CHAPTER I

INTRODUCTION

1.1

Background of the Study

This chapter presents the current research in e-government development and online
reverse auctions system, the problem and goals of the study and the significance of the
research. An overview of Malaysias e-government initiatives and Multimedia Super
Corridor (MSC) applications, particularly ePerolehan is provided in the chapter together
with an overview of reverse auctions system and eBidding benefits. A description of the
status of eBidding usage and the transactional value are also included.

1.2

Global e-Government Use

e-Government facilitates governments to provide services to business, government

agencies and citizens by leveraging on information communication technology (ICT)
and the Internet. e-Government is referred to as public service delivery to the public,
citizens and private sector via the internet (Ahmad and Othman, 2007). e-Government
is defined as the use of Internet as venue for more efficient administration and
governance (OECD, 2003). Moon (2005) argues governments employ the system as a
strategy to response to public expectations for enhanced and better public service
delivery. The government is facing increasing expectations by the public for fast and
efficient services similar to the quality offered by the private sector. The citizens who

had experienced of the ease and flexibility of using online banking services want similar
experience in government delivery systems. As a response to public expectations, the
government has embarked on the development of e-government to improve service
delivery (MAMPU, 2010).

The implementation of e-government is expected to bring many advantages and

benefits to the government and the public. The governments aspirations to implement
e-government are to improve the internal operations by transforming the processes,
reduce paperwork, increase the number of services to the public and reduce the
response time (Maniam, 2008). As e-government improves, the interaction quality and
work processes within and inter-agencies, the results are in terms of speedier policy
development, better coordination and enforcement (Kaliannan et al., 2007). In addition,
e-Government provides opportunities for citizens and business communities to access
government services and enhance the transactions between business and governments
(UN, 2005).

The e-Government supports good governance, strengthens existing relationships and

builds new partnerships within civil society. e-Government facilitates the introduction of
new enhanced services to the public and increases citizen participation in the policy
making (Gupta and Jana, 2003). Siddiquee (2008) argues that by increasing
consultations and collaborations among various parties and agencies, the citizens can
enjoy better quality services and increased business efficiency.

The global governments effectiveness in delivering online services is tracked in a few

studies, for example, e-Government initiatives in 22 nations were examined in the
Accenture Study in 2011. The study finds that leading world governments are modifying
the traditional models of service delivery to a higher level to strengthen their
relationships with citizens (Accenture, 2011). e-Government relationships with the
stakeholders can be briefly categorized as Government-to-Citizens (G2C), Governmentto-Businesses (G2B) and Government-to-Government (G2G) (Jeong, 2007).

In a Government-to-Citizens (G2C) relationship, user-friendly one-stop services centers

are employed to facilitate citizens interactions with the government. In a Governmentto-Businesses (G2B) relationship, government and the private sector communication
are improved to facilitate business transactions between parties. In a Government-toGovernment (G2G) category, collaborations between governmental agencies are
enhanced to increase data sharing and electronic transactions. MAMPU (2010)
contends that in G2G, these relationships cover more integrated agencies in terms of
collaboration between officials, departments, ministries and foreign countries.

1.3. e-Government Development

e-Government evolves through

a series of stages, from basic information,

communication feedback to conducting transactions and finally the interactive web

presence (Siau and Long, 2009). e-Government development stages can be depicted in
stages of development model as shown as Figure 1. The United Nations Five-Stage

Model (UN, 2005) summarizes the developmental stages any government may
experience as the e-government evolves. The first three stages cover automation and
digitization while the following stages cover government transformation, in terms of the
internal operations and citizens participation in policy-formulation and decision-making.

Figure 1. UNs 5-Stage Model of e-Government Development

iv
iii
ii
i

Source : (UN, 2005)

Various government researchers have developed numerous stage models for egovernment developments, for example, Layne and Lee, (2001) Four-Stage Model and
Moon (2002) Five-Stage Model, Layne and Lees (2001) Four-Stage Model and Moons
(2002) Five-Stage Model. These models are based on a combination of technical,
organization and managerial factors. IBM Four-Stage Model comprises automate,
enhance, integrate and on demand stage (IBM Business Consulting Services, 2003).

Belanger and Hiller (2006) Five-Stage Model considers political participation by citizens
in the highest stage by offering services such as online voting, online registration, or
posting comments on line. Lee (2010) study compares twelve different e-government
stage models and classify e-government stage models based on two themes: operation
and technology theme on one side and citizen services on the other. According to the
author, five distinct correlated metaphors, presenting, assimilating, reforming, morphing
and e-governance will dictate the relationship between each themes. These models
embrace the concepts of interaction, transaction, participation and involvement of the
citizen with integration, transformation and process management.

Table 1. Comparisons of Stage Models in e-Government Development

Model

Stages

UN (2005) 5-Stage Model

i. emerging web presence,

ii. enhanced web presence,
iii. interactive web presence,
iv. transactional web presence;
v. seamless integrated web presence

Layne and Lees 4-Stage

Model
(2001)

i. catalog;
ii. transaction;
iii. vertical integration;
iv. horizontal integration

Moons 5-Stage Model

(2002)

i. simple information dissemination;

ii. two-way communication;
iii. service and technical transaction;
iv. vertical and horizontal integration ;
v. political participation

IBM Study (2003)

i. automate ;
ii. integrate ;
iii. enhance ;
iv. on demand
5

Model

Stages

Belanger and Hiller (2006)

i. web presence;
ii. interaction;
iii. transaction;
iv. transformation ;
v. political participation
i.presenting;
ii. assimilating;
iii. reforming;
iv.morphing ;
v. e-governance

(Lee, 2010)

As seen in Table 1, there are leaps between each of the stages. The development
stages of e-government follow several phases from access level that allows citizens
and business access to government information; interaction level, which allows
interactions with government through email or download forms; transaction level that
allows users to conduct transactions online; and integration level, which integrates all
services in different e-government organizations and governance. Based on the models
developed by the researchers, e-government development occurs in stages from
access level to transaction and integration level and finally e-governance. Governmentto-Business (G2B) systems are grouped in the integration stages of development due
to the capabilities and services offered to the business community and citizens.

