Journal of Operations Management 30 (2012) 509520

Contents lists available at SciVerse ScienceDirect

Journal of Operations Management

journal homepage: www.elsevier.com/locate/jom

Effect of purchase volume exibility and purchase mix exibility on

e-procurement performance: An analysis of two perspectives
Sarv Devaraj a, , Ganesh Vaidyanathan b , Abhay Nath Mishra c
a

Management Department, University of Notre Dame, Notre Dame, IN 46556, United States
School of Business and Economics, Indiana University South Bend, South Bend, IN 46634, United States
c
Robinson College of Business, Georgia State University, Atlanta, GA 30302-3989, United States
b

a r t i c l e

i n f o

Article history:
Received 14 April 2011
Received in revised form 12 July 2012
Accepted 8 August 2012
Available online 23 August 2012
Keywords:
E-procurement
Frequency
Information sharing
Purchase volume exibility
Purchase mix exibility
Performance
Supplier customization
Trust

a b s t r a c t
Despite the widespread adoption of e-procurement by rms in recent years, academic research examining the mechanisms through which e-procurement applications lead to performance has been scarce.
Anecdotal evidence points to numerous situations where companies have failed to harness the potential of e-procurement. In this paper, we argue that online purchase volume and mix exibilities facilitated
by these applications play a signicant role in the ability of rms to benet from e-procurement. We
examine this tenet from both an economic as well as a social perspective. We propose that increased
online purchase volume exibility as well as online purchase mix exibility can be facilitated by two
mechanisms supplier customization as explained by transaction costs perspective, and information
sharing between supply chain partners using a social exchange theoretical perspective. The increased
purchase volume and mix exibility in turn leads to better performance along the dimensions of cost,
quality, and delivery. We present and test a nuanced perspective where we argue that (i) the effect of
supplier customization on both purchase volume and mix exibilities will be moderated by the frequency
of transactions conducted online, and (ii) the effect of information sharing on both purchase volume and
mix exibilities will be moderated by trust in the supplier. We estimate our research model using survey
data collected from 130 purchasing and procurement managers. We nd strong support for our proposed
research model with results indicating that purchase volume and mix exibilities play a vital mediating
role in impacting e-procurement performance. Theoretical and practical implications of the ndings are
discussed.
2012 Elsevier B.V. All rights reserved.

1. Introduction
The procurement function is considered the focal point of
effective strategic supply chain management in contemporary
organizations (Talluri and Sarkis, 2002). Advances in information
technology (IT) have enabled rms to streamline the procurement
process and value-chain activities to achieve substantial economic
benets (Devaraj et al., 2007; Gosain et al., 2005; Hill and Scudder,
2002). In particular, e-procurement the automation of a rms
procurement process is becoming increasingly recognized for
its potential to improve business operations and reduce expenses.
While a number of rms have been able to leverage e-procurement
applications to lower costs and increase overall protability, others have grappled with the intricacies of these technologies and
the mechanisms through which they can impact procurement
practices and economic benets (Olson and Boyer, 2003). In this

Corresponding author.
E-mail addresses: sdevaraj@nd.edu (S. Devaraj), gvaidyan@iusb.edu
(G. Vaidyanathan), amishra@gsu.edu (A.N. Mishra).
0272-6963/$ see front matter 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.jom.2012.08.001

paper, our focus is to examine how e-procurement inuences purchase volume exibility and purchase mix exibility in relation
to the traditional purchasing context, and how such exibilities
impact e-procurement performance. The key roles played by eprocurement applications in enabling rms to enhance purchase
volume and mix exibilities, and the consequent impact of such
enhancements, have been largely ignored in the existing literature.
In general, exibility in business processes has become increasingly important for rms to respond efciently to evolving
customer requirements, intense global competition, and rapid
technological advancements (Sambamurthy et al., 2003; Swafford
et al., 2006). Thus, exibility has been a topic of intense interest among operations management (OM) researchers (De Groote,
1994; Gerwin, 1993; Jordan and Graves, 1995; Koste and Malhotra,
1999; Suarez et al., 1996). However, prior research has focused
predominantly on examining the different dimensions of manufacturing exibility, the relationship between these dimensions, and
the impact of these dimensions of exibility on performance (Cao
and Dowlatshahi, 2005). Flexibility in other business processes and
the role that information technology can play in obtaining exibility has not been investigated in detail in extant literature.

510

S. Devaraj et al. / Journal of Operations Management 30 (2012) 509520

We conceptualize purchase volume and purchase mix exibilities as the ability of a rm to modify the mix and volume of input
goods dynamically using e-procurement with little penalty in time
and effort. We draw upon Duclos et al. (2003), who suggest that
exibility allows managers to change product mix and volume, relatively quickly and without much struggle. Grounding our research
in the theoretical perspectives of transactions costs economics
(TCE) and social exchange theory (SET), we propose that purchase
volume and mix exibilities in e-procurement can be facilitated
by two major factors information sharing and supplier customization. We further propose that two theoretically-grounded
important contextual factors trust and frequency accentuate the
relationships between purchase volume and mix exibilities and
their antecedents. This study provides a hitherto unexplored perspective on purchase volume exibility and purchase mix exibility
facilitated by e-procurement and the antecedents that facilitate
them.
The enhanced purchase volume and mix exibilities engendered
by e-procurement applications can also have substantial impacts
on procurement process performance. While the prior literature
in manufacturing strategy suggests that cost, quality, delivery, and
exibility are main competitive priorities and performance metrics,
we suggest that exibility may play a much more important role in
e-procurement than has been acknowledged in the prior research.
Our proposition is that other metrics cost, quality and delivery are driven by online volume and mix exibilities in the
context of procurement. While there is precedence in the literature of researchers analyzing rm performance based on exibility
(Devaraj et al., 2007; Jack and Raturi, 2002; Kekre and Srinivasan,
1990; Suarez et al., 1996; Vickery et al., 1999), to the best of our
knowledge, this is one of the rst studies in OM that theorizes
the role played by online purchase volume exibility and online
purchase mix exibility in the context of e-procurement and empirically validates it. Prior studies have examined the role of quality
on e-procurement satisfaction (Vaidyanathan and Devaraj, 2008),
buyer-side competence in B2B commerce (Rosenzweig and Roth,
2007), and organizational collaboration (Sanders, 2007). However,
our focus is on online purchase volume and mix exibilities, and
furthermore we propose that the use of exible purchase processes facilitated by e-procurement can enable organizations to
cope with changes in the marketplace to respond proactively to the
contingencies presented by uncertain business environments. Such
exibility in procurement process may result in signicant performance benets along cost, quality and responsiveness dimensions.
We test our research model with survey data collected from 130
purchasing and procurement managers from various industries.
Results obtained from employing structural equation modeling
provide strong support for our research model.
This paper makes four key contributions to the existing literature. First, we investigate the role of volume and mix exibilities
in e-procurement, which has been not been studied extensively
in the existing literature. In fact, scholars have issued several
research calls for a ne-grained understanding of purchase volume
and mix exibilities (Duclos et al., 2003). This study responds to
such calls. Our research model provides a nuanced understanding
of the key roles played by purchase volume and mix exibilities on e-procurement performance and the facilitating conditions
that enable rms to leverage these exibilities in the context of
e-procurement. We demonstrate that purchase volume and mix
exibilities can indeed impact other business process performance
metrics such as cost, quality, and delivery. Second, given that
organizational understanding of successful e-procurement use, the
mechanisms through which e-procurement use impacts business
processes, and the implications of such use are still lacking in the
literature (Bradley, 2005; Vaidyanathan and Devaraj, 2008) we propose that e-procurement can be used as a means for rms to gain

exibility. To the best of our knowledge, the exibility aspect of

e-procurement has been largely ignored in the existing literature.
This study provides a potential new direction for the examination
of e-procurement applications in rms. Third, we provide support
for and extend the literature that suggests that the use of information systems can enhance the exibility and agility of rms and
enable them to respond to market signals with ease (Sambamurthy
et al., 2003). Finally, by drawing on TCE and SET, this paper proposes, and nds empirical support for, two key enablers of purchase
volume and mix exibilities, as also moderators that impact the
inter-relationship between them.
The rest of the paper is organized as follows. We present the theoretical background along with the hypotheses in the next section.
In Section 3, we present the details of our empirical approach, and
discuss the adequacy of the measures used in the study. The results
of our analyses and ndings of our study are discussed in Section
4. Finally, in Section 5, we discuss the limitations and implications
of this study, and suggest areas of further research.

