Erlbaum, Mahwah, NJ.
Taute, H. A., McQuitty, S. and Sautter, E. P. (2011), “Emotional Information
Management and Responses to Emotional Appeals” in: Journal of Advertising,
Vol.40, No.3, 31-43.
Van Laer, T., De Ruyter, K., Visconti, L. M. and Wetzels, M. (2014), “The Extended
Transportation-İmagery Model: A Meta-Analysis of the Antecedents and
Consequences of Consumers’ Narrative Transportation” in: Journal of Consumer
Research, Vol.40, 797-817.
166 Şener, Suher, and Bir

Wells, W. D. (1988), “Lectures and Dramas”, in: Cafferata, P. and Tybout, A., Eds.
Cognitive and Affective Responses to Advertising, Lexington Books, Lexington,
13-21.
The Moderating Role of Congruence between Humor and Fun
Climate of the Company on the Effects of Humor in Internet
Job Ads

Eeva-Liisa Oikarinen

1 Introduction
Humor has been acknowledged as a common practice in advertising (Eisend,
2009; Weinberger and Gulas, 1992) and recognized as one of the most widely
studied advertising appeals (Voss, 2009). Humor has become more relevant in
the modern workplace, when new generations, who value fun at work (Romero
and Pescosolido, 2008; Lamm and Meeks, 2009) have entered job markets.
Thus, humor could be seen as congruent value promise of fun climate of the
company to the job seekers and humor usage might be a prominent tool for
recruiters, who are seeking job applicants who are emphasizing fun aspects at
work. Job adverts are important for companies: they offer way to promote their
employer brands (Backhaus, 2004) and manage organizational image (Rafaeli,
2000). However, the scientific discussion on recruitment advertising is still
emerging (Van Hoye and Lievens, 2005) and research related to humor in job
adverts is scarce. Moreover, the Internet offers a potent medium for humorous
online advertising practices, while humor is the most common advertising appeal
for viral ads (Eckler and Bolls, 2011; Porter and Golan, 2006; Golan and
Zaidner, 2008) and online environment has been recognized as a leading media
for companies when selecting and recruiting employees (Cappelli, 2001;
Backhaus, 2004). These issues may have motivated some companies to adopt
new online recruitment advertising tactics and using humor in Internet job ads
context, where it has not been widely utilized or investigated earlier. Humor is a
special type of incongruity element in advertising among others like irrelevance,
contradiction, unexpectedness or mismatch (Lee and Schumann, 2004: 80).
Traditionally, it is suggested, that congruent messages create more positive
responses when compared to incongruent ones (e.g. Kamins, 1990; Kamins and
Gupta, 1994). This study investigates the effectiveness of humor and especially
the role of congruence in Internet recruitment advertising context, where there is
a lack of research. The practical managerial challenge behind this study is, how
congruent humor usage, if included in a job ad should be related to the level of
fun climate of the company and the study sets the following research question:
How is the congruence level between humor and fun company climate
moderating the effects of humor in Internet job ads?

© Springer Fachmedien Wiesbaden 2017


G. Christodoulides et al. (Hrsg.), Advances in Advertising Research (Vol. VII),
European Advertising Academy, DOI 10.1007/978-3-658-15220-8_13
168 Oikarinen

2 Literature Review and Hypotheses


Congruence can be conceptualized in advertising in many ways. One
definition is that it relates to advertising as a match between stimulus element,
(e.g. product, brand or execution element in an ad) and one’s existing schema
related to that stimulus (Lee and Schumann, 2004). In online advertising context
the congruence can be defined also as thematic congruence between the media
context (e.g. a web page) and ad (web ad) (Janssens et al., 2012). On the one
hand, earlier studies have shown that congruent messages create more positive
responses than incongruent ones (e.g. Kamins, 1990; Kamins and Gupta, 1994).
On the other hand, incongruent information is able to create more attention than
congruent (see Lee and Schumann, 2004) and critical insights in some studies
favor the effects of incongruent brand communications (e.g. Dahlén et al., 2005;
Lange and Dahlén, 2003; Törn, 2009).
In this study understanding of humor is based on cognitive theories, which
are known as incongruity (I) and incongruity resolution (IR) types of humor
(Gulas and Weinberger, 2006). Incongruent type of humor has been much used
in advertising (Alden et al., 1993) and more specifically IR-type of humor
compared to incongruent seems to be more perceived as humorous by receivers
of humorous ads (Flaherty et al., 2004). The main characteristic of IR-type of
humor is that it contains surprise or inconsistency that demands processing by
the receiver to come to terms with the interpretation of the message (Spotts et
al., 1997). Moreover, personal temperament, age and gender affect the way that
different types of humor are understood (Ruch et al., 1990; Ruch et al., 1996;
Martin et al., 2003).
Humor in advertising has many widely accepted positive effects on consumer
reactions: attention to the ad (e.g. Duncan, 1979; Madden and Weinberger,
1984; Eisend, 2009) and more positive attitudes toward the ad and advertiser
(Weinberger and Gulas, 1992; Sternhal and Craig, 1973; Eisend, 2009).
However, there are also studies which have considered the risks and restrictions
related to humor usage (e.g. Beard, 2008; Madden and Weinberger, 1984).
Humor is possibly not suitable to be used in professional services settings
(Fugate et al., 2000), in high involvement setting (e.g. Fugate, 1998), in service
products with high personal relevance (e.g. health care) or with products relying
on a serious or high technology image (McCullough and Taylor, 1993, Scott et
al., 1990 in Fugate 1998), when usually strong arguments are expected rather
than humor.
The effects of humor in advertising depend on several moderating variables
and some commonly used in research are comparisons between fictious vs. real
brands, non-student vs. student samples and print vs. broadcast media (Eisend,
2011). Moreover, cultural orientation (Lee and Lim, 2008), involvement (Zhang
and Zinkhan, 2006) or prior brand evaluations (Chattopadhyay and Basu, 1990)
The Moderating Role of Congruence between Humour and Fun Climate 169

have also been described as moderating the effects of humor in advertising.


However, the studies in consumer advertising focusing on the moderating role of
congruence between humor and ‘funniness’ of products are still short in supply,
as an analogy to the moderating role of congruence between humor and fun
company climate in job advertising context in this study. Fun company climate is
a key aspect of the successful company culture because, for example, employees
experiencing fun in the workplace are more satisfied in their job (Karl and
Peluchette, 2006) and workplace fun is a strong predictor of applicant attraction
(Tews et al., 2012). All previously mentioned issues make using humor in job
adverts as a prominent tool for recruiters, who are seeking job applicants who
are emphasizing fun aspects at work.
Hypotheses H1-H3 are generated based on theoretical knowledge related to
the negative/positive effects of humorous advertising on attitudes/intentions.
Moreover, the final hypotheses include also potential moderating effect of
congruence between humor and fun climate on the effects of humor in job ads.
The assumption was made related to the nature of congruence: In low
congruence situation, the humorous job ad will have more negative responses
than the nonhumorous job ad. In high congruence situation, the humorous job ad
will have more positive responses than the nonhumorous job ad.

H1a-e: Humor - fun company climate congruence moderates the negative effects
of humor in Internet job ad content on attitudes to a) job ad, b) company, c) job,
d) project manager, e) industry

When there is a positive association between attitudes towards an object and


the intention to carry out behavior vis-à-vis this object (Ajzen, 1991; Cooke and
Sheeran, 2004), expecting that humor content in a job ad will lower intentions to
apply for the job in the ad. Hence, after including humor-company climate
congruence element as a potential moderator, it is possible to formulate the
second hypothesis:

H2: Humor – fun company climate congruence moderates the negative effects of
humor in Internet job ad content on intentions to apply for the job.

The reason why adverts in general are shared is very probably their ability to
spark emotion (Phelps et al., 2004) and viral aspects are more important for
advertisers (Bergers and Iyengar, 2013). Assuming that job ads with humor are
likely to behave similarly to consumer-related ads when it comes to viral aspect.
Further, more specifically, when congruence as an moderating factor is included:
170 Oikarinen

H3: Humor – fun company climate congruence moderates the positive effects of
humor in Internet job ad content on intentions to share a job ad.

3 Method

3.1 Procedure
The study is a 2 (humor: humor vs. nonhumour) x 2 (company climate: funny
vs. boring) between-subject experimental design (Croson, 2002; Söderlund and
Dahlén, 2010) with first factor humor (present/absent) and second factor
perceived level of company climate (funny/boring) in relation to a job (see Table
1).

Table 1: The design of the study


Congruence Humor in job ad, perceived funniness
of company climate
Low (1)No humor in job ad with Fun company climate
(2)Humor in job ad with Boring company climate
High (3)Humor in job ad with Fun company climate
(4)No humor in job ad with Boring company climate

A qualitative pretest (10 undergraduates business student respondents


answered literally to open-ended question) and A quantitative pretest (22
university business student respondents) were executed to find out what types of
companies or industries university business students as job seekers evaluate
being as the funniest and the most boring types of companies and industries to
work with.
Manipulation of climate of the company, was based on selection of industries
with high and low in fun climate (‘funny’ versus ‘boring’ companies) utilizing
the results of pretests. The ‘funny’ company was developed as ‘HealthVision’,
fictional small high-tech start-up company without an established reputation
operating in the e-health industry. IT company in a high tech industry was
considered relatively funny among business students in the quantitative pretest
(M=6.9) (10 point scale not fun at all 1 to very much fun 10). Job ad comprised
the typical elements of an online job ad and copy stated that the company was
looking for two trainees, targeting undergraduate students in economics,
software engineering and healthcare. The language of the ad and subsequent
questionnaire reflected the nationality of the participants, because humor is best
understood in its cultural context. The ‘boring’ company,‘TelConnect’,
operating telephone sales (telemarketing) was created, because in the
The Moderating Role of Congruence between Humour and Fun Climate 171

quantitative pretest ‘company operating telephone sales work’ was rated as the
lowest mean value (M=3.2) of all the 24 different company/industry types,
which were asked to be rated by respondents.
Companies with low fun climate (‘TelConnect’) and high fun climate
(‘HealthVision’) were modified to produce versions with incongruity resolution
type of humor as humorous stimuli, because the incongruent resolution type of
humor has been identified by Alden et al. (1993) as frequently used in
advertising in general. More specifically the personification type of humor was
used, which concerns attributing human characteristics to animals, plants and
objects (Catanescu and Tom, 2001). This is an example of humor type which
demands resolution and as an example of this was “HealthVision is a furry ICT
company in the wellness sector, currently employing five people whose bark is
fortunately worse than their bite.” Humor appeared only in the section
describing the company and humor manipulation was conducted by changing
individual words and the other content was constant between the ad versions.
The stimulus material (and English translations of the job ads) used in the study
are available from the author by request.
Participants (n=121, 58 men, aged 20-45, M=25.7 years) were volunteers
undergraduate students, which are representative sample for investigating
responses of younger job seekers to humorous stimuli used in job adverts as
most will soon be entering the job market (Berthon et al., 2005) and were the
target group seeking for a ‘trainee’ job. Participants were randomly allocated
one of the four versions of the job ad with different stimuli (see Table 1): 35
reviewed version 1, 27 reviewed version 2, 29 version 3, and 30 version 4. The
distribution of male and female between four treatment groups was not
significantly different (Pearson Chi-Square 5.811, df=3, p=0.121>0.05). The
objective of the study was informed to the participants as studying how people
perceive job ads with literal instructions how to fill the questionnaire forms.

3.2 Measures
Perceived humorousness was measured as for the manipulation check of
humor, with a set of adjective pairs (see Appendix 1, Table 2) scored on a 10-
point scale. All the hypotheses with related used measures (perceived funniness
of job, attitudes, intentions), items and Cronbach’s alphas are presented in Table
2.
172 Oikarinen

Table 2: Hypothesis with related measures, items and Cronbach’s alphas.


Hypothesis Measure Items Description of α
items and claims,
10 point scale,
mean of the items
calculated
Manipu- PercHum, Voss (2009), Nonhumorous – 0.92
lation check perceived Zhang humorous, Not
humorousness (1996) funny–funny, not
towards job ad playful–playful, not
amusing–amusing,
dull–not dull,
boring–not boring
Manipu- PercFun, Modified Not funny–funny, 0.92
lation check perceived from Voss not playful–playful,
funniness of (2009), not amusing–
job Zhang amusing, dull–not
(1996) dull, boring–not
boring
H1a Ajobad, attitude to Mitchell Bad–good, dislike– 0.95
job ad and Olson like, negative
(1981), impression–
Söderlund positive
and Dahlén impression,
(2010) uncomfortable–
comfortable
H1b Acomp, attitude to Mitchell Bad–good, dislike– 0.96
company and Olson like, negative
(1981), impression–
Söderlund positive
and Dahlén impression,
(2010 uncomfortable–
comfortable
H1c Ajob, attitude to Mitchell Bad–good, dislike– 0.96
job and Olson like, negative
(1981), impression–
Söderlund positive
and Dahlén impression,
(2010 uncomfortable–
comfortable
Apm, attitude to Mitchell Bad–good, dislike– 0.97
The Moderating Role of Congruence between Humour and Fun Climate 173

H1d project manager and Olson like, negative


(1981), impression–
Söderlund positive
and Dahlén impression,
(2010 uncomfortable–
comfortable
H1e Aindustry, attitude Mitchell Bad-good, dislike- 0.98
to industry and Olson like, negative
(1981), impression-positive
Söderlund impression,
and Dahlén uncomfortable-
(2010) comfortable
H2 Intsend, intention 3 items, Unlikely–likely, not 0.94
to apply probability probable–probable,
of sending impossible–
application possible
H3 Intshare, sharing Brown, Probability of 0.87
intentions Bhadury, passing on via
& Pope, social media,
2010, probability of
Huang et telling others about
al. (2012) job ad, probability
of talking about job
ad (1 very unlikely
and 10 very likely)

4 Results
Manipulation check of humor was done through Univariate analysis of
variance: Dependent variable was chosen perceived humorousness of job ad, and
fixed factors were humor group and congruence as categorical variables to check
interaction effects. F(1,116), F=32.2, p=0.000. MNohumour=3.2,
MHumour=5.0. There was no significant interaction effect between humor and
congruence on perceived humorousness, which indicated that manipulation of
humor was successful. Similarly, the manipulation check of the company fun
aspect was tested examining the level of the perceived funniness of the company
climate in funny and boring company climate situations. Table 3 represents the
main results of this study: Means (with standard deviations) of manipulation
assessment values and dependent variables for each experimental condition.
174 Oikarinen

Also the main effects of humor and interaction effects between humor and
humor/company climate congruence were presented.
The hypotheses related 1 a-e were supported when there was both significant
main effect and significant interaction effect. The interactive effects of humor
and the level of congruence between humor and company climate on attitudes
variables are presented in the Figure 1. As preliminary findings, the individual
figures show that there was a significant interaction effect between humor and
the level of congruence on all output variables. The hypotheses related 2 and 3
were not supported when there was not significant main effect of humor,
although there was significant interaction effect between humor and congruence.
The interactive effects of humor and the level of congruence between humor and
company climate on intention- variables are presented in the Figure 2.
The present study has examined How is the congruence level between humor
and fun company climate moderating the effects of humor in Internet job
ads?As a summary of empirical findings, the study has shown that incongruent
resolution type of humor in online job ads has significant negative main effects
on attitudes to the job ad, company and project manager. Humor has also
negative (but nonsignificant) main effect on attitude to job and attitude to
industry. Humor has slightly negative (but nonsignificant) main effect on
intention to apply a job. The only positive main effect of humor (nonsignificant)
is on intention to share a job ad. Moreover, the results show that there was
significant interaction between humor and the congruence level between humor
and fun climate of the company on all attitude variables (job ad, job, company,
project manager, industry) and intention variables (intention to share and
intention to apply). The results indicate that congruence moderates the effects of
humor in Internet job ads.
The Moderating Role of Congruence between Humour and Fun Climate 175

Table 3: The main results of the study.


Hypo- Level of No Humor Main effect of Interaction
thesis congru- Humor (M/SD) humor effect
ence (M/SD) between
between humor and
humor/ congruence
company
climate
Perceived Low 3.4(1.4) 4.9(2.3) F(1,116) F=32.2, Not significant
Humorousness High 2.9(1.6) 5.1(1.9) p=0.0, interaction
Manipulation MNohumour=3.2,
successful MHumour=5.0
Perceived Low 4.7(1.7) 4.7(1.9) *Main effect of fun *fun climate
funniness of the High 3.6(2.0) 5.5(1.7) climate F(1,116) and
job F=8.5, p=0.004 congruence
Manipulation MBoring Climate=4.1 F(1,116),
partially successful MFun Climate =5.0 F=8.7,
p=.0.004
H1a: Attitudejob ad Low 6.7(1.7) 4.1(2.3) F(1,116)=23.3, F(1,116)=7.0,
Supported High 5.8(1.9) 5.1(1.7) p=0.0 p=0.009
MNoHumour=6.3 vs.
MHumour=4.6
H1b: Attitudecomp Low 7.0(1.4) 4.4(2.1) F(1,115)=14.4, F(1,115)=17.7,
Supported High 5.4(1.9) 5.5(1.7) p=0.0 p=0.000.
MNoHumour=6.3
MHumour=5.0
H1c: Attitudejob Low 6.6(1.5) 3.6(2.3) F(1,115)=2.2, F(1,115)=50.3,
Not supported High 4.1(2.0) 6.1(1.8) p=0.14 p=0.000
MNoHumour =5.5 M
MHumour =4.8
H1d: Attitudepm Low 6.3(1.4) 4.5(1.9) F(1,111)=14.3, F(1,111)=3.3,
Supported High 5.7(1.6) 5.1(2.0) p=0.0 p=0.07(<0.10)
MNoHumour=6.0
MHumour=4.8
H1e: Attiduteindustry Low 7.4(1.5) 3.6(2.1) F(1,112)=0.3,p=0.5 F(1,112)=96.9,
Not supported High 3.0(2 6.3(2.1) 7 MNoHumour=5.4. p=0.000
MHumour=5.0,
(Main effect of
congruence)
F(1,112)=5.0,
p=0.03
H2: Intention to Low 4.4(2.8) 2.2(1.6) F(1,117)=1.8, F(1,117)=18.2,
apply for a job High 2.3(1.7) 3.5(2.2) p=0.19 p=0.000
Not supported MNoHumour=3.4
MHumour=2.9
H3: Intention to Low 3.7(2.5) 3.3(1.8) F(1,117)=2.2, F(1,117)=6.1,
share job ad High 2.4(1.9) 3.9(2.1) p=0.14MNoHumour=3. p=0.015
1 MHumour=3.6
176 Oikarinen

 
Figure 1: The interactive effects of humor and the level of congruence between humor
and company climate on attitudes variables.
The Moderating Role of Congruence between Humour and Fun Climate 177

   
Figure 2: The interactive effects of humor and the level of congruence between humor
and company climate on intention to send an application and intention to share job ad
variables.

5 Discussion, Limitations and Managerial Implications


This paper has investigated humor in recruitment advertising context and
extends advertising theory on the effects of humor in multiple ways (Van Hoye
and Lievens, 2005). First, the findings show that incongruent resolution type of
humor has significant negative main effects on job seekers attitudes to the job
ad, attitudes to the company and attitudes to the project manager. The results
differ from earlier studies in consumer advertising context, which have reported
positive effects of humor (Eisend, 2009; Weinberger and Gulas, 1992; Sternhal
and Craig, 1973). This study offers contribution to the discussion related to the
risks of using humor in advertising in general (Beard, 2008; Madden and
Weinberger, 1984) because results show that humor usage in job advertising
context seems indeed risky.
By identifying new specific moderator this study will give new insights into
broader discussion related to humorous advertising literature, where different
moderators have been identified, such as cultural orientation (Lee and Lim
2008), involvement (Zhang and Zinkhan, 2006) or recently interaction effects
between gender and type of humor (Schwarz et al., 2015). The findings inform
that congruence between humor and company climate has strong moderating
effects to the effects of humor in Internet job ads on job seekers responses and
the more favorable effects are induced in high congruency situation. For
managers, these results offer insights and more awareness about the risks related
humor usage in job advertising and suggest that it would be more favorable
seeking for congruency between humor usage in job adverts and actual
company climate.
178 Oikarinen

This study has some limitations which will give ideas for further research.
The choice of the population other than students as job seekers would be
beneficial to be able to generate more generalizable findings. Further studies
should be carried out in different cultural contexts to allow for cross-cultural
comparisons (Biswas et al., 1992; Hatzithomas et al., 2009; Laroche et al.,
2014) and age could be considered also as a potential moderator in future
studies. In this study, the mechanism how congruence is moderating the
negative effects of humor in Internet job ads seems to be different on attitudes
to job ad, company and project manager than attitudes to job and industry.
Understanding these differences could open new research ideas for further
studies. The effects of real-life companies’ job ads data would be worth
investigating with emphasis on different types of humor (e.g. more playful
humor types).

6 References

Ajzen, I. (1991), "The Theory of Planned Behavior," in: Organizational Behavior and
Human Decision Processes, Vol. 50, 2, 179-211.
Alden, D.L. and Hoyer, W.D. (1993), "An Examination of Cognitive Factors related to
Humorousness in Television Advertising," in; Journal of Advertising, Vol. 22, 2, 29-
37.
Alden, D.L.; Hoyer, W.D. and Lee, C. (1993), "Identifying Global and Culture-Specific
Dimensions of Humor in Advertising: A Multinational Analysis," in: Journal of
Marketing, Vol. 57, 64-75.
Backhaus, K.B. (2004), "An Exploration of Corporate recruitment descriptors on
monster.com," in: Journal of Business Communication, Vol. 41, 2, 115-136.
Beard, F.K. (2008), "Advertising and audience offense: The role of intentional humor,"
in: Journal of marketing communications, Vol. 14, 1, 1-17.
Berger, J. and Iyengar, R. (2013), "Communication Channels and Word of Mouth: How
the Medium Shapes the Message," in: Journal of Consumer Research, Vol. 40, 3,
567-579.
Berthon, P.; Ewing, M. and Hah, L.L. (2005), "Captivating company: dimensions of
attractiveness in employer branding," in: International Journal of Advertising, Vol.
24, 2, 151-172.
Biswas, A.; Olsen, J.E. and Carlet, V. (1992), "A Comparison of Print Advertisements
form the United states and France," in: Journal of Advertising, Vol. 21, 4, 73-81.
Brown, M.R.; Bhadury, R.K. and Pope, N.K.Ll. (2010), "The impact of comedic
violence on viral advertising effectiveness," in: Journal of Advertising, Vol. 39, 1, 49-
65.
Cappelli, P. (2001), "Making the most of on-line recruiting," in: Harvard Business
Review, Vol. 79, 2, 139-146.
Catanescu, C. and Tom, G. (2001), "Types of humor in television and magazine
advertising," in: Review of Business, Spring, 92-95.
Chattopadhyay, A. and Basu, K. (1990), "Humor in advertising: The moderating role of
prior brand evaluation," in: Journal of Marketing Research, Vol. 27, 4, 466-476.
The Moderating Role of Congruence between Humour and Fun Climate 179

Cooke, R. and Sheeran, P. (2004), "Moderation of cognition–intention and cognition–


behavior relations: A meta-analysis of properties of variables from the theory of
planned behavior," in: British Journal of Social Psychology, Vol. 43, 2, 159-186.
Croson, R. (2002), "Why and how to experiment: Methodologies from experimental
economics," in: University of Illinois Law Review, Vol. 2, 921-945.
Dahlén, M.; Lange, F.; Sjödin, H. and Törn, F. (2005), "Effects of Ad-Brand
Incongruency," in: Journal of Current Issues and Research in Advertising, Vol. 27,
2, 1-12.
Duncan, C.P. (1979), "Humor in advertising: A behavioral perspective," in: Journal of
the Academy of Marketing Science, Vol. 7, 4, 285-306.
Eckler, P. and Bolls, P. (2011), "Spreading the virus: emotional tone of viral advertising
and its effect on forwarding intentions and attitudes," in: Journal of Interactive
Advertising, Vol. 11, 2, 1-11.
Eisend, M. (2009), "A meta-analysis of humor in advertising," in: Academy of Marketing
Science Journal, in: Vol. 37, 2, 191-203.
Eisend, M. (2011), "How humor in advertising works: A meta-analytic test of alternative
models," in: Marketing Letters, Vol. 22, 115-132.
Flaherty, K.; Weinberger, M.G. and Gulas, C.S. (2004), "The Impact of Perceived
Humor, Product Type, and Humor Style in Radio Advertising," in: Journal of Current
Issues and Research in Advertising, Vol. 26, 1, 25-36.
Fugate, D.L. (1998), "The advertising of services: what is an appropriate role for humor,"
in: The Journal of Services Marketing, Vol. 12, 6, 453-472.
Fugate D.L.; Gotlieb, J.B. and Bolton, D. (2000), "Humorous services advertising: What
are the role of sex, appreciation of humor, and appropriateness of humor?," in:
Journal of professional Services Marketing, Vol. 21, 1, 9-22.
Golan, G.J. and Zaidner, L. (2008), "Creative strategies in viral advertising: An
application of Taylor’s six-segment message strategy wheel," in: Journal of
Computer-mediated communication, Vol. 13, 4, 959-972.
Gulas, C.S. and Weinberger, M.G. (2006), Humor in advertising. A Comprehensive
analysis. M.E Sharpe, Armonk, New York, London, England.
Hatzithomas, L.; Boutsouki, C. and Zotos, Y. (2009), "The Effects of Culture and
Product Type on the Use of Humor in Greek TV Advertising: An Application of
Speck’s Humorous Message Taxonomy," in: Journal of Current Issues and Research
in Advertising, Vol. 30, 1, 43-61.
Huang, J.; Su, S.; Zhou, L. and Liu, X. (2013), "Attitude Toward the Viral Ad:
Expanding Traditional Advertising Models to Interactive Advertising," in: Journal of
Interactive Marketing, Vol. 27, 36-46.
Janssens, W.; De Pelsmacker, P. and Geuens, M. (2012), "Online advertising and
congruency effects. It depends on how you look at it," in: International Journal of
Advertising, Vol. 31, 3, 579-604.
Kamins, M.A. (1990), "An Investigation to The “Match-Up” Hypothesis in Celebrity
Advertising: When Beauty May Be Only Skin Deep," in: Journal of Advertising, Vol.
19, 1, 4-13.
Kamins, M.A. and Gupta, K. (1994), "Congruence between Spokesperson and Product
Type: A Matchup Hypothesis Perspective," in: Psychology & Marketing, Vol. 11, 6,
569-586.
Karl, K. and Peluchette, J. (2006), "How Does Workplace Fun Impact Employee
Perceptions of Customer Service Quality?," in: Journal of Leadership &
Organizational Studies, Vol. 13, 2, 2-13.
180 Oikarinen

Lamm, E. and Meeks, M.D. (2009), "Workplace fun: the moderating effects of
generational differences," in: Employee Relations, Vol. 31, 6, 613-631.
Lange, F. and Dahlén, M. (2003), "Let’s be strange: brand familiarity and ad-brand
incongruency," in: Journal of Product and Brand Management, Vol. 12, 7, 449-461.
Laroche, M.; Nepomuceno, M.V. and Richard, M-O. (2014), "Congruency of humor and
cultural values in print ads. Cross-cultural differences among the US, France and
China," in: International Journal of Advertising, Vol. 33, 4, 681-705.
Lee, H.Y and Lim, E.A. (2008), "What’s funny and what’s not. The moderating role of
cultural orientation in ad humor," in: Journal of Advertising, Vol. 37, 2, 71-84.
Lee, E.J. and Schumann, D.W. (2004), "Explaining the special case of incongruity in
advertising: combining classic theoretical approaches," in: Marketing Theory, Vol.
41, 1/2, 59-90.
Madden, T.J. and Weinberger, M.G. (1984), "Humor in advertising: A practitioner
view," in: Journal of Advertising Research, Vol. 24, 4, 23-29.
Martin, R.A.; Puhlik-Doris, P.; Larsen, G.; Gray, J. and Weir, K. (2003), "Individual
differences in uses of humor and their relation to psychological well-being:
Development of the Humor Styles Questionnaire," in: Journal of Research in
Personality, 37, 48-75.
McGullough, L.S. and Taylor, R.K. (1993), "Humor in American, British, and German
ads," in: Industrial Marketing Management, Vol. 22, 1,17-29.
Phelps, J.E.; Lewis, R.; Mobilio, L.; Perry, D. and Raman, N. (2004), "Viral marketing
or electronic word-of-mouth advertising: Examining Consumer Responses and
Motivations to pass along email," in: Journal of advertising research, Vol. 44, 4, 333-
348.
Porter, L. and Golan, G. (2006), "From subservient chickens to brawny men: A
comparison of viral advertising to television advertising," in: Journal of Interactive
Advertising, Vol. 6, 2, 26-33.
Rafaeli, A. (2000), "Projecting an organizational identify: Lessons from employment
advertisements," in: Corporate Reputation Review, Vol. 3, 3, 218-239.
Romero, E. and Pescosolido, A. (2008), "Humor and group effectiveness," in: Human
Relations, Vol. 61, 3, 395-418.
Ruch, W.; Köhler, G. and VanThriel, C. (1996), "Assessing the "humorous
temperament": Construction of the facet and standard trait forms of the State-Trait-
Cherrfulness-Inventory–STCI," in: Humor: International Journal of Humor
Research, Vol. 9, 3-4, 303-339.
Ruch, W.; McGhee, P.E. and Hehl, F-J. (1990), "Age differences in the enjoyment of
incongruity-resolution and nonsense humor during adulthood," in: Psychology and
Aging, Vol. 5, 3, 348-355.
Schwarz, U.; Hoffmann, S. and Hutter, K. (2015), "Do Men and Women Laugh About
Different Types of Humor? A Comparison of Satire, Sentimental Comedy, and Comic
Wit in Print Ads," in: Journal of Current Issues & Research in Advertising, Vol. 36,
1, 70-87.
Scott, C.; Klein, D. and Bryant, J. (1990), "Consumer response to humor in advertising: a
series of field studies using behavioral observation," in: Journal of Consumer
Research, Vol. 16, March, 498-501.
Spotts H.E.; Weinberger, M.G. and Parsons, A.L. (1997), "Assessing the use and impact
of humor on advertising effectiveness: A contingency approach," in: Journal of
Advertising, Vol. 16, 3, 17-32.
The Moderating Role of Congruence between Humour and Fun Climate 181

Sternhal, B. and Craig, C.S. (1973), "Humor in advertising," in: Journal of Marketing,
Vol. 37, October, 12-18.
Söderlund, M. and Dahlén, M. (2010), "The “killer” ad: an assessment of advertising
violence," in: European Journal of Marketing, Vol. 44, 11/12, 1811-1838.
Tews, M. J.; Michel J.W. and Bartlett, A. (2012), "The Fundamental Role of Workplace
Fun in Applicant Attraction," in: Journal of Leadership & Organizational Studies,
Vol. 19, 105-114.
Törn, F. (2009), Challenging Consistency. Effects of Brand-Incongruent
Communications. Dissertation for the Degree of Doctor of Philosophy, Ph.D.
Stockholm School of Economics, 2009.
Van Hoye, G. and Lievens, F. (2005), "Recruitment-Related Information Sources and
Organizational Attractiveness: Can Something Be Done About Negative Publicity?,"
International Journal of Selection and assessment, Vol. 13, 3, 179-187.
Voss, K.E. (2009), "Using humor in conjunction with negative consequences in
advertising," Journal of Current Issues and Research in Advertising, Vol. 31, 2, 25-
39.
Weinberger, M.G. and Gulas, C.S. (1992), "The impact of Humor in Advertising: A
Review," in: Journal of Advertising, Vol. 21, 9, 35-59.
Zhang, Y. (1996), "Responses to Humorous Advertising: The Moderating Effect of Need
for Cognition," in: The Journal of Advertising, Vol. 25, 1, 15-32.
Zhang, Y. and Zinkhan, G.M. (2006), "Responses to humorous ads. Does Audience
Involvement Matter?," in: Journal of Advertising, Vol. 35, 4, 113-127.
Put It on the Right Side: The Effect of Print Advertisement
Location on Product Evaluation
Tanja Schneider and Heribert Gierl

1 Introduction

Until now, there are some studies that examined whether ads located on the
left versus the right side of a double page in magazines or newspapers are
associated with higher attention of the recipients. It can be presumed that
recipients automatically look at the right side of a double page when they flip
through a magazine or newspaper which may result in higher attention values.
However, empirical research on this issue found mixed results (Ferguson, 1935;
Finn, 1988; Walker and Cardillo, 1998; Smit et al., 2013). Moreover, the ad
location on the left versus right side may affect perceptions (Cai et al., 2012) and
subsequently product evaluations. To the best of our knowledge, there is no
research on investigating product evaluations in this context thus far. We
contacted two managers of companies that publish newspapers and magazines,
respectively, and asked them to report whether there are differences regarding
the price for taking out ads on the left versus right side and regarding the
demand of advertising companies and agencies in reference to these locations.
They told us that the price does not depend on whether the ad is located on the
left or right side but that there is a much higher demand for locating ads on right-
sided pages. In summation, advertising practice prefers ad locations on the right
side although little is known about the effects resulting from left/right-location
decisions. For instance, we looked at the first few pages (since most of the ads in
magazines are placed here) of the last 40 issues of a magazine (Grazia)
published in Germany and counted for ads on the left and the right side of the
double pages (double page ads were excluded). We found 16 ads on the left side
and 33 ads on the right side – what underlines the assumption that marketing
practice prefers right-side locations. To provide another example for the
decision task from the marketer’s view, we present two locations of the Mumm
brand (champagne) that actually have been located on different sides of a double
page in the Cosmopolitan magazine (Figure 1).
In the next section, we explain the fundamental effect of the left- versus
right-sided location of stimuli on responses toward these stimuli. Subsequently,
we apply this theory for predicting effects of left- versus right-sided locations of
advertisements on double pages in magazines on perceptions and product
evaluations. Then, we present the design and the findings of a new experiment

© Springer Fachmedien Wiesbaden 2017


G. Christodoulides et al. (Hrsg.), Advances in Advertising Research (Vol. VII),
European Advertising Academy, DOI 10.1007/978-3-658-15220-8_14
184 Schneider and Gierl

conducted to test the resulting presumptions. Finally, we give advertising


practice recommendations on how to deal with the issue considered here.