1.4 Malaysias e-Government Initiatives

The Government of Malaysia envisions a modern government enabled by technology

through its Vision 2020. The Government launches an ambitious e-government project
to increase the quality and public service delivery by leveraging on the capabilities
offered by Internet and multimedia technologies (Hazman et al., 2006). The eGovernment aspires to improve the ease of use, access and interaction quality with the
public and private sector by improving the process within the government. These factors
will result in speedier policy development, coordination and enforcement. Ramlah et al.,
(2007) contend the governments overall aim is to build an enabling environment to
achieve a knowledge-based economy. As such, e-government programs are introduced
to offer better services and harness the potential benefits of interactions with new IS
innovations with an increasingly well-informed population.

Multimedia Super Corridor (MSC) Flagship Applications were launched with egovernment as one of the key applications to spearhead the ICT industry. eGovernment application was developed with the aims to re-invent the government and
improve citizens and business interactions with the government. The overall vision is to
create an efficient administration and excellent service delivery in an enabling
environment. The aim is an enhanced collaboration between government, businesses
and citizens for the benefit of the country. The e-government aims are to re-invent

government and leverage the industrial capabilities of Multimedia Super Corridor (MSC)
by addressing the following areas (Siddiquee, 2008):

1) Improving accessibility between the stakeholders and government, that is to

enable the public, business and foreign governments have better access to
government services;
2) Enhancing public services quality;
3) Improving the processes or system work flows that are important in the
betterment of the government service deliveries;
4) Creating greater transparency and governance; and
5) Empowering government officials at the implementation level towards faster
decision-making by reducing bureaucracy in daily transactions with the public.

The Government of Malaysia has also lined up several flagship e-government projects
aimed at transformation from manual and paper-based process, standalone agencies to
a seamless and integrated government. The initiatives launched under the eGovernment Flagships since 1997 are explained as in Table 2.

Table 2. Modules the e-Government

No.

Projects

Characteristics

1.

Generic Office
Environment (GOE)

Provides a platform for collaboration for enhanced

communication, interactions and sharing of data for
the agencies.

2.

ePerolehan (eP)

Connects the registered suppliers and the public

sector via the internet where agencies could source
for goods and services offered by the suppliers.

3.

Project Monitoring
System (PMS)

Offers the capability for projects monitoring and

implementation with built-in knowledge depository
and management functions.

4.

Human Resource
Management
Information System
(HRMIS)
Electronic Services
(e-Services)

Facilitates government employee day to day human

resource development functions in an online
platform.

6.

Electronic Labour
Exchange (ELX)

Connects the labor market data to the public sector,

private sector and the public.

7.

E-Syariah

Reforms administrative services of the 102 Syariah

courts to increase the capabilities of the Islamic
Affairs Departments in terms of monitoring and
coordination of its branches.

5.

Provides direct transactions and communications

capabilities between the citizen, the public sector
and the service providers on an online platform.

Source: (Maniam et al., 2006)

Off all the above-mentioned projects, the focus of this study will be on ePerolehan,
followed by eBidding which is a reverse auction system module under ePerolehan.

1.5 Overview of ePerolehan

The Government aspires to re-invent the way in which it operates to be more

responsive to the needs of the business community and to increase transparency and
efficiency in procurement process. As a major purchaser of goods and services from
the private sector, the government needs an efficient procurement service. For example
in

2010

alone,

total

government

spending

was

RM206.2

billion

(http://www.treasury.gov.my/bajet2011/data/le/le11.html, 2012).

ePerolehan is defined as the official electronic procurement system and provides a

secure end-to-end Internet transaction process from the buyer to the sellers (MAMPU,
2009). ePerolehan aims to re-engineer and automate the traditional way of manual
procurement system by transforming the

system into an online marketplace for

suppliers and government agencies. ePerolehan promises increased value for

government spending, as well as transparency and accountability in the procurement
and sourcing processes (Kaliannan et al., 2007). ePerolehan is an innovation by reinventing government procurement processes to ensure better collaboration between
public and government for higher quality and more efficient public services. Ahmad and
Othman (2007) argue ePerolehan is an inter-organizational system (IOS) that has been

10

successfully deployed in the government as a venue to improve the services to the

public.

ePerolehan was launched in several phases in 1999, and thus far the application has
successfully allowed the government and suppliers to electronically conduct
procurement activities. The system allows the suppliers to offer their services, advertise,
transact and receive the payments from the departments through the web. ePerolehan
supports the entire procurement chain from submission of tender and contracts to
approval and payment including alert notifications to potential bidders. The objectives of
ePerolehan are as follows (Maniam, 2008):

1) To offer best returns for Government procurement;

2) To facilitate the suppliers to receive faster payment;
3) To enhance accountability and transparency in public sector procurement; and
4) To improve cooperation between the private sector and the public sector.

Table 3 outlines the description of the modules under ePerolehan. The following
sections will focus on electronic reverse auctions and eBidding, which is one of the
modules under ePerolehan.

11

Table 3. Types of Module Services Under ePerolehan

No.

Module

Services

1. Supplier registration

Register new contractors or consultants, renewal,

online updates of latest supplier information

2. Central contract

Facilitates in the requisition services from acceptance

and delivery of order to confirmation of invoice of
purchased items by the agencies.