2. Theory and hypotheses development

We rst present our research model (in Fig. 1) as a roadmap,
following which we discuss the theoretical underpinnings of the
model. As illustrated in Fig. 1, we suggest that online purchase
volume and mix exibilities are key constructs that determine
the performance of the e-procurement process in organizations.
Drawing on two different theoretical perspectives, we propose that
online purchase volume exibility and online purchase mix exibility can have two key antecedents. According to TCE, asset specicity
is a key driver of transaction costs and the relationship between
supply chain partners. Asset specicity has emerged as a reliable
and consistent predictor of inter-organizational relationships and
sourcing decisions (Poppo and Zenger, 1998). Asset specicity is
related to alternative uses of the asset involved in the transaction,
and it is measured by the lack of standardization. In our context,
asset specicity is operationalized as supplier customization. The
extent to which a supplier has customized its assets, processes
and tools to meet a buyers requirements is a signicant indicator of its commitment to meet the idiosyncratic requirements of
the buyer, and is a key economic factor that may impact online
purchase volume and mix exibilities.
According to SET, pure economic factors are insufcient
to explain inter-organizational relationships, and social factors
inform why rms engage in extensive coordination in the interests of maintaining long term relationships. In our context, we
have conceptualized information sharing as the key social factor
that may inuence online purchase volume and mix exibilities. Scholars have acknowledged communication as an essential
factor in e-procurement and suggested that product customization and information sharing constitute key characteristics in
e-procurement processes (Chen et al., 2004; Doney and Cannon,
1997; Muffato and Payaro, 2004; Subramani, 2004). Doney and
Cannon (1997) argue that supplier customization, and condential
information sharing are among the two most important characteristics in inter-organizational relationships and the coordination of
activities.
This paper has also conceptualized two moderating variables
grounded in TCE and SET frequency of transactions and trust that
are likely to impact the direct relationships between supplier customization and purchase volume and mix exibilities and between
information sharing and purchase volume and mix exibilities
respectively. The dependent constructs of interest in our model are
cost, quality, and responsiveness obtained through online purchasing in comparison to traditional modes of purchasing. We discuss
specic hypotheses related to buyersupplier characteristics,

S. Devaraj et al. / Journal of Operations Management 30 (2012) 509520

511

Fig. 1. Research model.

purchase volume and mix exibilities and procurement process

performance next.
2.1. Transaction cost economics (TCE) and supplier customization
Supplier customization refers to the dedicated assets, such as
tools and equipments, and tailored processes that a supplier are
devotes to the buyer rm, and is indicative of the willingness of
the supplier to conform to the needs of the buyer rm. For example, supplier customization includes initiatives on the part of the
supplier to acquire specialized equipments in order to manufacture
customized and tailored products and to modify internal processes
to conform to the needs of the buyer (Doney and Cannon, 1997).
From a transaction costs economics perspective, such customization initiatives constitute asset specicity. Because procedures,
assets and tools are customized to a specic buyer, their economic
value and salvage potential is considerably less if they are switched
to alternative transactions. Thus asset specicity tends to promote
supplierbuyer relationship on economic grounds.
From the buyer rms standpoint, supplier customization allows
it to modify the supply of products rapidly in response to changing
customer needs because the dedicated resources of the supplier can
accommodate buyer requests by modifying processes and employing customized tools and assets. Environmental changes, including
shifting customer needs, can create signicant uncertainty for a
rm. To mitigate this uncertainty, buyers minimize customization
at their end (Sia et al., 2008), and rather rely on suppliers with high
customization to facilitate product volume and mix exibilities.
Such accommodation by suppliers is common. For instance, Hsieh
et al. (2008) claim that market-oriented suppliers from Taiwan
emphasize on the customization of their processes to satisfy their
customers. Such a customization strategy requires a supply chain
that is both volume and mix exible (Salvador et al., 2007). This may
include exibility to increase or decrease the volume of customized
products or change the mix of customized products. In consumer
electronics sector for example, buyers welcome exibility because
of their need for high level of customization and frequent changes
in product mix and delivery.
Supplier arrangements and customizations enable buyer rms
to appropriately structure inter-organizational information ows
and coordinate procurement processes, thereby reducing the effort
necessary for resolving external uncertainties (Gosain et al., 2005).
Once supplier arrangements and customizations are in place, buyer
rms can use e-procurement applications effectively to procure
input materials that would serve customer requirements better.
The rich communication enabled by e-procurement applications

allows rms to transmit features, specications and requirements

much more succinctly and accurately to suppliers than would be
possible through traditional means, such as face-to-face meetings,
paper documents and phone calls, and still convey all the requirements so suppliers can manufacture the component. Additionally,
suppliers customization levels inuence the ability of buyer rms
to use technology in the entire procurement process, and conduct
business electronically.
Due to reasons outlined above lesser opportunistic behavior,
idiosyncratic assets losing value unless relationships are continued, evidence that the supplier rm cares for the relationship, it
is reasonable to believe that if a supplier rm engages in signicant customization, the focal rm (buyer rm) will leverage on this
idiosyncratic investment to gain more exibility from the relationship than in situations where the supplier amasses general purpose
assets and exploits them to supply several other buyers.
Based on the above discussion, we hypothesize
H1. Supplier customization (asset specicity) will be positively
associated with online purchase volume and mix exibilities.
In TCE, frequency is the notion of how often the parties transact with each other and is aimed at capturing the signicance or
importance of these transactions (Teo and Yu, 2005). The following
perspectives highlight the moderating role that frequency of transactions has on the relationship between supplier customization and
exibility performance:
(a) Frequency translates to recency of information The more frequent the transactions, the more recent will be the information
on volume and mix. With a customized setup and more recent
information, the supplier will be better positioned to deliver
on volume and mix exibility. For instance, ceteris paribus, the
quantity and mix of input materials a rm procures can be modied more easily if the materials are procured with a higher
frequency because of the recency of information. As the need
for dynamic adjustment is more acute for rms that transact
frequently with their suppliers, it is evident that the impact of
supplier customization on purchase volume and mix exibility
will be higher for such rms.
(b) Familiarity breeds success Given that the supplier has already
invested signicantly in customized assets to service the focal
customer, the more frequent exchanges between the supplier
and customer imply a level of familiarity and trust that enables
the supplier to provide a level of service (in this case exibility in
volume and mix) that would not have been possible otherwise.
Frequent transactions indicate the need of a closer relationship

512

S. Devaraj et al. / Journal of Operations Management 30 (2012) 509520

with suppliers (Garca-Dastugue and Lambert, 2003). Close

relationships between buyers and suppliers lead to better practices in business processes
(c) Frequency as a proxy for signicance of the relationship Frequent transactions between a supplier and a customer might
denote that in the suppliercustomer dyad, each of them is a
signicant trading partner for the other. From the suppliers
standpoint, the customer being a signicant trading partner and
the fact that the supplier has the ability to supply customized
service provide a catalyst for the supplier to offer higher degree
of exibility to the customer (the focal rm) who is a signicant
trading partner.
(d) Finally, frequency may be used as a basis to counter uctuations in
volume and mix faced by the buyer rm. This argument implies
that the buyer rm by interacting more frequently with the
supplier meets its own demand uncertainties. Further, the suppliers customization allows the seller rm to accommodate
such uctuations.
Based on these reasons, frequency moderates the relationship
between supplier customization and exibility. Thus, we hypothesize:
H2. The effect of supplier customization on purchase volume and
mix exibilities will be stronger when the frequency of transactions conducted between the exchange partners is high. That is,
frequency of transactions positively moderates the relationship
between supplier customization and purchase volume and mix
exibilities.
2.2. Social exchange theory and information sharing
In contrast to the economic rationale of transaction cost economics is the body of literature compiled by social exchange
researchers that presents various social rationales for why
exchange partners might continue to sustain a relationship. From
this theoretical perspective, exchange partners will engage in information sharing if this helps to foster their relationship for a variety
of reasons. First, attachment or prior history of the relationship is a
signicant driver of commitment to a relationship (Seabright et al.,
1992). If two rms have had a long history of successful collaboration, they are more likely to engage in activities such as information
sharing that benet the relationship. Second, shared values of
macro-culture among the exchange partners are an important
social dimension (Young-Ybarra and Wiersema, 1999). These values denote a common understanding of not only the inter-rm
setting but also about resolution of conicts and disputes if they
arise. Third, reputation is an important attribute in exchange relationships (Kollock, 1994). A rm is more willing to rely on and share
information with a partner that has a high reputation in the industry. Reputation, in a sense, is a safeguard against deceptive behavior
(Parkhe, 1993) and thus rms are likely to transact and share information more freely when the reputation of the exchange partners
is high.
Information transfer in a supply chain is dened as the regulated ow of information from one unit (e.g., rm, work group, or
individual) to another. In a systematic procurement arrangement
where regular transactions are expected to occur on a long-term
basis via negotiated contracts with qualied suppliers, organized
information transfer between the trading partners may be crucial to achieve efciency in recurring transactions (Kim et al.,
2005). Johnson et al. (2007) also emphasize the need of information exchange by integrating e-business technologies. According
to them, deploying such technologies with purchasing teams to
work effectively with critical supply chain partners provides competitive advantage to organizations. Inter- and intra-organizational