Figure 1: Example of different ad locations of the Mumm brand (Cosmopolitan Germany


June and July 2015)
Put It on the Right Side 185

2 Theoretical Considerations and Prior Research

We refer to the SNARC effect which indicates that people tend to construe a
spatial continuum when they elaborate on numerical information and to the
effect of left/right location which suggests that individuals tend to assign
numerical or quantitative values to a stimulus that is represented on a mental
continuum. Both effects describe the same phenomenon but differ regarding the
directions of the effects (Figure 2).

Figure 2: Basic theoretical relationships

2.1 SNARC Effect


There is a stream of fundamental research providing evidence to the
presumption that individuals tend to allocate numerical information on a left-
right continuum (e.g., Dehaene et al., 1993; Fias, 2001; Gevers et al., 2002). In
studies on the “spatial-numerical association of response codes” (SNARC),
individuals initially receive the information that they subsequently will see
numbers from a certain range, e.g., “from 0 to 9.” Then, they are exposed to
such numbers, e.g., “4” on a computer screen. They have two buttons, one on the
left and one on the right side of a keyboard. These buttons are labeled with
“odd” and “even,” respectively. The individuals have to push the correct button
for each number they are exposed to, and their reaction time is assessed. In
Figure 3, we present four conditions of this kind of experiment. In this example,
the displayed number is either “4” or “6” and the odd-button is either located on
the left-hand side of the individual or on her/his right-hand side. We illustrate
the outcome of this experiment by using results reported by Fias (2001, p. 253).
The results show that individuals react faster if they have to use their right hand
to indicate parity of higher compared to lower numbers. Moreover, individuals
186 Schneider and Gierl

respond faster if they have to use their left hand for indicating parity of lower
compared to higher numbers. These findings also appeared if verbal descriptions
(e.g., “four” or “six,” “March” or “November”) of numbers were presented.
These results indicate that individuals associate lower numbers more easily to
the left side and higher numbers to the right side of a mental continuum. The
authors explain this finding by the presumption that numbers are perceived
spatially when individuals have received information about the range of the
numbers they have to judge at the beginning of the experiment. Numerous
authors extended the basic setting of this experiment (e.g., Dehaene et al., 1990;
Brysbaert, 1995; de Hevia and Spelke, 2009).

Comments: The results indicate that individuals react faster with their left hand if they
have to judge a lower number (e.g., “4” from the interval 0-9) compared to judging a
higher number (e.g., “6”; .576 < .582). Moreover, they respond faster with their right
hand if they have to judge a higher number (e.g. “6” from that interval) compared to
judging a lower number (e.g., “4”.571 < .544).
Figure 3: Illustration of findings from a SNARC experiment

2.2 Effect of Left/Right Location

Other authors postulated that individuals tend to assign numerical


information to a stimulus depending on its location on a continuum that
individuals can add to the stimulus merely due to imagination. To illustrate this
phenomenon, we refer to the experiments conducted by Cai et al. (2011, p. 471;
2012, p. 721). These authors exposed individuals to a piece of paper which
showed a seaweed snack. The product was either located on the left side of the
Put It on the Right Side 187

paper or on its right side (the remaining part of the piece of paper was “empty”).
They asked the test participants to guess the product’s market price within a
range of 30 to 45 (HK$) and found that the estimated price was higher when the
product was displayed on the right side (Figure 4).

Figure 4: Illustration of findings from an experiment on the effect of location

The authors replicated this experiment by showing a fixed number of circles


and squares on the right versus the left side of a display; the test participants
estimated that there were more figures when these images were presented on the
right side (M = 19.87) compared to presenting the identical number of figures on
the left side (M = 16.22). The authors explained this effect by the existence of a
learned association between the left/right location and numerical magnitude.
They argue that this relation has been learned because individuals are frequently
exposed to “number lines” (e.g., horizontal x-axes) or rulers with the smaller
numbers displayed on the left side. We add some additional examples in favor of
this principle where people apparently rely on the relation between numerical (or
quantitative) information and the stimulus location on a left/right-side of a
continuum (Figure 5).

2.3 Hypotheses
We use these approaches and presume that individuals infer numerical
information from an ad’s location on the left versus right side on a double page.
The location on the left versus right side can indicate a position on a latent
continuum (although the continuum itself is not depicted). We hypothesize that
the location which mirrors a position on a continuum is used to infer on the
advertised product’s price and quality:
188 Schneider and Gierl

H1: Perceptions of the product’s expensiveness are lower if the ad is located on


the left side of a double page compared to an ad location on the right side.
H2: Perceptions of the product’s quality are lower if the ad is located on the left
side of a double page compared to an ad location on the right side.

Figure 5: Some real-life examples indicating a relation between the left/right location
and numerical information

However, these presumptions are not sufficient for deriving a hypothesis


about which ad location results in higher product evaluations because the left
location may be associated with both a lower price and lower quality and the
Put It on the Right Side 189

right location is likely to be associated with a higher price as well as higher


quality. Thus, we introduce an additional variable into our approach that could
predict which aspect (low price versus high quality) is more important. There are
some authors who postulated a consumer aversion against paying too much for
less expensive products and a consumer aversion against receiving too little
quality for expensive products (e.g., Tversky and Kahneman, 1991, p. 1054;
Hardie et al., 1993, p. 388; Simonson et al., 1993, p. 360). Heath and Chatterjee
(1995, p. 282) argue that consumers intend to use expensive products for a
longer period of time what will inhibit them from replacing the products if the
quality turns out to be dissatisfying. To avoid a mispurchase (which would be
apparent for a longer period of time), consumers are expected to focus on the
product’s quality for avoiding the risk of low product performance. On the
contrary, consumers are likely to intend to use less expensive products for a
shorter period of time and, thus, are more willing to replace them if the quality
turns out to be poor. Thus, they are more likely to focus on the price of these
products in order to avoid the financial risk of paying too much (Heath, et al.,
2000, p. 304; Simonson and Tversky, 1992, p. 292). If consumers focus on the
(low) price, the effect of left/right location likely results in lower perceptions of
expensiveness when the ad is located on the left side – resulting in higher
evaluations of products of less expensive brands. If consumers focus on (high)
quality, the effect of location predicts higher perceptions of quality when the ad
is shown on the right side – resulting in more favorable evaluations of products
of expensive brands. From these additional considerations we conclude the
following:

H3a: For less expensive brands, attitudes toward the promoted product are
higher if the ad is located on the left side of a double page compared to an ad
location on the right side (due to lower perceptions of expensiveness).
H3b: For expensive brands, attitudes toward the promoted product are higher if
the ad is located on the right side of a double page compared to an ad location
on the left side (due to higher perceptions of quality).

In summation, we expect that the brand’s price level moderates the effect of
the ad’s location on product evaluations. In the case of less expensive brands, we
expect an effect of the left/right location on evaluations via perceptions of
expensiveness. In the case of expensive brands, we presume the existence of an
effect of the ad’s location via perceptions of quality.
190 Schneider and Gierl

3 Experiment

3.1 Experimental Design


We investigated the effect of the left/right location of a full-page
advertisement on a double page of a magazine on perceptions of quality and
expensiveness and on the attitudes toward the promoted product depending on
the brand’s price level. We used brands of six categories to test this effect. Thus,
our study was based on a 2 (ad location: left or right side on a double page) × 2
(price level of the brand the promoted product belongs to: less expensive or
expensive) × 6 (product category: mascara, shoes, coats, trench coats, hand-
bags, and bikinis) factorial between-subjects design. The latter factor served as a
replicate factor to prove stability of the findings. We chose these categories of
fashion goods because they can be mainly characterized by price and quality.

3.2 Manipulation of the Ad Location

We used issues of well-known magazines (e.g., Cosmopolitan) to create the


test stimuli. These stimuli consisted of a sequence of pages. First, the stimuli
showed a title page of the magazine. Then, the sequence showed a few double
pages of the magazine that contained editorial reports. On one of these double
pages, an ad was shown. This ad was either located on the left side or on the
right side of one of the double pages. The sequence of pages only contained one
target ad. In Figure 6 and Figure 7, we provide examples showing what the
double pages containing the ad looked like.
Put It on the Right Side 191

Figure 6: Double pages that show target ads for mascara depending on the ad’s
location (left vs. right side) and the brand’s price level (low vs. high)

3.3 Manipulation of the Price Level


We selected a less expensive and an expensive brand from the categories of
mascara, shoes, coats, trench coats, handbags, and bikinis. For instance, from the
mascara category, we chose the Manhattan brand as a representative of less
expensive brands and Dior as an expensive brand. In the survey, we asked the
test participants to estimate the prices of the promoted products in Euro. We
found that the selected brands were associated with highly different price
estimates for each category. In any case, the price for the product of the
expensive brand was estimated at least twice as high than the price for the
product of the less expensive brand (Table 1).
192 Schneider and Gierl

Figure 7: Double pages that show target ads for bikini depending on the ad’s location
(left vs. right side) and the brand’s price level (low vs. high)

3.4 Procedure
We conducted online-surveys with the use of SoSci Survey. The test
participants were exposed to the sequence of pages on the computer screen as
described above and, then, were asked to imagine running over the pages of the
magazine. Subsequently, they indicated their perceptions of the promoted
product’s quality and expensiveness as well as their attitudes toward this
product.
Put It on the Right Side 193

Table 1: Manipulation check results of the brands’ price level

3.5 Measures and Sample

The questionnaire contained four statements that indicated attitudes toward


the product (attractive, interesting, appealing, and likeable). The test participants
agreed or disagreed to these items on a seven-point scale (α = .949). To assess
perceptions of quality and expensiveness, we used single-item measures; the
participants agreed or disagreed to “Within its category, this product has a high
quality” and “Within its category, this product is expensive” on a seven-point
scale. As noted above, we also assessed the estimated price. In total, 809 female
consumers participated in our survey (Mage = 25.39 years, SD = 7.013, 66.7%
students).

3.6 Results
The findings of our study are summarized in Table 2.
For less expensive brands, the results indicate that the products are
associated with lower perceptions of expensiveness when they are advertised on
the left side of a magazine’s double page (Mleft = 3.02, Mright = 3.71, F(1, 400) =
37.855, p < .001). The ad location did not affect the perceptions of product
quality (Mleft = 3.95, Mright = 3.76, F(1, 400) = 1.768, NS). With regards to the
attitudes, the left location turned out to be advantageous (Mleft = 4.75, Mright =
4.13, F(1, 400) = 14.572, p < .001).
On the contrary, for expensive brands, the ad location affected the
perceptions of product quality (Mleft = 5.16, Mright = 5.80, F(1, 405) = 30.520, p
< .001) but had no impact on the perceptions of expensiveness (M left = 5.53,
Mright = 5.60, F(1, 405) = .228, NS). Regarding the attitudes, the right location
outperformed the left location (Mleft = 4.53, Mright = 5.38, F(1, 405) = 28.918, p
< .001).
194 Schneider and Gierl

Table 2: Effects of the ad’s location on perceptions of quality, perceptions of


expensiveness, and attitudes toward the product depending on the brand’s price level

To gain deeper insights into the mechanism underlying these results, we


estimated a dual-mediation model. We used the ad’s location as a binary
independent variable (0 = left side, 1 = right side). We included the perceptions
of quality and expensiveness as mediating variables. These variables were
moderately correlated (R = .682) which indicates that the analysis suffers from
collinearity problems. The attitudes served as the dependent variable. We split
the data into two subsamples and estimated the model for the less expensive
brands and the expensive brands separately by using a procedure suggested by
Preacher and Hayes (2004). These analyses (Figure 8) showed that the ad
location affects attitudes via perceptions of expensiveness in the case of less
expensive brands (products appear even less expensive when the ad is located on
the left side) while the ad location influences attitudes via perceptions of quality
in the case of expensive brands (products are associated with even higher quality
perceptions when the ad is located on the right side). In summation, H1 and H2
are partly supported and H3a and H3b are in line with our data.
Put It on the Right Side 195

Figure 8: Results of a dual-mediation model used to explain the effect of the ad’s
left/right location on attitudes toward the promoted product

3.7 Interpretation
We found that the location of an ad on a double page affects attitudes toward
the product. Moreover, we found that the sign of the effect depends on the price
level of the promoted product. For products of less expensive brands, the left
location affects attitudes positively via lower perceptions of expensiveness. For
products of expensive brands, the right location increases attitudes via higher
perceptions of quality. However, the results of the mediation analyses indicate
the relevance of additional processes beneath the indirect effects via perceptions
of expensiveness and quality because there are high residual direct effects and,
admittedly, the estimated size of the indirect effects is rather low. In summation,
we found evidence for the location effect depending on the brand’s price level
but our interpretation of the effect based on considerations about the divergent
relevance of price vs. quality only reveals a part of the mechanism.

3.8 Limitations
Basically, we can find an effect of location depending on the brand’s price
level. However, we cannot perfectly explain the effect thus far. The weak
indirect effects via perceptions of quality and expensiveness indicate that we
neglected another mechanism that actually causes the effect of location on
product evaluations. Probably, the single-item measures used for assessing the
perceptions were poor. Moreover, the concept of “quality” itself might have
been too general. For instance, the evaluation of shoes and clothing may be
based on perceptions of social conspicuousness; the used term “quality” may
have caused connotations such as durability and persistence that are less
important for evaluating the shoes and clothing. Another crucial aspect is the
196 Schneider and Gierl

choice of the products considered in our experiment (i.e., cosmetics, shoes, and
clothing). For instance, if we had considered groceries such as diet products,
consumers might infer the amount of the calories of the products depending on
the left/right location of the ad. If we had considered fruits, the ad’s location
might be used to infer conclusions about the vitamin C content (Cai et al., 2012).
Thus, for other product categories, different perceptions (besides perceptions of
expensiveness and quality) may be affected by choosing the location of an ad on
a double page.

4 Implications for Practice

In general, we recommend marketers to pay attention to location effects. You


may validate the effect described in this paper yourself through numerous
replications. Take the coins from your purse and ask your colleague at lunch “to
sort the coins.” Most probably, s/he will sort them from less valuable coins on
the left to the most valuable coins on the right side. Give her/him a folding
yardstick and ask your colleague “to measure the length of the table.” According
to the expectations, this person will locate the beginning of the yardstick on the
table’s left side from her/his perspective. Ask her or him “to count with one’s
fingers from one to five.” Probably, a right-handed person turns her/his left hand
and uses her/his thumb to indicate the number “one.” Through these experiences,
marketers may trust in the existence of the general tendency in individuals to
construe a virtual left-right continuum for sorting and evaluations tasks.
However, deciding about problems such as locating an advert on the left or
right side of a magazine needs more knowledge. From our study, we conclude
that the optimum decision about whether the print ad is located on the left or
right side of a double page depends on the product’s price and quality. When the
(low) price is to be emphasized, the left location is appropriate. When (high)
quality is to be signaled, the right location helps achieving the marketer’s goal.
We recommend using the left location to promote products of less expensive
brands and choosing the right location for promoting products of expensive
brands. However, we restrict these recommendations to the categories of fashion
goods because we focused on these categories in our experiment.
Put It on the Right Side 197

5 References

Brysbaert, M. (1995), “Arabic Number Reading: On the Nature of the Numerical Scale
and the Origin of Phonological Recoding,” in: Journal of Experimental Psychology:
General, 124(4), 434-452.
Cai, F.; Shen, H. and M. K. Hui (2011), “The Number-Location Association and its
Marketing Implications,” in: R. Ahluwalia; T. L. Chartrand and R., K. Ratner (eds.),
Advances in Consumer Research (Vol. 39), Duluth, MN: Association for Consumer
Research, 471-472.
Cai, F.; Shen, H. and M. K. Hui (2012), “The Effect of Location on Price Estimation:
Understanding Number-Location and Number-Order Associations,” in: Journal of
Marketing Research, 49(3), 718-724.
De Hevia, M. D. and E. S. Spelke (2009), “Spontaneous Mapping of Number and Space
in Adults and Young Children,” in: Cognition, 110(2), 198-207.
Dehaene, S.; Bossini, S. and P. Giraux (1993), “The Mental Representation of Parity and
Number Magnitude,” in: Journal of Experimental Psychology: General, 122(3), 371-
396.
Dehaene, S.; Dupoux, E. and J. Mehler (1990), “Is Numerical Comparison Digital?
Analogical and Symbolic Effects in Two-Digit Number Comparison,” in: Journal of
Experimental Psychology: Human Perception and Performance, 16(3), 626-641.
Ferguson, L. W. (1935), “The Importance of the Mechanical Features of an
Advertisement,” in: Journal of Applied Psychology, 19(5), 521-526.
Fias, W. (2001), “Two Routes for the Processing of Verbal Numbers: Evidence from the
SNARC Effect,” in: Psychological Research, 65(4), 250-259.
Finn, A. (1988), “Print Ad Recognition Readership Scores: An Information Processing
Perspective,” in: Journal of Marketing Research, 25(2), 168-177.
Gevers, W.; Reynvoet, B. and W. Fias (2002), “The Mental Representation of Ordinal
Sequences is Spatially Organized,” in: Cognition, 87(3), B87-B95.
Hardie, B. G. S.; Johnson, E. J. J. and P. S. Fader (1993), “Modeling Loss Aversion and
Reference Dependence Effects on Brand Choice,” in: Marketing Science, 12(4), 378-
394.
Heath, T. B. and S. Chatterjee (1995), “Asymmetric Decoy Effects on Lower-Quality
versus Higher-Quality Brands: Meta-Analytic and Experimental Evidence,” in:
Journal of Consumer Research, 22(3), 268-284.
Heath, T. B.; Ryu, G.; Chatterjee, S.; McCarthy, M. S.; Mothersbaugh, D. L.; Milberg, S.
and G. J. Gaeth (2000), “Asymmetric Competition in Choice and the Leveraging of
Competitive Disadvantages,” in: Journal of Consumer Research, 27(3), 291-307.
Preacher, K. J. and A. F. Hayes (2004), “SPSS and SAS Procedures for Estimating
Indirect Effects in Simple Mediation Models,” in: Behavior Research Methods,
Instruments, & Computers, 36(4), 717-731.
Simonson, I.; Nowlis, S. and K. Lemon (1993), “The Effect of Local Consideration Sets
on Global Choice between Lower Price and Higher Quality,” in: Marketing Science,
12(4), 357-376.
Simonson, I. and A. Tversky (1992), “Choice in Context: Tradeoff Contrast and
Extremeness Aversion,” in: Journal of Marketing Research, 29(3), 281-295.
198 Schneider and Gierl

Smith, E. G.; Neijens, P. C. and R. Heath (2013), “The Differential Effects of Position,
Ad and Reader Characteristics on Readers’ Processing of Newspaper Ads,” in:
International Journal of Advertising, 32(1), 65-84.
Tversky, A. and D, Kahneman (1991), “Loss Aversion in Riskless Choice: A Reference-
Dependent Model,” in: The Quarterly Journal of Economics, 106(4), 1039-1061.
Walker, J. and D. Cardillo (1998), “Does Size Matter (or Position, or Colour, or Context
…)?” in: Admap, January, 26-28.
The Influence of Majority Agreements on Attitudes
Franziska Oefele and Heribert Gierl

1 Introduction

There are two types of consensus information which express the viewpoint of
the majority (Mackie, 1987, p. 42; Freling and Dacin, 2010, p. 163): An
individual can be informed that the majority of people are in favor of a product,
support a topic, or are proponents of an idea. Alternatively, the individual can
get the information that the majority of people dislike a product, reject a topic,
or are opponents of an idea. The first case of consensus can be denoted as
majority agreement and the latter case as majority disagreement. Numerous
companies use information about majority agreement for promoting their
products. In Figure 1, we present some examples to illustrate what information
about majority agreement looks like in print advertisements or on company
websites. For instance, in one of its ads, the Nivea brand company highlights the
results obtained from 2,347 readers of the Freundin magazine that eight out of
ten women would recommend “Nivea Pure and Natural Facial Creme” to their
friends. To provide another example, in one of its ads, the Clearasil brand
company states that 541 readers of the Glamour magazine tested “Clearasil
Stayclear” and 81% of them found it “surprisingly soft to the skin.” These
examples illustrate how companies refer to surveys and how they include the
results consisting of high portions of the promoted product’s supporters into
their advertising campaigns (i.e., majority agreement). In the following, we
summarize findings from prior research on the effects of majority-agreement
information and derive a gap of research. In our experiment, we investigated the
effectiveness of the presence of majority-agreement information on the attitudes
toward the promoted product depending on whether an argument in favor of this
product is presented in the advertisement or not. Our conceptual model is shown
in Figure 2.

© Springer Fachmedien Wiesbaden 2017


G. Christodoulides et al. (Hrsg.), Advances in Advertising Research (Vol. VII),
European Advertising Academy, DOI 10.1007/978-3-658-15220-8_15
200 Oefele and Gierl

Figure 1: Examples of majority-agreement information contained in advertisements


The Influence of Majority Agreements on Attitudes 201

Figure 2: Conceptual model

2 Previous Research

Until now, there have been some studies that investigated the effectiveness of
consensus information. This research can be separated into three streams.

2.1 Comparison of Majority Agreement vs. Disagreement in the HSM


Model
Research findings: In the tradition of the heuristic-systematic model of
persuasion, researchers compared the effect of majority disagreement to a
product or to a topic to the effect of majority agreement on attitudes toward that
product or topic, respectively, and included argument strength and personal
relevance of the product/topic as additional variables. In the following, we
present the findings of Maheswaran and Chaiken (1991), Darke et al. (1998),
and Martin et al. (2007). We summarize the study details in Table 1 and the
findings in Figure 3. Similar studies have been conducted by Martin and
Hewstone (2003), Erb et al. (2006), and Martin et al. (2006). Overall, this
research found that information about majority agreement is advantageous
compared to information about majority disagreement. Moreover, this research
revealed that the effect of majority agreement on attitudes is not contingent on
the consumer’s personal relevance of the product/topic.
Explanations: The latter finding is explained as follows: Basically, when
individuals are exposed to an evaluation or decision task, they want to arrive at
an accurate evaluation or a correct decision which implies high cognitive effort
while collecting and processing information and – simultaneously – wish to
invest little cognitive effort for conducting this task (“economic concern”).
When a product or a topic is highly relevant for the individual, achieving high
accuracy is more important than reducing one’s cognitive effort. When a product
or a topic has low personal relevance, reducing cognitive effort is more
important than achieving high evaluation accuracy. Under the condition of high
personal relevance, individuals are willing to spend a greater deal of effort and,
thus, possess abilities to process arguments as well as heuristic cues such as
202 Oefele and Gierl

consensus information. In conditions of low personal relevance, due to the


“sufficiency principle,” individuals are likely to rely on easy-to-process
information. Because consensus information can easily be processed, individuals
use this piece of information in conditions of low and high personal relevance of
the product/topic.

Table 1: Experimental designs of authors testing the effect of majority-agreement vs.


majority disagreement information in the stream of the heuristic-systematic model
The Influence of Majority Agreements on Attitudes 203

2.2 Comparison of the Majority-Agreement to the Consensus-Information-


Absent Condition
Research findings: Other researchers (e.g., Erb et al., 1998; Codaccioni and
Tafani, 2011) compared the impact of majority agreement to a condition where
consensus information was absent. The presence of information about majority
agreement turned out to be beneficial for promoting a product or a project
compared to the consensus-information-absent condition.
204 Oefele and Gierl

Figure 3: The effect of majority agreement/disagreement in combination with argument


strength and the consumer’s personal relevance of the product/topic
The Influence of Majority Agreements on Attitudes 205

Explanations: Literature on this topic discusses several reasons why the


presence of majority-agreement information positively affects evaluations. First,
Moscovici (1980) stated that there is a general need by individuals for social
recognition and acceptance. If people adopt the opinion of the majority they can
optimize social recognition when the relevant opinion is expressed in public or
when the behavior is visible to others. This response is also denoted as
compliance to the majority. Second, Mackie (1987, p. 42) hypothesized that the
opinion of the majority is regarded “as reflecting objective reality.” Individuals
are likely to believe that the true value of an issue is mirrored by the impression
of the majority about that issue. This belief is also denoted as the consensus-
implies-correctness heuristic. Similarly, Cialdini (2007, p. 105) argued that
people believe that they are able to reduce the danger of making a false decision
when they adopt the behavior of the majority: “The greater the number of people
who find an idea correct, the more a given individual will perceive the idea to be
correct.” Thirdly, the positive effect of majority agreement can be the result of a
process of misattribution. Generally, individuals are interested in looking for
causes for phenomena they observe. In particular, when consumers gain
information about the preferences of the majority, the cause and the consequence
could be mistaken. If something is desirable, then numerous people may strive
for it. Misattribution occurs if consumers infer that an item is desirable from the
information about a high rate of people preferring it. Fourthly, Mackie (1987, p.
42) argued as follows: If people disagree with the position of the majority, they
would feel the need to think about why they refuse to agree, which requires
cognitive effort. On the contrary, when people agree with the position of the
majority, they are unlikely to feel the need to think about why they should accept
this position. Thus, this task needs less cognitive activity. Consequently, if
people wish to evaluate a product/topic without investing much cognitive effort,
they are likely to follow the position of the majority. Fifthly, individuals may
have learned through a socialization process that they receive social approval
when they adopt good manners that are practiced by many other people.
Otherwise they received social disapproval if they refused to adopt these good
manners. Thereby, individuals may have learnt that “the majority is right” and
transfer this relation to different fields such as product evaluation.