3. Direct purchase
Performs requisition processing that starts when
(procurement value up to procuring officials choose goods or services or
RM50,000.00)
products and the cycle is complete when the supplier
receives the purchase order.
4. Quotation system
Provides quotation process for procurement from
(procurement value up to RM100, 000 to RM200,000. Invites are forwarded to
RM200,000.00)
suppliers for their replies to the requests.
5. Tender
(more than
RM200,000.00)

Performs tendering process for procurement from

RM200, 000 or higher.

6. eBidding (from RM50,000 Provides services to qualified suppliers to bid online, in

and above)
a specified pre-determined period. eBidding allows the
bidders to see their status and other bidders offers
online.

Source: (Kassim et al., 2010)

12

1.6

Online Reverse Auctions System

An online reverse auctions system is an innovative and emerging procurement auctions

system in e-government services. Online reverse auctions are beginning to replace
traditional paper-based and manual transactions due to its popular benefits (Beall et al.,
2003). An online reverse auction is defined as an online real-time bidding event hosted
by a single buyer to outsource products and/or services (Emiliani and Stec, 2004).
Carter et al., (2004) define reverse auctions system as real-time online auctions
between a purchaser and multiple suppliers. In an auction, suppliers compete against
each other online using specialized software by submitting successively lower-priced
bids during a specified period.

Online reverse auctions system is expected to be widely embraced by global

governments due to its benefits and the success of its use in the private sector (Wyld,
2012). There are many factors why the international governments are beginning to use
the auctions system. This is due to the fact that it is more efficient to manage
transactions between public sector and their suppliers in online auctions environment.
For the government, reverse auctions offers many compelling benefits, in terms of price
reduction, cost and time savings and increased offerings from various types of
suppliers. These factors contribute to a more effective and transparent procurement
transactions. Looking from the perspective of the suppliers, they can expand their

13

market reach to increase market penetration and decrease transaction costs by using
the electronic channel, (Geyskens et al., 2002).

In a public sector reverse auctions market, public procuring agency enjoys lower prices
when pre-qualified bidders compete with other suppliers in successive lowering of bids
to win the contract. Jap (2007) contends that studies show buyers can save from 5
percent to 40 percent (with an average of 20 percent) on the cost of goods and services
by allowing multiple bids in comparison to typical traditional procurement currently used
in government transactions. Beall et al., (2003) argue that governments find the
auctions attractive from the benefit of price reductions, price visibility and reduced
transaction cost.

1.7 Overview of Malaysias eBidding

eBidding is government online procurement reverse auctions system under ePerolehan.

eBidding allows the sellers to advertise and offer their goods and services to the public
sector by connecting the sellers with buyers communities via a business model that has
been successfully implemented under ePerolehan (Pekeliling Perbendaharaan, 2009).
eBidding adopts the reverse auctions concept unlike forward auctions or English
auctions (an ascending price auction). eBidding was introduced in 2006 under the
Ministry of Finance (MoF) to realize the vision to innovate and increase efficiency in
public sector procurements. To date, all Ministries have utilized eBidding for procuring
supplies, products and services. Each agency can procure goods and services provided

14

by registered suppliers via eBidding. The procuring authority for eBidding is under the
responsibility of respective Secretary Generals of the Ministries and Heads of
Departments of the agencies and monitored by MAMPU and ePerolehan Unit.

eBidding is one of the ePerolehan modules that allow an option for the procuring
agencies, either to employ conventional procurement process or to use eBidding as the
venue to source for goods and services. eBidding is suited for procuring agencies which
are sourcing for goods and services with costs of RM50,000 and above without the
need for complex technical evaluations. Public sector procuring officials will have the
option, either to source the goods and services via tender evaluation process or
eBidding, which provide the benefits of shorter procurement process, time-saving and
lower prices. (Pekeliling Perbendaharaan, 2009)

1.7.1 The Supplier Community

Treasury registered suppliers who offer products, goods and services are the supplier
community. There are approximately 120,000 Treasury registered suppliers under four
classifications, namely the Central Contract, Direct Purchase, Quotation and Tender
Purchase (http://home.eperolehan.com.my/v2/index, 2012). They are responsible to
actively coordinate and participate in the procurement by working closely with the eprocurement division under the purview of the Treasury. It is the responsibility of the
suppliers to continuously update the information regarding the products and services
they are offering to the government.

15

1.7.2 The Buyer Community

The buyer community is the Malaysian public sector. The public sector is a major
purchaser of goods and services from the suppliers spending a total of RM206.2 billion
in 2010 alone, (http://www.treasury.gov.my/bajet2011). The manual transaction and
management cost is high with ever increasing demands from the public for a more
effective and transparent procurement transaction. The buyer should have all the
necessary details in place in order to transact and respond to the suppliers bids in an
online procurement (http://home.eperolehan.com.my/v2/index, 2012).

1.7.3 The Service Provider

In general, the Ministry of Finance particularly the ePerolehan Unit and eBidding module
provide the services, in terms of monitoring, maintenance, data operation center, user
training and procedure implementation. The modules of the system are supplier
registration, central contract, direct purchase, quotation system, tender system and
eBidding (http://home.eperolehan.com.my/v2/index, 2012).

16

1.8 eBidding Processes

The eBidding procurement process is geared towards procuring services and supplies
with pre-determined scope and clear specifications from participating bidders and there
should be not less than five suppliers which will be screened by pre-qualification panel
(Pekeliling Perbendaharaan, 2009). There are four key criteria in order to be eligible for
procurement of goods and services under eBidding as follows:-

1)

There must be standard and clear specifications (ready for delivery) for the list of
goods and services, which do not require complex technical evaluations or
screenings;

2)

The goods and services are managed under the Treasurys approved panel
system;

3)

Bids for goods and services costing from RM50,000 and above is allowed subject
to approval by the Controlling Officer of the Ministries (namely, the Secretary
General or Heads of Departments);

4)

Agencies must ensure there are at least five (5) bidders participating before the
pre-qualification selection is undertaken. After the pre-qualification selection, as
pre-requisite at least 3 bidders are required to participate in eBidding; and

5)

If there are less than 3 bidders, the agencies must revert to tender evaluation
process again (Pekeliling Perbendaharaan (SPP) Bil. 3, February 2009).