collaboration is shown to have a strong direct impact on organizational performance (Sanders, 2007).
Organizations use their sense and respond capabilities to
acquire more information about the potential changes and opportunities and to respond to those possibilities. The information
sharing capabilities include the integration of the systems, decisions, and processes (Hsu et al., 2008). The availability of
information allows rms to sense the need for change in their current process conguration and to develop mechanisms for dealing
with change. Information sharing is needed to allow an enterprise to sense the needs of the partners and communicate its own
needs to them. Knowledge obtained through sharing of information
enables rms to react to unanticipated change and to be continually
attuned to change. Firms that exchange information with suppliers
extensively can respond to demand changes with agility. Frequent
information exchange between a rm and its suppliers encourages
the rm to share both tactical information (e.g., engineering change
orders, reject rates, inventory positions), and strategic information
(e.g., product roadmap, demand forecast, cost curves) with suppliers. Such detailed information sharing enables the buyer and its
suppliers to coordinate their design efforts, production plans and
shipping schedules more effectively.
Changes in demands give rise to uncertainty among suppliers
and customers. To resolve this uncertainty, suppliers and customers need to share information about customer demands, and
supplier capacities and schedules. Supply chain coordination suffers when decision makers have incomplete information about
latest customer demands (Sahin and Robinson, 2002, 2005). In
contrast, Cachon and Fisher (2000) have found that the sharing
of real-time demand information provides signicant operational
improvements. In addition to providing operational savings, rms
have come to realize that sharing information with business partners can facilitate meeting customers needs in a timely manner.
The exchange of information with suppliers enables buyer rms
to communicate customer requirements to suppliers, and aids
in procuring parts and components that are required to satisfy changing customer requirements. As customers needs evolve
dynamically, rms can gather information about them, share this
information with suppliers and respond to such events faster
through the use of e-procurement applications in comparison to
traditional modes. Firms that share extensive information with
suppliers are likely to leverage their existing supplier base and
exploit e-procurement solutions for enhanced collaboration, exibility and transactional efciencies (Jap and Mohr, 2002). These
rms are more open to using e-procurement solutions for coordinating production plans and shipping schedules on the Internet
with extant suppliers, and to transacting online (Premkumar and
Ramamurthy, 1995). The implementation of an e-procurement
application makes it possible to digitalize delivery plans and share
extensive information with suppliers, which lead to greater exibility and better control of the goods supplied (Muffato and Payaro
(2004). Based on these arguments, we hypothesize,
H3. Information sharing facilitated by e-procurement will be positively associated with purchase volume and mix exibilities.
Young-Ybarra and Wiersema (1999) state that trust between
organizations will have a positive impact on the desire and ability
of partners to adjust to changing environmental demands. Higher
levels of trust in business partners indicate that an organization
believes in the competence or expertise of its partners to deliver.
The trusting organization also expects that the trusted party will
be benevolent and not take advantage of the new conditions by
willful exploitation (Young-Ybarra and Wiersema, 1999). Trust is
considered to be the central tenet of interorganizational exchanges,
and has been posited to provide outcomes that are superior to
the ones that rms can obtain if they act solely in self-interest

S. Devaraj et al. / Journal of Operations Management 30 (2012) 509520

(Anderson and Narus, 1990). Additionally, while increased information sharing enables suppliers and buyers to be exible to customer
demands (Rosenzweig et al., 2003), Handeld and Bechtel (2002)
argue that rms are more willing to be responsive to partners who
have exhibited trustworthiness.
The trust between exchange partners can strengthen the relationship between information sharing and purchase volume and
mix exibility through many reinforcing impacts. Higher levels of
trust encourage organizations to share accurate and frequent information with partners because of the belief that such information
will not be misused but used to benet both partners. The use
of new information for mutual benets becomes easier when the
trust between partners is high. The cooperative behavior facilitated
by higher trust enables rms to survive greater stress and display
adaptability under changing conditions (Doz, 1996; Lorenz, 1988).
Higher trust between partners lowers transactions risks and information abuse online by partners and increases the chances that
contractual terms will be honored online, and hence is likely to
enhance the relationship between information sharing and exibilities facilitated through e-procurement.
Based upon these arguments, we propose:
H4. The effect of information sharing on purchase volume and mix
exibilities will be stronger when the trust between the exchange
partners is high. That is, trust between exchange partners positively moderates the relationship between information sharing and
purchase volume and mix exibilities.
2.3. Performance implications of purchase volume and mix
exibilities in e-procurement
Koste and Malhotra (1999) conducted a comprehensive review
of antecedents and consequences of exibility in the literature and
called for future studies to examine the relationship between exibility and other performance outcomes. Although prior research
has documented the impact of exibility on rm performance and
competitiveness (Reichhart and Holweg, 2007; Anand and Ward,
2004; De Groote, 1994; Gerwin, 1993; Jordan and Graves, 1995;
Upton, 1997), we provide a unique perspective on how procurement exibilities can impact performance at the process level. We
contend that exibility in business processes is a means to an end,
and in fact, may inuence other dimensions of performance signicantly. In other words, exibility may be a predictor of other
dimensions of performance, such as cost, quality, and delivery, the
three main drivers of performance in the supply-chain literature
(Thomas and Grifn, 1996). There are several reasons that might
explain why exibility might lead to better outcomes of cost, quality, and delivery.
The rst rationale is that material ow strongly impacts business performance, and organizational purchase volume and mix
exibilities may enable timely modication of the volume and mix
of products sourced from suppliers, and thereby can have a significant impact on such ows. This is consistent with the precepts of
the theory of swift and even ow (TSEF), which suggest that swift
ow of materials will result in improved performance in terms of
cost and quality (Schmenner and Swink, 1998).
The second reason is that exibility in the purchasing process
allows rms to be able to be responsive to changing customer needs
and requirements. In todays marketplace marked with uncertainty in customer demands, rms that are able to better relay
the customer-induced changes to the supplier side (via exible
purchasing agreements) will perform better on quality, costs, and
responsiveness. Also, the ability to adjust the volume and mix
of input goods dynamically using e-procurement enables companies to lower transaction costs and employ competitive sourcing
opportunities. Such efforts, typically, result in cost reductions of