2.3 Variables Moderating the Effectiveness of Majority-Agreement


Information
Literature also investigated factors that increase the effect of majority-agreement
information. Factors that increase the effect of majority-agreement information
are also investigated through different literature.
Susceptibility for interpersonal influence: Chang (2012) investigated the
moderating role of the individual’s susceptibility for interpersonal influence and
206 Oefele and Gierl

found that this variable amplified the effect of majority agreement; however, in
another study, this author found no moderating effect of the individual’s general
predisposition as being interdependent or independent from the position of
relevant others, which is rather similar to the concept of susceptibility (Chang,
2010).
Level of majority agreement: As another moderating variable, the level of
majority agreement was considered. For instance, in advertisements, the
company could mention that 90%, 95%, or 100% of the group whose opinion is
reported recommended the product. Regarding this issue, prior research
provided mixed results (Beltrami and Evans, 1985; Freling and Dacin, 2010).
Beltrami and Evans (1985) mentioned that consumers are highly skeptical about
the truth of very high percentages of majority agreement, consider them as not
believable, and use this impression for devaluating the promoted product.
Peer-group-based vs. general-public-based majority: For instance, a reader
of the Glamour magazine could interpret consensus information derived from a
survey under readers of this magazine as peer-group information. On the other
hand, this reader might interpret consensus information derived from a survey
among a representative sample of consumers as the general-public agreement.
Chang (2012) considered this aspect as a moderating variable for the
effectiveness of majority-agreement information but did not find an effect.
Sample size for assessing majority agreement: Another moderator is the size
of the sample which is used to infer majority agreement. As our examples in
Figure 1 show, the reported sample size could be rather small (e.g., 291 in the
case of the ad of the Dove brand) or very large (e.g., 5,000 in the case of one of
the shown ads of the Nivea brand). Beltrami and Evans (1985) and Darke et al.
(1998) focused on this aspect and reported mixed results. The firstly cited
authors found that the sample size increased the effect of majority agreement in
the condition of high personal relevance of the promoted product.

3 Gap in Research and Hypotheses

Based on these findings, we looked for a gap in research which further studies
could close. The findings from the studies in the tradition of the heuristic-
systematic model of persuasion compare majority agreement to strong majority
disagreement. Although these studies are highly valuable for theory development
and testing, considering majority disagreement as the reference for assessing the
effect of majority agreement is unsuitable for deriving recommendations whether
to use this piece of information or not. Advertisers are likely interested to gain
insight into the effect of providing information about majority agreement
compared to not providing such information. Thus, majority agreement should
be compared to the absence of such information.
The Influence of Majority Agreements on Attitudes 207

In the studies that actually compared the effectiveness of majority-agreement


information to a control condition (where consensus information is absent), the
researchers did not consider the relevance of simultaneously presented
arguments and did not compare low- to high-relevancy conditions. Because
majority-agreement information is easy to process (at least in the cases of real
ads shown in Figure 1, in the cases of studies summarized in Figure 3, and in the
cases of the stimulus material we intended to use), considering different levels of
personal relevancy is of seemingly less importance. Thus, we focus on
additionally considering the presence vs. absence of arguments in favor of the
product. We test:
H1: The presence of majority-agreement information in advertisements
compared to its absence improves attitudes toward the promoted product when
arguments in favor of the product are absent.
H2: The presence of majority-agreement information in advertisements
compared to its absence improves attitudes toward the promoted product when
arguments in favor of the product are present.

4 Study
Experimental design: We conducted an experiment and used a 2 (majority-
agreement information: present or absent) × 2 (argument in favor of the product:
present or absent) factorial between-subjects design. We replicated this
experiment for six products (Samsung laptop, Opel Corsa, Garnier cleansing
toner, Nivea body lotion, Kasai makeup, and Biorepair toothpaste). Because
prior research had shown that the individuals’ predisposition to be generally
interdependent vs. independent from the opinion of others and that peer-group-
based vs. general-public-based majority agreements are no relevant moderators,
we did not include these factors.
Test stimuli: We created four versions of print advertisements for each of
these products. All ad versions contained an image of the promoted product and
a slogan. The versions differed regarding the presence or absence of the
majority-agreement information and regarding the presence or absence of a
product argument. We considered different types of arguments (e.g., search
claims such as “Always a clear view due to the high resolution screen” in the
case of a laptop, experience claims such as “The makeup with nourishing mat
effect. Covers perfectly for twelve hours” in the case of a makeup, and credence
claims such as “Protects the environment due to low emissions” in the case of a
car; see Table 2).
208 Oefele and Gierl

Table 2: Arguments included in the print advertisements

However, we did not systematically manipulate the type of argument. We


used the findings from prior research to decide on further aspects. For instance,
as noted above, the level of majority agreement should not be extremely high.
Otherwise, consumers might be skeptical about the validity of the information.
That is why we used figures around 90% to indicate the existence of a majority
of people who shared the same favorable opinion toward the promoted product.
We referred to sizes ranging between 300 and 500 consumers for assessing
majority agreement. In Figure 4 and Figure 5, we show the created ad versions
for the Nivea brand and the Kasai brand.
The Influence of Majority Agreements on Attitudes 209

Figure 4: Examples of test stimuli for the Nivea brand


210 Oefele and Gierl

Figure 5: Examples of test stimuli for the Kasai brand

Sample: In total, 760 female students from a university located in the South
of Germany (Mage = 23.58 years, SD = 3.703) took part in our experiment. Data
collection took place between 2011 and 2015. However, for each product, the
data was collected within a small span of time.
Procedure: For the Opel Corsa and the Samsung laptop, data was collected
with the use of an online survey tool. The data for the other products was
gathered through face-to-face interviews. The test persons were exposed to one
version of the 24 advertisements and could view it for as long as they wished.
The Influence of Majority Agreements on Attitudes 211

Measures: After viewing, the participants indicated their attitudes toward the
promoted product by agreeing or disagreeing to “attractive,” “likeable,” “good,”
and “appealing” on a seven-point scale (α = .923). Imhoff and Erb (2009)
recommended considering the consumer’s need for uniqueness when the effect
of majority agreement is investigated. Thus, we adopted some statements from
Tepper Tian et al. (2001, p. 55) to assess this construct in the questionnaire. The
test persons agreed or disagreed to “I collect unusual products as a way of telling
people I am different,” “I rarely act in agreement with what others think are the
right things to buy,” and “When products and brands I like become extremely
popular, I lose interest in them” on a seven-point scale (α = .687)). Consumer
differences regarding this control variable did not affect our findings at all.
Manipulation check: We analyzed whether including the argument (arg) into
the advertisement actually increased product attitudes. We considered the
conditions where majority-agreement information (mai) was absent and found
that this manipulation worked as intended (Marg absent, mai absent = 3.76, Marg present, mai
absent = 4.46, F(1; 389) = 25.214, p < .001). We refrained from conducting a
manipulation check for the majority-agreement information because the
manipulation was obvious to the test persons.
Results and hypotheses tests: We calculated the mean values and the
standard deviations for each test condition and show the results in Table 3. We
collapsed data across the products because the effects of the majority-agreement
information and the presence of an argument in favor of the product did not
differ systematically across the products. In Hypothesis 1, we expected a
positive effect of the presence (vs. absence) of majority-agreement information
in the argument-absent condition. The results are in line with this presumption
(Marg absent, mai absent = 3.76, Marg absent, mai present = 4.49, F(1; 355) = 22.532, p <
.001). In Hypothesis 2, we also predicted this effect for the argument-present
condition. However, we did not find this effect (Marg present, mai absent = 4.46, Marg
present, mai present = 4.43, F(1; 401) = .042, NS) and thus have to reject this
presumption.
212 Oefele and Gierl

Table 3: The effect of majority-agreement information depending on the presence vs.


absence of an argument in favor of the promoted product

Interpretation: We presume that our experimental setting did not motivate


the test persons to deal intensely with the promoted product. Thus, they simply
scrutinized whether there is or is no information in favor of the product to justify
a more positive evaluation and did not care about the kind and the number of
favorable pieces of information.
Limitations: The findings are based on data from female students and, thus,
are not generalizable to other consumer segments. Moreover, we only
considered few products in our survey. Finally, we did not manipulate argument
strength or the consumer’s personal relevance of the product to gain deeper
insights into the conditions of effectiveness of majority-agreement information.

5 Implications for Practice

Our study indicates that the use of majority-agreement information is only


effective if arguments in favor of the product are not present in advertisements.
When any argument is provided, majority-agreement information is ineffective.
Surprisingly, individuals don’t add up both pieces of information. One
reason could be the low believability of high consensus information: Individuals
use such information only when no other information is available. Therefore,
when companies have difficulties to explicitly express benefits of their products
compared to competitor products (e.g., suppliers of all-purpose glue), the use of
majority-agreement information seems to be helpful for persuasion purposes.
Next and also surprising, the effect of the presence of majority-agreement
information does not depend on whether the presented argument can be
classified as search, experience, or credence claim. Economics of information
would suggest that majority-agreement information is effective when it
The Influence of Majority Agreements on Attitudes 213

accompanies experience claims (because then it validates the argument).


Contrarily to this position, we recommend using majority-agreement information
for any type of product when effective arguments are not available
independently of whether the main product’s benefit can be seen before
consumption, after consumption, or not seen even after consumption. We learnt
from our results that individuals like to get a cue (independently of whether it is
an argument or a heuristic cue such as majority-agreement information in favor
of the product) that enables them justifying a favorable evaluation of the
promoted product. This effect exists for fast-moving-consumer goods such as
toothpaste as well as for expensive categories such as automobiles.

6 References
Beltrami, R. F. and K. R. Evans (1985), “Perceived Believability of Research Results
Information in Advertising,” in: Journal of Advertising, 14 (3), 18-31.
Chang, C. (2010), “Making Unique Choices or Being Like Others: How Priming Self-
Concepts Influences Advertising Effectiveness,” in: Psychology and Marketing, 27
(4), 399-415.
Chang, C. (2012), “Effectiveness of Consensus Information in Advertising: The
Moderating Roles of Situational Factors and Individuals,” in: Journal of Business and
Psychology, 27 (4), 483-494.
Cialdini, R. (2007), “The Psychology of Persuasion,” New York: Collins.
Codaccioni, C. and E. Tafani (2011), “Advertising Effectiveness as a Function of
Numerical Support: From Majority Compliance to Minority Conversion,” in:
European Review of Applied Psychology, 61 (2), 77-87.
Darke, P. R.; Chaiken, S.; Bohner, G.; Einwiller, S.; Erb, H.-P. and D. Hazlewood
(1998), “Accuracy Motivation, Consensus Information, and the Law of Large
Numbers: Effects on Attitude Judgment in the Absence of Argumentation,” in:
Personality and Social Psychology Bulletin, 24 (11), 1205-1215.
Erb, H.-P.; Bohner, G.; Hewstone, M.; Werth, L. and M.-A. Reinhard (2006), “Large
Minorities and Small Majorities: Interactive Effects of Inferred and Explicit
Consensus on Attitudes,” in: Basis and Applied Social Psychology, 28 (3), 221-231.
Erb, H.-P.; Bohner, G.; Schmälzle, K. and S. Rank (1998), “Beyond Conflict and
Discrepancy: Cognitive Bias in Minority and Majority Influence,” in: Personality and
Social Psychology Bulletin, 24 (6), 620-633.
Freling, T. H. and P. A. Dacin (2010), “When Consensus Counts: Exploring the Impact
of Consensus Claims in Advertising,” in: Journal of Consumer Psychology, 20 (2),
163-175.
Imhoff, R. and H.-P. Erb (2009), “What Motivates Nonconformity? Uniqueness Seeking
Blocks Majority Influence,” in: Personality and Social Psychology Bulletin, 35 (3),
309-320.
Mackie, D. M. (1987), “Systematic and Nonsystematic Processing of Majority and
Minority Persuasive Communications,” in: Journal of Personality and Social
Psychology, 53 (1), 41-52.
214 Oefele and Gierl

Maheswaran, D. and S. Chaiken (1991), “Promoting Systematic Processing in Low-


Motivation Settings: Effect of Incongruent Information on Processing and Judgment,”
in: Journal of Personality and Social Psychology, 61 (1), 13-25.
Martin, R. and M. Hewstone (2003), “Majority versus Minority Influence: When, Not
Whether, Source Status Instigates Heuristic or Systematic Processing,” in: European
Journal of Social Psychology, 33 (3), 313-330.
Martin, R.; Hewstone, M. and P. Y. Martin (2007), “Systematic and Heuristic Processing
of Majority- and Minority-Endorsed Messages: The Effects of Varying Outcome
Relevance and Levels of Orientation on Attitude and Message Processing,” in:
Personality and Social Psychology Bulletin, 33 (1), 43-56.
Martin, R.; Martin, P. Y.; Smith, J. R. and M. Hewstone (2007), “Majority versus
Minority Influence and Prediction of Behavioral Intention and Behavior,” in: Journal
of Experimental Social Psychology, Vol. 43 (5), 763-771.
Moscovici, S. (1980), “Toward a Theory of Conversion Behavior”, in: L. Berkovitz (ed.),
Advances in Experimental Social Psychology (Vol. 13), New York: Academic Press,
209-239.
Tian, K. T.; Bearden, W. O. and G. L. Hunter (2001), “Consumers’ Need for
Uniqueness: Scale Development and Validation,” in: Journal of Consumer Research,
28 (1), 50-66.
The Effects of Message Framing and Reference Points of
Public Service Announcements on Bystander Intervention in
College Students’ Binge-Drinking

Kang Li and Nora Rifon

1 Introduction
Binge drinking, a serious social issue, refers to excessive drinking in a short
period of time (i.e., five or more alcoholic beverages for men, and four or more
for women, in about 2 hours) (Pilling and Brannon 2007, www.cdc.gov). A great
deal of social harm and health problems are associated with binge drinking.
Besides the diseases caused by excessive drinking (e.g., liver disease,
neurological damage), it has been found that there is a high correlation between
binge drinking and other risky acts such as drunk driving, unprotected or
unplanned sexual activities, and marijuana use (Lee and Bichard 2006, Pilling
and Brannon 2007). Pilling and Brannon (2007) pointed out that there are more
problem drinkers in the college population than in any other age group in the
United States. More than half of the college students in the US qualify as binge
drinkers, and 23% of undergraduates participate in binge drinking more than
once a week (Wechsler et al. 2000). As a risky alcohol abuse behavior, binge
drinking continually alarms parents, schools, and lawmakers (Lee and Bichard
2006).
There are a lot of Public Service Announcements (PSAs) designed to reduce
binge drinking. However, usually the designs of anti-alcohol PSAs focus on
persuading problem drinkers to stop drinking or drink less; few of the messages
are designed to encourage bystanders to intervene in binge drinking when they
see others drinking too much. This study will explore which message designs are
more effective in encouraging intervention behaviors in binge drinking, known
as a behavior of “bystander intervention” (White and Malkowski 2014).
Bystander intervention stresses the importance of getting bystanders to notice
risky behaviors and provide help when it is necessary (White and Malkowski
2014). Banyard et al. (2007) suggested that bystander intervention is important
in dealing with high-risk situations. And research also showed that risks can be
truly reduced by intervention (White and Malkowski 2014). White and
Malkowski (2014) claimed that college students taking responsibility to keep
each other safe would have positive effects on campus climate, including
reducing problem activities such as binge drinking.

© Springer Fachmedien Wiesbaden 2017


G. Christodoulides et al. (Hrsg.), Advances in Advertising Research (Vol. VII),
European Advertising Academy, DOI 10.1007/978-3-658-15220-8_16
216 Li and Rifon

This study investigates how to make effective anti-binge-drinking PSAs from


the angle of encouraging bystander intervention behaviors among college
students, rather than from the traditional angle of purely discouraging binge
drinking behaviors by targeting the drinkers themselves. According to the theory
of planned behavior (Ajzen 1991), behavioral control can be strongly predicted
by behavioral intention. Therefore, this study will specifically investigate the
effects of message framing (loss vs. gain) and reference points (self-other
reference vs. other reference) on people’s intentions to perform bystander
intervention behaviors.
There are several expected scholarly contributions of this study. First,
message framing will be examined from a different perspective in the anti-binge-
drinking context. Most literature on message framing regarding anti-drinking ads
examines the effectiveness of message framing on reducing an undesired
behavior (e.g., binge drinking) by targeting drinkers themselves, no matter
whether the PSAs address the harms of excessive drinking or the benefits of not
drinking excessively. In contrast, with the same goal of preventing binge
drinking, this study draws from prospect theory to examine the effects of
message framing on reducing the undesired behavior by targeting bystanders
instead of the drinkers themselves, thus encouraging a desired behavior (i.e.,
bystander intervention) instead of discouraging the undesired one. This enriches
the message framing literature in anti-alcohol contexts. Second, reference points
in PSAs on health or risk issues usually refer to self-reference versus self-other
reference because issues such as drinking and smoking always primarily affect
oneself, (e.g., Cheng, Woon and Lynes 2011, Loroz 2007). This study will fill a
research gap by investigating the different types of reference points which have
been rarely researched before (i.e., other reference vs. self-other reference)
because intervention behavior will primarily affect others. Third, this study will
present an exploration the interaction effects of message framing and reference
points, which have rarely been considered in previous studies. Fourth, this study
will contribute further understandings of participants’ involvement with the
message and attitude toward ads, which will be used as two mediators. Finally,
gender differences will be considered in this study to explore whether or not
gender influences individuals’ involvement with ads.
Furthermore, this study will provide practical suggestions for making
effective anti-binge-drinking PSAs from the perspective of encouraging
bystander intervention, and thereby, increasing college students’ awareness of
taking responsibility for keeping each other safe and intervening in excessive
drinking. This will help reduce binge-drinking among college students and have
positive impacts on students’ health and safety. The current research will also
benefit parents, educators, policy makers, and society in general because binge
The Effects of Message Framing and Reference Points 217

drinking reduction will help reduce other risky behaviors among college students
and also decrease crime rates.

2 Literature Review and Hypotheses

2.1 Message Framing


The “right message” is critical to an effective and successful campaign
(Cheng, Woon and Lynes 2011). One strategy for developing a successful
message is message framing (Cheng, Woon and Lynes 2011), which has often
been considered in terms of gain-framed and loss-framed approaches. Gain
framing stresses the benefits or positive consequences of conducting a behavior
in a certain decision context, while loss framing emphasizes the negative
outcomes of refusing to undertake a behavior (Schneider 2006). According to
prospect theory (Kahneman and Tversky 1979), people’s preference for an
option is influenced by how that option is framed. Kahneman and Tversky
(1979) suggested that people tend to be risk-averse to secure gains and risk-
seeking to avoid losses (Loroz 2007). Specifically, in health- and risk-related
messaging, a loss- or negative-framed message may be more effective in
promoting detection behaviors that may invoke risks because people fear to lose
(e.g., doing a breast examination, which is perceived to be risky because it can
lead to the discovery of cancer) (Loroz, 2007). Whereas a gain-framed message
may be more effective in advocating prevention behaviors to maintain the status
quo due to people’s need of securing gains (e.g., sunscreen use, watching one’s
diet, smoking cessation, and seat belt use, all of which can be considered as
behaviors aimed at maintaining one’s current health condition) (Rothman and
Salovey 1997, Schneider 2006, Cheng, Woon and Lynes 2011, Gallagher and
Updegraff 2012). This detection versus prevention paradigm is generally
supported by most health- or risk-related message framing research (Schneider
2006).
In the binge drinking case, although intervening in binge drinking cannot be
called a detection behavior, but it is still primarily risk-seeking (e.g. breaking
local social norms, possible physical attack, potential loss of friendship or
personal reputation) to avoid losses (e.g., harms associated with binge drinking).
Moreover, from another perspective, intervening in binge drinking is an
immediate behavior aimed at altering the current drinking status instead of
maintaining the status on the occasion of binge drinking. Therefore, loss-framed
messages should be more effective in promoting bystander intervention
behaviors in binge drinking contexts.
218 Li and Rifon

H1: A loss-framed message is more effective in leading to a stronger intention


of bystander intervention in binge drinking than a gain-framed message.

2.2 Reference Points in Anti-Drinking Messages


Reference points here refer to who the behaviors are intended to impact
according to the emphasis in the message. In most health-related ads, self-
referencing (effects on self) is the main approach (e.g., you may get lung cancer
if you keep smoking), which addresses how the behaviors influence the recipient
(Cheng, Woon and Lynes 2011). Self-other referencing (effects on both self and
others) is another approach in health-related ads (e.g., smoking may harm your
own health as well as your family’s health because of the second-hand smoke),
which indicates that the behaviors influence both the recipients and others
(Cheng, Woon and Lynes 2011). A number of studies have investigated the
impacts of self-reference on persuasion (e.g., Burnkrant and Unnava 1989,
Debevec and Romeo 1992, Burnkrant and Unnava 1995, Meyers-Levy and
Peracchio 1996). Messages associated with the self have been found to be highly
involving, which may be due to the fact that the network of associations with the
self is one of the most developed and richest networks in human memory
(Rogers, Kuiper and Kirker, 1977, Kuiper and Rogers 1979). Anderson,
Glassman and Gold (1998) suggested that mental representations of self are
more accessible, distinctive, and rich in memory than mental representations of
others. Hence, a self-referencing message is more likely to be highly involving
than a self-other referencing message, which in turn results in more available
cognitive resources for processing the message (Loroz 2007).
As for the present study, since the intervention behaviors always involve
others, other-reference (effects on others) would be a necessary approach, which
emphasizes how the intervention behaviors can affect others’ lives. Meanwhile,
self-other reference would also be an appropriate approach in messages
advocating bystander intervention behaviors because intervention will definitely
involve the two sides of both self and others. According to the literature, people
have higher involvement with and are more likely to be persuaded by
associations with self than associations with others (Loroz 2007), so self-other
referencing points should be more persuasive than a reference point only to
others.

H2: A message with self-other referencing is more effective in leading to a


stronger intention of bystander intervention in binge drinking than a message
with other-referencing.

Loroz (2007) demonstrated that there is an interaction effect of reference


points and message framing in prosocial persuasion contexts. Specifically, in a
The Effects of Message Framing and Reference Points 219

negatively-framed message, a self-referencing point is more persuasive than a


self-other referencing point, whereas in a positively-framed message, the
opposite is true. According to this literature, there might be an interaction as
well between message framing (loss vs. gain) and reference points (others
referencing vs. self-other referencing) in this study. However, Loroz’s (2007)
research focuses on recycling issues in environmental contexts and it
investigated different reference points from the present study, thus, his/her
findings might not completely apply to the current binge drinking intervention
case. Moreover, few other studies have been found to examine the interaction
effects between reference points and message framing (Loroz 2007) so the
interaction effects of message frames and reference points still need more
research.

RQ1: Is there an interaction effect between message framing (loss vs. gain) and
reference points (other-referencing vs. self-other referencing) on people’s
intentions of bystander intervention in binge drinking?

2.3 Involvement with Ads and Attitude Toward Ads


Involvement refers to the level of association an individual has toward a
thing based on his/her needs, interests, and values (Zaichkowsky 1985, Tsai and
Tsai 2006). Though there are multiple types and definitions of involvement
across disciplines (Paek et al. 2011), involvement is prevalently viewed as
perceived personal relevance, which refers to a self-related perception (Celsi and
Olson 1988). In other words, consumers’ level of involvement with an action or
situation depends on the degree to which they perceive the action or situation to
be personally relevant, or whether the action or situation influences them in
some way to achieve their personal goals or values (Celsi and Olson 1988).
It has been found that involvement plays an important role in message
processing and persuasion (Paek et al. 2011). Celsi and Olson (1988)
demonstrated that different levels of involvement could influence consumers’
attention and comprehension processes differently. The Elaboration Likelihood
Model (ELM) suggested that there are two routes of persuasion: central and
peripheral (Petty and Cacioppo 1986). Under a high-involvement condition,
people may take the central route to process the message and pay more attention
to message arguments; while under a low-involvement condition, people may
take a peripheral route and change their opinions based on some cues or source
characteristics (Paek et al. 2011). When sound arguments are designed in a
persuasive message, higher levels of involvement lead to greater message-
relevant thinking and more effective persuasion (Petty, Cacioppo and Goldman
1981, Geary et al. 2008).
220 Li and Rifon

Celsi and Olson (1988) proposed a concept of “felt involvement,” which was
defined as “a consumer’s overall subjective feeling of personal relevance”
(p.211). They suggested that there are mainly two sources of personal relevance:
“intrinsic sources of personal relevance” (ISPR), which is stored in people’s
long-term memory and derived from past experiences; and “situational sources
of personal relevance” (SSPR), which depends on situational factors (e.g.,
example & example) (Celsi and Olson 1988).
Since ISPR are relatively more stable and related to personal relevant
knowledge and experiences (Celsi and Olson 1988), targeting SSPR may be a
better way to increase the felt involvement since it is easier to change situational
factors such as the design of a PSA or cues in the message than the individual
experiences of each viewer. Hence, in this study, the felt involvement will be
researched in terms of involvement with the ads other than involvement with the
issue in order to evaluate the effects of the message design.
Involvement with ads refers to the level of psychological concern a consumer
has for an ad (Zaichkowsky 1985, Tsai and Tsai 2006). Previous studies
indicated that reference points could influence the involvement with the
message; and messages with a reference point with regard to self are likely to be
highly involving (Loroz, 2007). Thus, the following hypothesis is posited:

H3: The PSA messages regarding binge drinking intervention with self-other
referencing will lead students to have higher involvement with ads than the ones
with only other-referencing.

Involvement with ads impacts consumers’ attention and ad comprehension


processes (Celsi and Olson 1988), and prior studies show that different levels of
attention to ads can lead to different attitudes toward them (e.g., Thorson, Chi
and Leavitt 1992). Therefore, involvement with ads may further influence the
consumers’ attitudes toward ads.
Previous research suggested that attitudes predict behaviors (Hansen 1976).
Attitude was defined as “a psychological tendency that is expressed by
evaluating a particular entity with some degree of favor or disfavor” (Eagly and
Chaiken 1993, p.1). The theory of planned behavior (Ajzen 1991) proposed that
attitudes toward a behavior influence the behavioral intention, and then have
impacts on the behavior performance. In the field of brand advertising, prior
research shows that attitude toward ads can affect the intention of brand choice
(e.g. Biehal, Stephens and Curlo 1992). However, the test of whether or not
attitude toward ads has predictive power for behavioral intentions is still needed
in the PSA context (Jung and Villegas 2011; Hennrikus, Jeffery and Lando
1995).
Based on the above discussion, the following hypothesis is developed:
The Effects of Message Framing and Reference Points 221

H4: Higher involvement with a PSA message leads to a more positive attitude
toward ads, and further results in stronger intentions of bystander intervention in
binge drinking.

Furthermore, the interaction effects between message framing and reference


points on involvement with ads and on attitude toward ads will be also explored:

RQ2: In a PSA regarding bystander intervention in binge drinking, is there an


interaction effect between message framing (loss vs. gain) and reference points
(other-referencing vs. self-other referencing) on people’s involvement with ads
(RQ2a) and attitude toward ads (RQ2b)?

In addition, the following research question with regard to the direct effects
of message framing or reference points on involvement or on ad attitudes will be
also investigated:

RQ3: In a PSA regarding bystander intervention in binge drinking, is there a


direct effect of message framing on involvement with ads (RQ3a) or on ad
attitude (RQ3b)? Is there a direct effect of reference points on ad attitude
(RQ3c)?

2.4 Gender Differences in Binge Drinking


Little literature has been found to investigate whether or not there are
differences of involvement between men and women when they are reading anti-
binge-drinking messages. Also few studies examine gender differences in
intervention intention on occasions of binge drinking. According to information
provided by the Centers for Disease Control and Prevention (CDC), the binge
drinking prevalence among men is twice the as high as among women
(www.cdc.gov). And prior studies show that it is more likely for men to
intervene because it can reveal their competence or strength (Becker and Eagly
2004, White and Malkowski 2014). Therefore, it is posited that men might be
more involved in a bystander intervention PSA regarding in binge drinking, and
thereby, they may have more positive attitudes toward the PSA and stronger
intention of intervention.

H5: Gender is a predictor of involvement with ads (H5a). Specifically, men have
higher involvement with ads regarding binge drinking than women (H5b), more
positive attitude toward ads (H5c), and stronger intervention intention (H5d).
222 Li and Rifon

Based on all the proposed hypotheses and research questions, a hypothesized


model is developed for this study in order to show all of the hypotheses and
research questions more clearly (see Figure 1).

Figure 1: Hypothesized Model

3 Method
This study employed a 2 (message framing: loss vs. gain) × 2 (reference
points: other referencing vs. self-other referencing) online experiment.

3.1 Participants and Procedures


There were 175 college students recruited from a large Midwestern
university in the US. The recruitment information was posted in the college’s
online system where students can enroll to participate in research to earn class
points. After removing the incomplete questionnaires, a total of 169 responses
were qualified for analysis. Among the 169 participants, 24.9% were men and
75.1% were women. Their average age is 21 years old (M = 21, SD = 1.39).
An online survey tool, Qualtrics, was used to administer the online
experimental survey. The participants were randomly assigned to one of the four
experimental conditions to view a version of the PSA message. After viewing
the ads, they were required to fill out a follow-up online questionnaire.
The Effects of Message Framing and Reference Points 223

3.2 Stimuli and Manipulation Check


Four print PSAs were created as stimuli in this study. The stimuli were
designed to reach the expected effects that were required by this study (loss-
framed/other referencing, loss-framed/self-other referencing, gain-framed/other
referencing, and gain-framed/self-other referencing). The backgrounds of these
four PSAs were designed consistently. All of the four messages have the same
title of “When you see your friends binge drinking, step up! Get involved! Stop
them!” Below this title, different persuasive texts were designed for different
conditions.
A manipulation check was conducted to make sure the stimuli work. Results
of one-way analysis of variance showed that the manipulation is satisfactory.

3.3 Measurement
Both involvement with ads and attitude toward ads were measured by seven-
point bipolar scales. Intention of intervention behavior was measured by a
seven-point Likert-type scale ranging from strongly disagree (one point) to
strongly agree (seven points).
The measures of involvement with the message were adapted from
Zaichkowsky (1994). The participants were asked to answer the question “To
me, the presented message is….” Seven items were included in this scale, such
as: relevant/irrelevant, means nothing/means a lot to me, etc. (α = .87).
The measures of attitude toward ads were adapted from Mackenzie and Lutz
(1989). The participants were asked to rate the message using two items:
good/bad and favorable/unfavorable (α =.79, lower point indicated more
positive attitude).
The measures of intention of intervention behaviors were adapted from Kim
and Chung (2011). The participants were asked to indicate their degree of
agreement with five items, such as “I will intervene if I see my friend binge
drinking,” and “It’s my responsibility to stop a friend when binge drinking.” (α =
.80).