17

Figure 2. eBidding Auction Process Flow Adapted From ePerolehan Malaysia,

(2009)

Source : (Pekeliling Perbendaharaan, 2009)

The eBidding process comprises four key steps namely pre-qualification, online bidding
in 14 days, selection and appointment of supplier and contract administration. As
illustrated in Figure 2, the public sector sourcing officials in the procuring agencies will
advertise the bidding specifications in an online and printed media. This is followed by
processing and evaluating received applications by pre-qualification committee;
conducting online bidding process via eBidding and finally selecting the lowest bidder
and issuing appointment letter to the winner. Following this, contract management is
executed with delivery of services and making payment to the appointed supplier.

18

1.9 Benefits of eBidding

The eBidding creates an opportunity for suppliers to sell goods or services directly to
the government procuring agencies without third party or middle-men. The supplier
community consists of 110,000 registered suppliers with the Treasury. The government
spent RM20.5 billion on procurement via ePerolehan in 2006 and has increased the
amount to RM23.2 billion in for 2007 (Ucapan Bajet 2009, 2009). eBidding has
contributed to government savings by reducing government expenditure with savings of
between 20 per cent and up to 40 per cent of the actual allocated budget ( www.
tender2u.blogspot.com, 2011).

The compelling factors for the government to employ reverse auctions are efficiency in
price negotiation and cost savings. The online competitive bidding drives the pricing
lower which enables the public sector to reduce costs through efficient procurement
(Gershon, 2004). The supplier community benefits in terms of higher accessibility to
many procurement offers. eBidding makes it easy for new supplier to be introduced into
e-commerce by giving them an access point for e-commerce transactions. Marketing
and offering goods and services become much easier and quicker providing a much
broader offerings and varieties to the buyer. In addition, by simplifying the work flows
and paper-based work operational costs will be decreased. Maniam (2008) contends
that eBidding benefits the buyer due to the highly automated process, lower operational
costs and faster turnaround time.

19

Suppliers benefit from more accurate information, reduced transaction and operational
cost with online submission of offerings and tender information. The eBidding will allow
faster payments to the sellers as soon as the transaction is concluded via an online and
secured network. The eBidding also allows the suppliers to employ a single electronic
catalogue for all government ministries and thus, the suppliers will be able to extend
their offerings to a much wider market at the same time (www.home.ePerolehan.gov.my,
2011).

In terms of social benefits, implementation of eBidding promises socio-economic gains

such as, more effective and user-friendly public sector through efficient processing of
information

and

knowledge. These

values will

contribute

towards enhanced

competitiveness leading to higher economic growth of the country and promote the
image of the country as having a proactive governance, rather than reactive governance
(MAMPU, 2010). eBidding was implemented based on the public demand for increased
transparency in government procurement. By using the system, the government aspires
to combat corrupt practices, remove the middle-man in negotiations and promote
transparency in procurement transactions. The bidding is transparent as the reverse
auctions hide the bidders from the buyer and allows real-time pricing (Jap, 2007). The
bidders will know instantly the status his bids in relation to other prices in the
negotiation. Reverse auction is ideally suited for transactions between procuring officials
and their suppliers, since it enables temporal and geographical conveniences, reduce
administration costs and provide prompt responses (Beall et al., 2003). However,
reverse auctions do not allow the expression of non-price attributes such as quality,

20

service and reliability, thus hindering collaboration in buyer-supplier relations (Emiliani

and Stec, 2004). As such, one of the possible reasons of low usage is that the sourcing
officials feel they are no longer contributing to the bidding process as negotiator and
service provider while losing the supplier-buyer relationships normally present in manual
transactions.

1.10 eBidding Adoption Issues

It has been six years since eBidding was rolled-out in 2006 and based on the statistics
from MOFs ePerolehan team project report, the eBidding system adoption is
experiencing a low adoption problem (www.home.ePerolehan.gov.my, 2010). There are
only 6,000 suppliers who are active users out of the approximately 50,000 suppliers that
are capable of conducting ePerolehan transactions (The Star, 2 April 2010). Until April
2012, there are a total of 419,688 transacting procurement units involving all ministries
and agencies in the Peninsular of Malaysia. However the number of transacting
procurement units via eBidding is only 315 transactions out of 419,688 transactions of
procurement units. The balance amount was transacted using the manual system
(www.home.ePerolehan.gov.my, 2012).

The governments approach to address the problem of low adoption of the system are
by using infrastructure and system perspectives (increasing Internet band-with, updating
system and hardware requirements), rather than looking at user acceptance of the
system. The ePerolehan support team only addresses issues related to training,

21

enablement of government responsible centers, site readiness, user complaints and

revenue generation from the use of the system. The team does not examine the factors
that influence eBidding adoption. There has been no empirical study undertaken by the
ePerolehan Project Team or Commerce Dot Com Sdn. Bhd. (the concessionary) on
eBidding to address the issue of low adoption by the government users. Implementation
issues were previously mitigated by coordination efforts by the concessionary company
and respective agencies monitored by MAMPU.