513

65% compared to traditional procurement transactions (Croom,

2000).
Finally, one of the major benets of exibility in business processes is the responsiveness it affords to the members of the
value-chain (Fisher, 1997). Such exibility allows companies to use
substitutable input goods and dynamically allocate different volume and product mixes among suppliers, taking into consideration
their competence and prior performance as well as the demand of
nal products from the end-consumers.
Based on the above arguments, exibility in volume and mix is
hypothesized to have a benecial impact on process performance
assessed along the dimensions of cost, quality, and responsiveness.
We hypothesize:
H5 (a, b, c). Purchase volume and mix exibilities facilitated by
e-procurement will be positively associated with (a) lower costs,
(b) higher quality, and (c) better responsiveness and delivery performance.
3. Sample and measures
We describe the sample, measures, and the psychometric properties of the scales used in this section.
3.1. Sample
Our sample respondents consisted of procurement managers
from the aerospace, automotive, electronics, and consumer product manufacturing industries in US. We solicited their responses
primarily through mail and based on membership directories and
mailing lists which included purchasing and procurement managers from Midwest US. Out of 400 individuals contacted, 141
participated in the study yielding a response rate of 35.3%. Of these,
only 131 responded to the complete survey, giving us a usable or
effective response rate of 32.8%. We informed the respondents that
a benchmarking report will be made available to them at the conclusion of the study. Participants were rst introduced to the study
requirements and then directed to the survey questionnaire in the
mail or directed to a website created for the purpose. There was
no statistical difference among the constructs used in this study
between the responses obtained through the two methods. The
items used in the survey are listed in Appendix A. Mailing was
done in two waves and we conducted a standard two-sample T
test examining the differences between the early and late waves as
a way of testing non-respondent bias. We did not observe significant differences between the two waves of data collection at the
0.05 level of signicance.
3.2. Construct validation
The measures employed in this study are adapted from extant
literature. Table 1 presents the denitions of the constructs along
with their references. Appendix A provides the scale items, and reliabilities of the constructs used in this study. Appendix B presents
the descriptive statistics and correlation matrix. We used AMOS 5.0
to implement a conrmatory factor analysis (CFA) described later
in this section.
3.2.1. Reliability
Reliability is an indication of measurement accuracy, which is
the extent to which instrumentation produces consistent or errorfree results. In our study, all the developed scales demonstrated
reliabilities above the recommended threshold of 0.7 for cronbachs
alpha (Nunnally and Bernstein, 1994). Appendix A illustrates reliabilities, factor loadings, and the percent of variation extracted for
all the constructs. We rst conducted a factor analysis to get a

514

S. Devaraj et al. / Journal of Operations Management 30 (2012) 509520

Table 1
Denitions, constructs, and references.

Construct
Supplier customization
Information sharing
Frequency
Trust
Purchase volume exibility
Purchase mix exibility

Performance
Cost
Delivery
Quality

Denitions

References

This entails suppliers conforming to the needs of buyers and fullling them
Involves the extent to which suppliers and customers share private information with each
other
The percent of transactions done online
Trusting suppliers online
The ability of buyers to change the volume of products purchased with little penalty in time,
effort, cost, or performance
The ability of buyers to change the mix of products purchased with little penalty in time,
effort, cost, or performance
Both volume and mix exibility enable rms to dynamically purchase products in response to
changes in market demands during the short life cycle of the product
Performance comprises of relative cost, delivery, and quality
The extent to which online purchases provide cost advantages relative to traditional purchases
The extent to which online purchases provide faster and more reliable delivery relative to
traditional purchases
The extent to which online purchases meet conformance quality requirements

Doney and Cannon (1997)

Doney and Cannon (1997)

preliminary check for convergent and discriminant validities. As

can be seen from the table in Appendix C, all items loaded onto their
respective factors with high loadings and there were no signicant
cross-loadings.
3.2.2. Content validity
Content validity is the degree to which items in an instrument
reect the content generalization (Cronbach, 1951). As indicated
in Table 1, the items used in this survey were extracted from validated scales in the extant literature. Furthermore, all the items were
presented in a pilot study to purchasing and procurement managers and validated by checking the content for their managerial
real-world meaning. Feedback was sought from three researchers
actively involved in e-procurement and supply-chain management
research. Next, we shared the questionnaire with three procurement managers. In both situations, we had a detailed conversation
with the individuals, who made an assessment of the clarity of our
questions. In some instances, the questions were modied based
upon their comments. The review of literature as well as the results
from the pilot study analyzed by a group of experts in the eld
provided further reassurance about the adequacy of the content
validity of our instrument (Boudreau et al., 2001).
3.2.3. Convergent validity
The convergent validity of a scale can be checked using the
BentlerBonett coefcient () (Bentler and Bonnet, 1980). The
BentlerBonnet coefcient is the ratio of the difference between the
chi-squared value of the null measurement model and the specied measurement model to the chi-squared value of the null model.
The null model has no hypothesized factor loading on a common
construct in the conrmatory factor analysis. The BentlerBonnet
coefcient value between 0.80 and 0.90 is acceptable and a value
greater than 0.90 indicates strong convergent validity (Bentler and
Bonnet, 1980). The BentlerBonnet coefcient for our model is
0.916, which indicates strong convergent validity.
3.2.4. Discriminant validity
Discriminant validity explores the extent to which a concept and
its indicators differ from another concept and its indicators (Bagozzi
et al., 1991). This is achieved when measures of each construct
converge on their corresponding true scores (which are unique
from other constructs) and can be tested by comparing the correlations between the pairs of dimensions (Venkatraman, 1989). This
requires a comparison of two CFA models, one with the correlation
constrained to equal one and another with an unconstrained model.
A signicantly lower 2 value for the model with the unconstrained
correlation, when compared with the constrained model, provides

Doney and Cannon (1997)

Duclos et al. (2003)
Devaraj et al. (2004)

Doney and Cannon (1997)

Doney and Cannon (1997)
Mentzer et al. (2001)

support for discriminant validity (Venkatraman, 1989). A 2 difference value with an associated p-value less than 0.05 (Joreskog and
Sorbom, 1989) supports the discriminant validity criterion. This
procedure is repeated for all pairs of scales in the instrument (not
presented in the interests of brevity). The signicances of all differences are less than 0.05, which indicates discriminant validity of
the constructs employed in our study.
3.2.5. Overall model t
Finally, we conducted a CFA. Results of the CFA yielded a
goodness-of-t index (GFI) of 0.928, adjusted goodness-of-t
(AGFI) of 0.801, BentlerBonnet delta of 0.916, root mean square
error of approximation (RMSEA) of 0.04, and a chi-square to degree
of freedom ratio of 1.83. A GFI over the cut-off 0.9 and the AGFI over
the cut-off 0.8 imply good absolute t. A RMSEA value means a low
residual variance and hence a good parsimonious t. RMSEA is close
to the cut-off value of 0.06 as suggested by Browne and Cudeck
(2003). Thus, the CFA points to evidence of a good measurement
model.
3.2.6. Common method variance
Common method variance can be a potential source of bias in
survey research. We examined this issue in several ways. First, consistent with prior studies, we conducted Harmans one-factor test
(Podsakoff and Organ, 1986). A factor analysis performed on the
variables did not yield a single-factor solution. Hence, based on
this test the threat of common method bias is not signicant. The
second test of common method variance was based on partial correlations (Podsakoff et al., 2003). We computed the rst factor from
a principal factor analysis of all the constructs and included this as
a control variable in the analysis, since this factor is assumed to
be a good approximation of the common method bias. The addition of the factor score as a control variable did not signicantly
change the variation explained in the performance variables suggesting that common method variance is not a serious concern in
our study (Podsakoff et al., 2003). In summary, based on tests of
reliability, validity, and overall model t, there is strong support
for the suitability of the constructs employed in this study.
4. Analyses and results
We employed structural equation modeling (SEM) to estimate
the proposed research model. To model the interactions between
supplier customization and frequency of transactions as well as
between information sharing and trust, we employed the cross
product of the items of the separate constructs as the latent variable

S. Devaraj et al. / Journal of Operations Management 30 (2012) 509520

Table 2
Validity statistics.
Goodness of t indices
Absolute t
Chi-square/DF
Goodness of Fit Index (GFI)
Adjusted Goodness of Fit Index (AGFI)
Incremental t
Normed Fit Index (NFI)
Incremental Fit Index (IFI)
Non-Normed Fit Index (NNFI)
Comparative Fit Index (CFI)
Parsimonious t
Root Mean Square Error of Approximation (RMSEA)