4 Results
Structural Equation Modeling (SEM) was applied to analyze the data by
using Mplus 7 software (Muthén & Muthén, 2012). The results show the model
fit well, χ2 (155) = 196.932, p <.05; CFI =.980; TLI =.977; RMSEA =.041 (see
Figure 2).
The results show that message framing is a significant predictor of bystander
intervention intention, and loss framing leads to stronger bystander intervention
intention than gain framing, γ = -.53, p < .05. Thus, the data are consistent with
hypothesis 1.
224 Li and Rifon

Self-other referencing was found to lead to a stronger intention of bystander


intervention in binge drinking than a message with only other-referencing;
however, the difference is not statistically significant, γ = .39, p = .09.
Therefore, the data are not consistent with hypothesis 2 and reference point is
not a directly significant predictor of bystander intervention intention in binge
drinking.
The data are consistent with both hypotheses 3 and 4. That is, reference point
is a significant predictor of involvement with ads regarding binge drinking, and
self-other referencing leads students to have higher involvement with ads than
PSAs with only other-referencing, γ = .36, p < .05. Furthermore, higher
involvement with ads leads to a more positive attitude toward ads (β = -.94, p <
.001), and more positive ad attitudes further result in stronger intentions of
bystander intervention in binge drinking (β = -.97, p < .001).
Gender was found as a significant predictor of involvement with ads, γ = .47,
p < .001. Thus, the data are consistent with hypothesis 5a. But there are no
significant effects of gender on ad attitudes (γ = .40, p = .06) and intervention
intention (γ = .38, p = .10). However, opposite results were found for hypotheses
5b, 5c, and 5d. Based on the results, women had higher involvement with the
binge drinking messages, which in turn led to more positive attitudes toward the
ads, and stronger intention of intervention than men.
For research questions 1 and 2, the results show that there is an interaction
effect between message framing and reference points on people’s involvement
with ads (γ = -.24, p < .05); in a loss-framed message, self-other reference points
worked better to enhance people’s involvement with ads, while in a gain-framed
message, other-reference points were more likely to make people feel involved
with the ads. However, there is no such interaction effect on ad attitudes (γ =
.17, p = .06) or on intervention intention, γ = .07, p = .49.
As for research question 3, message framing does not affect involvement
with ads significantly (γ = .19, p = .25) but it has a significant effect on ad
attitudes (γ = -.44, p < .05). There is no direct effect of reference points on ad
attitudes (γ = -.14, p = .43).
The Effects of Message Framing and Reference Points 225

Figure 2: Final Model

5 Discussion and Managerial Implications


Rather than approaching the topic of binge drinking prevention from the
traditional perspective, which only seeks to discourage such behaviors, this
study chooses to approach the topic from the angle of preventing binge drinking
by investigating how to make effective PSAs to encourage bystander
intervention behaviors.
The findings of this study show that in a PSA regarding binge drinking,
message framing is a significant direct factor that affects people’s attitudes
toward ads and intervention intentions. Loss framing leads to more positive ad
attitudes and stronger bystander intervention intention than gain framing. This
finding is consistent with the prediction drawn from prospect theory that
suggests loss framing works better in the case of risk-seeking behaviors by
reinforcing people’s desire to avoid losses.
In the previous research, some studies investigated the role of reference
points (e.g., Burnkrant and Unnava 1989, Debevec and Romeo 1992, Burnkrant
and Unnava 1995, Meyers-Levy and Peracchio 1996, , Loroz 2007, Cheng,
Woon and Lynes 2011). But most of these studies focused on self-referencing
and self-other referencing, not other-referencing. The present study contributes
in this area and proposes that there can be another type of reference points in
PSAs, i.e., other-referencing, especially for messages advocating intervention
behaviors. The findings of this study suggest that reference point is a significant
indirect factor of intervention intention. That is, reference point exerts effects on
226 Li and Rifon

intervention intention via the mediation of involvement with ads and attitude
towards ads. And other-referencing is less effective than self-other referencing in
promoting an intervention behavior, which may be because people’s mental
representations of self are more accessible, distinctive and rich than their mental
representations of others (Anderson, Glassman and Gold 1998). When people
face a risky situation involving someone else, they can choose either intervene to
help or not intervene. If they know their behavior of intervention will not only
affect the person who is in risk but also will affect themselves (i.e., they may
also have negative consequences if they do not intervene, or they may have
benefit if they intervene), then they may be more likely to intervene in
someone’s risky situation and offer help. The findings of the present study
demonstrate the important role of reference points in PSA message designs.
Advertisers can emphasize the consequences of both sides (i.e., self and others),
rather than only addressing the consequences of others, to increase the readers’
intention of intervention in binge drinking situations.
Moreover, this study found that there is an interaction effect between
message framing and reference points on involvement with ads, which is
consistent with the findings of Loroz’s (2007) study. Specifically, in a loss-
framed message, self-other reference points work better to enhance people’s
involvement with ads; while in a gain-framed message, other-reference points
were more likely to lead to higher involvement with ads.
The above findings indicate that there is a moderated mediation relationship
among the abovementioned variables. In advertising literature, based on the
ELM model (Petty and Cacioppo 1986), involvement was often tested as a
moderator (e.g., Paek et al. 2011, Tsai and Tsai 2006). However, in the present
study, we investigated whether or not involvement can be a mediator between
several factors and ad attitudes. There are multiple types of involvement (Paek et
al. 2011); involvement with ads was chosen for this study. Involvement with ads
was found as a significant mediator between reference points and ad attitudes.
Furthermore, message framing moderates the relationship between reference
points and involvement with ads. Specifically, self-other referencing, rather than
other-referencing, leads to higher involvement with ads, which further leads to
more favorable ad attitudes and stronger intervention intention. And a loss-
framed message with self-other reference points is the most effective message
type for encouraging bystander intervention behaviors in binge drinking
contexts.
Another interesting finding of this study is that gender is found to be a
significant predictor of involvement with ads, but in an opposite direction of the
hypothesis, i.e., women, rather than men, are more highly involved in the binge
drinking intervention PSAs, and furthermore have more positive ad attitudes and
stronger intervention intentions. One reason for this finding may be that women
The Effects of Message Framing and Reference Points 227

are more likely to be in danger during or after binge drinking. For example,
besides injury and death, women are more likely to be victims of sexual assault
than men. A campus sexual assault study conducted in 2009 showed that by the
time college women are seniors, almost 20% of them will become victims of
sexual assault, and more than 80% of campus sexual assaults are related to
alcohol (Yoffe 2013). This threat of being victims of assault may make women
feel more involved in a binge drinking intervention PSA and result in stronger
intervention intention than men. This finding has some implications for policy
makers, advertisers and scholars. First, women might be a group of people that
need to be paid more attention in binge drinking, even though the previous data
always show that men drink more than women. Second, intervention PSAs
regarding binge drinking can be designed to include women characters or
women-related consequences so that they increase women’s intention to
intervene in binge drinking. Third, men might be a group of people that are
harder to be persuaded to intervene in binge drinking, therefore, the factors that
lead men to intervene should be further researched.

6 References
Andersen, S. M., Glassman, N. S., & Gold, D. A. (1998). Mental representations of the
self, significant others, and nonsignificant others: Structure and processing of private
and public aspects. Journal of Personality and Social Psychology, 75(4), 845-861.
Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human
Decision Processes, 50, 179 –211.
Banyard, V. L., Moynihan, M. M., & Plante, E. G. (2007). Sexual violence prevention
through bystander education: An experimental evaluation. Journal of Community
Psychology, 35(4), 463-481.
Becker, S. W., & Eagly, A. H. (2004). The heroism of women and men. American
Psychologist, 59, 163–178.
Biehal, G., Stephens, D., & Curlo, E. (1992). Attitude toward the ad and brand
choice. Journal of Advertising, 21(3), 19.
Burnkrant, R. E., & Unnava, H. R. (1989). Self-referencing: A strategy for increasing
processing of message content. Personality and Social Psychology Bulletin, 15(4),
628.
Burnkrant, R. E., & Unnava (1995). Effects of self-referencing on persuasion. Journal of
Consumer Research, 22(1), 17-26.
Celsi, R. L., & Olson, J. C. (1988). The role of involvement in attention and
comprehension processes. Journal of Consumer Research, 15(2), 210-224.
Cheng, T., Woon, D. K., & Lynes, J. K. (2011). The use of message framing in the
promotion of environmentally sustainable behaviors. Social Marketing Quarterly,
17(2), 48-62.
228 Li and Rifon

Chia-Ching Tsai, & Ming-Hung Tsai. (2006). The impact of message framing and
involvement on advertising effectiveness - the topic of oral hygiene as an
example. Journal of American Academy of Business, Cambridge, 8(2), 222-226.
Debevec, K., & Romeo, J. B. (1992). Self-referent processing in perceptions of verbal
and visual commercial information.Journal of Consumer Psychology, 1(1), 83-102.
Eagly, A. H., & Chaiken, S. (1993). The psychology of attitudes. Fort Worth, TX:
Harcourt, Brace, Jovanovich.
Eagly, A. H., & Chaiken, S. (2007). The advantages of an inclusive definition of attitude.
Social Cognition, 25(5), 582-602.
Fact sheets-binge drinking. Centers for Disease Control and Prevention. Available at:
http://www.cdc.gov/alcohol/fact-sheets/binge-drinking.htm
Gallagher, K. M., & Updegraff, J. A. (2012). Health message framing effects on
attitudes, intentions and behavior: A meta-analytic review. Annals of Behavioral
Medicine, 43, 101-116.
Geary, C. W., Burke, H. M., Johnson, L., Liku, J., Castelnau, L., Neupane, S., & Niang,
C. (2008). Personal involvement of young people in HIV prevention campaign
messages: The role of message format, culture, and gender. Health Education &
Behavior, 35(2), 190-206.
Hansen, F. (1976). Psychological theories of consumer choice. Journal of Consumer
Research, 3(3), 117-142.
Hennrikus, D. J., Jeffery, R. W., & Lando, H. A. (1995). The smoking cessation process:
Longitudinal observation in a working population. Preventive Medicine, 24, 235–
244.
Jung, W. S., & Villegas, J. (2011). The effects of message framing, involvement, and
nicotine dependence on anti-smoking public service announcements. Health
Marketing Quarterly, 28(3), 219-231.
Kahneman, D., & Tversky, A. (1979). Prospect theory: An analysis of decisions under
risk. Econometrica, 47, 263-291.
Kim, H. Y., & Chung, J. (2011). Consumer purchase intention for organic personal care
products. Journal of Consumer Marketing, 28(1), 40-47.
Kuiper, N. A., & Rogers, T. B. (1979). Encoding of personal information: Self–other
differences. Journal of Personality and Social Psychology, 37(4), 499-514.
Lee, M. J., & Bichard, S. L. (2006). Effective message design targeting college students
for the prevention of binge-drinking: Basing design on rebellious risk-taking
tendency. Health Communication, 20(3), 299-308.
Lei, & Wu, Q. (2007). An NCME instructional module on introduction to structural
equation modeling: Issues and practical considerations. Educational Measurement,
Issues and Practice, 26(3), 33-43.
Loroz, P. S. (2007). The interaction of message frames and reference points in prosocial
persuasive appeals. Psychology & Marketing, 24, 1001-1023.
MacKenzie, S. B., & Lutz, R. J. (1989). An empirical examination of the structural
antecedents of attitude toward the ad in an advertising pretesting context. Journal of
Marketing, 53(2), 48-65.
Meyers-Levy, J., & Peracchio, L. A. (1995). Moderators of the impact of self-reference
on persuasion. Journal of Consumer Research, 22(4), 408-423.
The Effects of Message Framing and Reference Points 229

Myers, R.E. (2010). Moderating the effectiveness of messages to promote physical


activity in type 2 diabetes (Doctoral dissertation), Retrieved from University of South
Florida Scholar Commons, Theses and Dissertations. Paper 1719.
http://scholarcommons.usf.edu/etd/1719
Paek, H., Hove, T., Jeong, H. J., & Kim, M. (2011). Peer or expert? the persuasive
impact of YouTube public service announcement producers. International Journal of
Advertising: The Quarterly Review of Marketing Communications,30(1), 161-168.
doi:http://dx.doi.org/10.2501/IJA-30-1-161-188
Petty, R. E., Cacioppo, J. T., & Goldman, R. (1981). Personal involvement as a
determinant of argument-based persuasion. Journal of Personality and Social
Psychology, 41(5), 847-855.
Petty, R. E., Cacioppo, J. T. (1986). The elaboration likelihood model of persuasion.
Advances in Experimental Social Psychology, 19, pp. 123–205.
Pilling, V. K., & Brannon, L. A. (2007). Assessing college students' attitudes toward
responsible drinking messages to identify promising binge drinking intervention
strategies. Health Communication, 22(3), 265-276.
Rogers, T. B., Kuiper, N. A., & Kirker, W. S. (1977). Self-reference and the encoding of
personal information. Journal of Personality and Social Psychology, 35(9), 677-688.
Schneider, T. R. (2006). Getting the biggest bang for your health education buck.
American Behavioral Scientist, 49, 812-822.
Thorson, E., Chi, A., & Leavitt, C. (1992). Attention, Memory, Attitude, and Conation:
A Test of the Advertising Hierarchy. Advances in Consumer Research, 19, 366–379.
Available: http://www.acrwebsite.org/search/view-conference-
proceedings.aspx?Id=7323
Wechsler, H., Lee, J. E., Kuo, M., & Lee, H. (2000). College binge drinking in the
1990s: A continuing problem: Results of the harvard school of public health 1999
college alcohol study. Journal of American College Health, 48(5), 199-210.
White, C. H., & Malkowski, J. (2014). Communicative challenges of bystander
intervention: Impact of goals and message design logic on strategies college students
use to intervene in drinking situations. Health Communication, 29(1), 93.
Yoffe, E. (2013). College women: stop getting drunk. Available at:
http://www.slate.com/articles/double_x/doublex/2013/10/sexual_assault_and_drinkin
g_teach_women_the_connection.html
Zaichkowsky, J. L. (1985). Measuring the involvement construct. Journal of Consumer
Research, 12(3), 341.
Zaichkowsky, J. L. (1994). The personal involvement inventory: Reduction, revision,
and application to advertising. Journal of Advertising, 23(4), 59-70.
How Customers Engage with Brands: A New Framework

Ewa Maslowska, Edward C. Malthouse, and Tom Collinger

1 Introduction

Marketers’ and advertisers’ thinking about processes is linear (Schultz,


2015). For a long time marketers have used the AIDA Model (attention, interest,
desire and action) to explain how customers process advertising messages and
make purchase decisions. In 2009, McKinsey introduced an alternative view, the
Loyalty Loop, which acknowledges two-way character of brand-customer
relationship, but still assumes that a customer follows some kind of pathway
(i.e., “consumer decision journey”). But does such a pathway exist?
Digital media and technologies have created a new marketing environment
that is nonlinear, interactive, and driven by empowered and interconnected
customers. This environment is forcing brands to modify the way they think
about customers and how to communicate with them. Instead of simply sending
ad messages, now they should engage with customers. Engagement is discussed
in the social context; marketers focus on the ability of brand communication to
elicit affective responses and co-creative experiences (Gambetti et al., 2012) and
stress that engagement is a result of delivering messages that are meaningful,
resonate with customers, and make them passionate about and involved with the
brand (Smith, 2014). As the ARF describes engagement in terms of a mental
activation process (Plummer et al., 2006), engagement in advertising means a
process that occurs in response to communication and underlies customer
behavior (Mast and Zaltman, 2006).
This growing relevance of engagement has been recognized not only by
marketing practitioners but also scholars, and has been discussed and
conceptualized in different disciplines. In the marketing and consumer behavior
literatures, engagement has been approached in two different ways—focusing on
the psychological versus behavioral component—although some definitions
include both aspects (Table 1). On the psychological side, Hollebeek (2011, p.
790) claims that engagement is “the level of customer’s motivational, brand-
related and context-dependent state of mind” and Bowden (2009) defines
engagement as a psychological process driving loyalty. Sprott et al. (2009)
develop a theory of brand engagement in self-concept. Finally, Heath (2007)
argues in favor of the affective nature of engagement, driven by the feelings
evoked by an advertisement.
The behavioral aspects of engagement have been recognized by Verhoef et
al. (2010, p. 248), who describe engagement as “a behavioral manifestation

© Springer Fachmedien Wiesbaden 2017


G. Christodoulides et al. (Hrsg.), Advances in Advertising Research (Vol. VII),
European Advertising Academy, DOI 10.1007/978-3-658-15220-8_17
232 Maslowska, Malthouse, and Collinger

toward the brand or firm that goes beyond transactions.” Similarly, Van Doorn
et al. (2010) focus on the behavioral aspects of the relationship between a
customer and a firm that does not include transactions, “resulting from
motivational drivers” (p. 254). Kumar et al. (2010) add purchases to the
definition of engagement, while Bijmolt et al. (2010) are very specific and focus
on three key behavioral manifestations of engagement, WOM, co-creation, and
complaining behavior, occurring at different stages of the customer life cycle
(i.e., acquisitions, development, and retention). Some researchers tried to
combine both perspectives. Vivek et al. (2010) focus on customers’ participation
in activities related to an organization, but acknowledge the cognitive and
affective elements of engagement, while Hollebeek et al. (2014; p. 154) define
engagement as “a consumer's positively valenced brand-related cognitive,
emotional and behavioral activity during or related to focal consumer/brand
interactions.”

Table 1: Most relevant previous definitions of brand engagement


Authors Definition
Brodie et al. Customer engagement (CE) is a psychological state that occurs by
(2011, p. 260) virtue of interactive, cocreative customer experiences with a focal
agent/object (e.g., a brand) in focal service relationships. It occurs
under a specific set of context dependent conditions generating
differing CE levels; and exists as a dynamic, iterative process
within service relationships that cocreate value. CE plays a central
role in a nomological network governing service relationships in
which other relational concepts (e.g., involvement, loyalty) are
antecedents and/or consequences in iterative CE processes. It is a
multidimensional concept subject to a context- and/or
stakeholder-specific expression of relevant cognitive, emotional
and/or behavioral dimensions.
Bowden A psychological process that models the underlying mechanisms
(2009, p. 65) by which customer loyalty forms for new customers of a service
brand as well as the mechanisms by which loyalty may be
maintained for repeat purchase customers of a service brand.
Calder et al. Engagement is a multilevel construct that emerges from the
(2016) thoughts and feelings about one or more rich experiences involved
in reaching a personal goal.
Higgins and Engagement is a state of being involved, occupied, fully
Scholer (2009, absorbed, or engrossed in something — sustained attention.
p. 102)
How Customers Engage with Brands: A New Framework 233

Hollebeek et A consumer's positively valenced brand-related cognitive,


al. (2014, p. emotional and behavioral activity during or related to focal
154) consumer/brand interactions. Engagement has three dimensions:
cognitive, emotional, and behavioral.
Ksiazek et al. Engagement is a broad phenomenon that describes all sorts of
(2014, p. 3) user attention and involvement with media.
Kumar et al. Active interactions of a customer with a firm, with prospects and
(2010, p. 297) with other customers, whether they are transactional or
nontransactional in nature
Mollen and Online engagement is a cognitive and affective commitment to an
Wilson (2010, active relationship with the brand as personified by the website or
p. 923) other computer-mediated entities designed to communicate brand
value. It is characterized by the dimensions of dynamic and
sustained cognitive processing and the satisfying of instrumental
value (utility and relevance) and experiential value (emotional
congruence with the narrative schema encountered in computer-
mediated entities).
Patterson et al. Customer engagement describes the level of a customer’s various
(2006) “presence” in their relationship with the organization. The
presences include physical presence, emotional presence and
cognitive presence. Customer engagement is conceived as a
higher-order construct which consists of four components,
namely, vigor, dedication, absorption, and interaction.
Van Doorn et Customer engagement behaviors go beyond transactions, and may
al. (2010, p. be specifically defined as a customer’s behavioral manifestations
254) that have a brand or firm focus, beyond purchase, resulting from
motivational drivers. The behavioral manifestations, other than
purchases, can be both positive (i.e., posting a positive brand
message on a blog) and negative (i.e., organizing public actions
against a firm).
Vivek et al. Customer engagement is the intensity of an individual’s
(2012, p. 133) participation in and connection with an organization’s offerings or
organizational activities, which either the customer or the
organization initiates. The individuals may be current or potential
customers. CE may be manifested cognitively, affectively,
behaviorally, or socially. The cognitive and affective elements of
CE incorporate the experiences and feelings of customers, and the
behavioral and social elements capture the participation by current
and potential customers, both within and outside of the exchange
situations.

Theories on engagement disagree on antecedents and consequences, and


various authors indicate the need to advance both theoretical and empirical
234 Maslowska, Malthouse, and Collinger

understanding of customer engagement (e.g., Chandler and Lusch, 2015;


Gummerus et al., 2012; Hollebeek, 2011). MSI includes: “How should
engagement be conceptualized, defined, and measured? How do social media
and other marketing activities create engagement?” as one of the highest
priorities for 2014-2016. Plummer et al. (2006) gives quotes from different
industry leaders and shows that the term has even more meanings. Because
engagement is commonly used to refer to so many distinct phenomena, we will
recommend that the term should be avoided, and that more specific terms be
used. Building on previous research on engagement, consumer behavior and
service marketing, we intend to reconcile the differences in the current literature
by introducing a new framework that reflects the new realities of environments
in which companies operate—dynamic, nonlinear, real-time, and reflective of
the inter-relationships between brands, customers, and one another. The
framework recognizes the growing role of customer non-purchase engagement
behaviors that influence purchase behaviors, whether triggered by individuals,
the brand, or other customers, and attempt to categorize them. By doing this, we
make a unique contribution to literature on engagement, and potentially, to the
practice.

2 The New Engagement Framework

Responding to the complexity of today’s marketing environment, we propose


the engagement ecosystem framework, which includes brand actions, customer
motivations, purchase behaviors, brand consumption and brand dialogue
behaviors. The framework also acknowledges the role of other actors, and
recognizes that interactions are non-linear, inter-connected and reactive;
meaning that each action causes a reaction of not only the intended recipient of
the message. Hence, we illustrate the elements of engagement ecosystem as
gears that all influence each other whenever one of them is in motion (Figure 1).
In other words, when one thing changes, the ecosystem changes. We postulate
that all the elements of our engagement framework contribute to the multi-level
concept of engagement, and therefore, should be called what they actually are,
while the term engagement should be reserved to describe the dynamic
ecosystem emerging from the interactions between these concepts. We will now
briefly discuss each element of the framework.
How Customers Engage with Brands: A New Framework 235

Figure 1: Engagement ecosystem

2.1 Company Actions


Traditionally, brands addressed customers with marketing actions aimed at
increasing purchases. Today companies also influence motivations driving other
behaviors than purchases, such as word of mouth or product co-development.
Moreover, it is not always brands that start and control the conversation: Brands
are no longer only senders of a message, but also recipients of direct and indirect
communication (Hennig-Thurau et al., 2010, Figure 1), while being privy to
customer commentary in the social graph. Brand actions include all prompts
from the company, such as advertising, marketing initiatives, promotions, but
also responses from the company such as webcare.
All of these actions can not only be directed at existing customers but also at
other actors, such as other customers or media. For example, a company’s
response to a customer’s post on Facebook may be read by non-customers,
company employees and other stakeholders. In addition, a company can write a
press release or respond to comments made on a TV show. Hence, we need to
explore engagement not only as customer engagement but also considering other
actors such as suppliers, manufacturers, and retailers (Chandler and Lusch,
2015; Van Doorn et al., 2010).

2.2 Customer Motivations


Engagement is driven by the individual’s needs, motives, and goals (Ashley
and Tuten, 2015; Calder et al. 2016). Customers may be aware of their motives
or can act upon motives they are not conscious of (Converse et al., 1958). In
addition, although motivations are at an individual level, they may reflect wider
social values, beliefs and conventions (Cofer and Appley, 1964). Motivations
236 Maslowska, Malthouse, and Collinger

can also be cognitive, so related to goals, or of emotional nature. The type of


motivation that is to be activated may depend on customers’ experiences. We
define experiences as the thoughts and feelings that individuals have about an
object that reflect the individual’s interaction with the brand over time as a way
of achieving personal goals (Calder et al., 2008; 2016). When an object is
experienced positively, an individual feels attracted to it, while when it is
experienced negatively, an individual feels repulsed by it (Calder and
Malthouse, 2008; Higgins, 2006).

2.3 Purchase Behaviors


Many brand actions are directed at persuading customers to buy brand’s
products. Shopping to buy is a goal-focused and transaction-oriented activity
that can occur either offline or online (Wolfinbarger and Gilly, 2001). However,
shopping can also go beyond the goal-directed utilitarian experience and can be
used as means to lessen feelings of loneliness and boredom, fantasize and escape
from everyday life (Koufari, 2002). When a customer’s goal is to buy, it is a
purchase behavior, yet, when the goal is to get knowledge about a product
category or socialize, it may be categorized as a brand dialog behavior.

2.4 Customer Brand Dialog Behaviors


Brand dialog behaviors (BDBs) represent another way in which customers’
engage with brands (Malthouse and Calder, 2011). By BDBs we mean what Van
Doorn et al. (2010) call customer engagement behaviors (CEBs), which are
focused on a brand or firm and go beyond transactions. These behaviors include
a vast array of actions (e.g., see Li and Bernoff, 2008; Muntinga et al., 2011),
which can be positive, negative or neutral, take place offline and online, and be
targeted at various audiences including self, other and potential customers, the
brand, general public, firm’s employees, etc. (Van Doorn et al., 2010).
We extend engagement conceptualizations proposed by Schaufeli et al.
(2002) and Ksiazek et al. (2014), which suggest that engagement is a continuum
of resources invested in, respectively, work-related or media-related activities,
and propose that BDBs constitute a continuum based on the invested resources
leading from passive observation to active co-creation. The more interactive the
behavior becomes, the more customers show initiative and the more resources
(e.g., cognitive, emotional, time, monetary) they invest in the behavior. Such an
approach is in line with other theories in customer research, for example, the
behavioral hierarchy proposed by Park et al. (2010). In addition BDBs can be
described in terms of goal relevance. Some BDBs, for example, seeing a spam
advertisement, are irrelevant to customers’ brand-related goals, while others, for
example, sharing a photo taking a break and eating Kit Kat, are very relevant
(Malthouse et al. 2016).
How Customers Engage with Brands: A New Framework 237

Learning from the work of Muntinga et al. (2011), Shao (2009) and
Malthouse et al (2013, p. 272), we place three stages on this continuum:
observing, participating, and co-creating (Figure 2). The ordering along the
continuum is determined by the level of cognitive elaboration done by the
customer (Malthouse et al. 2016). Observing represents the lowest level of
BDBs and means being exposed or exposing oneself to a brand-related stimulus,
for example, seeing an advertisement or reading a review. It is close to what
Muntinga et al. (2011) and Shao (2009) call consuming in the context of online
brand-related activities to describe watching and reading content shared by other
customers or the company. We use a different term to recognize the passive
character of this stage. Participating (Shao 2009) is what Muntinga et al. (2011)
call contributing. At this level customers respond to stimuli by, for example,
commenting on a picture posted by a brand or another customer on Facebook.
Finally, we call the third level Co-creating. At this stage customers take even
more initiative and generate their own content or participate in creating
something new. Co-creating is related to creating (Muntinga et al., 2011) and
producing (Shao 2009). We feel that this term, co-creating, describes this stage
better since the customer does not have to necessarily create or produce anything
new, but may participate in, for example, product development or tech support,
when a customer responds in a forum to queries from other customers about,
say, a software product. The Dorito’s “Do us a Flavor” promotion to stimulate
user generated videos about the brand dramatize this active form of co-creating.
Concerning the effects generated by specific BDBs, previous research is
limited. Based on the elaboration likelihood model and research into
interactivity, we can speculate that behaviors that are more goal-relevant and
require more interaction and resources affect a brand more. More interaction
means more involvement and deeper cognitive processing. However, the
outcome does not always have to be positive. A negative experience with a
brand can lead to either withdrawal from the relationship with the brand or
display of negative BDB (e.g., writing a negative review). Both withdrawal and
negatively-valenced engagement can have undesirable consequences for the
brand (Kim et al. 2015, 2016).
238 Maslowska, Malthouse, and Collinger

Figure 2: Categories of non-purchase behaviors

2.5 Brand Consumption


Brand consumption is arguably the most important and possibly intimate way
in which individuals engage with a brand. Consuming means that people make
use of product or service in a variety of ways (Holt, 1995). Consumption does
not require control over or possession of an object (Vivek, 2012), but occurs via
the use of an object or service (Humphreys and Geisler, 2007). For example, in
case of subscription services (e.g., Netflix), car rental (e.g., Zip Car) or branded
destinations (e.g., The Wizard World of Harry Potter), customers do not own the
product, but use it for some time. Products can be consumed in two ways:
actions, in which customers use objects, and interactions with other people, in
which consumption objects serve as central resources. Hence, consumption can
serve different purposes; it can be both an aim in itself and mean to some further
aims (Holt, 1995). Many of the industry definitions of engagement cited in
Plummer et al. (2006) focus on consumption, such as the number of page views,
time spent in a branded environment, etc.

3 Conclusions

Customer engagement is transforming the dynamics that have historically


informed marketing decisions, but the term has grown to mean too many
How Customers Engage with Brands: A New Framework 239

different things. We propose a new customer engagement framework that


accounts for the new realities of customer engagement, which is dynamic
(Brodie, et al. 2011) and reflects the inter-relationships between the elements of
the framework. We recommend using the names of the component parts instead
of the catchall term engagement, while referring to the collection of all parts as
the engagement ecosystem. By introducing the new engagement framework, this
paper reconciles previous engagement conceptualizations and makes both
theoretical and practical contribution, as well as implies directions for future
research and theorizing. It also recognizes the relevance of brand dialogical
behaviors, which are customer non-purchase behaviors. Finally, we describe
those behaviors using two dimensions, i.e., brand-related goal-relevance and
interactivity and categorize them on the continuum leading from lack of
engagement to high engagement (Brodie et al. 2011). We also acknowledge that
their valence can be both positive and negative (Van Doorn 2011). This is an
important contribution considering that such behaviors, although does not
include purchases, are important drivers of customer satisfaction, loyalty, and
lifetime value.