The non-adoption of eBidding by government users and their suppliers pose serious
implications for the continuous usage and development of the innovative system. The
suppliers could not participate in eBidding if the system is not chosen by the officials for
transactions. The low rate of adoption by the users indicates inherent officials problem
with eBidding rather than with the suppliers. There is an example where eBidding has
been pointed as the cause for failure in goods and services procurement in a
government agency, due to a mistake in the transaction caused by inexperienced and
untrained officials. As a result, the procurement was canceled and re-tendered via
manual transaction (Laporan Audit 2010, page 201). This example is one of the possible
causes why officials are not employing eBidding, preferring the alternative manual
procurement methods, although eBidding has been touted as the best tool for costsaving for the government.

The losses from the low adoption is in terms of the investment poured into the planning
and development of eBidding which would be going to waste if there is low acceptance

22

by the supplier community. It is a fact that huge investment has been spent on the
development, implementation and maintenance of the system and this problem would
defeat the governments aspirations of increasing efficiency, productivity and
transparency in its operations. The losses are also in terms of the cost to purchase and
develop the technology including the recurring costs to maintain and conduct the
reverse auctions. CommerceDotCom Sdn. Bhd. (CDCSB) has invested RM180 million
to develop the ePerolehan system including the eBidding module. Due to the low usage
of the system, the contract between the government and CDCSB has been extended
until 2012. A high level of usage by suppliers is important as the business model is feebased and depends on the number of transactions and it is important for CDCSB to
maintain a high number of transactions for the continuity of their business.

Notwithstanding, the government also loses in terms of expected cost saving benefits
from eBidding use if the system is not adopted by the users. The Government would be
able to enjoy 20 to 40 percent savings from the operational budget (BERNAMA, 2009).
From the saving, the funding can be channeled to other beneficial development
projects. Furthermore, the sourcing officials would have to revert to conventional
procurement processes that require the tender documents to be downloaded, printed
and deliberated at tender meetings, which will consume more time and administrative
costs. These processes would have incurred higher operating costs which could be
channeled to other activities in delivering services to the public.

23

1.11 Problem Statement

The low usage reflected the possible inclination of the sourcing officials to procure
goods and services via conventional methods (direct purchase or manual quotation
system) rather than using eBidding. The use of eBidding by the government users is
voluntary and the sourcing officials have the options, either to conduct a manual
transaction or use eBidding (Pekeliling Perbendaharaan, 2009). The eBidding system
offers compelling benefits to improve administrative efficiency and quality of service
delivery. However, these benefits may be wasted by officials unwillingness to use the IS
for many reasons that is worth investigating.

The focus of previous e-government studies has always been on the IS adoption by
suppliers rather from the sourcing officials perspective. George (2007) argues despite
increased trainings and promotions, low e-government adoption remains a continuous
problem for the e-government stakeholders. The system is designed to lessen officials
burden and improve efficiency in public service, as such the questions regarding
government users adoption of eBidding still remain unanswered. Questions are still
being asked by the stakeholders whether the officials are using eBidding and reasons
the system is not being used. It is unclear how far is the eBidding system being used by
the officials and what are the users perspectives about the information system. As
such, the research aims are to investigate the constructs that influence officials
adoption of eBidding.

24

1.12

Objectives of the Study

The objectives of this study are to examine the factors influencing eBidding adoption by
procuring officials in the public sector and to understand the users behavior in adopting
eBidding. The specific purposes of this study are as follows:

1. To identify the variables that may influence the adoption of eBidding by government
users;
2. To examine the effects of the variables on the adoption of eBidding by government
users;
3. To examine if some of the variables have moderating or mediating effects on the
relationships established as stated in objective 2; and
4. To propose a framework to analyze the adoption of eBidding by government users.

25

1.13 Practical Implications

The study is expected to contribute towards improving policy implementations for policy
makers and practitioners and helps to improve eBidding adoption by public sector
procurement officials. This is accomplished by providing a deeper understanding on the
critical factors that leads to behavior to adopt the system. The benefits of the reverse
auctions system would only be realized if the government officials employ it, thus
ensuring future implementation of the system. For example, MAMPU could leverage on
the significance of system performance and facilitating conditions of eBidding to
encourage its usage. This knowledge can help the Government improve eBidding
based on the users feedback and the lessons learned from this study could facilitate
sourcing officials to introduce appropriate measures to mitigate the problems of low
adoption of eBidding

1.14 Theoretical Contributions

From the perspective of theoretical contribution, the study will include constructs from
two research streams (the literature on innovation diffusion and information system
success) which will add to the comprehensiveness of the proposed research model.
From the literature, area of behavioral acceptance of IS has been well investigated. eGovernment adoption research has been less than adequate because earlier IS
acceptance studies have mainly focused on business-for-profit organizations. Murali et

26

al., (2007) contend these studies have not addressed the user acceptance of
technology in government and public agencies which are different in terms of market
exposure, legal and social perspectives. The proposed research model can be used as
an initial investigation for other e-government acceptance studies. An inclusive
framework has the potential to more fully understand the impact of IS on e-government
participation (Tolbert and McNeal, 2003). It is anticipated that the findings of the study
would contribute towards building the foundation for future studies in e-government IOS
adoption.

Reverse auctions has been widely studied in the private sector settings and some
authors now argue that it should be able to produce similar results for the government.
The benefit of reverse auctions to the government is by improving the contribution of egovernment to organizational efficiency and effectiveness (Jap, 2007). The research
would also contribute to the literature on public sector reverse auction systems by
providing insight into the value of the system for the users. The following section on
online auctions in Malaysia will illustrate that previous studies on auctions deployment in
Malaysia are qualitative in nature and focus on factors that motivates suppliers to adopt
auctions rather from the government users perspectives. As the study will also
introduce new variables in a quantitative study on eBidding, the study might shed some
light into how user acceptance determinants from various theoretical background and
models might affect the acceptance of eBidding.