1.34
0.96
0.84
0.91
0.91
0.92
0.92
0.05

capturing the interaction. Table 2 provides several goodness-oft statistics to assess how well the specied model explains the
observed data from three aspects: absolute t, incremental t, and
parsimonious t (Maruyama, 1998, Tanaka, 1993). The absolute t
measures of GFI, AGFI, and the Normed chi-square, which is the
ratio of the chi-square divided by degrees of freedom indicate a
good t of the model to the data (Bentler and Bonnet, 1980; Bagozzi
and Yi, 1988).
Four incremental t indices are also listed in Table 2. (i) Normed
Fit Index (NFI), represents the proportion of total covariance among
observed variables explained by a target model using a baseline null
model, (ii) Incremental Fit Index (IFI) is the ratio of the discrepancy
of the proposed and baseline models over the difference of their
respective degrees of freedom (Bollen, 1989), (iii) Non-Normed Fit
Index (NNFI), also known as TuckerLewis Index (TLI), examines
moment structures (Bentler and Bonnet, 1980; Bollen, 1989), and
(iv) Comparative Fit Index is one of the indices that determines
incremental ts and estimates non-centrality parameters of the
model (Gefen et al., 2000). The values shown in the table suggest
that our proposed model ts the data well. Additionally, a small
RMSEA value means a low residual variance and hence a good parsimonious t and RMSEA. These values provide strong evidence of
good model t.
4.1. SEM results
A summary of the hypotheses test results is presented in
Table 3. The results of our analysis provide support for H1 (p < 0.01).
Supplier customization has a positive and signicant relationship
with both purchase volume and mix exibility.
We nd partial support for the moderating effects of frequency
of transactions on the relationship between supplier customization and purchase volume and mix exibilities. The relationship is
statistically signicant for purchase volume exibility but not mix
exibility. This nding suggests that while supplier customization

515

in itself might have a benecial impact on purchase volume exibility, its effect is amplied when there is increased frequency of
online transactions between supply chain partners. Thus, H2 is only
partially supported.
We nd statistical support for H3 (p < 0.01) conrming that
information sharing has a positive association with both purchase
volume and mix exibility.
We also found evidence of positive moderating effects of trust
on the relationship between information sharing and both purchase
volume and mix exibilities. In other words, when investments in
information sharing are supported with trust between the partners, we see evidence of greater purchase volume exibility and
purchase mix exibilities. Thus, H4 is statistically supported.
Finally, H5 (a, b, and c) are all supported (p < 0.01), demonstrating that purchase volume and mix exibilities facilitated
by e-procurement enable procurement cost, delivery and quality
improvements in comparison to traditional modes of procurement.
Finally, the control for size of the organization was statistically
signicant with quality performance.
4.2. Alternate model testing: mediating role of purchase volume
and mix exibilities
Fig. 2 illustrates the comparison of alternative models. We tested
for the mediating effect of purchase volume exibility and purchase mix exibility by implementing Baron and Kennys (1986)
procedure to test for mediation.
Step 1: Use e-Procurement Performance as the dependent variable and Customization/Information Sharing as the independent
variable, test the statistical signicance of this relationship.
Step 2: Use Mix/Volume exibility ow as the dependent variable and Customization/Information Sharing as the independent
variable to test for the statistical signicance of this relationship.
Step 3: Use e-Procurement Performance as the dependent variable
and Customization/Information Sharing as independent variables
to test for the statistical signicance of this relationship.
Step 4: Test the effect of Customization/Information Sharing (independent variable) on e-Procurement Performance (dependent
variable) after controlling for Mix/Volume exibility (independent variable). For Mix/Volume exibility to completely mediate
the relationship between Customization/Information Sharing
and e-Procurement Performance, the coefcient for Customization/Information Sharing should be statistically non-signicant.
If all four steps are signicant then the data are consistent with
complete mediation. If steps 13 hold and step 4 does not, then the
data are consistent with partial mediation.

Table 3
Summary of the structural model.
Hypothesis

Expected relationship

Path coefcient

Supported?

H1: Supplier customization volume exibility

Supplier customization purchase mix exibility
H2: Interaction of supplier customization and frequency purchase volume exibility
Interaction of supplier customization and frequency purchase mix exibility
H3: Information sharing purchase volume exibility
Information sharing purchase mix exibility
H4: Interaction of information sharing and trust purchase volume exibility
Interaction of information sharing and Trust Purchase mix exibility
H5A: Purchase volume exibility cost
Purchase mix exibility cost
H5B: Purchase volume exibility quality
Purchase mix exibility quality
H5 C: Purchase volume exibility delivery
Purchase mix exibility delivery

+
+
+
+
+
+
+
+
+
+
+
+
+
+

0.26
0.17
0.16
0.07
0.48
0.31
0.34
0.14
0.33
0.24
0.30
0.22
0.26
0.11

Yes; p < 0.01

Yes; p < 0.05
Yes; p < 0.05
No; p > 0.1
Yes; p < 0.01
Yes; p < 0.01
Yes; p < 0.01
Yes; p < 0.05
Yes; p < 0.01
Yes; p < 0.01
Yes; p < 0.01
Yes; p < 0.01
Yes; p < 0.01
Yes; p < 0.05

516

S. Devaraj et al. / Journal of Operations Management 30 (2012) 509520

Fig. 2. Comparison of alternative models.

Table 4
Model comparisons.

(2 /DF)
GFI
AGFI

4.3. A general path analytic model combining mediation and

moderation

Hypothesized mediated model

Direct effects model

1.34
0.962
0.842

2.77
0.884
0.741

Consistent with these steps we nd evidence for complete

mediation. In other words, information sharing and supplier customization provide no direct improvement on e-procurement
performance; their impacts are felt completely through purchase
volume and mix exibilities. In testing structural equation models,
Marsh (1994) suggests comparing the research model to alternative models. Thus, we also tested a direct effects model to show
the direct effects of e-procurement process characteristics on eprocurement performance.
A comparison of the two models is presented in Table 4.
Based on the comparative model t criteria, we observe that
the hypothesized mediating model ts the data signicantly better than the other model. Furthermore, the Normed chi-square (i.e.
the (2 /DF) for our hypothesized mediating model is less than the
cut-off value of 2 as suggested by Byrne (1989). These results, along
with better goodness-of-t indices for the mediated model, lead us
to believe that purchase volume and mix exibilities play an important mediating role in explaining e-procurement performance.

There are many studies that combine moderation and mediation

within the same research model. Normally, these are presented
as moderated mediation or mediated moderation. Recent work
by Edwards and Lambert (2007) suggest that these analytical
approaches might have important shortcomings that conceal the
true form of the moderated and the mediated effects being studied.
Their approach involves combining moderated regression analysis and path analysis to obtain better estimates of the moderating
and mediating coefcients. The general approach can be described
below:
1. Mediation is modeled as a path model with relationships among
variables in the model captured as regression equations.
2. Moderation is modeled by supplementing the above regression
equations with the moderator variable and its product with the
independent variable and the mediator variable.
3. By using reduced form equations, obtained by substituting the
regression equation for the mediator variable into the equation
for the dependent variable, direct, indirect, and total effects can
be computed.
4. The reduced form equations contain products of regression
coefcients; therefore ordinary OLS cannot be used. Constrained
nonlinear regression (CNLR) is used to generate coefcients from
bootstrap samples.

S. Devaraj et al. / Journal of Operations Management 30 (2012) 509520

We used IBM SPSS 19 to conduct the above analysis. Results

of this analysis showed an average coefcient for the moderating effect of frequency on the relationship between supplier
customization and volume exibility and mix exibility was 0.24
(signicant) and 0.08 (not signicant) respectively. Further, the
average coefcient for the moderating effect of trust on the
relationship between information sharing and volume exibility
and mix exibility was 0.32 (signicant) and 0.16 (signicant)
respectively. These results afrm those obtained through SEM and
reported in Table 3.
Taken together, the statistical support for all the hypotheses
presented in this study point to the important role that purchase
volume and mix exibilities play in the context of e-procurement.