4 Managerial Implications

We believe that our framework clarifies what engagement is, which elements
of engagement can be tracked and measured, and which should be inspired. Our
model suggests that firms’ actions should be focused on creating an experience
living at the intersection of personal goals and the value created by the brand,
because the experience stimulates brand dialogical behaviors (BDBs), shopping
and consumption. And, it suggests that firms must no longer be only
communicating but also listening and responding to the growing voices of
engaged brand experiencers.
Our framework also posits that engagement behaviors may take place with
the use of different technologies, including mobile devices. A common belief is
that mobile does not convert. However, this belief does not hold if we take into
account the whole framework. To illustrate, even if a person does not make a
purchase on a mobile device, the positive experience can still translate into
benefits, as the customer can buy the product offline or on a computer with a
larger screen (Wang et al. 2015). Hence, marketers should try to answer the
question, what is the role of consumption and BDBs. For example, is a customer
using a free Spotify account less valuable than the one using a paid account? If
we only look at the first contact, the answer would be yes, but if we take into
account the whole engagement ecosystem, including other behaviors, the answer
may change. That would mean that consumption and BDBs can constitute a
multiplier of benefits. Marketers should investigate whether that is the case and,
240 Maslowska, Malthouse, and Collinger

if so, how big the multiplier is and whether it differs across behavior categories.
By investigating these issues, marketers should be able to answer questions
about the best engagement strategy for their brand.
In a similar vein, we think companies should go beyond ROI. Commonly,
measurement of marketing communications impact has been in service to
understanding investment returns. While valuable, this practice, alone, neglects
to value the growing influence of customer driven BDBs on driving transactions
of new customers with no direct investment (e.g., Kumar et al. 2010).
Building on that, we recommend that companies should execute experiments
to understand which types of engagement have the greatest impact in growing
customer value of engagers, as well as customer value of those engaged by other
customers. It is understood, and well researched, that customers consider, shop
and buy differently across categories. It will be possible, however, to build
experiments that demonstrate the most valuable forms of engagement unique to
that category. As the learning from the experiments grows, a prioritization and
categorization will be possible to improve strategic decisions about the customer
engagement most effective for the enterprise.
To summarize we propose several tips what brands should do. First, we
recommend to use our framework to inform how well brand’s customers are
driving relevant and valuable engagement not merely what the brand is doing to
drive it. Second, we suggest that brands should invest heavily in customers and
their experiences, so customers can become brands’ advertising. Third, brands
should focus on stimulating only the kinds of customer engagement that drive
relevant connections to the brand. Fourth, companies should focus on generating
co-creation opportunities that best connect the brand’s value and promise to the
customer. Fifth, brands should collect big data and monitor social media to
prove how the most relevant brand posts drive the most customer value. Finally,
we think the brand dialogue hierarchy of observation, participation and co-
creation should be used to rank and then create the most valuable and effective
engagement strategies.

5 References

Ashley, C., Tuten, T. (2015), “Creative Strategies in Social Media Marketing: An


Exploratory Study of Branded Social Content and Consumer Engagement,” in:
Psychology & Marketing, 32(1), 15-27.
Bijmolt, T. H., Leeflang, P. S., Block, F., Eisenbeiss, M., Hardie, B. G., Lemmens, A.,
Saffert, P. (2010), “Analytics for Customer Engagement,” in: Journal of Service
Research, 13(3), 341-356.
Bowden, J. L. (2009), “The Process of Customer Engagement: A Conceptual
Framework,” in: Journal of Marketing Theory and Practice, 17(1), 63–74.
How Customers Engage with Brands: A New Framework 241

Brodie, R. J., Hollebeek, L. D., Juric, B., Ilic, A. (2011), “Customer Engagement:
Conceptual Domain, Fundamental Propositions & Implications for Research in Service
Marketing,” in: Journal of Service Research, 14, 3, 252–71.
Calder, B. J., Malthouse, E. C. (2008), “Media Engagement and Advertising
Effectiveness,” in: Kellogg on advertising and media, 1-36.
Calder, B. J., Isaac, M., and Malthouse, E. C. (2016), “Capturing Consumer Experiences:
A Context Specific Approach to Measuring Engagement,” in” Journal of Advertising
Research, March, 1-14.
Chandler, J. D., Lusch, R. F. (2015), “Service Systems: A Broadened Framework and
Research Agenda on Value Propositions, Engagement, and Service Experience,” in:
Journal of Service Research, 18(1) 6-22.
Cofer, C. N., Appley, M. H. (1964), Motivation: Theory and research. London: Wiley.
Converse, P. D., Huegy, H. W. Mitchell, R. V. (1958). Elements of Marketing.
London:Sir Issac Pitman and Son.
Gambetti, R. C., Graffigna, G., Biraghi, S. (2012), “The Grounded Theory approach to
consumer-brand engagement,” in: International Journal of Market Research, 54(5),
659-687.
Giese, J. L., Cote, J. A. (2000), "Defining Consumer Satisfaction," Academy of
Marketing Science Review [Online] 00 (01) Available: http://www.amsreview.org/
articles/giese01-2000.pdf
Gummerus, J., Liljander, V., Weman, E., Pihlström, M. (2012), “Customer Engagement
in a Facebook Brand Community,” in: Management Research Review, 35(9), 857 –
877.
Heath, R. (2007), “How do we predict advertising attention and engagement?” Working
Paper. UNSPECIFIED, Bath. (Unpublished).
Hennig-Thurau, T., Malthouse, E. C., Friege, C., Gensler, S., Lobschat, L.,
Rangaswamy, A., Skiera, B. (2010), “The Impact of New Media on Customer
Relationships,” in: Journal of Service Research, 13(3), 311-330.
Higgins, E. T., Scholer, A. A. (2009), “Engaging the Consumer: The Science and Art of
the Value Creation Process,” in: Journal of Consumer Psychology, 19(2), 100-114.
Higgins, E. T. (2006), “Value from Hedonic Experience and Engagement,” in:
Psychological Review, 113(3), 439.
Hollebeek, L. D. (2011), “Demystifying Customer Brand Engagement: Exploring the
Loyalty Nexus,” in: Journal of Marketing Management, 27(7-8), 785-807.
Hollebeek, L. D., Glynn, M. S., Brodie, R. J. (2014), “Consumer Brand Engagement in
Social Media: Conceptualization, Scale Development and Validation,” in: Journal of
Interactive Marketing, 28(2), 149-165.
Holt, D. B. (1995), “How Consumers Consume: A Typology of Consumption Practices,”
in: Journal of Consumer Research, 22(1), 1-16.
Humphreys, A., & Geisler, M. (2007), “Access versus Ownership in Consumer
Research,” in: Advances in Consumer Research, 34, 696.
Kim, S.J., Wang, R.J. and Malthouse, E.C. (2015), “The Effects of Adopting and Using a
Brand’s Mobile Application on Customers’ Subsequent Purchase Behavior,” in:
Journal of Interactive Marketing, 31, 28-41.
242 Maslowska, Malthouse, and Collinger

Kim, S.J., Wang. R.J., Maslowska, E. and Malthouse, E.C. (2016), “‘Understanding a
fury in your words’”: The Effects of Posting and Viewing Electronic Negative Word-
of-Mouth on Purchase Behaviors,” in: Computers in Human Behavior, 54, 511-521.
Koufaris, M. (2002), “Applying the Technology Acceptance Model and Flow Theory to
Online Consumer Behavior,” in: Information Systems Research, 13(2), 205-223.
Ksiazek, T. B., Peer, L., Lessard, K. (2014), “User Engagement with Online News:
Conceptualizing Interactivity and Exploring the Relationship between Online News
Videos and User Comments,” in: New Media & Society, 1-19.
Kumar V., Aksoy, L., Donkers, B., Venkatesan, R., Wiesel, T., Tillmanns, S. (2010),
“Undervalued or Overvalued Customers: Capturing Total Customer Engagement
Value,” in: Journal of Service Research, 13(3), 297–310.
Li, C., Bernoff, J. (2008) Groundswell: Winning in a World Transformed by Social
Technologies. Boston, MA: Harvard Business Press.
Malthouse, E. C., Calder, B. J. (2011), “Comment: Engagement and Experiences:
Comment on Brodie, Hollenbeek, Juric, and Ilic (2011),” in” Journal of Service
Research, 14(3), 277-279.
Malthouse, E. C., Haenlein, M., Skiera, B., Wege, E., Zhang, M. (2013), “Managing
Customer Relationships in the Social Media Era: Introducing the Social CRM House,”
in: Journal of Interactive Marketing, 27(4), 270-280.
Malthouse, E.C. Vandenbosch, M., Calder, B. J. and Kim, S. J. (2016), “Evidence that
User Generated Content that Produces Engagement Increases Purchase Behaviors,” in:
Journal of Marketing Management.
Mast, F., Zaltman, G. (2006). MI4 Committee Working Definition of Engagement”
“Engagement is turning on a prospect to a brand idea enhanced by the surrounding
context.” New York: The Advertising Research Foundation.
MSI — Marketing Science Institute (2014), 2014–2016 Research Priorities, Available:
http://www.msi.org/research/2014-2016-research-priorities/
Mollen, A., Wilson, H. (2010), “Engagement, Telepresence, and Interactivity in Online
Consumer Experience: Reconciling Scholastic and Managerial Perspectives,” in:
Journal of Business Research, 63, 9/10, 919–25.
Muntinga, D. G., Moorman, M., Smit, E. G. (2011), “Introducing COBRAs: Exploring
Motivations for Brand-Related Social Media Use,” in: International Journal of
Advertising, 30(1), 13-46.
Park, C. W., MacInnis, D. J., Priester, J., Eisingerich, A. B., Iacobucci, D. (2010),
“Brand Attachment and Brand Attitude Strength: Conceptual and Empirical
Differentiation of Two Critical Brand Equity Drivers,” in: Journal of Marketing, 74(6),
1-17.
Patterson, P., Ting, Y., De Ruyter, K. (2006). Understanding Customer Engagement in
Services. Brisbane: AZMAC Proceedings.
Plummer, J., Cook, B., Diforio, D., Schachter, B., Sokolyanskaya, I., Korde, T. (2006),
Measures of Engagement, New York: The Advertising Research Foundation.
Schaufeli, W. B., Salanova, M., González-Romá, V., Bakker, A. B. (2002), “The
Measurement of Engagement and Burnout: A Two Sample Confirmatory Factor
Analytic Approach,” in: Journal of Happiness Studies, 3(1), 71-92.
Schmitt, B. (2012), “The Consumer Psychology of Brands,” Journal of Consumer
Psychology, 22, 8-17.
How Customers Engage with Brands: A New Framework 243

Schultz, D. E. (2015). 5 Myths of Consumer Behavior. Available:


https://www.ama.org/publications/MarketingInsights/Pages/the-path-not-taken.aspx
Shao, G. (2009), “Understanding the Appeal of User-Generated Media: A uses and
Gratification Perspective,” Internet Research, 19(1), 7-25
Sprott, D., Czellar, S., Spangenberg, E. (2009), “The Importance of a General Measure
of Brand Engagement on Market Behavior: Development and Validation of a Scale,”
Journal of Marketing Research, 46(1), 92-104.
Wang, R. J.-H., Malthouse, E. C., & Krishnamurthi, L. (2015), “On the Go: How Mobile
Shopping Affects Customer Purchase Behavior,” in: Journal of Retailing, 91(2), 217-
234.
Wolfinbarger, M., Gilly, M.C. (2001), “Shopping Online for Freedom, Control, and
Fun,” in: California Management Review, 43(2), 34-55.
Van Doorn, J., Lemon, K. N., Mittal, V., Nass, S., Pick, D., Pirner, P., Verhoef, P. C.
(2010), “Customer Engagement Behavior: Theoretical Foundations and Research
Directions,” in: Journal of Service Research, 13(3), 253-266.
Verhoef, P. C., Reinartz, W. J., Krafft, M. (2010), “Customer Engagement as a New
Perspective in Customer Management,” in: Journal of Service Research, 13(3), 247-
252.
Vivek, S. D., Beatty, S. E., Morgan, R. M. (2012), “Customer Engagement: Exploring
Customer Relationships Beyond Purchase,” in: The Journal of Marketing Theory and
Practice, 20(2), 122-146.
Part IV. Culture and Advertising
Hofstede Cultural Framework and Advertising Research: An
Assessment of the Literature

Salman Saleem and Jorma Larimo

1 Introduction
The Hofstede’s cultural framework has been frequently used to investigate
the influences of culture on advertising (Chang et al., 2009). For instance, the
model has been used to explain the cross-cultural differences of advertising
appeals (Albers-Miller and Gelb 1996) and the portrayal of gender roles in
advertising (Paek et al., 2011). Some scholars have reviewed Hofstede-inspired
research in business, management and applied psychology disciplines in order to
assess the relevance of Hofstede’s framework in a cross-cultural context
(Kirkman et al., 2006; Taras et al., 2010). Similarly, Soares et al. (2007) focused
on Hofstede-inspired studies in marketing research but not specifically in
advertising and this work is almost ten years old. Therefore, a strong and
practical need is an assessment of literature and discussions to improve the use
of Hofstede’s cultural framework in advertising. However, this is not the first
attempt to assess previous advertising research similar efforts have been made
before with different emphasis (e.g. Okazaki and Mueller, 2007; Taylor and
Bowen, 2012). Despite their usefulness, these reviews have some limitations:
analyzed articles published in SSCI indexed journals only; accessed advertising
literature for short periods at different points in time; only very limitedly
examined methodological aspects; and seldom discussed the distinction between
various facets of culture and their relative significance in predicting the
relationship between culture and advertising.
This research considers Hofstede-inspired advertising literature from 1980 to
2012 with an aim to identify possible gaps for future research and pitfalls in the
extent of the literature. Thus, we attempt to make several contributions. First, the
study draws advertising researchers’ attention to the discrepancy between the
cultural values and practices. Second, the present study evaluates cultural
dimensions, research topics, and the geographic scope of earlier studies. This
undertaking will help to identify which cultural dimensions; thematic areas of
advertising, countries/regions have received most research attention and which
have been ignored. Finally, the review evaluates the methodologies used in
earlier advertising research. Major marketing, advertising, psychology and
cross-cultural communication journals were examined systematically to identify
Hofstede-inspired advertising articles. Second, we performed a bibliographic

© Springer Fachmedien Wiesbaden 2017


G. Christodoulides et al. (Hrsg.), Advances in Advertising Research (Vol. VII),
European Advertising Academy, DOI 10.1007/978-3-658-15220-8_18
248 Saleem and Larimo

search of computerized databases (ABI/INFORM, EBSCOHOST Business


Source, Emerald, Taylor & Francis and JSTOR) by using a variety of Hofstede’s
terms (e.g. IND, Individualistic, Individualism) and advertising-related terms
(e.g. cross-cultural and international advertising). The criterion for article
selection is: the study (a) has to include one or more of Hofstede’s cultural
dimensions to examine the relationship between culture and advertising and (b)
has to be empirical. The search resulted in a total of 57 studies published in
several journals between 1992 and 2012 (for details see Table 1).

Table 1: Reviewed journals, with the corresponding number of stu dies


Topic area Number of Percentage
Studies
Journal of Advertising 20 35
International Marketing Review 6 10.5
Sex Roles 4 7.0
Journal of Advertising Research 3 5.3
Journal of Current Issues and Research in Advertising 3 5.3
Journal of International Consumer Marketing 3 5.3
Other 18 31.6
Total 57 100.0

2 Hofstede’s Cultural Framework and Advertising Research


Many researchers have found that Hofstede’s dimensions support the
presence of culture-specific content in print and electronic advertisements (e.g.
Albers-Miller and Gelb, 1996; Saleem et al., 2015). Similarly, some researchers
conclude from their experiments that advertising that is congruent with
Hofstede’s cultural dimensions is more persuasive than non-congruent
advertising (Choi and Miracle, 2004). However, a few scholars applying
Hofstede’s cultural framework did not find support for the suggestion that
culturally consistent advertising content and values were reflected and were
persuasive (Mortimer and Grierson, 2010). Advertising and marketing scholars
consider Hofstede’s model as outdated (Paek et al., 2011), and thus, challenge
its predictive value; rather they pay attention to the conceptual analysis of
cultural dimensions when formulating hypotheses (De Mooij, 2013a). However,
researchers need to understand the differences between various facets of culture
and their usefulness in the research context (Sun et al., 2014).

3 Cultural Values: The Desirable and the Desired


Most human values have opposing elements - this refers to the distinction
between the desired (cultural values) and the desirable (cultural practices) (De
Mooij, 2013a). Hofstede measured individual behavioral preference, which is
the desired aspect of cultural values; the statements about the desired values do
Hofstede Cultural Framework and Advertising Research 249

not necessarily correspond to the way people behave in reality and the actual
practices in a society (Hofstede et al., 2010; De Mooij, 2013b). Scholars
generally assume that the desired values are the core elements that drive other
cultural expressions - including practices; therefore, we expect these two
different facets of culture to correlate positively (Taras et al., 2010; Fischer,
2006). Studies in management and social psychology reveal the inconsistency
between values and practices (Fischer, 2006; House et al., 2004) and point out
that often they are in opposition to each other (Sun et al., 2014). The research by
Fischer (2006) provides evidence of a discrepancy between values and practices
as he found very little overlap (7.84%) between seven of Schwartz’s cultural
values and practices scales. Also, the GLOBE study shows the discrepancy
between values and practices, as they found surprisingly a significant negative
correlation between values and practices for seven of the nine dimensions
(House et al., 2004). Similarly, De Mooij (2013a) has emphasized that due to the
contrast between values and practices, every society has its own opposing
values, referred to as the value paradox. This indicates that researchers have to
be very careful about interpreting individual behavioral preferences in terms of
norms. Thus, we can say that the use of cultural values alone is not enough to
investigate the effect of culture on advertising. Among reviewed studies, an
overwhelming majority (55 out of 57 studies) proposed a positive link between
cultural values and advertising. Such prevalence of knowledge, that there is
congruence between cultural values and advertising leaves no room for value
and practice discrepancy or De Mooij’s (2013a) value paradoxes. For content
analysis studies multi-measure approach suggested by Soares et al. (2007) may
be appropriate, by combining evidence from ethnography, regional affiliation,
and Hofstede’s dimension scores, a researcher may draw inferences about
discrepancies between values and practices. For an experimental investigation,
the researcher should measure the cultural values (by using Hofstede’s cultural
values scale) as well as the cultural practices (by using the referent shift model -
for details see Fischer, 2006; House et al., 2004) to examine to what extent both
facets impact the attitude and/or behavior toward advertising.

4 Studied Cultural Dimensions


Hofstede’s three cultural dimensions (individualism/collectivism,
masculinity/femininity, and uncertainty avoidance) were included most
frequently among reviewed studies; perhaps scholars assume that these values
have stronger predictive power than other. However, the links between other
cultural dimensions and advertising are equally plausible. Future advertising
research may focus on the least studied dimension of Long Term orientation and
the newly introduced sixth dimension - indulgence/restraint. Further, none has
examined the interaction effect of cultural values on advertising. There are no
250 Saleem and Larimo

strong theoretical reasons that cultural values would not interact and that they
operate independently to influence the outcome (Gibson et al., 2009). Therefore,
it is likely that interactions among values might provide insights into their
complex effects on individuals (Kirkman et al., 2006). De Mooij (2015)
emphasized that often it is the configuration of dimensions that explains the
variation in values across cultures. For example, horizontal individualism (HI) is
the configuration of individualism and low power distance and it includes more
equality and lower achievement needs, whereas vertical individualism (VI) is the
combination of individualism with high power distance and it places more
emphasis on achievement values (De Mooij 2013b). The study by Shavitt et al.
(2011) documents the interaction effect of HI and VI on advertising. They found
that advertisements from a VI country (USA) rated significantly higher in terms
of their emphasis on status than those from an HI country (Denmark). The
interaction effect may also exist among other cultural dimensions. We urge
future research to examine interaction effects among cultural dimensions to
investigate the role of culture in advertising.

5 Geographic Scope of Studies


To date, the Hofstede cultural framework has been utilized to study the role
of culture in advertising in 36 countries. A large share of the Hofstede-inspired
advertising research focused on East and Southeast Asian countries followed by
North America. Considering the size and significance of Europe in the global
market, it appears to be under-investigated. The Middle East was examined in
eight articles followed by Latin America which was examined in six articles.
Similarly, Russia, Australia and East and West Africa are clearly under-studied,
as each was examined in one out of 57 studies.
Hofstede Cultural Framework and Advertising Research 251

Table 2: Frequency of countries/regions (Classification based on CIA World Fact Book)


Country/Region Frequency Percentage
North America (the USA 43; Canada 3; Mexico7) 53 29
Latin America (Brazil 6; Chile 4) 10 5.5
East/Southeast Asia (China 14; Hong Kong 5; Japan 5; South 57 31.1
Korea 16; Taiwan 8; Indonesia 1; Thailand 7; Vietnam 1)
South Asia (India7) 7 3.8
North Asia (Russia1) 1 0.5
Middle East (Egypt 3; Israel 4; Kuwait 2; Lebanon 3; Saudi 18 9.8
Arabia 2; Turkey 1; UAE 3)
Oceania (Australia1) 1 0.5
South Africa (2) 2 1.1
Eastern and Western Africa (Kenya1; Ghana1) 2 1.1
Eastern Europe (Poland 1; the Czech Republic1) 2 1.1
Northern Europe (Finland 4; Sweden 3; the United Kingdom 13 7.1
6)
Western Europe (Belgium 1; France 8; Germany 6; 14 7.7
Netherlands 1)
Southern Europe (Spain 2; Greece 1) 3 1.6
Total 183 100

Among the reviewed studies, two-thirds examined two or three countries,


somewhat more than 10% of the studies analyzed four to six countries and a
similar proportion studied seven or more countries and five studies analyzed one
country. But for cross-cultural comparative research, the use of two or three data
points is not enough for generalization, and thus, not appropriate for theory
testing (Cadogan, 2010; Franke and Richey, 2010). For instance using statistical
analysis, Franke and Richey (2010) demonstrate that a minimum of 7–10 or
more countries may support a credible generalization of multi-country research.
However, Taylor (2014) favored the use of two or more countries and
emphasized that in the case of comparing few countries, scholars should present
compelling theoretical and practical reasons why specific countries need to be
compared. To sum up, future research should use a large number of countries
and when few countries are used, the argument regarding country selection
should be on a sound theoretical basis. Furthermore, the majority of ‘two or
three country’ articles compared culturally dissimilar countries, as did all articles
dealing with ‘four to six countries’ and ‘seven or more countries’. We argue that
examining culturally similar countries can provide some insights how a smaller
cultural difference impacts on advertising, and the applicability of Hofstede’s
framework. Because almost all of Hofstede-inspired advertising research has
focused on culturally dissimilar countries, we urge future research to compare
also advertising in culturally similar countries.
252 Saleem and Larimo

Table 3: Number of countries in selected articles


Number of countries Number of studies Percentage
One country 5 8.8
Two or Three countries 37 65
Four to Six countries 7 12.2
Seven or more countries 8 14
Total 57 100

6 Thematic Areas in Reviewed Studies


Among reviewed studies a clear majority of Hofstede-inspired advertising
research examined advertising appeals, and gender portrayal, perhaps scholars
preferred these topic areas because of their close theoretical association with
culture. Further, Table 4 also shows that research examining the effect of culture
on the basic formats of advertising is very limited.

Table 4: Frequency of topics


Topic area Number of Studies Percentage
Cultural Values/Appeals/ Themes 35 61.4
Character Portrayal 9 15.8
Comparative Advertising 4 7
Multiple Topics 4 7
Celebrity Endorsement 3 5.3
Visual Characteristics of Advertising 2 3.5
Total 57 100%

De Mooij (2013b) emphasized that the use and effectiveness of advertising


formats vary with culture; for example, an entertainment format is effective in
Japan but less so in the United States. Future research should investigate the use
and effectiveness of various advertising forms across cultures. Another relatively
ignored is green advertising, especially in the light of the evidence that pro-
environmental value orientation differs substantially among individuals holding
individualistic versus collectivistic values (Soyez, 2012). It would be useful to
shed more light on celebrity endorsements because the significance of celebrity
related aspects such as their creditability, trustworthiness, attractiveness,
likability and celebrity/product fit also varies across cultures.

7 Methodological Considerations
As can be seen in Table 5, the clearly dominant methodology was content
analysis followed by experiments/surveys, one study used both experiments and
content analysis and another employed both experiments and ethnographic
interviews. Qualitative techniques are valuable in advertising research and prior
incorporation of them may help the advertiser/researcher to develop advertising
Hofstede Cultural Framework and Advertising Research 253

for subsequent empirical testing (Craig and Douglas, 2012). Similarly


descriptive findings of content analysis can supplement the research by
incorporating the consumer perceptive through survey/experiments or qualitative
methods. This would allow researchers to combine evidence from different
sources to ensure the analytical enrichment and triangulation of findings.
However, only one study employed a mixed method and one other used content
analysis and experiments together, indicating that there is room for more of these
types of studies.

Table 5: Data collection method


Research Method Number of studies Percentage
Content Analysis 34 59.6
Experiment 16 28.1
Survey 5 8.8
Content Analysis + Experiments 1 1.8
Experiment + Ethnographic Interviews 1 1.8
Total 57 100

7.1 Content Analysis Based Studies


Despite wide-spread prevalence and popularity, scholars have identified
several methodological limitations of the content analysis method. For instance,
it is difficult to establish its reliability and validity: there is limited
generalizability of its results and an objective measure, so does not give the
consumer’s subjective experience with advertising (Okazaki and Mueller, 2007).
The remedy for the methodological limitations of traditional content analysis is
Lerman and Callow’s (2004) narrative coding approach (Okazaki and Mueller,
2007). Lerman and Callow (2004) argue that consumers’ interpretation of the
same advertising image may differ across cultures: for instance, the achievement
appeal in one country may be interpreted as taking care of oneself in another
country. Their analysis shows that narrative coding worked better than the
traditional content analytic technique. Consequently, they challenged the
findings of previous cross-cultural advertising studies (for details see Lerman
and Callow, 2004). Among reviewed studies all content analysis type advertising
appeal studies employed the traditional content analysis method; therefore, the
results of these studies could be simply the artifact of the method employed.
Future advertising appeal research may perhaps use a narrative coding method.
254 Saleem and Larimo

7.1.1 Sample Characteristics


As can be seen in Table 6, more than half of the studies analyzed magazine
advertisements, ten studies examined TV commercials, two studies each
examined internet advertising and newspaper advertising, and one article
analyzed both magazine ads and TV commercials. Because target audiences may
vary across various media vehicles, so does advertising. Therefore, there is room
for studies comparing and contrasting different media and attempting to uncover
potential media differences in the use of various advertising contents. Also,
magazine and TV advertising receive more research attention than other media
vehicles.

Table 6: Media focus in content analysis studies


Media Number of studies Percentage
Magazines advertisements 20 57.1
TV advertisements 10 28.6
Newspaper advertisements 2 5.7
Internet Advertising 2 5.7
Magazine + TV advertisements 1 2.9
Total 35 100

A recent industry report by Zenith Optimedia (2013) shows that global


advertising spending for internet advertising will increase by 15% during 2014-
2016, while the share of magazine and newspaper advertisements will decrease
at an average of 1%-2% per year. Future research may examine the effect of
culture on social media communications, mobile, and internet advertising.

7.1.2 Data Collection


All content analysis studies relied on university students or bilingual
individuals to code advertisements because of cost, convenience, and
accessibility. Employing professional advertisers or translators who are familiar
with the culture as judges to analyze advertising could provide some variation.
Further, like the review by Okazaki and Mueller (2007), our analysis also
revealed that most of the content analysis studies (except for only a few)
employed a nominal scale to decide the presence of certain elements in
advertising. There is an inherent danger in using a nominal scale for quantifying
some latent constructs such as advertising appeals because some critical
information on the variables is lost while measuring them through this type of
scale. To better address the coding issues, researchers may employ ordinal or
interval scales to code advertising appeal (for example, see Saleem et al., 2015).
An additional benefit of using ordinal or interval scales is that they give
researchers the opportunity to perform more sophisticated analysis like multiple
Hofstede Cultural Framework and Advertising Research 255

correspondence analysis or multidimensional scaling (Craig and Douglas, 2012).


Furthermore, advertising appeals studies frequently employed Pollay’s (1983)
list of values as a coding instrument to measure the reflection of the culture in
advertising. Scholars have criticized and questioned the applicability of Pollay’s
(1983) list of values for cross-cultural comparison of advertising because it
mainly contains American values (De Mooij 2015). To remedy this problem,
future research may use description and definition of culture by Hofstede as
coding categories.

7.1.3 Reliability
Among reviewed studies, thirteen studies used Perreault and Leigh’s (1989)
index (P&L), ten studies used a percentage of agreement, and five studies
adopted Kassarjian’s (1977) process of inter-coder reliability index.
Furthermore, two studies each employed Holsti’s (1969) and Cohen’s Kappa
and one study used Scott’s pi Formula.

Table 7: Use of reliability measures in content analysis studies


Inter-coder reliability index Number of Percentage
Studies
Perreault and Leigh (1989) 13 37
Percentage of agreement 10 28.6
Kassarjian (1977) 5 14.3
Holsti (1969) 2 5.7
Cohen’s Kappa 2 5.7
Other 3 8.7
Total 35 100

The inter-coder reliability index based simply on a percentage of agreement


between coders such as Holsti’s (1969) may not be regarded as valuable because
it does not take into account the possibility of agreements by chance.
Consequently, agreements by chance inflate the percentages of agreements and
reliability (Grayson and Rust, 2001). For this reason, Scott’s pi formula is
regarded as superior because it integrates the chance agreements. Scott’s pi
formula adjusts for chance agreements by manipulating the pooled ex-ante
coding decisions of the judges. But Scott’s pi formula has been criticized for
being conservative because it acknowledges only the agreements beyond chance.
The P&L index integrates chance agreements in a way that does not depend on
the marginal frequencies. Therefore, we argue that P&L index is a better
reliability index because it is straightforward and fits into many research
circumstances (Grayson and Rust, 2001). To sum up, the assumptions of each
inter-coder reliability index are different. Therefore, advertising researchers
should look at the appropriateness of various indexes in the context of the study
and data being analyzed. A preferable approach is to calculate and report two (or
256 Saleem and Larimo

more) indexes by establishing reasons that take into account the assumptions
and/or weaknesses of each.