27

1.15 Organization of the Study

The study comprises four chapters. The first chapter introduces the study by presenting
the context of current research in e-government development and online reverse
auctions system, the problem and goals of the study, the relevance and significance of
the research. An overview of Malaysias e-government initiatives and a preview of
ePerolehan is provided in the chapter. In addition, an overview of the reverse auction
system concept, eBidding benefits to the stakeholders are provided.

The second chapter examines the empirical literature of the e-government G2B
services, and literature on the individuals' technology acceptance in a working
environment. The empirical studies and framework related to factors that influence
technology adoption is also examined including the comparisons of the theories related
to the technological adoption. Published literatures on public sector e-procurement and
auctions adoption are also discussed.

The third chapter of the study describes the theoretical framework and hypotheses
development. The chapter attempts to propose an empirical model of technology
adoption in an EG context along with the proposed hypotheses. The chapter provides
an overview of the related theoretical framework for the development of the conceptual
model to explain the factors influencing the behavioural adoption of eBidding.

28

Chapter four outlines the research methodology employed in the study. The chapter
explains the design and method of study in areas of population of study, sampling,
sample size, research instruments, and measurements of variables, validity and
reliability tests, data collection and data analysis for hypotheses testing.

Chapter five presents discussions on the data analysis which include the feel for data,
testing the goodness of data, testing the hypothesis developed for the research and the
model fit. In chapter six, discussions and conclusions are made on the research findings
with some recommendations. Research issues and objectives are examined and
analyzed including the results, outcomes and discussion of findings. In addition, study
limitations, theoretical and practical contributions are discussed in the chapter.

29

CHAPTER 2

LITERATURE REVIEW
2.1 Introduction

This chapter will review individuals' technology acceptance literature including the
theoretical framework and empirical literature concerning the factors that influence
technology adoption of information system (IS). Published literature on public sector eprocurement and reverse auctions adoption are also discussed. The chapter also
explains the development of the study's theoretical framework by reviewing the overall
trends of e-government development and technology adoption. This will help readers to
have a greater understanding on the processes of technology adoption within a
government setting.

2.2 Past Research on Technologies Related to e-Procurement Adoption

Many researches on e-procurement adoption have been undertaken which covers a

wide variety of research areas (Vaidya et al., 2006). Currently, many studies attempt to
explain the impacts of adopting an e-procurement on public sector agencies activities in
terms of increasing efficiency and effectiveness of their procurement activities
(Chadwick and May 2003; West, 2004).

30

The success factors influencing adoption of public sector e-procurement tools have
been studied by Moon (2005). The author investigates the diffusion of state
procurement tools, namely procurement systems (purchasing and automated procuring
systems),

internet

bidding,

reverse

auctions

and

e-procurement

institutional

infrastructure (digital signatures and establishment of statutes for online bidding). Moon
(2005) argues that procurement unit empowerment and managerial innovation are
important factors for an e-procurement adoption among state officials. Vaidya et al.,
(2006) identify several success factors for e-procurement initiatives. According to the
authors, active suppliers involvement, business process re-engineering and integrated
system are key factors affecting the adoption of the system. Gunasekaran and Ngai
(2008) argue that accountability, IS specialists, streamlined work-flow systems and top
management support are key antecedents for the auctions system adoption.

There are several studies in Malaysia that examine the factors influencing ePerolehan
adoption and usage by the public sector (George, 2007; Murali et al., 2007). Murali et
al., (2007) argue e-procurement is a well investigated area of research. However, they
contend there is not a single generic model developed for e-procurement adoption by
the public and business organizations. Thus, the focus in this research is to offer insight
into what compels e-government adoption of reverse auctions system, by the
government users themselves, as well as to understand what constitutes acceptable
system and service quality.

31

2.3 Public Sector Reverse Auction

Online reverse auction is defined as a type of an e-auction that is conducted in real-time

between a single buying organization and pre-qualified suppliers in an online
environment (Jap, 2007). In a public sector reverse auctions setting, the suppliers
compete in a bidding process to the government agency for the supply of goods or
services with clearly defined specifications. Public sector reverse auctions research is at
its infancy (Hawkins et al., 2009). Their empirical study on reverse auctions have
contributed to the understanding of the antecedents (motivations to use) public sector
sourcing professionals. Reverse auctions is an innovative sourcing strategy in the area
of supply chain management (SCM) and as a strategic sourcing tool, it is employed to
facilitate firms to enhance the organizations competitive advantage (Emiliani and Stec,
2004). Hawkins et al., (2009) define sourcing strategy as a holistic plan that determines
purchasing strategy includes evaluations of antecedents, such as specifiability,
competitiveness and selection criteria. Tully (2000) contends reverse auctions provides
an opportunity for organizations to achieve competitive advantage by reducing the price
and the transaction costs in transactions for goods and services.

In the case of Malaysia, the governments reverse auctions system, the eBidding was
launched in 2006 as a module under ePerolehan. It is an innovative online procurement
service with the promise of increasing procurement efficiency and transparency and
saving time and cost for both procuring managers and suppliers (Unit ePerolehan,
2009). eBidding is an online reverse auction with a single buyer (the Government

32

agency) and a group of pre-selected sellers in an online bidding for the supply of goods,
services or products. As a module, it is an innovative feature of the ePerolehan system
that allows an option for the procuring agencies either to employ conventional
procurement process or to use reverse auctions as the venue to source for goods and
services (Pekeliling Perbendaharaan Bil 3., 2009). As such, the Government believes
that reverse auctions should appeal to suppliers due to the benefits offered by the
system. However, many suppliers are reluctant to adopt the innovation as shown in the
case of eBidding.