5. Implications, limitations, and future research

There are several academic and managerial implications stemming from the ndings of this study. From a theoretical standpoint,
rst, there has been a need to examine the mechanisms through
which business value of e-procurement is derived. We proposed
and showed that using transaction cost economics and social
exchange theory, two mechanisms that drive exibility are supplier
customization and information sharing.
Second, while purchase volume and mix exibility is critical
to supply chain performance, it is the least studied in academic
research and points to a gap in the literature. In todays dynamic
environment, the concept of exibility is especially important.
We dene, and operationalize purchase volume exibility and
purchase mix exibility derived from e-procurement. Third, we
propose and empirically test a nuanced perspective of the contexts under which the two causal pathways from information
sharing and supplier customization lead to exibility. We nd evidence for moderating effects of frequency and trust, two important
constructs steeped in the transaction cost economics and social
exchange theory literature. While these two constructs have been
found signicant for supply chain management, their role has been
largely unexplored in the context of purchase volume and mix exibilities.
Finally, there is still a need for academic research on the performance impacts of e-procurement. We show that purchase volume
and mix exibilities serve the important role of mediators through
which eventual effects are realized on cost, quality, and delivery
performance. Thus, exibility in e-procurement can be viewed not
as an end in itself, but rather a means to the end objectives of cost,
quality, and responsiveness or delivery performance. In summary,
while extant literature contends that we still lack understanding of
the coordinated capabilities of an inter-organizational infrastructure that will enable exible supply chains and theory to guide
its development (Gosain et al., 2005; Weill and Broadbent, 1998),
our study responds to this mandate by illustrating how purchase
volume and mix exibilities impact e-procurement performance.
The only result that was not signicant in our analysis was
the hypothesized moderating role of frequency on the relationship between supplier customization and purchase mix exibility.
This result indicates that once a supplier has made the investment
in tools, assets and modied procedures, the recency of information made possible through higher frequency of transaction may
not provide any additional benets as the a priori customization enables rms to modify the mix of products easily through
online purchasing. From a managerial perspective, the relationships between the e-procurement characteristics and performance
can provide useful insights to procurement managers, supply
chain specialists, and practitioners. First we offer directions on the
identication and measurement of key characteristics that affect
purchase volume and mix exibility. Further, we demonstrate

517

the business impact of enhancing purchase volume exibility and

purchase mix exibility. While there has been considerable attention in practitioner publications on the value of these exibilities,
we assess the magnitude of its effect on measures of cost, quality, and delivery by examining the coefcients in the estimation
model. Thus, managers can decide if an investment in increasing
purchase volume and mix exibilities is worthwhile in their contexts. Furthermore, we demonstrate that purchase volume and mix
exibilities are instrumental for obtaining value in other procurement process dimensions, and thus managers must spend sufcient
resources to facilitate them from a coordination perspective.
While there has been much discussion about agility in practitioner publications, there has been little empirical evidence of
the business value of agility. To the extent that one of the key
manifestations of agility is exibility, we demonstrate the business
value of agility by documenting the effect of exibility on business
performance such as cost, quality, and delivery. Based on information sharing and relying on each others strengths, electronic
manufacturing companies build relationships with their contract
manufacturers and this leads to strong and agile purchasing strategy (Mason et al., 2002). Consequently, collaborative systems that
foster information sharing between buyer and suppliers provide
large companies with gains in exibility and the ability to alter
purchasing capacity levels (Humphreys et al., 2001).
Finally, our study helps practitioners in understanding the
impact of the levers driving purchase volume and mix exibilities,
but more importantly, the contexts under which these levers work
best. That is, supply chain managers who nd their organizations
transacting on a signicant basis (frequency) with their customers
would do well to invest in customization, whereas their investment
in information sharing is more suitable when there is a high degree
of trust in the suppliercustomer relationship.
The study has a few limitations as well which present opportunities for future research in this area. First, respondents to our survey
were primarily from the aerospace, automotive, electronics, and
consumer products manufacturing industries. As such, our results
are generalizable to such a population and further research might
be warranted to demonstrate the efcacy of our research model
in a wider cross-section. Second, it is conceivable that there are
many other dimensions of procurement performance than those
that we included in the study such as strategic importance of
e-procurement to the rms overall strategy. To the extent that
industry type and size of the rm might serve as proxies for strategic necessity of e-procurement, our analysis accounts for them in
the form of control variables. Our selection depicts a parsimonious
collection of constructs based on extant literature as well as our
discussions with procurement managers. It is not our intent to
examine the various forms of e-procurement systems in practice
but address the more fundamental issue of what underlying characteristics affect performance and how they do it. Future studies
might examine the various forms of procurement systems. This is
again an opportunity for future research to expand the domain of
the relationships examined in this study. Finally, we have considered the buyer side perspective only in this empirical study. Further
studies might explore the buyerseller dyad and its implications for
e-procurement performance.
In conclusion, while e-procurement technologies have become
pervasive in business, there has been little academic inquiry
to understand the process by which they confer performance
gains. We proposed and empirically examined how two key
mechanisms supplier customization, and information sharing impact purchase volume and mix exibilities, which in
turn affects e-procurement performance. We hope our study
spurs further interest and research to examine other pathways through which e-procurement technologies provide business
value.

518

S. Devaraj et al. / Journal of Operations Management 30 (2012) 509520

Appendix A.
Scale items and reliabilities of constructs
Factor loading
Supplier customization (adapted from Doney and Cannon, 1997) ( = 0.742, variation explained = 66.42%)
CUST1
0.690
Just for us my online supplier is willing to customize its products
CUST2
0.858
Just for us my online supplier is willing to change its process
Just for us my online supplier is willing to customize services, tools and equipment
CUST3
0.884
Information sharing (adapted from Doney and Cannon, 1997) ( = 0.736, variation explained = 65.84%)
0.852
Our online suppliers share proprietary information with our rm
INFO1
0.794
Our online suppliers will share condential information to help us
INFO2
INFO3
0.787
Our online suppliers make online catalog information available
Trust (adapted from Doney and Cannon, 1997) ( = 0.936, variation explained = 73.95%)
Our online suppliers keep promises it makes to our rm
TRST1
0.704
TRST2
0.915
Our online suppliers are not always honest with us (reverse coded)
0.961
We believe the information that our online suppliers provide us.
TRST3
0.701
Our online suppliers are genuinely concerned that our business succeeds
TRST4
0.961
When making important decisions, our online suppliers consider our welfare as well as its own
TRST5
We trust that our online suppliers keep our best interests in mind
TRST6
0.915
0.783
Our online suppliers are trustworthy
TRST7
0.892
We nd it necessary to be cautious with online suppliers (reverse coded)
TRST8
Cost (adapted from Doney and Cannon, 1997) ( = 0.907, variation explained = 84.30%)
0.949
I save money when I purchase online compared to traditional purchases
COST1
0.920
Online purchasing is costlier than traditional purchasing (reverse coded)
COST 2
COST 3
0.884
Our online cost per transaction relative traditional purchasing has been reduced
Delivery (adapted from Doney and Cannon, 1997) ( = 0.725, variation explained = 72.69%)
0.853
By ordering online, delivery is faster than traditional purchasing
DELV1
0.853
By ordering online, delivery is more reliable than traditional purchasing
DELV2
Quality (adapted from Mentzer et al., 1997; Parasuraman et al., 1988) ( = 0.707, variation explained = 72.69%)
0.879
Products ordered online meet technical requirements compared to traditional purchasing
QUAL1
Products ordered online are rarely nonconforming compared to traditional purchasing.
QUAL2
0.607
QUAL3
0.800
I believe that what I ask for is what I get in online purchasing compared to traditional purchasing
Purchase volume exibility and purchase mix exibility (adapted from Devaraj et al., 2004) ( = 0.723, 0.708) variation explained = 62.46% and 60.05%)
PVF11
Online purchasing increases our purchase volume exibility relative to traditional purchasing
Online services or suppliers provide me with exibility in purchasing
PVF12
Online purchasing increases our purchase mix exibility relative to traditional purchasing
PMF13
Online purchasing provides me with exibility in purchasing
PMF14
Note: A seven point Likert scale from 1 to 7 (strongly disagree to strongly agree) was used for all study constructs.

Appendix B.
Correlation matrix and descriptive statistics for constructs
Construct

Mean

SD

(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

(9)

(1) Customization
(2) Information sharing
(3) Frequency
(4) Trust
(5) Purchase volume exibility
(6) Purchase mix exibility
(7) Cost
(8) Quality
(9) Delivery
(10) Size

6.48
6.28
16.71
6.20
6.47
6.32
6.38
6.46
6.18
54938.09

0.56
0.625
6.41
0.43
0.48
0.42
0.57
0.41
0.52
48647.15

0.559**
0.176*
0.108
0.318**
0.187*
0.323**
0.216*
0.352**
0.116

0.480
0.023
0.410**
0.318**
0.457**
0.408**
0.345**
0.237

0.021
0.158
0.078
0.176*
0.277*
0.220*
0.171

0.081
0.098
0.047
0.001
0.217*
0.148

0.259**
0.322**
0.304**
0.393**
0.164

0.246**
0.294**
0.180*
0.011

0.329**
0.108
0.100

0.180*
0.228

0.122

*
**

Correlation is signicant at the 0.05 level (2-tailed).