7.1.4 Data Analysis Techniques


The content analysis studies focus on answering the question of differences
or similarities in the usage of appeals within or between countries. These types
of analysis are usually straightforward; therefore, more than one-third of studies
used Chi-square to establish the difference in use of appeals. Other commonly
employed statistical methods were descriptive statistics, MANOVA, ANOVA, t-
test, z-test, Pearson correlation, regression and some studies employed a
combination of various statistical techniques. Use of the above-mentioned
statistical techniques is quite logical and natural because a majority of the
studies were comparative or descriptive and utilized dichotomous variables.
Future research, however, may use alternative coding methods to make use of
continuous or interval type variables. This could possibly give the researcher an
opportunity to use some more sophisticated techniques like multiple
correspondence analysis.

7.2 Experimental/ Survey-Based Studies

7.2.1 Sample Characteristics


More than three-quarters of experiment/survey type studies used students and
only five studies used a non-student sample. The use of student samples has been
justified based on accessibility and homogeneity of the group. A recent
investigation shows that the predictive power of cultural values is stronger for
older males and working managers than for students (Taras et al., 2010). Since
various demographics moderate the effect of culture for a specific outcome,
therefore, scholars urge that cross-cultural research should observe caution
before adopting the common practices of sample matching by using university
and college students(Taras et al., 2010; Taylor and Bowen, 2012). As the
majority of reviewed articles used student subjects, the findings of these studies
may only reflect the impact of sample characteristics rather than real cross-
cultural differences or similarities. Future research is urged to ensure matched
samples by using more generalizable samples or a combination of student and
non-student populations.
Hofstede Cultural Framework and Advertising Research 257

7.2.2 Method of Culture Assessment


Among reviewed studies, more than half relied on Hofstede’s cultural
dimensions scores to assess culture. But Hofstede dimensions are national rather
individual-level constructs, and can only be used to study country level
phenomenon (e.g. level of literacy) (Hofstede et al., 2010). Also, the study by
Fischer et al. (2010) found that the structure of values at individual and country-
level are not similar. These findings confirm scholar’s assertion that cultural
dimension scores cannot be used to compare individuals across cultures
(Sharma, 2010). Furthermore, studies that measured cultural orientation can also
criticize for using instruments that lack in conceptualizing culture. For instance,
cultural values scale by Furrer et al. (2000) can be criticized for using
inconsistent and vaguely related items(Sun et al., 2014) and treating individual
cultural values as a bi-polar construct (Sharma, 2010). Recently De Mooij
(2015) criticized CVSCALE by Yoo et al. (2011) that for making Hofstede
questionnaire less work-related they have changed the content thus, causing
conceptual in-equivalence. Thus, future experiment/survey type advertising
research, should not only measure culture but also need to use refined measures
of culture. For example, the researcher may use a 40-item personal cultural
orientations scale developed by Sharma (2010), which measures Hofstede’s five
bi-polar cultural dimensions as ten personal cultural values at an individual
level.

7.2.3 Back-Translations of Survey and Experiment Instruments


Among reviewed studies, majority utilized Brislin’s (1980) back-translation
process and few studies did not mention/use any translation. As the meanings,
associations and interpretation of constructs often vary across the cultures and it
is critical to ensure that the constructs carry the equivalent meanings in different
contexts (He et al., 2008). Therefore, the translation process should help the
researcher to determine whether constructs have the equivalence of meaning in
different countries (Craig and Douglas, 2012). The traditional back translation
does not address the issues related to the comprehension and meaning of the
instrument by the respondent (Douglas and Craig, 2007). A direct and literal
translation of idioms and expressive statements, which are often used as
advertising appeals, by moving from English to another language and back
again, may produce the same wording, which would suggest that the translation
was accurate. The remedy to this problem is Douglas and Craig’s (2007)
‘collaborative and iterative translation’ approach, that cross-cultural team
members should translate and test the instrument and revise the process until a
satisfactory translation is achieved. Lastly, the non-verbal stimuli are also part of
the advertising (Craig and Douglas, 2012), but none of the studies have paid
258 Saleem and Larimo

attention to their translation. In order to avoid the miscommunication or wrong


interpretation of advertising, we suggest that future research should also translate
the non-verbal stimuli to ensure that they evoke the desired image.

Table 8: Translation/back-translation process in survey and experiment studies


Number of Percentage
studies
Brislin (1980) 12 52.2
Werner and Campbell (1970) 1 4.3
McGorry (2000) 1 4.3
Not mentioned 7 30.4
Not Applicable 2 8.7
Total 23 100

7.2.4 Reliability and Validity


Among reviewed studies, more than half studies used Cronbach’s Alpha, four
studies utilized confirmatory factor analysis (CFA), one study used exploratory
factor analysis, and five studies did not use or mention any measure of
invariance. All these methods of measurement have some inherent weaknesses.
For instance, an item may be meaningful in one culture but a misfit in another. In
this case, the CFA approach usually discards the differently functioning item. In
Rasch approach, an item parameter can be estimated easily for one group where
it fits well while the same item is discarded for the other groups where it is
meaningless (Ewing et al., 2005). In other words, CFA model relies on one
particular item for defining latent variables, and the Rasch model allows for
testing whether an item follows the model or not. A comparison of the CFA and
Rasch models with real data shows that Rasch measures have higher precision
and accuracy and that it provides a stronger justification of equivalent of
measure related to individual respondents (Salzberger and Sinkovics, 2006). As
Rasch model is sound on theoretical and practical basis compared to CFA; we
urge experiment/survey type cross-cultural advertising research to use this
method for measuring equivalence.

7.2.5 Data Analysis Technique


The studies based on experiments and surveys deal with the cross-cultural
differences in the effectiveness of various advertising contents such as appeals in
different countries. Therefore, the natural choice of statistical analysis in
reviewed studies was ANOVA, MANOVA, and Chi-square. Other commonly
used statistical methods were regression analysis, t-test, and only one study
employed structure equation modeling (SEM). Clearly Hofstede-inspired
advertising research is lacking in the use of sophisticated statistical techniques.
Hofstede Cultural Framework and Advertising Research 259

Therefore, future research should use the second generation statistical analysis
like SEM to achieve psychometric standards (i.e. reliability and validity).

7 Conclusion and Managerial Implications


We reviewed 57 advertising articles that used Hofstede’s framework for
advertising research that appeared between 1992 and 2012 in various journals.
Hofstede-inspired advertising research seldom paid attention to the discrepancy
between the cultural values and practices. Further, Hofstede-inspired advertising
research is overly reliant on certain topic areas, cultural dimensions, and
geographical regions. Based on the review, it can be concluded that Hofstede-
inspired advertising research has placed insufficient attention on methodological
aspects and that more attention needs to be paid to the accurate and effective
utilization of Hofstede’s framework in advertising research.

Table 9: Summary of recommendations


Distinction between the desirable and the desired
Explore the overlap or discrepancy between the cultural values and practices
Content analysis studies may combine evidence from ED, RA, and Hofstede scores to
examine discrepancy between values and practices
Experimental studies should measure the cultural values as well as practices to determine the
extent to which both facets impact the attitude toward advertising
Cultural dimension
Need to focus on the effect of long-term orientation and indulgence/restraint dimension
on advertising
Need to focus on interaction effect among cultural dimensions to investigate the link
between culture and advertising
Topic areas
Need to focus on cross-cultural difference of various advertising formats, green
advertising, and the celebrity-related factors across cultures
Geographic Scope
Use larger number of countries - a minimum 7-10 or more for credible generalization
In case of using few countries provide compelling rationales for country selection
Need to compare culturally similar countries
More studies on Eastern and Southern Europe, BRIC, Latin America, and Africa
Methodology
Content analysis
Use mixed method, ordinal and ratio scale to code ads and narrative coding
Need to perform cross-media comparisons, and analyze social media and mobile
advertising
Experiment and Survey
Use of iterative and collaborative translation technique
Measure individual’s value orientation while analyzing his/her preference of advertising
Translate the non-verbal stimuli
Use more generalizable samples of respondents
For equivalence use the Rasch model
Need to use second generation statistical techniques like SEM
260 Saleem and Larimo

IVI (indirect value inference i.e. Hofstede’s dimension scores); DVI (direct value inference);
ED (ethnographic description); RA (regional affiliations)

From managerial perceptive our findings suggest that cultural values alone
are insufficient to predict the influence of culture on advertising, rather cultural
practices should also be taken into account. The implication is that understating
of various facets of culture might help managers in deciding advertising content
across culture. Therefore, advertisers are encouraged to resort on the
ethnographic description, regional affiliation, and Hofstede’s dimensions scores,
to explore values that might be appealing to the consumer. Besides thinking the
culture in terms of values and practices, managers may have to understand that
predictive power of cultural values is stronger for non-student samples.
Therefore, only student samples are inadequate. Last but not least advertising
managers need to pay attention to a number of other aspects such as accuracy in
translation of advertising copy, the effect of non-verbal stimuli across cultures
and use of various analytical techniques. To sum up, we identified several
shortcomings and outlined suggestions for scholars, journal editors, reviewers,
and managers responsible for international advertising research (see Table 9).

9 References

Albers-Miller, N. D. and Gelb, B. D. (1996), "Business Advertising Appeals as a Mirror


of Cultural Dimensions: A Study of Eleven Countries," in: Journal of Advertising,
Vol. 25, 4, 57-70.
Brislin, R.W. (1980), "Translation and Content Analysis of Oral and Written Material,"
in: Triandis, H. and Berry, J.W., ed. Handbook of Cross-Cultural Psychology,
Boston, Allyn and Bacon, 389-444.
Cadogan, J. (2010), "Comparative, Cross-Cultural, and Cross- National Research: A
Comment on Good and Bad Practice," in: International Marketing Review, Vol. 27,
6, 601-05.
Chang, T. K.; Huh, J.; McKinney, K.; Sar, S.; Wei, W. and Schneeweis, A. (2009),
"Culture and Its Influence on Advertising: Misguided Framework, Inadequate
Comparative Design and Dubious Knowledge Claim," in: International
Communication Gazette, Vol. 71, 8, 671-692.
Choi, Y.K. and Miracle, G.E. (2004), "The Effectiveness of Comparative Advertising in
Korea and the United States: A Cross-Cultural and Individual-level analysis," in:
Journal of Advertising, Vol. 33, 4, 75-87.
Craig, C.S. and Douglas, S.P. (2012), "International Advertising Research: Conceptual
and Methodological Issues," in: Okazaki, S., ed. Handbook of Research on
International Advertising. Cheltenham, UK, Edward Elgar Publishing, Inc. 201-230.
De Mooij, M. (2015),"Cross-Cultural Research in International Marketing: Clearing Up
Some of the Confusion," in: International Marketing Review, Vol. 32, 6, 646 - 662.
De Mooij, M. (2013a),"On the Misuse and Misinterpretation of Dimensions of National
culture," in: International Marketing Review, Vol. 30, 3, 253-261.
Hofstede Cultural Framework and Advertising Research 261

De Mooij, M. (2013b), Global Marketing and Advertising: Understanding Cultural


Paradoxes. Thousand Oaks, CA: Sage Publications.
Douglas, S.P. and Craig, C.S. (2007), "Collaborative and Iterative Translation: An
Alternative Approach to Back Translation," in: Journal of International Marketing,
Vol. 15, 1, 30-43.
Ewing, M.T.; Salzberger, T. and Sinkovics, R.R. (2005), "An Alternate Approach to
Assessing Cross-Cultural Measurement Equivalence in Advertising Research," in:
Journal of Advertising, Vol. 34, 1, 17-36.
Fischer, R. (2006), "Congruence and Functions of Personal and Cultural Values: Do My
Values Reflect My Culture’s Values?," in: Personality and Social Psychology
Bulletin, Vol. 32, 11, 1419-1431.
Fischer, R.; Vauclair, C.M.; Fontaine, J.R. and Schwartz, S.H. (2010), "Are Individual-
level and Country-level Value Structures Different? Testing Hofstede’s Legacy with
the Schwartz Value Survey," in: Journal of Cross-Cultural Psychology, Vol. 41, 2,
135-51.
Franke, G.R. and Richey, R.G. (2010), "Improving Generalizations from Multi-country
Comparisons in International Business Research", in: Journal of International
Business Studies, Vol. 41, 8, 1275-93.
Furrer, O.; Liu, B. S. C. and Sudharshan, D. (2000), "The Relationships between Culture
and Service Quality Perceptions Basis for Cross-cultural Market Segmentation and
Resource Allocation," in: Journal of Service Research, Vol. 2, 4, 355-371.
Gibson, C.B.; Maznevski, M. and Kirkman, B. (2009), "When Does Culture Matter? " in:
Bhagat, R. S. and Steers, R. M., ed. Cambridge Handbook of Culture,
Organizations, and Work, Cambridge, Cambridge University Press, 46-70.
Grayson, K. and Rust, R. (2001), "Interrater Reliability," in: Journal of Consumer
Psychology, Vol. 10, 1, 71-73.
He, Y.; Merz, M. A. and Alden, D. L. (2008), "Diffusion of Measurement Invariance
Assessment in Cross-national Empirical Marketing Research: Perspectives from the
Literature and a Survey of Researchers," in: Journal of International Marketing, Vol.
16, 2, 64-83.
Hofstede, G.; Hofstede, G.J. and Minkov, M. (2010), Cultures and Organizations:
Software of the Mind. Third edition, London: McGraw-Hill.
Holsti, O.R. (1969), Content Analysis for the Social Sciences and Humanities. Reading:
Addison- Wesley.
House, R. J.; Hanges, P. J.; Javidan, M.; Dorfman, P. W. and Gupta, V. (2004), Culture,
Leadership, and Organizations: The Globe study of 62 Societies. Thousand Oaks,
CA: Sage publications.
Kassarjian, H.H. (1977), "Content Analysis in Consumer Research," in: Journal of
Consumer Research, Vol. 4, 2, 8-18.
Kirkman, B. L.; Lowe, K. B. and Gibson, C. B. (2006), "A Quarter Century of Culture’s
Consequences: A Review of Empirical Research Incorporating Hofstede’s Cultural
Values Framework," in: Journal of International Business Studies, Vol. 37, 3, 285-
320.
Lerman, D. and Callow, M. (2004), "Content Analysis in Cross-cultural Advertising
Research: Insightful or Superficial?," in: International Journal of Advertising, Vol.
23, 4, 507-21.
262 Saleem and Larimo

Mortimer, K. and Grierson, S. (2010), "The relationship between Culture and


Advertising Appeals for Services," in: Journal of Marketing Communications, Vol.
16, 3, 149-62.
Okazaki, S. and Mueller, B. (2007), "Cross-cultural Advertising Research: Where we
have been and Where we Need to Go," in: International Marketing Review, Vol. 24,
5, 499-518.
Paek, H.J.; Nelson, M.R. and Vilela, A.M. (2011), "Examination of Gender-role
Portrayals in Television Advertising across Seven Countries," in: Sex Roles, Vol. 64,
3-4, 192-207.
Perreault, W.D. and Leigh, L.E. (1989), "Reliability of Nominal Data Based on
Qualitative Judgments," in: Journal of Marketing Research, Vol. 26, 2, 135-148.
Saleem, S.; Larimo, J. A.; Ummik, K. and Kuusik, A. (2015), "Cultural and Paradoxical
Values in Advertising in Eastern Europe: Evidence from Estonia," in: Baltic Journal
of Management, Vol. 10, 3, 313-330
Salzberger, T. and Sinkovics, R.R. (2006), "Reconsidering the Problem of Data
Equivalence in International Marketing Research: Contrasting Approaches Based on
CFA and the Rasch Model for Measurement," in: International Marketing Review,
Vol. 23, 4, 390-417.
Sharma, P. (2010), "Measuring Personal Cultural Orientations: Scale Development and
Validation," in: Journal of the Academy of Marketing Science, Vol. 38, 6, 787-806.
Shavitt, S.; Johnson, T.P. and Zhang, J. (2011), "Horizontal and Vertical Cultural
Differences in the Content of Advertising Appeals," in: Journal of International
Consumer Marketing, Vol. 23, 3-4, 297-310.
Soares, A.M.; Farhangmehr, M. and Shoham, A. (2007), "Hofstede’s Dimensions of
Culture in International Marketing studies," in: Journal of Business Research, Vol.
60, 3, 277-84.
Soyez, K. (2012), "How National Cultural Values Affect Pro-environmental Consumer
Behavior," in: International Marketing Review, Vol. 29, 6, 623-46.
Sun, G.; D’Alessandro, S.; Johnson, L. W. and Winzar, H. (2014), "Do We Measure
What We Expect to Measure? Some Issues in the Measurement of Culture in
Consumer Research," in: International Marketing Review, Vol. 31, 4, 338-362.
Taras, V.; Steel, P. and Kirkman, B. L. (2010), "Negative Practice–Value Correlations in
the GLOBE data: Unexpected Findings, Questionnaire Limitations and Research
Directions," in: Journal of International Business Studies, Vol. 41, 8, 1330-1338.
Taylor, C.R. (2014), "Conducting State-of-the-Art Cross-Cultural Research," in:
International Journal of Advertising, Vol. 33, 4, 627-31.
Taylor, C.R. and Bowen, C.L. (2012), "Best Practices for Cross-Cultural Advertising
Research: Are the Rules Being Followed?," in: Okazaki, S., ed. Handbook of
Research on International Advertising, Cheltenham, UK, Edward Elgar Publishing,
Inc. 3-19.
Yoo, B.; Donthu, N. and Lenartowicz, T. (2011), "Measuring Hofstede’s Five
Dimensions of Cultural Values at the Individual Level: Development and Validation
of CVSCALE," in: Journal of International Consumer Marketing, Vol. 23, 3-4, 193-
210.
Hofstede Cultural Framework and Advertising Research 263

Zenith Optimedia. (2013), "Executive Summary: Advertising Expenditure Forecasts


December." http://www.zenithoptimedia.com/wp-content/uploads/2013/12/Adspend-
forecasts-December-2013-executive-summary.pdf/
The Effects of Different Ad Appeals in Non-Prescription Drug
Advertising: A Cross-Cultural Investigation

Isabell Koinig, Sandra Diehl, and Barbara Mueller

1 Introduction
Concerns regarding health dominate the 21st century (Krisberg, 2004).
Recent years have been characterized by a “health communication renaissance”
(Bernhardt, 2004: 2051). Health communication refers to “any type of
communication whose content is concerned with health” (Rogers, 1996: 15),
addressing individuals and organizations with the goal of preventing illness and
fostering health (Thompson et al., 2011). This also encompasses pharmaceutical
ads, which “can be defined as messages created by marketers of pharma
products that attempt to inform, persuade and even entertain the target audience
with the goal of influencing recipient’s attitudes – and ultimately behavior – in a
favorable manner” (Diehl et al., 2008: 100). Ads are utilized to promote
prescription and non-prescription drugs; the latter can be categorized as self-
medication preparations, which are directly promoted to physicians and
consumers alike and are the only kind of promotion allowed by law in the E.U.
In the United States, in contrast, both prescription and non-prescription drugs
can be advertised directly to consumers (Buckley, 2004).
The bulk of academic research to date has focused on direct-to-consumer
prescription drug ads (DeLorme et al., 2010). Only minimal research has been
conducted in the area of non-prescription drug ads (also known as over-the-
counter or OTC advertising), suggesting that our understanding of the impact of
such messages might be outdated as well as incomplete. Apart from very few
contributions (Main et al., 2004; Diehl et al., 2008; DeLorme et al., 2010),
neither consumer evaluations of OTC drug ads, nor the topic of empowerment
have been addressed. Given that a more thorough understanding of the
effectiveness of different ad appeals in pharmaceutical advertising can benefit
both academics and practitioners, the present study will attempt to contribute to
this gap in the academic literature. The countries examined in this study are the
U.S., the largest single drug market, Germany, the largest European medications
market, Austria, a second Euro market, as well as Brazil, an emerging OTC drug
market as well as the largest South American pharmaceutical market
(MarketLine, 2014). In all four countries, OTC drug sales are on the rise; in
addition, many former prescription drugs (Rx) are now available as over-the-
counter (OTC) preparations, as their patents expire.
An extensive content analysis (Koinig and Diehl, 2013) revealed strong
potential for standardized ad executions in the U.S. and the two E.U. markets

© Springer Fachmedien Wiesbaden 2017


G. Christodoulides et al. (Hrsg.), Advances in Advertising Research (Vol. VII),
European Advertising Academy, DOI 10.1007/978-3-658-15220-8_19
266 Koinig, Diehl, and Mueller

(all generally individualistic in orientation). This study extends previous


research by incorporating a quantitative consumer survey and adding a Latin
American country, which represents a significantly more collectivistic
orientation. The present investigation is a cross-cultural endeavor and anticipates
differences in ad evaluation and degree of consumer self-empowerment, which
might be attributed to Individualism/Collectivism (Hofstede, 1980; House et al.,
2004). This dimension is expected to influence ad reception, as written
information is highly valued in individualistic cultures such as Austria,
Germany, and the U.S. (Hofstede, 1980), where messages are communicated in
a clear, comprehensible fashion (Wells et al., 2006). Communication in a
collectivistic nation such as Brazil is marked by significant differences;
respondents value emotions more than facts, often seek entertainment and prefer
visuals over factual data (de Mooij, 2011). As a consequence, different ad
appeals are predicted to lead to varying degrees of attitude toward the ad and
empowerment.

2 Literature Review and Hypotheses


On the Relationship between Health, Culture and Communication:
Health, often referred to as a fundamental human right (WHO, 2013), is an
intangible and abstract concept that is surprisingly hard to define (Earle, 2007).
According to the WHO’s definition, health refers to “a state of complete
physical, mental and social well-being”, implying the lack of disease (WHO,
2006). Communication is, without question, a valuable asset in the health
domain: “Health communication is an approach which attempts to change a set
of behaviors in a large-scale target audience regarding a specific problem in a
predefined period of time” (Clift & Freimuth, 1995: 68). Whether employed by
(commercial) manufacturers or (non-commercial) public service providers, such
messages draw attention to the factors that impact health, appealing to
individuals’ behaviors and capacities (Green & Tones, 2010). By raising the
awareness of, and familiarity with. medical conditions, the mass media assist
patients in reducing uncertainties about health issues, enabling them to gather
useful information in order to come to qualified decisions (Wright et al., 2008).
Linking the two areas of health and communication, health communication
refers to the practice of communicating information regarding health topics to a
widely dispersed mass audience (U.S. Department of Health and Human
Services, 2014). It is predominantly utilized for both educational and
commercial purposes, encompassing a variety of activities, e.g., public health
campaigns, health education materials, as well as doctor-patient interactions
(Schiavo, 2007). The provision of (mass-mediated) information is singular in its
goal: to foster individual health literacy and influence personal decisions with
regard to health and well-being (U.S. Department of Health and Human
The Effects of Different Ad Appeals in Non-Prescription Drug Advertising 267

Services, 2014). As such, the media have the ability to influence and even alter
people’s health behaviors (Wright et al., 2008), functioning as a leveler by
elevating lay persons to a more equal footing with professionals (Parrott, 2003).
As part of the commercialization of health, health has been turned into a
commodity: health products are heavily advertised and can be acquired through
financial means. The term “patient” – referring to a passive individual in the
hands of medical experts – has been substituted with the more pro-active
concept of a “health consumer” (Porter, 1985), who is actively involved in
his/her health-care and does not hesitate to question medical opinions (Nettleton,
1997). Influenced by technological advances such as the Internet, consumers
have begun to take a hands-on approach to their health by looking for, and
demanding, information relevant to their well-being (Auton, 2004).
Pharmaceutical ads present one such source of information, with the potential to
contribute to consumer education (Morse, 1993) by involving consumers’ very
own skills in the process (Morse, 1993).
Both health and communication are embedded in cultural settings, as
individuals are socialized in a particular cultural context that exposes them to, as
well as familiarizes them with, particular values, norms and beliefs that are
expressive of individuals’ respective cultures. Culture functions as “the
collective programming of the mind that distinguishes the members of one group
or category of people from other [cultures and/or nations]” (Hofstede, 1984: 51)
and, thus shapes responses to promotional messages. Yet, the influence of
culture is said to be diminishing. Consumers are becoming increasingly global,
driven by similar values throughout the world (van lttersum & Wong, 2010),
making standardized advertising a fruitful strategy.
The concept of advertising standardization originated from experts’ on-going
claims of an emerging global culture with unified values (Melewar &
Vemmervik, 2004). Over the years, ad standardization has become one of the
most commonly researched yet controversial topics in the marketing realm
(Okazaki et al., 2006). As research on the perception and evaluation of
standardized ad messages in different cultures is rather limited (Hudson et al.,
2002; Callow & Schiffman, 2004), the present paper will try to add to the body
of research in this area.

Ad Appeal Categories: A distinguishing aspect that separates successful


from unsuccessful promotions is the ad appeal, the “life giving spark of an
advertisement” (Kleppner, 1979). It refers to any kind of promotional message
that pursues the goal of motivating consumers to seek (and purchase) the product
or service mentioned therein (Mueller, 1987). Hence, it describes the way in
which advertisers hope to trigger attention in recipients, influencing and shaping
their interests and feelings towards a particular product (Belch & Belch, 1993).
268 Koinig, Diehl, and Mueller

Among the various categories of appeals are hard-sell or informative versus soft-
sell or emotional advertising appeals. Informative appeals are very direct,
invoking people’s rational thinking by utilizing reason. Explicit message features
are characteristic of hard-sell appeals (Okazaki et al., 2010). Emotional appeals,
in contrast, take a more subtle approach (Okazaki et al., 2010), working with
atmospheric stimuli and telling stories to create favorable impressions in the
minds of consumers (Beard, 2004). These two approaches blur at times,
resulting in a mixed advertising appeal, which is characterized by both product
specifics combined with emotional statements (Kroeber-Riel and Esch, 2011).
Ad effectiveness is also heavily dependent upon the product category being
promoted. As they impact people’s well-being, pharma products are considered
to be particularly highly involving: they are purchased on an irregular basis,
require extensive information searching, and the perceived risk of selecting the
wrong brand is high (Diehl et al., 2007). With regard to ad appeals, this product
class benefits from rational arguments and factual information stressing product
benefits (Okazaki et al., 2010; 2013). Consequently, when promoting OTC
drugs and similar high-involvement products, rational appeals are commonly
employed in individualistic societies such as Austria, Germany and the U.S.
(Okazaki et al., 2010; 2013). As high-involvement goods, advertisers’ claims are
subject to serious questioning and high levels of skepticism (Mueller, 2006;
Obermiller et al., 2005), which can best be allayed via the use of information-
oriented content. Thus, the following hypothesis is proposed:

H1: Ad evaluation of the informative OTC drug ad will be more favorable than
that of the mixed appeal ad, which will be more favorable than that of the
emotional appeal ad in all countries.

Advertising as a Source of Consumer Self-Empowerment: Advertising


messages can provide information that empowers consumers to arrive at
qualified decisions (Assael, 2004). In the health context, empowerment involves
“wrestling control for the consumer” (Worcester and Whately, 1992: 23), and
increasing individual control and responsibility for health-related matters
(Grace, 1991). Put simply, “to empower means to give power” (Thomas and
Velthouse, 1990: 667), and it is also concerned with strengthening individual’s
self-efficacy and self-control (Rose, 2000), which is said to result in a more
exact match between consumers’ needs and products offerings (Kreps, 1979).
Empowerment is directly related to the health-care sector; today’s patient is
no longer passive but has emerged as a health-care consumer (Aspect, 2011),
who cares for himself/herself through consumption as a “rational and self-
interested agent” (Denegri-Knott et al., 2006: 963). The empowered patient
“takes an active role in their disease management and understands […] the
The Effects of Different Ad Appeals in Non-Prescription Drug Advertising 269

condition and the treatment options available” (Kantar Health, 2014). In order to
make sense of health matters and become empowered, information which is
perceived as relevant and ‘enabling’ by consumers is essential (HAI, 2006).
As such, empowerment is a subjective experience which presupposes
consumer self-confidence (Eylon, 1998). Consumers are called upon to believe
in their own capabilities and skills – a process referred to as self-empowerment
(Tones & Tilford, 2001). Self-empowerment supposes a relatively high level of
self-esteem along with a repertoire of (health-promoting) capabilities that enable
individuals to exercise power over their lives and health (Tones & Tilford,
2001). In this context, the availability of information is regarded as maximizing
consumer benefits (Davies & Elliott, 2006) and, consequently, also their degree
of (self-) empowerment. The level of individual empowerment is, thereby,
subject to (1) the amount of information available to the consumer/patient and
(2) their ability to make sense of this information (Newholm et al., 2006).
Empowerment and the selection of OTC drugs are facilitated by education;
consumers’ lay knowledge is supplemented with information received from a
credible (expert) source, may it be a doctor, pharmacist, or promotion (Wilson,
2001). Though heavily criticized for their biased claims and superficial
discussion of risks and side effects (HAI, 2009), ads can serve to involve
patients: “Informed patient participation in health care choices is impossible
without access to accurate, comprehensive, unbiased information about the pros
and cons of all available treatment options” (Coulter, 2003: 39). For this reason,
pharmaceutical companies are called upon to disseminate “information that is
accurate and medically sound” (Kantar Health, 2014) or provide links to such
sources. Thus, the following hypothesis is proposed:

H2: A higher degree of self-empowerment will result from the informative OTC
drug ad than from the mixed appeal ad, which will be higher than from the
emotional appeal ad in all countries.

3 Method

3.1 Stimulus Ads


Three full-page ads were developed, which differed in their message
appeal: informative, emotional, and mixed. The informative ad featured the
advertised drug in a dominant visual and was paired with a short body copy
emphasizing the product’s effectiveness, unique composition, plus its tolerance.
The slogan included the product name in a problem/solution manner: “Getting
the best out of life despite pain – with Senza!” The emotional version used the
270 Koinig, Diehl, and Mueller

same slogan, combined with a visual depicting a happy-looking couple that


occupied 90 % of the total page. In terms of an emotional appeal, happiness was
employed - the only state that is universally identifiable (Russel, 1994). Health is
also regarded as a pre-requisite for happiness (Rod & Saunders, 2004). The
mixed appeal presented a combination of the informative and emotional
versions: while it employed the latter’s prominent visual, it also incorporated the
textual information on product specifics. To avoid the impact of an established
and well-known brand name (Yin, 1999), a fictitious brand name (‘Senza’,
alluding to ‘without’ pain) was incorporated. The ads were pre-tested to ensure
they were indeed perceived as portraying the three appeal categories. The ads
were translated into English, German and Portuguese via the translation/back-
translation method (please see Appendix 1 for the English version of the ads).
The study was conducted with a non-student sample of 967 respondents
recruited from March to August 2014 in Austria (n= 240), Germany (n= 244),
the United States (n= 242), and Brazil (n= 241). All interviewers received
extensive training and were provided with a standardized text to employ when
approaching subjects. Respondents were between 18 and 93 years of age, with
an average age of 35.9 years across the four countries examined. In all countries,
subjects were recruited in mid-size cities and represented a broad range of
professions. Female/male participation was almost equally distributed (f =
50.6%; m = 49.4%). A structured questionnaire was developed and translated
into the three languages using the translation/back-translation technique.
Participants were approached in public places. After completing questions
regarding their health as well as their general attitude towards pharmaceutical
advertising, participants were presented with one of the three stimulus ads. After
exposure to the ad, subjects were asked to answer questions related to ad
evaluation, message involvement, attitude towards the brand, as well as the
empowerment garnered from the message, and finally purchase intention.
Responses to each question were measured on a 7-point Likert scale ranging
from (1) ‘I do not agree at all’ (7) ‘I fully agree’.