The supply chain management and procurement literature have covered many cases
documenting how leading global firms employed reverse auctions to increase
competitiveness, cost efficiency and manage their supply chain (Wyld, 2011). Research
on reverse auctions in recent years has focused on the antecedent factors of online
auctions adoption in private sector and business organizations at organizational level
(Radhakrishnan and Davis, 2004). The authors contend that studies on government
reverse auctions adoption have shown low percentage adoption by users and suppliers.
The following section will show the trends of Malaysian online auctions.

33

2.4 Online Auctions Adoption in Malaysia

There is little empirical research examining the antecedents of reverse auctions use by
the public sector (Jap, 2007) and a similar trend is shown in the case of Malaysia. An
empirical study on the factors influencing the adoption of reverse auctions in Malaysia
will enhance the boundary of knowledge in terms of broadening the understanding on
the use of reverse auctions as sourcing strategy in Malaysia.

Previous studies on online auctions are as follows; Jamaludin et al., (2011);

Kamarulzaman and Rahman, (2009) ; Settoon and Wyld, (2003) ; Mukhtar et al.,
(2009);Eu-Gene, (2010) ; P. Antony, (2005) ; Anthony, P (2009);Anthony P., and Law E.
(2012);Gan, K.S. et al., (2009) ; Abdullah, R. and Zahadi E. D.(2011) ; Akmaliah A. K,
(2007); Abdulhadi, (2013) ; Haque et al., (2010) ; Jailani N.et al., (2008) and Tan M.H.,
(2001).

A qualitative study on the key factors of e-auctions use in the Malaysian agricultural
sector has identified the advantages and drawbacks of e-auction use in Malaysia
(Kamarulzaman and Rahman, 2009). The authors argue that e-auction use in the
agricultural sector could reduce purchasing costs of materials and assists suppliers in
negotiating prices and improving competitive bidding. However, the inability of suppliers
to understand detailed understanding of costing structure and prices potentially harm

34

supplier-buyer relationship and sow distrust. The authors also contend that specifiability
and detailed description of the services are important in determining auction success.

A qualitative study on top auctions sites, namely lelong.com.my and eBay was
undertaken by Eu-Gene, (2010). The author proposes four critical success factors in
determining successful adoption of e-auctions, namely security, costs, market volume
and website performance. Another empirical study by Abdulhadi, (2013) on eBay
Malaysia evaluates the customer satisfaction by focusing on the influence of the various
factors of satisfaction among the Malaysian eBay users. The study proposes a model of
framework for the research study on online customer satisfaction. The factors
investigated are merchandise, transaction, response time, security/privacy, payment
and delivery with response time found not to be significant in the study.

In a case study by Akmaliah A.K. (2007) on Lelong.com.my the author sets out to
investigate entrepreneurship and innovation factors in the development of new online
based business. The study outlines the history from its infancy to the successful online
auction market in Malaysia. The author contends that service innovation is derived from
collaborations among technology, organisation, managerial and market ecosystem by
focusing on customer loyalty and satisfaction.

An integrated e-marketplace framework in Malaysian wood based products industry was

proposed by Mukhtar et al., (2009). The authors propose a development of an emarketplace which allows online auctions and purchasing with an interface to employ

35

short-messaging-services in the e-negotiation transactions. A case study on online

auctions found many online auctions are not Syariah-compliant (Jamalludin et al.,
2011). The study proposes a Syariah-compliant electronic auction that can potentially be
the model for Syariah-compliant online auctions for commercial and public use.

In a case study by Abdullah, R., and Zahadi E. D. (2011), they propose a framework of
an online bidding system for registration numbers of new registered vehicle. The
framework is designed to help users to obtain a number plate through an open bidding
process and bidding results notifications. The framework comprises invitations to bid
functions based on highest price and on-time bid placements. Jailani et al., (2008)
propose a framework for modelling an auditable e-marketplace for mobile users that
provide mobile users with simple, transparent and unobtrusive user interface for bidding
in market auctions. These factors include capacity for matching buyers and sellers,
facilitating transactions, adequate institutional infrastructure, secure and auditable emarketplace, secure agent-based marketplace for accessibility and standardized
protocols.

Hashim M. et al., (2008) evaluate the e-marketplace in Malaysia. A content analysis of

e-marketplace websites including, the market structure, B2B features and B2B stages of
growth. e-Market makers were interviewed and comprehensive factors of strategic,
functional, technology, build liquidity, create value and others are discussed to analyse
the success of the e-marketplaces. Based on the content analyses, a marketplace
framework was proposed based on Extensible Mark-up Language (XML).

36

An empirical study by Tan M.H., (2011) on the development of an e-procurement auction

system was undertaken for small medium enterprise (SMEs) in Malaysia. The author
proposes an online auctions model that caters to business-to-business auctions using a
Unified Modeling Language (UML) as a modelling tool. The study also provides an initial
insight into how the online auction marketplace is used and developed.

In an exploratory study by Antony P. (2005) on an English Auction (Robosar), the author

investigates sellers strategy for online auctions for Malaysia and global marketplace.
He proposes a design of a seller agent that recommends a reservation price of a given
item to be auctioned off by the seller that comprised successful failed-sell period,
competitors and profit functions. In another empirical study by Anthony P. (2009) on
electronic auctions, the author outlines types of auctions benefits to consumers. The
author also proposes bidding strategies for bidders and sellers to maximize participation
in electronic bidding including employing buy-it-now, sniping tactics and participation in
multiple auctions. For sellers, the author suggest they employ shilling tactic by
participating in the bidding to push prices downward and by offering reasonable price
consistent with bidders requirements.