Correlation is signicant at the 0.01 level.

S. Devaraj et al. / Journal of Operations Management 30 (2012) 509520

519

Appendix C.
Factor analyses showing unidimensionality, and convergent and discriminant validity
Component

TRST1
TRST2
TRST3
TRST4
TRST5
TRST6
TRST7
TRST8
COST1
COST2
COST3
QUAL1
QUAL2
QUAL3
CUST1
CUST2
CUST3
INFO1
INFO2
INFO3
DELV1
DELV2
FLEX1
FLEX2
FLEX3
FLEX4

0.652
0.937
0.942
0.673
0.952
0.883
0.742
0.883
0.038
0.092
0.038
0.086
0.006
0.098
0.161
0.123
0.086
0.092
0.040
0.096
0.134
0.134
0.116
0.053
0.055
0.072

0.054
0.115
0.006
0.096
0.006
0.034
0.019
0.134
0.903
0.849
0.863
0.016
0.104
0.111
0.256
0.027
0.112
0.202
0.273
0.045
0.047
0.006
0.281
0.123
0.177
0.125

0.085
0.036
0.040
0.126
0.040
0.034
0.076
0.078
0.022
0.187
0.122
0.901
0.875
0.563
0.178
0.046
0.078
0.121
0.101
0.050
0.108
0.073
0.193
0.065
0.033
0.074

0.161
0.221
0.103
0.111
0.103
0.145
0.028
0.110
0.093
0.135
0.079
0.049
0.025
0.209
0.551
0.684
0.623
0.158
0.152
0.129
0.084
0.140
0.135
0.034
0.138
0.092

0.064
0.040
0.033
0.031
0.033
0.040
0.161
0.047
0.110
0.149
0.031
0.034
0.178
0.220
0.091
0.135
0.153
0.641
0.467
0.848
0.172
0.063
0.041
0.137
0.062
0.114

6
0.115
0.072
0.062
0.087
0.062
0.067
0.092
0.086
0.039
0.092
0.015
0.202
0.059
0.020
0.142
0.198
0.194
0.094
0.190
0.133
0.826
0.787
0.161
0.089
0.121
0.142

7
0.050
0.035
0.096
0.062
0.096
0.008
0.139
0.075
0.152
0.141
0.049
0.042
0.081
0.179
0.027
0.078
0.029
0.104
0.040
0.165
0.143
0.126
0.727
0.902
0.341
0.186

8
0.087
0.127
0.084
0.021
0.073
0.023
0.049
0.061
0.143
0.168
0.069
0.073
0.063
0.038
0.051
0.083
0.034
0.126
0.069
0.153
0.182
0.137
0.349
0.391
0.734
0.631

Extraction method: Principal Component Analysis. Rotation method: Varimax with Kaiser Normalization.

References
Anderson, J., Narus, J., 1990. A model of distributor rm and manufacturer rm
working relationships. Journal of Marketing 54 (1), 4258.
Anand, G., Ward, P.T., 2004. Fit, exibility and performance in manufacturing: coping
with dynamic environments. Production and Operations Management 13 (4),
369385.
Bagozzi, R.P., Yi, Y., 1988. On the evaluation of structural equation models. Academy
of Marketing Science 16 (1), 7494.
Bagozzi, R.P., Yi, Y., Phillips, L.W., 1991. Assessing construct validity in organized
research. Administrative Science Quarterly 36 (3), 421458.
Baron, R.M., Kenny, D.A., 1986. The moderatormediator variable distinction in
social psychological research: conceptual, strategic and statistical considerations. Journal of Personality and Social Psychology 51, 11731182.
Bentler, P.M., Bonnet, D.G., 1980. Signicance tests and goodness of t in the analysis
of covariant structures. Psychological Bulletin 88 (3), 588606.
Bollen, K.A., 1989. A new increment r index for general structural equation models.
Sociological Methods and Research 17, 303316.
Boudreau, M., Gefen, D., Straub, D.W., 2001. Validation in IS research: a state-of-theart assessment. MIS Quarterly 25 (1), 124.
Bradley, A., 2005. E-procurement: a long way to go. Supply Management 10 (20),
15.
Browne, M., Cudeck, R., 2003. Alternative ways of assessing model t. In: Bollen, K.A.,
Long, J.S. (Eds.), Testing Structural Equation Models. Sage Publications, London,
U.K.
Byrne, B.M., 1989. A Primer of LISREL: Basic Applications and Programming for
Conrmatory Factor Analytic Models. Springer-Verlag, New York, NY.
Cachon, G., Fisher, M., 2000. Supply chain inventory management and the value of
shared information. Management Science 46 (8), 10321048.
Cao, Q., Dowlatshahi, S., 2005. The impact of alignment between virtual enterprise
and information technology on business performance in an agile manufacturing
environment. Journal of Operations Management 23, 531550.
Chen, I.J., Paulraj, A., Lado, A.A., 2004. Strategic purchasing, supply management and
rm performance. Journal of Operations Management 22 (5), 505523.
Cronbach, L.J., 1951. Coefcient alpha and the internal structure of tests. Psychometrika 16, 297334.
Croom, S., 2000. The impact of Web-based procurement on the management of
operating resources supply. The Journal of Supply Chain Management 36 (1),
413.
De Groote, X., 1994. The exibility of production processes: a general framework.
Management Science 40 (7), 933945.
Devaraj, S., Wei, J., Krajewski, L., 2007. Impact of eBusiness technologies on operational performance: the role of production information integration in the supply
chain. Journal of Operations Management 25 (6), 11991216.
Devaraj, S., Hollingworth, D.G., Schroeder, R.G., 2004. Generic manufacturing strategies and plant performance. Journal of Operations Management 22 (3), 313333.

Doney, P.M., Cannon, J.P., 1997. An examination of the nature of trust in buyerseller
relationships. Journal of Marketing 61, 3551.
Doz, Y.L., 1996. The evolution of cooperation in strategic alliances: initial conditions
or learning processes. Strategic Management Journal 17, 5583.
Duclos, L.K., Vokurka, R.J., Lummus, R.R., 2003. A conceptual model of supply chain
exibility. Industrial Management & Data Systems 103 (6), 446457.
Edwards, J.R., Lambert, L.S., 2007. Methods for integrating moderation and
mediation: a general analytical framework using moderated path analysis. Psychological Methods 12 (1), 122.
Fisher, M.L., 1997. What is the right supply chain for your product? Harvard Business
Review 75 (2), 105116.
Gefen, D., Straub, D.W., Boudreau, M.C., 2000. Structural equation modeling and
regression: guidelines for research practice. Communications of AIS 4 (7), 7178.
Gerwin, D., 1993. Manufacturing exibility: a strategic perspective. Management
Science 39 (4), 395410.
Gosain, S., Malhotra, A., El Sawy, O.A., 2005. Coordinating for exibility in e-business
supply chains. Journal of Management information Systems 21 (3), 746.
Garca-Dastugue, S.J., Lambert, D.M., 2003. Internet-enabled coordination in the
supply chain. Industrial Marketing Management 32 (3), 251263.
Handeld, R.B., Bechtel, C., 2002. The role of trust and relationship structure in
improving supply chain responsiveness. Industrial Marketing Management 31,
367382.
Hill, C.A., Scudder, G.D., 2002. The use of electronic data interchange for supply
chain coordination in the food industry. Journal of Operations Management 20,
375387.
Hsieh, Y., Chiu, H., Hsu, Y., 2008. Supplier market orientation and accommodation of
the customer in different relationship phases. Industrial Marketing Management
37, 380393.
Hsu, C., Kannan, V.R., Tan, K., Leong, G.K., 2008. Information sharing, buyersupplier
relationships, and rm performance: a multi-region analysis. International Journal of Physical Distribution & Logistics Management 38 (4), 296310.
Humphreys, P.K., Shiu, W.K., Chan, F.T.S., 2001. Collaborative buyersupplier relationships in Hong Kong manufacturing rms. Supply Chain Management: An
International Journal 6 (4), 152162.
Jack, E.P., Raturi, A., 2002. Sources of volume exibility and their impact on performance. Journal of Operations Management 20 (5), 519548.
Jap, S., Mohr, J.J., 2002. Leveraging Internet technologies in B2B relationships.
California Management Review 44 (4), 2438.
Johnson, P.F., Klassena, R.D., Leendersa, M.R., Awaysheha, A., 2007. Utilizing ebusiness technologies in supply chains: the impact of rm characteristics and
teams. Journal of Operations Management 25 (6), 12551274.
Jordan, W.C., Graves, S.C., 1995. Principles on the benets of manufacturing: process
exibility. Management Science 41 (4), 577594.
Joreskog, K.G., Sorbom, D., 1989. LISREL 7: A Guide to the Program and Application.
SPSS Inc., Chicago.
Kekre, S., Srinivasan, K., 1990. Broader product line: a necessity to achieve success?
Management Science 36 (10), 12161231.