3.2 Measures
For Ad Evaluation, established scales were consulted (Zuckermann, 1988;
McKenzie & Lutz, 1989) and comprised of the following items: “interesting”,
“pleasant”, “favorable”, and “positive”. For consumer self-empowerment,
however, no suitable scale could be transferred to the advertising realm. Thus, a
scale was developed for purposes of this investigation, adopting three items from
self-efficacy and consumer self-confidence scales (Schwarzer & Jerusalem,
1995; Bearden et al., 2001). Principal component analyses revealed that the
three items measuring ad evaluation (Cronbach α = .889) and the three items
The Effects of Different Ad Appeals in Non-Prescription Drug Advertising 271

measuring self-empowerment (Cronbach α = .886) each loaded on one single


component. The authors computed the mean values for each construct.

4 Results
Ad Evaluation: Of the numerous variables that were measured in this
research project, one variable was selected for analysis in this chapter, namely
attitude towards the ad or ad evaluation, proclaimed to be the strongest predictor
of ad effectiveness (MacKenzie et al., 1986). Hypothesis 1 sought to explore
which ad appeal yielded the most favorable responses in terms of attitude
towards the ad, proposing informative appeals to outrank not only mixed, but
also emotional appeals. The four items “interesting”, “pleasant”, “favorable”,
and “positive” constituted the factor Ad Evaluation (Cronbach α: .889). Results
revealed that in terms of ad liking (ad evaluation), the mixed appeal version,
which combined rational arguments with atmospheric images (Kroeber-Riel and
Esch, 2011) outperformed both the informative and emotional versions. The
mixed approach received the most favorable ratings amongst respondents in
three out of four countries (T: 4.534; AUT: 4.371; GER: 4.303; USA: 4.167;
BRA: 5.308) followed by the purely informative appeal (T: 4.344; AUT: 4.212;
GER: 4.311; USA: 4.036; BRA: 4.817). The purely emotional ad scored lowest
in all four countries (T: 3.884; AUT: 3.831; GER: 3.970; USA: 3.229; BRA:
4.506).
ANOVAs revealed that in terms of ad evaluation, significant differences
were only found in the U.S. (F = 3.279, p = 0.040) and in Brazil (F = 5.108, p =
0.007). In Austria (F = 0.096, p = 0.908) and Germany (F = 1.654, p = 0.194),
the differences were not significant. Contrast tests in all countries revealed no
significant differences between the evaluation of the informative and mixed ad in
any of the four countries. The same applied to the differences between
informative and emotional ads. These parts of the hypothesis H1 must then be
rejected. With regard to the comparison of the mixed and emotional ads
however, the mixed ad received significantly more positive evaluations in the
U.S. and Brazil than the emotional ad. At least in these two countries, this aspect
of the hypothesis was supported (see Appendix 1). Overall however, the
differences in evaluation of the three different ad appeals in the four countries
were not so distinct, therefore H1 must be rejected.
272 Koinig, Diehl, and Mueller

Figure 1: Ad evaluation across countries

Consumer Self-Empowerment: Hypothesis 2 sought to test the degree of


consumer self-empowerment derived from each ad, postulating that informative
ads would provide consumers with the greatest degree of self-empowerment,
followed by the mixed and emotional advertisements. Three questions addressed
message empowerment as derived from the message, which were factorized into
the factor Message Empowerment (Cronbach α: .886): (1) The ad allows me to
objectively evaluate Senza, (2) I can now accurately compare Senza with other
comparable brands on matters that are important to me, and (3) I would have
more confidence in using Senza now than before I saw this commercial In line
with findings related to ad evaluation, the mixed appeal ad received the highest
scores (T: 4.1274; AUT: 3.7389; GER: 3.7833; USA: 3.5433; BRA: 4.8508).
The second highest scores were reported for the informative ad (T: 3.4141;
AUT: 3.1567; GER: 3.1049; USA: 3.3600; BRA: 4.0400), while the least
degree of message empowerment was generated by the emotional ad (T: 2.6376;
AUT: 2.4603; GER: 2.3672; USA: 2.0935; BRA: 3.6194).
With regard to the variable self-empowerment, differences between the
four ads were more distinct in the four countries. ANOVAs revealed significant
differences in Germany (F = 5.635, p = 0.004), the US (F = 8.696, p = 0.000)
and Brazil (F = 10.202, p = 0.000). In Austria (F = 2.798, p = 0.064) the
The Effects of Different Ad Appeals in Non-Prescription Drug Advertising 273

differences were only significant at the 10% level. For self-empowerment,


contrast tests in each country explored differences between the various ad
appeals. Looking at the three individualistic countries (Austria, Germany and
US), perceived self-empowerment did not differ regardless of whether an
informative or mixed ad was used. However, in terms of the informative and
emotional ad, differences were significant in all three countries, indicating that
an informative ad led to higher self-empowerment when compared to an
emotional ad. The same was true for the degree of self-empowerment induced by
a mixed vs. an emotional ad. Self-empowerment was higher for mixed than for
emotional ad (the difference was not significant for Austria, however). The
collectivistic country Brazil revealed a somewhat different pattern: here the
mixed ad led to a higher degree of self-empowerment as compared with both the
informative and emotional ads. However, no significant differences were found
for the comparison between informative and emotional ads. In all countries, the
assumption that the informative ad would lead to a higher level of self-
empowerment than the mixed ad could not be confirmed. The second part of the
hypothesis, which suggested informative appeals to be superior to emotional
appeals when it comes to self-empowerment, was confirmed for three of the four
countries. The same was true for the third part of the hypothesis, which assumed
the superiority of the mixed appeal when compared with the emotional appeal;
here again the mixed appeal outperformed the emotional appeal in three of four
countries. Therefore our data supports the second and third parts of hypothesis
H2. Overall, however, H2 can only be considered as partially supported (see
Appendix 2 for the detailed results of the contrast tests). The assumption that
information contributes to consumers’ self-empowerment is supported in so far
as all ad versions containing factual text blocks outranked their exclusively
emotional counterpart1.3

1 All differences point in the right direction, but 2 out of 8 differences are not significant
(see Appendix 2).
274 Koinig, Diehl, and Mueller

Figure 2: Consumer self-empowerment across countries

5 Discussion, Limitations and Directions for Future Research


The purpose of this investigation was to add to the body of knowledge
regarding the influence of culture on consumer responses to standardized OTC
drug ads in a cross-cultural setting. Study results demonstrate that with regard to
ad evaluation only limited differences were found, but in two countries (Brazil
and the U.S.), the mixed appeal turned out to be the most successful, outranking
both the informative and the emotional approach. Reasons for the mixed
appeal’s favorable evaluation might lie in the fact that in addition to product
information, the desired end result (the emotion of happiness) possibly triggered
positive responses. Study results further suggest that a well-designed,
standardized drug ad can influence consumers’ sense of self-empowerment.
With regard to the three individualistic countries (Austria, Germany and US),
perceived self-empowerment did not differ regardless of whether an informative
or mixed ad was used, however it was lower when an emotional ad appeal was
used. In Brazil, a collectivistic country, the mixed appeal was also superior to
the emotional ad when it came to fostering self-empowerment and in contrast to
the other countries also superior to the informative ad. This again suggests that
the mixed appeal is the most fruitful approach.
Brazilian citizens, as collectivistic citizens are said to be more visually
oriented and favor implicit and image-centered messages. However these
The Effects of Different Ad Appeals in Non-Prescription Drug Advertising 275

preferences were not reflected in the data collected: The overall lowest values
were accounted for the emotional ad which was neither able to create positive ad
evaluations nor high levels of consumer self-empowerment, occupying the last
rank in three out of four countries. Taking all these results together, the mixed
appeal appears to be the most promising candidate for cross-cultural application,
when ad evaluation and self-empowerment are considered. Advertisers might be
well advised to refrain from using emotional OTC drug ads, as they led to lower
self-empowerment than informative and mixed ads (in three out of four
countries) . They are also more negatively evaluated particularly when compared
to mixed appeals (especially in Brazil and the U.S.). Neither the assumed
preference for an emotional ad in collectivistic countries, nor the assumed
preference for an informative ad in individualistic countries, could not be
confirmed in the present study. Instead, the mixed appeal turned out to be most
fruitful in all countries.
The findings of this investigation are of relevance to various audiences. For
pharmaceutical advertisers and international marketers, the data suggest that a
standardized advertising campaign might be of advantage, and established that
certain tactical approaches (utilizing predominantly informative appeals) might
be of value. The findings are important to health-care providers and
governmental parties, as well. In a time of increasing health expenses and
decreasing reimbursements (Paasche-Orlow and Wolf, 2007; Brems et al.,
2011), individuals might refrain from seeking medical treatment altogether for
fear of accruing medical bills. Hence, governments and pharmaceutical
manufacturers should team up to improve health messages’ overall acceptance
and positively shape demand for OTC drugs (UTexas, 2011). The study also
offers insights to communication researchers by addressing the concept of
empowerment and applying it to the area of health communication and
pharmaceutical advertising.
There are a number of limitations to the present investigation. Future
examinations of the effectiveness of OTC drug advertising should include
countries characterized by even more differences in their level of collectivism,
such as Asian markets (as expressed by their scores on the GLOBE or Hofstede
cultural scales; House et al., 2004; Hofstede, 1980). Limitations with regard to
the use of a fictitious product must be pointed out: even though its use has been
encouraged by previous researchers, as it allows subjects to evaluate the
advertisement untainted by previous experiences (Yin, 1999), skepticism
towards an unfamiliar brand is also of concern and can negatively impact both
product evaluations and purchase intentions, particularly for high-involvement
products (Diehl et al., 2007). Moreover, promotions for drugs treating different
medical symptoms (e.g., heart burn or digestive problems) might yield different
results.
276 Koinig, Diehl, and Mueller

While the present study provides a generalized discussion of ad evaluations


and empowerment, results may well vary based on consumer demographics,
including sex, and age as well as educational background. Future research
should not only focus on print advertising, but include the Internet, via which
both medications and treatment options are increasingly being promoted (Harms
et al., 2008), allowing consumers to actively engage in their health care
decisions.

6 References
Aspect. (2011). Connecting to the Empowered Customer: How the right customer
strategy can help companies succeed in the age of the empowered consumer.
Retrieved from http://in.aspect.com/Documents/Papers/Aspect_ConnectingEmpower
edCust_WP.pdf (accessed 15.04.2014)
Assael, H. (2004). Consumer Behavior: A Strategic Approach. Boston: Houghton
Mifflin.
Auton, F. (2004). The advertising of pharmaceuticals direct to consumers: A critical
review of the literature and debate. In: International Journal of Advertising. Vol.
23(1). 5-52.
Beard, F.K. (2004). Hard Sell 'Killers' and Soft Sell 'Poets': Modern Advertising's
Enduring Message Strategy Debate. In: Journalism History. Vol. 30(3). 141-149.
Bearden, W.O./Hardesty, D.M./Rose, R.L. (2001). Consumer Self-Confidence:
Refinements in Conceptualization and Measurement. In: Journal of Consumer
Research. Vol. 28(1). 121-134.
Belch, G.E. and Belch, M.A. (1993). Introduction to Advertising and Promotion: An
integrated marketing communications perspective. Homewood, IL: Irwin.
Bernhardt, J. M. (2004). Communication at the Core of Effective Public Health. In:
American Journal of Public Health. Vol. 94(2). 2051-2053.
Boruchovitch, E. and Mednick, B. (2002). The meaning of health and illness: some
considerations for health psychology. In: Psico-USF. Vol. 7(2). 175-183.
Buckley, J. (2004). Pharmaceutical Marketing – Time for Change. In: Electronic Journal
of Business Ethics and Organization Studies. Vol. 9(2). 4-11.
Callow, M. and Schiffman, L.G. (2004). Sociocultural Meanings in Visually
Standardised Print Ads. In: European Journal of Marketing. Vol. 38(9/10). 1113-
1128.
Clift, E. and Freimuth, V. (1995). Health Communication: What is it and what can it do
for you? In: Journal of Health Education. Vol. 26(2). 68-74.
Coulter, A. (2003). Slander: Liberal Lies about the American Right. New York, NY:
Three Rivers Press.
Davies, A. and Elliott, R. (2006). The evolution of the empowered consumer. In:
European Journal of Marketing. Vol. 40(9/10). 1106-1121.
de Mooij, M. (2011). Consumer Behavior and Culture: Consequences for Global
Marketing and Advertising. Los Angeles, CA: SAGE.
DeLorme, D.E./Huh, J./Reid, L.N./An, S. (2010). The state of public research on over-
the-counter drug advertising. In: International Journal of Pharmaceutical and
Healthcare Marketing. Vol. 4(3). 208-231.
The Effects of Different Ad Appeals in Non-Prescription Drug Advertising 277

Denegri-Knott, J./Zwick, D./Schroeder, J.E. (2006). Mapping consumer power: an


integrative framework for marketing and consumer research. In: European Journal of
Marketing. Vol. 40(9/10). 950-973.
Diehl, S./Mueller, B./Terlutter, R. (2007). Skepticism toward pharmaceutical advertising
in the U.S. and Germany. In: Advances in International Marketing. Vol. 18. 31-60.
Diehl, S./Mueller, B./Terlutter, R. (2008). Consumer responses towards non-prescription
and prescription drug advertising in the US and Germany: They don’t really like it,
but they do believe it. In: International Journal of Advertising. Vol. 27(1). 99-131.
Earle, S. (2007). Exploring Health. In: Earle, S./Lloyd, C.E./Sidell, M./Spurr, S. (eds.).
Theory and Research in Promoting Public Health. London: Sage. 37-66.
Earle, S. (2007). Promoting public health: Exploring the Issues. In: Earle, S./Lloyd,
C.E./Sidell, M./Spurr, S. (eds.). Theory and Research in Promoting Public Health.
London: Sage. 1-36.
Eylon, D. (1998). Understanding empowerment and resolving its paradox. In: Journal of
Management History. Vol. 4(1). 16-28.
Grace, V. M. (1991) The marketing of empowerment and the construction of the health
consumer: a critique of health promotion. In: International Journal of Health Services.
Vol. 21. 329-343.
Green, J. and Tones, K. (2010). Health Promotion: Planning and Strategies. 2nd ed.
Thousand Oaks, CA: SAGE.
HAI. (2009). Understanding and Responding to Pharmaceutical Promotion. Retrieved
from http://www.haiweb.org/11062009/drug-promotion-manual-CAP-3-090610.pdf
(accessed 26.11.2014)
Harms, F./Gänshirt, D./Rumler, R. (2008, eds.). Pharmamarketing.
Gesundheitsökonomische Aspekte einer innovativen Industrie am Beispiel von
Deutschland, Österreich und der Schweiz. Stuttgart: Lucius & Lucius.
Hofstede, G. (1980). Culture's Consequences: International Differences in Work-Related
Values. Thousand Oaks, CA: Sage.
Hofstede, G. (1984). National cultures and corporate cultures. In: Samovar, L.A. and
Porter, R.E. (eds.). Communication Between Cultures. Belmont, CA: Wadsworth.
House, R. J./Hanges, P.J./Javidan, M./Dorfman, P.W./Gupta, V. (eds.) (2004). Culture,
leadership, and organizations. The GLOBE study of 62 societies. Thousand Oaks,
CA: Sage.
Hudson, S./Hung, C.L./Padley, L. (2002). Cross-National Standardization of
Advertisements: A Study of the Effectiveness of TV Advertisements targeted at
Chinese Canadians in Canada. In: International Journal of Advertising. Vol. 21. 345-
366.
Kantar Health. (2014). Healthcare Consumer OTC/DTC Study (MARS). Retrieved from
http://kantarmedia.us/product/healthcare-consumer-otcdtc-study-mars (accessed
22.11.2014)
Kleppner, O. (1979). Advertising Procedure. 7th ed. New Jersey: Prentice Hall.
Kreps, D. (1979). A Representation Theorem for Preference for Flexibility. In:
Econometrica. Vol. 41. 565-577.
Kroeber-Riel, W. and Esch, F.-R. (2011). Strategie und Technik der Werbung. 7th ed.
Stuttgart: Kohlhammer.
MacKenzie, S.B. and Lutz, R.J. (1989). An Empirical Examination of the Structural
Antecedents of Attitude toward the Ad in an Advertising Pretesting Context. In:
Journal of Marketing. Vol. 53. 48-65.
278 Koinig, Diehl, and Mueller

MacKenzie, S.B./Lutz, R.J./Belch, G.E. (1986). The role of attitude toward the ad as a
mediator of advertising effectiveness: a test of competing explanations. In: Journal of
Marketing Research. Vol. 23 (May). 130-143.
Main, K.J./Argo, J.J./Huhmann, B.A. (2004). Pharmaceutical Advertising in the USA:
information or influence? In: International Journal of Advertising. Vol. 23. 119-112.
MarketLine. (2014). OTC Pharmaceuticals: Global Industry Profile. Datamonitor Group.
Melewar, T.C. and Vemmervik, C. (2004). International advertising strategy: A review,
reassessment and recommendation. In: Management Decision. Vol. 42(7). 863-881.
Morse, R.L. (ed.) (1993). The Consumer Movement: Lectures by Colson E. Warne.
Manhattan, KA: Family Economics Trust.
Mueller, B. (2006). The role of Product involvement in advertising message perception
and believability. In: Diehl, S. and Terlutter, R. (eds.). International Advertising and
Communication: Current Insights and Empirical Findings. Wiesbaden: Gabler. 3-22.
Nettleton, S. (1997). Governing the risky self: How to become healthy, wealthy and wise.
In: Petersen, A. and Bunton, R. (eds.). Foucault, Health and Medicine. London:
Routledge. 207-222.
Newholm, T./Laing, A./Hogg, G. (2006). Assumed empowerment: consuming
professional services in the knowledge economy. In: European Journal of Marketing.
Vol. 40(9/10). 994-1012.
Obermiller, C./Spanberg, E.R./MacLachlan, D.L. (2005). Ad skepticism: the
consequences of disbelief. In: Journal of Advertising. Vol. 34(3). 7-17.
Okazaki, S./Mueller, B./Taylor, C.R. (2010). Global Consumer Culture Positioning:
Testing Perceptions of Soft-Sell and Hard-Sell Advertising Appeals Between U.S. and
Japanese Consumers. In: Journal of International Marketing. Vol. 18(2). 20-34.
Okazaki, S./Mueller, B./Taylor, C.R. (2010). Measuring Soft-Sell Versus Hard-Sell
Advertising Appeals. In: Journal of Advertising. Vol. 39(2). 5-20.
Okazaki, S./Mueller, B./Diehl, S. (2013). A Multi-Country Examination of Hard-Sell and
Soft-Sell Advertising: Comparing Global Consumer Positioning in Holistic- and
Analytic-Thinking Countries, in: Journal of Advertising Research, Vol. 53 (3),.258-
272.
Okazaki, S./Taylor, C. R./Zou, S. (2006). Advertising Standardization’s Positive Impact
on the Bottom Line. A Model of When and How Standardization Improves Financial
and Strategic Performance. In: Journal of Advertising. Vol. 35(3). 17-33.
Parrott, R. (2003). Media Issues. In: Thompson, Teresa L./Dorsey, Alicia M./Miller,
Katherine I./Parrott, Roxanne. (eds.). Handbook of Health Communication. London:
Lawrence Erlbaum Associates. 445-448.
Porter, R. (1985). Introduction. In: Porter, R. (ed.). Patients and Practitioners: Lay
Perceptions of Medicine in Pre-Industrial Society. Cambridge: Cambridge University
Press. 1-22.
Rod, M. and Saunders, S. (2004). The informative and persuasive components of
pharmaceutical promotion: An argument for why the two can coexist. In: International
Journal of Advertising. Vol. 28(2). 313-349.
Rogers, E.M. (1996). The field of health communication today: An up-to-date report. In:
Journal of Health Communication. Vol. 1. 15-23.
Rose, N. (2000). Tod des Soziden? Eine Neubestimmung der Grenzen des Regierens. In:
Bröckling, U./ Krasmann, S./Lemke, Th. (eds.), Gouvernementalität der Gegenwart.
Studien zur Ökonomisierung des Sozialen. Frankfurt: Suhrkamp. 72-109.
Rossiter, J.R. and Bellman, S. (2005). Marketing Communications: Theory and
Applications. French Forest: Pearson Education Australia.
The Effects of Different Ad Appeals in Non-Prescription Drug Advertising 279

Russell, J.A. (1994). Is there universal recognition of emotion from facial expression? A
review of the cross-cultural studies. In: Psychological Bulletin. Vol. 115. 102–141.
Schiavo, R. (2007). Health Communication: From Theory to Practice. San Francisco:
Wiley and Sons.
Schwarzer, R. and Jerusalem, M. (1995). Generalized Self-Efficacy scale. In: Weinman,
J./Wright, S./Johnston, M. (eds.). Measures in health psychology: A user’s portfolio.
Causal and control beliefs. Windsor, UK: NFER-NELSON. 35-37.
Thompson, T./Parrott, R./Nussbaum, J. (2011). The Routledge Handbook of Health
Communication. 2nd ed., New York: Routledge.
Tones, K. and Tilford, S. (2001). Health Promotion: effectiveness, efficiency and equity.
3rd ed. Cheltenham, UK: Nelson Thornes.
U.S. Department of Health and Human Services. (2014). What is health literacy.
Retrieved from http://www.health.gov/communication/literacy/ (accessed 10.02.2013)
van lttersum, K., and Wong, N. (2010). The Lexus or the olive tree? Trading off between
global convergence and local divergence. In: lnternational Journal of Research in
Marketing. Vol. 27. 107-118.
van Raaij, W.F. (1993). Postmodern Consumption: Architecture, Art, and Consumer
Behavior. In: Advances in Consumer Research. Vol. 1. 550-558.
Wells, W./Moriarty, S./Burnett, J. (2006). Advertising – Principles and Practice. 7th ed.
Upper Saddle River, NJ: Pearson Prentice Hall.
WHO. (2006). Constitution of the World Health Organization. Basic Documents. 45th
ed. Geneva, Switzerland. Retrieved from
http://www.who.int/governance/eb/who_constitution_en.pdf (accessed 11.06.2014)
WHO. (2013). The right to health. Retrieved from
http://www.who.int/mediacentre/factsheets/fs323/en/ (accessed 01.03.2014)
Wilson, P. (2001). A Policy Analysis of the Expert Patient in the United Kingdom: Self-
Care as an Expression of Pastoral Power? In: Health and Social Care in the
Community. Vol. 9(3). 134- 142.
Wright, K.B./Sparks, L./O’Hair, H.D. (2008). Health Communication in the 21st century.
Oxford: Blackwell.
Yin, J. (1999). International Advertising Strategies in China: A Worldwide Survey of
Foreign Advertisers. In: Journal of Advertising Research. Vol. 39. 25-35.
Zuckerman, M. (1988). Behavior and biology: Research on sensation seeking and
reactions to the media. In: Donohew, L./Sypher, H.E./Higgins, E.T. (eds.).
Communication, social cognition, and affect. Hillsdale, NJ: Erlbaum. 173–194.
280 Koinig, Diehl, and Mueller

7 Appendix

Appendix 1: Results of contrast tests between the three ad appeals for ad evaluation

Appendix 2: Results of contrast tests between the three ad appeals for self-empowerment
Examining the Mediating Role of Integrated Marketing
Communication on the Relationship between Adhocracy
Culture and Brand Advantage

Lucia Porcu, Salvador del Barrio-García, Juan Miguel Alcántara-Pilar, and


Esmeralda Crespo-Almendros

1 Introduction
Despite over twenty years of intense academic research on Integrated
Marketing Communication (IMC), this concept “continues to stir debate,
discussion and, in some cases, confusion” (Kliatchko and Schultz, 2014, p.
373). Several authors (Kliatchko, 2008; Porcu, Del Barrio-García and Kitchen,
2012) have highlighted the lack of consensus among academics over the
definition of IMC. Since its very emergence, the IMC concept has expanded
from a narrowly focused marketing communications approach to a more holistic
‘firm-wide’ organisational perspective (Duncan and Moriarty, 1998; Cook,
2004; Christensen, Firat and Torp, 2008; Muñoz-Leiva, Porcu and Del Barrio-
García, 2015). Consequently, this concept has been referred-to variously over
the last decade as several authors have opted to drop the term ‘marketing’
naming the concept as ‘Integrated Corporate Communication’ (e.g. Pickton,
2004), ‘Integrated Communication’ (e.g. Christensen, Firat and Torp, 2008),
‘Integrated Communication Management’ (Einwiller and Boenigk, 2012) and
‘Strategic Integration’ (Kerr and Patti, 2013). The lack of conceptual clarity also
emerged among practitioners. In fact, a recent study found that the majority of
both agency-based and client respondents did not use the term IMC within their
organizations, but rather other terms such as “integrated thinking, integrated
planning, integrated marketing, full services, 360, or simply integration”
(Kliatchko and Schultz, 2014, p. 380). Despite this, ‘IMC’ is still the most
popular label and is utilized even in the conceptualizations taking the broader
firm-wide approach.
The theoretical ‘confusion’ has hindered the development of valid and
reliable measurement scales, preventing scholars from conducting substantial
empirical research. Several researchers (Mulhern, 2009; Taylor, 2010; Kliatchko
and Schultz, 2014) are still calling for empirical efforts to understand how
integration works. In this regard, further research is needed to identify the
factors that promote or hinder IMC and assess its beneficial effects on
performance, especially on brand-related outcomes. In fact, the paucity of
empirical evidence of the beneficial effects of integration on performance

© Springer Fachmedien Wiesbaden 2017


G. Christodoulides et al. (Hrsg.), Advances in Advertising Research (Vol. VII),
European Advertising Academy, DOI 10.1007/978-3-658-15220-8_20
282 Porcu, Barrio-García, Alcántara-Pilar, and Crespo-Almendros

represents a barrier to its “broad practice by organizations” (Mulhern, 2009),


while Ots and Nyilasi (2015) pointed out that integration is often problematic in
practice due to the organisational context, highlighting the need to analyse the
influence of organisational culture (especially the role of leadership) on
integration processes. Organisational culture has been broadly examined in the
management literature, while prior research regarding the role of organisational
culture on the implementation of IMC has mostly led to conceptual
developments, leadership style being the variable that has attracted the greatest
attention from academics. Given the dearth of empirical evidence in this
research area, the analysis of the effect of this organisational variable on IMC is
likely to provide a key contribution to the existing body of knowledge. With
these premises, this study aims to: (1) develop and validate a new scale to
measure the firm-wide IMC, conceptualized by taking a more holistic
organisational approach; (2) analyse the effect of competitive intensity on both
the adoption of an adhocracy culture and the implementation of IMC; (3)
examine the mediation role of IMC on the relationship between adhocracy
culture and brand advantage.

2 Literature Review and Hypotheses


A comprehensive literature review enabled us to develop a new IMC
theoretical framework that embraces the more holistic organisational approach,
which conveys that the locus of integration is the entire organization (and not
just the marketing function). In this study, this broader ‘firm-wide’ perspective is
adopted to define IMC as the stakeholder-centred interactive process of cross-
functional planning and alignment of organisational, analytical and
communication processes that allows for the possibility of continuous dialogue
by conveying consistent and transparent messages via all media to foster long-
term profitable relationships that create value. Four IMC dimensions have been
identified: message consistency, interactivity, stakeholder-centred strategic focus
and organisational alignment. The ‘message consistency’ dimension is the most
basic level of integration and refers to the communication of coherent and clear
positioning. This captures the “coordination and coherence of messages and
channels” that has been considered as the major point of convergence of most
conceptualizations (Kliatchko, 2005). The ‘interactivity’ dimension is the crucial
condition for dialogue between an organization/brand and all stakeholders. In
this study, interactivity is intended to refer to the context of general human social
experience, with reciprocity (which characterizes the relationship between the
interlocutors), speed of response (to received messages) and (interlocutors')
responsiveness being the three facets of this dimension (Johnson, Bruner and
Kumar, 2006). The ‘stakeholder-centred strategic focus’ dimension relates to the
need for the whole organization to acknowledge that its main strategic goal
Examining the Mediating Role of Integrated Marketing Communication 283

should be to create added value for stakeholders so as to establish and maintain


long-term relationships with them. Finally, the ‘organisational alignment’
dimension refers to internal integration (vertical and horizontal) at the
organisational level. The wide range of communication activities must take into
account the company as a whole, hence aligning organisational processes,
spanning departmental boundaries and eliminating functional silos are of
paramount importance to achieving the highest level of integration (Duncan and
Moriarty, 1998; Gulati, 2007; Kliatchko and Schultz, 2014).
Undoubtedly, IMC has recently gained momentum among both academics
and practitioners due to the increasingly dynamic technological and competitive
environments that greatly affects the marketing and communication processes
(Taylor, 2010). In this regard, the extant literature suggests that a market
environment characterized by a high level of competitive intensity promotes the
implementation of IMC (Low, 2000; Reid, 2005). More specifically, Reid
(2005) empirically tested this relationship, arguing that the desire of being
competitive in such turbulent environments may lead organizations to make
greater message coordination efforts to maximize the brand performance.
Likewise, Low (2000) found a positive relationship between competitive
intensity and IMC and concluded that a greater experience in the IMC planning
and implementation would contribute to enhance the skills that are needed to
compete and respond to competitors’ marketing actions. Thus, we posit that:

H1: The level of competitive intensity is positively related to the implementation


of Integrated Marketing Communication.

Organisational culture is “a pattern of shared basic assumptions […] that


have worked well enough to be considered valid and therefore to be taught to
new members as the correct way to perceive, think and feel in relation to those
problems” (Shein, 1985, p. 4). In this study, we used the Competing Values
Framework (CVF, Cameron and Quinn, 1999) and we focused on the adhocracy
culture. This culture type has been described as a dynamic and creative working
environment, where innovativeness, agility and flexibility are the bonding
materials within the organization and both employees and leaders are risk-takers.
The management literature (e.g. Zhou, Yim and Tse, 2005) indicates that
competitive intensity is a crucial antecedent of innovation-oriented
organisational culture, thus a positive link between competitive intensity and the
adoption of an adhocracy culture type is to be expected. In fact, the
organizations operating in a dynamic market environment characterized by high
competitive intensity are more likely to adopt an adhocracy culture to better
respond to environmental changes. Based on this rationale:
284 Porcu, Barrio-García, Alcántara-Pilar, and Crespo-Almendros

H2: The level of competitive intensity is positively related to the adoption of


adhocracy culture.

Since IMC involves the whole organization, a key issue is how the
integration process should be organized; hence, the organisational system factors
are expected to be relevant antecedents of IMC (Christensen et al., 2008). More
specifically, several scholars (e.g. Ots and Nyilasy, 2015) have recently included
the study of the role of organisational culture amongst the suggested future
research directions. Zheng, Yang and McLean (2010) suggested that
organisational culture can be a source of sustained competitive advantage and a
key-factor to enhance business performance; however, its influence is mediated
and exerted by shaping the behavior of the organisational members. Therefore,
we expect the IMC implementation to mediate the effect of the adoption of an
adhocracy culture type on brand advantage. The extant literature suggests that
the adoption of a consensus-based decision-making model (Gronstedt, 1996;
Gulati, 2007; Shin, 2013), the organisational flexibility (Christensen et al.,
2008), the level of risk tolerance among senior managers (Smith, 1998;
Madhavaram et al., 2005) and their commitment to innovation (Schultz and
Schultz, 1998) are all positively related to IMC. All these elements match the
definition of adhocracy culture type proposed by Cameron and Quinn (1999) in
their CVF. The positive relationship between the implementation of IMC and the
achievement of a superior performance has been supported by several scholars,
who highlighted the beneficial effects exerted by IMC on brand-related
outcomes (e.g., Duncan and Moriarty, 1998; Low, 2000; Reid, 2005; Einwiller
and Boenigk, 2012; Luxton et al., 2015). Therefore, we posit that:

H3: The adoption of adhocracy culture is positively related to the


implementation of IMC.

H4: The implementation of IMC is positively related to brand advantage.

3 Method

3.1 Data Collection and Sample


The data for this research were gathered via an online self-administered
survey conducted among services businesses operating in Spain with at least 40
employees. The key informant method was used, with CEOs, senior marketing
and communication managers and other senior managers being targeted as key
informants. As a sampling frame, a commercial listing of 969 businesses was
Examining the Mediating Role of Integrated Marketing Communication 285

drawn from the Bureau van Dijk SABI database, based on Standard Industrial
Classification (SIC) classification systems. Four SIC codes were covered in the
study (701, 702, 703, 704). The modus operandi for the data collection
consisted of three steps. As a preliminary step, a set of 180 businesses was
randomly extracted from the commercial listing to purify the proposed IMC
scale via a pilot study conducted with the same procedures utilized for the main
survey. The 39 respondents of the pre-test were dropped from the sampling
frame for the second study. Second, a telemarketing firm was employed to
contact the sample by telephone to ask for the key informants’ and firms’
availability to participate in this study, verify the key informants’ name, position
and collect their e-mail address (not included in the database due to privacy
protection legal issues). This procedure resulted in the removal of other 406
from the sampling framework. The most common reasons were firms’ resistance
to collaborating in external investigation and key informants’ limited time or
lack of interest in such studies. Third, a customized link to the online
questionnaire was emailed to the individuals who agreed to participate (n = 524).
Following the initial e-mail, a follow-up e-mail was sent to encourage response.
There were 180 fully completed valid responses for an effective response rate of
18.6% (of the total population of 969 CEOs and other senior managers) and of
34.4 % of the 524 who had agreed to participate. The response rate is consistent
to prior literature and the makeup of respondent pool is excellent. In fact, almost
50% of the respondents are CEO and over 40% stayed at the organization for
over 10 years, and as such they should have a broad view of IMC practices in
their organization. The non-response bias has been addressed as a deviation
between sample and population distributions via comparing the organisational
details of the companies included in the sample and the population from which
the sample was drawn. No significant differences were found between the two
groups in terms of business size, age and type (SIC code), thus non-response
bias cannot be considered a major concern in the current study. In addition, the
Harman’s single factor test (McFarlin and Sweeny, 1992) suggested that the
common method bias in the data was relatively limited. Table 1 provides a
socio-demographic description of the population and the sample.

Table 1: Sample and population characteristics


Sample Population
Organization n % n %
Business size (number of employees)
40-50 45 25.00 270 27.86
51-249 114 63.33 590 60.89
250 + 21 11.67 109 11.25
Total 180 100 969 100
286 Porcu, Barrio-García, Alcántara-Pilar, and Crespo-Almendros

Business type (SIC code)


701 153 85.0 883 91.12
702 17 9.4 45 4.64
703 3 1.7 17 1.75
704 7 3.9 24 2.48
Total 180 100 969 100
Business age
< 10 years 30 16.67 135 13.93
10-20 years 44 24.44 303 31.27
21-30 years 30 16.67 205 21.16
30 + years 76 42.22 326 33.64
Total 180 100 969 100
Key-informant
Gender
Male 113 62.78
n.a.
Female 67 37.22
Total 180 100
Age
Under 25 6 3.33
25-35 53 29.44
36-45 67 37.22 n.a.
46-55 35 19.44
55+ 19 10.56
Total 180 100.0
Education
Primary School 1 0.56
High school graduate 12 6.67
Professional training 13 7.22
n.a.
University Undergraduate 62 34.44
Universit Graduate 52 28.89
Master’s level graduate/Ph.D. 40 22.22
Total 180 100
Experience
Fewer than 5 years 55 30.56
5-10 years 50 27.78
n.a.
11-20 years 42 23.33
Over 20 years 33 18.33
Total 180 100
Position
CEO 85 47.22
Senior marketing and communication n.a.
72 40.00
managers
Other senior managers 23 12.78
Total 180 100
Note: n.a. = not available.

3.2 Measures
As regards the measures implemented, we opted to develop a new
instrument to assess IMC based on the broad theoretical approach presented
Examining the Mediating Role of Integrated Marketing Communication 287

above, while the remaining constructs were measured by using scales drawn
from the extant literature. To develop the IMC scale, a multi-stage research
design modeled after Churchill (1979) has been executed. First, a
comprehensive literature review was conducted to identify the conceptual
domain and generate an initial set of 59 items. Second, a two-round Delphi study
was carried out using an international expert panel for content evaluation of the
proposed IMC theoretical framework and items. These procedures enabled to
refine the content and reduce the duplication resulting in a Likert type scale
including 25 items scored from ‘1, strongly disagree’ to ‘7, strongly agree’ (4
items for ‘message consistency’; 7 for ‘interactivity’; 7 for ‘stakeholder-centred
strategic focus’; 7 for ‘organisational alignment’). To assess competitive
intensity, the scale proposed by Jaworski and Kohli (1993) and later modified by
Song, Droge, Hanvanich and Calantone (2005) was used. The adhocracy culture
measure was derived from the Organisational Culture Assessment Instrument
(OCAI), developed by Cameron and Quinn (1999) using the Competing Values
Framework. The OCAI has been widely validated in previous research
(Cameron and Quinn, 1999) and used in over 10,000 organizations worldwide
(Shih and Huang, 2010). The adhocracy culture scale is composed of six items,
scored on a 7-point Likert type scale ranging from ‘1, strongly agree’ to ‘7,
strongly disagree’. The scale to measure brand advantage has been drawn from
Reid (2005). This scale is composed of three items measured using a seven-point
Likert type scale from ‘1 = much less compared with the closest competitor in
the last three years’ to ‘7 = much more compared with the closest competitor in
the last three years’. Finally, a set of measures was included to ascertain the
characteristics of the respondents and companies for sample description
purposes.

4 Data Analysis and Results


In this study, the Structural Equation Modeling method has been used to test
the research hypotheses. Two types of SEM can be distinguished: covariance-
and variance-based SEM, the latter being recently developed and considered as a
powerful analytical method. Among variance-based SEM methods, partial least
squares (PLS; Chin 1998; Lohmöller 1989) path modeling is regarded as the
“most fully developed and general system” and called a “silver bullet” (Hair et
al., 2011). On this basis, it is rational to apply the PLS SEM in this research
using SmartPLS (Ringle, Wende and Becker, 2015) for the following reasons.
First, this technique has been widely used in marketing research (Hair et al.,
2011) and in recent IMC research (e.g. Luxton et al., 2015). Second, the use of
PLS has been recommended when theoretical knowledge about a topic is scarce
(Petter et al., 2007). Third, Dijkstra and Henseler (2015) found in simulation
studies that PLSc was only slightly lower in power than full information
288 Porcu, Barrio-García, Alcántara-Pilar, and Crespo-Almendros

maximum likelihood (FIML) SEM, however it had advantages in handling non-


normally distributed data. For a long time, analysts have criticized the use of
PLS as an estimator for structural equation model mainly due to a key
disadvantage: the PLS estimates – especially path coefficients and loadings- are
only consistent at large. The lack of consistency is a serious disadvantage due to
its adverse impact on substantial research findings, meaning that the estimates do
not approach true values as sample size increases. Dijkstra and Henseler (2015)
have recently introduced an important advancement to PLS: consistent PLS
(PLSc). Henseler et al. (2016) has suggested that these improvements enable to
consider PLS “a full-fledged structural equation modeling approach”. PLSc
provides key improvements: first, estimates do approach true values
asymptotically; second, it allows non-recursive models to be estimated; third, it
provides a global assessment of goodness-of-fit. With these premises, the
proposed model has been estimated via using PLSc, the results showing an
adequate overall fit (SRMR = .07). IMC was specified as second-order reflective
construct, while the remaining variables were specified as first-order reflective
constructs. The bootstrap technique is well known and commonly used in PLS
analysis to estimate the significance of weights, loadings and path coefficients.
In the current study, to obtain the level of statistical significance of indicator
loadings, dimension weights and loadings we performed a consistent bootstrap
analysis with 5000 subsamples. First, the measurement model (Table 2) was
assessed to validate the scales; second, the proposed hypotheses were tested by
examining the structural model (Figure 1).
The results of the measurement model indicated that all standardized
coefficients were statistically significant (p < .01) and exceed the more
conservative .707 cutoff (Chin, 2010), except for ADHO_3 (β ADHO_3 = .61),
ALIN_3 (β ALIN_3 = .54), ICOM_1 (β ICOM_1 = .51) and ICOM_2 (β ICOM_2 = .60),
and ICOM_3 (β ICOM_3 = .44). While β ADHO_3, β ALIN_3, β ICOM_1 and β ICOM_2 were
above the less conservative threshold of .5, β ICOM_3 was slightly below the
recommended cutoff, thus ICOM_3 became a prime candidate for deletion.
However, we decided to keep this indicator to preserve the content validity of
the competitive intensity construct. In addition, the Cronbach’s α scores exceed
the threshold of .8 in all cases, thus confirming the internal consistency of the
measures, while the construct Composite Reliability (CR) exceed the
recommended cutoff of .7 for all cases. These findings provide evidence for the
reliability of the measures. To test convergent validity, the construct Average
Variance Extracted (AVE) was calculated, the results showing that exceeded the
recommended cutoff of .5 for all cases, except for competitive intensity
construct (AVE = .470, slightly below the threshold of .5). These results provide
evidence of an adequate convergence validity. To complete the validation
process, the criterion proposed by Fornell and Larcker (1981) to test the
Examining the Mediating Role of Integrated Marketing Communication 289

discriminant validity was applied, the results showing that the square root of
AVE was greater than the correlation shared among the constructs. In addition,
the Heterotrait-Monotrait (HTMT) ratio (Henseler, Ringle and Sarstedt, 2015),
has been computed for each pair of constructs on the basis of the item
correlations. The computation yelded values between .129 in respect of HTMT
(brand advantage, competitive intensity) and .779 in respect of HTMT
(interactivity, organisational alignment), thus satisfying the most conservative
criterion (HTMT < .85) and providing evidence for discriminant validity.
Once the measures were validated, the structural (inner) model has been
assessed to test the proposed model and hypotheses. More specifically, the
results indicate that the level of competitive intensity is positively and
significantly (βicomadho = .300; t = 3.700; p < .01) related to the adoption of
adhocracy as the dominant organisational culture type, providing empirical
support to H1. In addition, a strong, positive and significant relationship was
found between the level of competitive intensity and IMC (βicomimc = .109; t =
1.964; p < .05), thus H2 is statistically supported. Moreover, H3 and H4
predicted that IMC fully mediates the relationship between adhocracy
organisational culture and brand advantage. To test the mediation effect, two
approaches have been applied. First, the method proposed by Baron and Kenny
(1986) has been adopted. The findings indicate that the effect exerted by the
adoption of adhocracy culture on brand advantage is not significant (t = 1.593; p
> .05), while strong, positive and significant relationships were found between
adhocracy culture and IMC (β adhoimc = .733; t = 13.824; p < .01) and between
IMC and brand advantage (β imcbrad = .491; t = 3.778; p < .01). These findings
are in full support of H3 and H4. Second, we executed the procedure suggested
by Zhao et al. (2010) and based on the estimation and analysis of the indirect
effects. The results revealed that the indirect (β = .361; p < .01) and total (β =
.546; p < .01) effects of adhocracy culture on brand advantage are positive and
significant. These results demonstrate the condition for the full mediation of
IMC on the adhocracy culture-brand advantage relationship. Finally, the
magnitude of the R2 was evaluated as a criterion of explanatory power. R2 values
of .67, .33 and .19 indicate substantial, moderate and weak explanatory power
respectively (Chin, 2010). Except for the R2 value for the adhocracy culture
construct (R2 = .09), the R2 values suggest a substantial and moderate
explanatory power. The Stone-Geisser’s Q2 test has been used to assess the
predictive relevance, the Q2 values being obtained by using the blindfolding
procedure. As a criterion of predictive relevance, Q2 values greater than zero
suggest a good predictive relevance/quality. The results (see Figure 1) show that
all the Q2 values were greater than zero, thus suggesting a relatively good
explanatory power for the model.
290 Porcu, Barrio-García, Alcántara-Pilar, and Crespo-Almendros

Table 2: Results of the measurement model


Standardised
Itemsa Constructs t Alpha AVE CR
Coefficients
CONS_1 .90 *** 34.674
CONS_2 Message .77 *** 18.927
.925 .757 .925
CONS_3 Consistency .88 *** 30.672
CONS_4 .92 *** 44.951
INTE_1 .86 *** 22.530
INTE_2 .78 *** 15.436
INTE_3 .87 *** 29.270
INTE_4 Interactivity .79 *** 17.494 .943 .702 .943
INTE_5 .82 *** 17.313
INTE_6 .89 *** 24.269
INTE_7 .84 *** 19.146
STAK_1 .83 *** 21.370
STAK_2 .79 *** 20.628
Stakeholder-
STAK_3 .90 *** 37.808
centred strategic
STAK_4 .81 *** 21.978 .937 .680 .937
focus
STAK_5 .83 *** 20.012
STAK_6 .81 *** 20.078
STAK_7 .79 *** 18.401
ALIN_1 .79 *** 14.396
ALIN_2 .88 *** 32.293
ALIN_3 Organisational .54 *** 8.055
ALIN_4 Alignment .82 *** 18.914 .933 .679 .936
ALIN_5 .92 *** 24.092
ALIN_6 .86 *** 19.574
ALIN_7 .88 *** 26.816
BRAD_1 .86 *** 15.120
BRAD_2 Brand Advantage .75 *** 11.934 .808 .589 .809
BRAD_3 .68 *** 7.831
ADHO_1 .75 *** 11.676
ADHO_2 .78 *** 13.750
ADHO_3 Adhocracy .61 *** 8.202
.898 .592 .896
ADHO_4 Culture .79 *** 17.450
ADHO_5 .81 *** 14.308
ADHO_6 .85 *** 17.774
ICOM_1 .51 *** 2.952
ICOM_2 .60 *** 3.361
Competitive
ICOM_3 .44 *** 2.378 .816 .470 .804
Intensity
ICOM_4 .87 *** 7.151
ICOM_5 .88 *** 6.647
Notes: a = the complete set of items is not listed here due to space constraint, however
they could be provided on request by the authors ;*** = p < .01
Examining the Mediating Role of Integrated Marketing Communication 291

Figure 1: Estimated structural model

5 Discussion, Limitations and Future Research


The major contributions of this study are: the development of a valid and
reliable scale to measure IMC taking a broad organisational approach; the
empirical evidence regarding the influence of competitive intensity on both the
adoption of an adhocracy culture and on the implementation of IMC; the very
first empirical demonstration of the mediating role of IMC on the relationship
between the adoption of adhocracy culture and brand advantage. In fact, the
results revealed that IMC is positively affected by adhocracy, a culture type
characterized by creative-orientation, flexibility, and innovation- driven
leadership. Thus, this culture type emerges as an important internal driver of
IMC that contributes to build an IMC-friendly environment. Competitive
intensity was found to promote both the adoption of such culture type and the
implementation of IMC. Finally, this study empirically proves that IMC is
positively associated to a greater brand advantage and mediates the relationship
between adhocracy culture and brand advantage. The present research has a
number of limitations that can be taken into account by future researchers. First,
this study focused solely on one culture type, thus future research should analyse
how the adoption of the other organisational culture types (e.g., clan, market and
hierarchical) affect IMC. Second, the use of self-reported data to measure brand
advantage must be recognised as a limitation, and future research should use
objective measures to corroborate the relationship between IMC and the ‘actual’
brand-related outcomes.
292 Porcu, Barrio-García, Alcántara-Pilar, and Crespo-Almendros

6 Managerial Implications
Several managerial implications can be drawn from these results. First, the
proposed scale can be used by managers as an IMC audit tool to assess and
monitor IMC implementation and assist them in their decision-making
processes. Second, the findings indicate that adhocracy culture constitutes a
favorable culture type to promote IMC. Thus, managers should be aware of the
fact that organizations characterized by flexibility, a high commitment to
creativity and innovation, and risk-taking leadership are more likely to
implement IMC. We strongly recommend managers to evaluate whether the
culture adopted by the organization is compatible with the desired organisational
culture type for promoting integration. Cameron and Quinn (1999) emphasized
that cultural change in organizations is crucial to their survival in the current
rapidly evolving market environment, and suggested that organisational culture
needs to be monitored over time. Undoubtedly, leadership is a key element of
organisational culture since “the important changes needed to support cultural
change are the behaviors of managers […] (‘they must walk the talk’)”.
Therefore, we suggest managers to use a leadership diagnosing tool, such as the
Management Skills Assessment Instrument (MSAI) to identify their own current
managerial strengths and weaknesses as well as the competencies that will help
the organization move toward the preferred future culture.
Finally, these findings provide evidence of the strong positive effect of
IMC on brand advantage and shed light on the mediating role of IMC on the
relationship between adhocracy culture and brand advantage, suggesting that the
achievement of an effective implementation of IMC is associated with how well
adhocracy cultural values are translated into value to the organization in terms of
brand advantage. Consequently, managers and practitioners should make greater
efforts to enhance the integration mechanisms (for example, the vertical and
horizontal communication) and eliminate the functional silos and organizational
barriers.

7 References
Baron, R. M. and Kenny, D. A. (1986), “The Moderator-Mediator Variable Distinction
in Social Psychological Research: Conceptual, Strategic, and Statistical
Considerations,” in: Journal of Personality and Social Psychology, Vol. 51, 6, 1173-
182.
Cameron, K. S., Quinn R. E. (1999). Diagnosing and changing organisational culture:
Based on the Competing Values Framework. Addison-Wesley: Reading,
Massachusetts.
Chin, W. W. (1998), “Issues and opinion on structural equation modelling,” in: MIS
Quarterly, Vol. 22, 1, 8-15.
Examining the Mediating Role of Integrated Marketing Communication 293

Chin, W. W. (2010), “How to Write Up and Report PLS Analyses,” in: Handbook of
Partial Least Squares: Concepts, Methods and Applications, V. Esposito Vinzi, W.
W. Chin, J. Henseler, and H. Wang (eds.), Berlin, Germany: Springer Handbooks of
Computational Statistics, 655-690.
Christensen, L. T., Firat, A. F., Torp, S. (2008), “The Organization of Integrated
Communications: Toward Flexible Integration,” in: European Journal of Marketing,
42, 3, 423-452.
Churchill, G. A. (1979), “A paradigm for developing better measures of marketing
constructs,” in: Journal of Marketing Research, 16, 64-73.
Cook, W. A. (2004), “IMC's Fuzzy Picture: Breakthrough or Breakdown?”, Journal of
Advertising Research, Vol. 44, 1, 1-2.
Dijkstra, T. K. and Henseler, J. (2015), “Consistent partial least squares path modeling,”
in: MIS Quarterly, Vol. 3, 2, 297-316.
Duncan, T. R. and Moriarty, S. (1998), “A Communication-Based Marketing Model for
Managing Relationships,” in: Journal of Marketing, 62, 2,1-13.
Einwiller, S. and Boenigk, M. (2012), “Examining the link between integrated
communication management and communication effectiveness in medium-sized
enterprises”, Journal of Marketing Communications, Vol. 18, 5, 335-366.
Fornell, C., and Larker, D. F. (1981), “Evaluating Structural Equation Models with
Unobservable Variables and Measurement Error,” in: Journal of Marketing Research,
18, 39-50.
Gronstedt, A. (1996). “How agencies can support Integrated Marketing
Communications,” in: Journal of Business Research, Vol. 37, 3, 201-206.
Gulati, R. (2007). “Silo busting,” in: Harvard Business Review, 85, 5, 98-108.
Hair, J. F., Ringle, C. M. and Sarstedt, M. (2011), “PLS-SEM: Indeed a Silver Bullet”,
in: Journal of Marketing Theory and Practice, Vol. 18, 2, 139-152.
Henseler, J., Hubona, G. and Ray, P. A. (2016), “Using PLS Path Modeling in New
Technology Research: Updated Guidelines,” in: Industrial Management and Data
System, in press.
Henseler, J., Ringle, C. M. and Sarstedt, M. (2015), “A new criterion for assessing
discriminant validity in variance-based structural equation modeling,” Journal of the
Academy of Marketing Science, Vol. 43, 1, 115-135.
Jaworski, B.J. and Kohli, A.K. (1993), “Market Orientation: Antecedents and
Consequences”, Journal of Marketing, Vol. 57, July, 53-70.
Johnson, G. J., Bruner II, G. C., Kumar, A. (2006), “Interactivity and its facets revisited:
Theory and empirical test,” in: Journal of Advertising, Vol. 35, 4, 35-52.
Kerr, G., Patti, C. (2013), “Strategic IMC: From abstract concept to marketing
management tool,” in: Journal of Marketing Communications, 1-23 (in press), doi:
http://dx.doi.org/10.1080/13527266.2013.786748.
Kliatchko, J. (2005), “Towards a new definition of Integrated Marketing
Communications (IMC),” in: International Journal of Advertising, 24, 1, 313-348.
Kliatchko, J. (2008), “Revisiting the IMC construct. A revised definition and four
pillars,” in: International Journal of Advertising, 27, 1, 133–160.
Kliatchko, J., Schultz, D.E. (2014), “Twenty years of IMC: a study of CEO and CMO
perspectives in the Asia-Pacific region,” in: International Journal of Advertising, 33,
2, 373-390.
Lohmöller, J. B. (1989), Latent variable path modeling with partial least squares.
Heidelberg: Physica-Verlag.
294 Porcu, Barrio-García, Alcántara-Pilar, and Crespo-Almendros

Low, G. (2000), “Correlates of Integrated Marketing Communications,” in: Journal of


Advertising Research, Vol. 40, 3, 27-39.
Luxton, S., Reid, M., Mavondo, F. (2015), “Integrated Marketing Communication
Capability and Brand Performance,” in: Journal of Advertising, Vol. 44, 1, 37-46.
Madhavaram, S., Badrinarayanan, V., McDonald, R. E. (2005), “Integrated marketing
communication (IMC) and brand identity as critical components of brand equity
strategy - A conceptual framework and research propositions,” in: Journal of
Advertising, 34 (4), 69-80.
McFarlin, D. B., and P. D. Sweeney (1992), “Research Notes. Distributive and
Procedural Justice as Predictors of Satisfaction with Personal and Organizational
Outcomes,” in: Academy of Management Journal, Vol. 35, 3, 626-37.
Mulhern, F. (2009), “Integrated marketing communications: From media channels to
digital connectivity,” in: Journal of Marketing Communications, Vol. 15, 2/3, 85-
101.
Muñoz-Leiva, F., Porcu, L., Del Barrio-García, S. (2015), “Discovering prominent
themes in integrated marketing communication research from 1991 to 2012: a co-
word analytic approach,” in: International Journal of Advertising: The Review of
Marketing Communications, Vol. 34, 4, 678-701.
Ots, M. and Nyilasi, G. (2015), “Integrated Marketing Communication (IMC): Why does
it fail?”, in: Journal of Advertising Research, Vol. 55, 2, 132-145.
Petter, S., Straub, D., and Rai, A. (2007), “Specifying formative constructs in
information systems research,” in: MIS Quarterly, Vol. 31, 4, 623-656.
Pickton, D. (2004). “Assessing Integrated Corporate Communication”, in: Oliver, S. M.,
(2004), Handbook of Corporate Communication and Public Relations, Pure and
Applied, Routledge: New York, USA, 227-241.
Porcu, L., Del Barrio-García, S., Kitchen, P.J. (2012), “How Integrated Marketing
Communications (IMC) works? A theoretical review and an analysis of its main
drivers and effects,” in: Communication and Society, Vol. 25, 1, 313-348.
Reid, M. (2005), “Performance Auditing of Integrated Communication (IMC) actions
and outcomes,” in: Journal of Advertising, Vol. 34, 4, 41-54.
Ringle, Christian M., Wende, S., Becker, J.M. (2005), SmartPLS 2.0.M3. Hamburg:
SmartPLS. Retrieved from http://www.smartpls.com.
Schultz, D. E., Schultz, H. (1998), “Transitioning marketing communications into the
twenty-first century,” in: Journal of Marketing Communications, 4, 9-26.
Shein, E. (1985), Organisational Culture and Leadership, San Francisco: Jossey Bass.
Shih, C. C., Huang, S. J. (2010), “Exploring the relationship between organisational
culture and software process improvement deployment”, in: Information and
Management, Vol. 47, 5, 271-281.
Song, M., Droge, C., Hanvanich, S. and Calantone, R. (2005), “Marketing and
technology resource complementarity: an analysis of their interaction effect in two
environmental contexts,” in: Strategic Management Journal, Vol. 26, 3, 259-276.
Smith, P. R. (1998), Marketing Communications: An Integrated Approach, second
Edition, London, United Kingdom: Kogan Page Limited.
Taylor, C.R. (2010), “Integrated Marketing Communications in 2010 and Beyond,” in:
International Journal of Advertising, 29, 2, 161-164.
Zheng, W., Yang, B. Y., McLean, G. N. (2010), “Linking organisational culture,
structure, strategy, and organisational effectiveness: Mediating role of knowledge
management,” in: Journal of Business Research, Vol. 63, 7, 763-771.
Examining the Mediating Role of Integrated Marketing Communication 295

Zhou, K. Z., Yim, C.K. and Tse, D.K. (2005), “The effects of strategic orientations on
technology and market based breakthrough innovations,” in: Journal of Marketing,
Vol. 6, 2, 42–6.
Profiling Lead Researchers in Advertising Research

Terri H. Chan and Caleb H. Tse

1 Introduction

Regardless of academic discipline, output productivity in top journal papers


has gone beyond just assessing individual scholarship and career decisions, to
also affecting how researchers’ home universities are ranked. The pressure to be
prolific is salient at the university level, where competition for operating grants,
donations, and research funds is escalating along with governance demands and
scrutiny over fund usage and research outcomes (Geuna and Martin, 2003;
Siemens, Burton, Jensen, and Mendoza, 2005). Every year, research productivity
benchmarks are updated and publicized (e.g., QS World University Rankings
and Financial Times). Partly driven by these intense public interests, researchers
are investigating the factors that drive research output, productivity and other
performances in various disciplines (Judge, Cable, Colbert, and Rynes, 2007;
Stremersch, Verniers, and Verhoef, 2007).
Over the years, there is an emerging literature on advertising research
productivity. These scholars have examined researchers, their output and how
their home institutes rank based on research output in journals and other outlets
(e.g., Russell and Martin, 1976; Reid, 2014). Recent work has tended to focus
on top journal publications as well as particular issues such as internationalism
in advertising. Zou (2005) looked at author and institutional contributions to
international advertising research while Polonsky and Carlson (2009) studied the
extent of globalism in advertising research authorship. They assert that a North
American dominance in authorship remains, but increased authorship from
around the world is trending upwards.
Two gaps in the literature on advertising research productivity remain. To
begin, the impacts of new media which have caused unprecedented changes in
media platform (from offline to online) and content (from firm-based to public-
based content) on advertising has yet to be documented. More importantly, an
updated profiling and examination of the current lead researchers (i.e., top 1%
scholars) who affect research productivity and the future of advertising academia
has yet to be conducted. These lead researchers represent key talents that many
institutions want to recruit and build upon. As prolific researchers they provide
more output than average researchers. In addition, these lead scholars help build
research clusters that enable their home institution to stand out among competing
institutions. Many of them have international working links and occupy

© Springer Fachmedien Wiesbaden 2017


G. Christodoulides et al. (Hrsg.), Advances in Advertising Research (Vol. VII),
European Advertising Academy, DOI 10.1007/978-3-658-15220-8_21
298 Chan and Tse

gatekeeping roles in the advertising academia such as journal ou