520

S. Devaraj et al. / Journal of Operations Management 30 (2012) 509520

Kim, K.K., Umanath, N.S., Kim, B.H., 2005. An assessment of electronic information
transfer in B2B supplychannel relationships. Journal of Management Systems
22 (3), 293320.
Kollock, P., 1994. The emergence of exchange structures: an experimental study
of uncertainty, commitment and trust. American Journal of Sociology 100,
313345.
Koste, L.L., Malhotra, M.K., 1999. A theoretical framework for analyzing the dimensions of manufacturing exibility. Journal of Operations Management 18 (1),
7593.
Lorenz, E., 1988. Neither friends nor strangers: informal networks of subcontracting
in French Industry. In: Gambetta, D. (Ed.), Trust: Making or Breaking Cooperative
Relations. Basil-Blackwell, New York, pp. 194210.
Marsh, H.W., 1994. Conrmatory factor analysis models of factorial invariance: a
multifaceted approach. Structural Equation Modeling 1 (1), 534.
Mason, S.J., Cole, M.H., Ulrey, B.T., Yan, L., 2002. Improving electronics manufacturing supply chain agility through outsourcing. International Journal of Physical
Distribution & Logistics Management 32 (7), 610620.
Maruyama, G.M., 1998. Basics of Structural Equation Modeling. Sage Publishers,
Thousand Oaks, CA.
Mentzer, J.T., Flint, D.J., Hult, G.T.M., 2001. Logistics service quality as a segmentcustomized process. Journal of Marketing 65 (4), 82104.
Mentzer, J.T., Rutner, S.M., Matsuno, K., 1997. Application of the means-end value
hierarchy model of understanding logistics service quality. International Journal
of Physical Distribution and Logistics Management 27 (9/10), 230243.
Muffato, M., Payaro, A., 2004. Implementation of e-procurement and e-fulllment
processes: a comparison of cases in the motorcycle industry. International Journal of Production Economics 89, 339351.
Nunnally, J.C., Bernstein, I.H., 1994. Psychometric Theory, 3rd ed. McGraw-Hill, New
York, NY.
Olson, J.R., Boyer, K.K., 2003. Factors inuencing the utilization of Internet purchasing in small organizations. Journal of Operations Management 21 (2), 225245.
Parasuraman, A., Zeithaml, V.A., Berry, L.L., 1988. SERVQUALA multiple-item scale
for measuring consumer perceptions of service quality. Journal of Retailing 64
(1), 1240.
Parkhe, A., 1993. Strategic alliance structuring: a game theoretic and transaction
cost examination of interrm cooperation. Academy of Management Journal
36, 794829.
Podsakoff, P.M., MacKenzie, S.B., Jeong-Yeon, L., Podsakoff, N.P., 2003. Common
method biases in behavioral research: a critical review of the literature and
recommended remedies. Journal of Applied Psychology 88 (5), 879903.
Podsakoff, P.M., Organ, D.W., 1986. Self reports in organizational research: problems
and prospects. Journal of Management 12 (4), 531544.
Poppo, L., Zenger, T., 1998. Testing alternative theories of the rm: transaction cost, knowledge-based, and measurement explanations for make-orbuy decisions in information services. Strategic Management Journal 19,
853877.
Premkumar, G., Ramamurthy, K., 1995. The role of interorganizational and organizational factors on the decision mode for the adoption of interorganizational
systems. Decision Sciences 26 (3), 303336.
Reichhart, A., Holweg, M., 2007. Creating the customer-responsive supply chain:
a reconciliation of concepts. International Journal of Operations & Production
Management 11 (27), 11441172.
Rosenzweig, E.D., Roth, A.V., 2007. B2B seller competence: construct development
and measurement using a supply chain strategy lens. Journal of Operations
Management 25, 13111331.
Rosenzweig, E., Roth, A.V., Dean, J.W., 2003. The inuence of an integration strategy
on competitive capabilities and business performance: an exploratory study of

consumer products manufacturers. Journal of Operations Management 21 (4),

437456.
Sahin, F., Robinson, E.P., 2005. Information sharing and coordination in make-toorder supply chains. Journal of Operations Management 23, 579598.
Sahin, F., Robinson, F.P., 2002. Flow coordination and information sharing in supply chains: Review, implications, and directions for future research. Decision
Sciences 33 (4), 505536.
Sambamurthy, V., Bharadwaj, A., Grover, V., 2003. Shaping agility through digital
options: reconceptualizing the role of information technology in contemporary
rms. MIS Quarterly 27 (2), 237263.
Sanders, N., 2007. An empirical study of the impact of e-business technologies on
organizational collaboration and performance. Journal of Operations Management 25 (6), 13321347.
Salvador, F., Rungtusanatham, M., Forza, C., Trentin, A., 2007. Mix exibility
and volume exibility in a build-to-order environment: synergies and tradeoffs. International Journal of Operations & Production Management 11 (27),
11731191.
Schmenner, R.W., Swink, M.L., 1998. On theory in operations management. Journal
of Operations Management 17 (1), 97113.
Seabright, M., Levinthal, M., Fichman, M., 1992. Role of individual attachments in
the dissolution of interorganizational relationships. Academy of Management
Journal 35, 122160.
Sia, S.W., Kjoh, C., Tan, C.X., 2008. Strategic maneuvers for outsourcing exibility:
an empirical assessment. Decision Sciences 39 (3), 407443.
Suarez, F.F., Cusumano, M.A., Fine, C.H., 1996. An empirical study of manufacturing exibility in printed circuit board assembly. Operations Research 44 (1),
223240.
Subramani, M., 2004. How do suppliers benet from information technology use in
supply chain relationships? MIS Quarterly 28 (1), 4573.
Swafford, P.M., Ghosh, S., Murthy, N., 2006. The antecedents of supply chain agility of
a rm: scale development and model testing. Journal of Operations Management
24, 170188.
Talluri, S., Sarkis, J., 2002. A model for performance monitoring of suppliers. International Journal of Production Research 40 (16), 42574269.
Tanaka, J.S., 1993. Multifaceted conceptions of t in structural equation models. In:
Bollen, K.A., Long, J.S. (Eds.), Testing Structural Equation Models. Sage Publishers,
Newbury Park, CA.
Thomas, D.J., Grifn, P.M., 1996. Coordinated supply chain management. European
Journal of Operations Research 94 (1), 115.
Teo, T.S.H., Yu, Y., 2005. Online buying behavior: a transaction cost economics perspective. Omega 33, 451465.
Upton, D.M., 1997. Process range in manufacturing: an empirical study of exibility.
Management Science 43 (8), 10791092.
Vaidyanathan, G., Devaraj, S., 2008. The role of quality in e-procurement performance: an empirical examination. Journal of Operations Management 26,
407425.
Venkatraman, N., 1989. Strategic orientation of business enterprises: the construct,
dimensionality and measurement. Management Science 35 (8), 942962.
Vickery, S., Drge, C., Germain, R., 1999. The relationship between product customization and organizational structure. Journal of Operations Management 17
(4), 377391.
Weill, P., Broadbent, M., 1998. Leveraging the New Infrastructure: How Market Leaders Capitalize on Information. Harvard Business School Press, Boston,
MA.
Young-Ybarra, C., Wiersema, M., 1999. Strategic exibility in information technology alliances: the inuence of transaction cost economics and social exchange
theory. Organization Science 10 (4), 439459.