In a quantitative study on online auctions, Gan, K.S. et al., (2009) conduct an empirical
evaluation of self-adaptive genetic algorithm for enhanced online auction strategies. The
authors argue the system produce higher success rate and increase in revenue in
auctions marketplace. In Anthony P. and Law E. (2012) empirical work on auctions
strategy, they analyse the performance of the selling agents with varying pricing

37

strategy when offering item for auctions. The study shows that the use of these
strategies produced significant drawbacks with negative impacts towards the overall
selling upshots. To counteract these shortcomings, they developed an autonomous
seller agent using a heuristic decision making framework. In Haque et al., (2010)
empirical work on auctions modelling, the authors find that the proposed algorithm can
be employed to examine economic models autonomously. The agent-oriented model is
developed by comparing the agent-oriented model with traditional auction model which
is proven profitable for the users.

A study on the potential impacts of Government-to-Business (G2B) reverse auctions

was undertaken in Malaysia and four other regional countries (Settoon and Wyld, 2003).
The authors contend that the e-sourcing strategies could yield significant benefits for
governmental budget in terms of immediate cost savings, time and temporal efficiencies
and deficit reduction. They conclude that the governments could improve efficiency in
state expenditure, reduce subsidies and reliance on external financing by adopting
electronic reverse auctions.

In sum, the research trends in Malaysia are predominantly studies that can be
described as descriptive and case studies that focused on private sector and business
areas with analysis at organizational level. The trend also indicates that online reverse
auctions in Malaysia are a very recent and an emerging area of research that warrant
more investigations especially auctions adoption in public sector domain.

38

2.5 Overview of Established Theories on User Acceptance

Major theories of technology adoption have been proposed by studies on computer

adoption within an organizational context, where the purpose is to improve employees'
acceptance of IT in work place and increase their performance (Venkatesh and Davis,
2000). e-Government technology adoption research has been mostly concerned with
the adoption of online government services using web technologies (Murali et al., 2007).
The officials perception and satisfaction with the government system must be taken into
consideration to identify the antecedents affecting the employment of new technologies
in an e-government setting.

The early technology acceptance theories emerge from the disciplines of psychology
and sociology, for example the Social Cognitive Theory that focused on knowledge
acquisition process through observation (Bandura, 1986, cited in Venkatesh et al.,
2003). Theory of Reasoned Action (TRA) and Theory of Planned Behavior (TPB) are
early acceptance theories which are the basis of newer adoption models. TPB is the
extension of TRA by adding perceived behavior control constructs (Matthew et al.,
2008). In 1991, the Model of Personal Computer Utilization (MPCU) was proposed as
an alternative to TRA and TPB while Motivational Model (MM) was developed to help
identify the factors or antecedents of the acceptance behaviour of the Internet (Davis,
1992). However, these underlying theories will not be discussed because they are no
longer considered up-to-date with the current technology adoption literature.

39

The current research on information system (IS) adoption and use has been
continuously updated and several theories have been proposed to explain users
intention to use the IS. Oliveira et al., (2011) contend these models originate from
integration of various research streams such as psychology, sociology and information
systems. Rosen (2004) argues that a number of research have been undertaken with
various frameworks to examine the adoption scenario of IS applications in the market,
such as e-procurement and internet banking. The studies successfully identify the
factors motivating the adoption of technology by the consumers. There are also
numerous studies which investigated technology adoption in working environment at
individual level of analysis. For example, TAM (Davis, 1989), DOI (Rogers, 2003) and
the UTAUT (Venkatesh et al., 2003) successfully identify the antecedents which
influence user intention and actual adoption of an IS. Literature review on individual
usage of new technology in a working environment will also be outlined.

2.5.1 Technology Acceptance Model (TAM)

The model is widely used in the study of individual's adoption of technology. TAM
focuses on individual perceptions about technology use. TAM argues that individuals'
beliefs and perceptions about technology use form their attitudes toward the technology,
and those attitudes determine their intention to adopt or not to adopt an innovation. TAM
as shown in Figure 3, has been commended by various researchers on its applications
in understanding user behavior. Introduced in 1989, TAM is still being employed in
various IS adoption studies (Jeyaraj et al., 2006). The model is used to examine factors

40

influencing individual adoption of a technology. Han (2003) argues that since its
introduction, TAM is empirically employed in various IS innovation studies from
communication technologies, database systems to Internet-based services (for
example, online libraries, online services and virtual workstations). As depicted in Figure
3, when users are introduced with an innovation, a host of factors will influence their
behaviour in relation to the way they will employ the innovation. There are two attributes
under TAM, namely perceived usefulness (PU) and perceived ease of use (PEOU).

a. Perceived usefulness (PU) is referred to as whether the users job performance

will be enhanced when the system is employed; and
b. Perceived ease of use (PEOU) is referred to as whether the use of the IS will be
free from effort by the user (Davis, 1989).

Figure 3. Technology Acceptance Model

Source: Davis et al., (1989)

41

TAM has been tested in various empirical studies with different research settings that
results in consistency and reliability, which confirms the validity of the theory. Oliviera et
al., (2011) contend that TAM is a useful theory to examine the motivations influencing
the adoption of new technology in a working environment.

The TAM model has been extended as the Technology Adoption Model 2 (TAM2) with
additional constructs to measure several new social influence dimensions. The new
elements added to the theory include constructs such as, subjective norm, voluntary of
use and user experience (Venkatesh and Davis, 2000).

The subjective norm refers to the influence of important individuals surrounding the user
whether he or she should perform the behaviour. Voluntariness is referred as the effect
of mandatory and voluntary usage of the technology. Image refers to how a technology
may influence the user status. Experience is defined as the subjective norm which
decrease or increase with increased system use by the user. These constructs help to
explain the Perceived Usefulness (PU) in TAM. However, TAM2 only explores the basis
of perceived usefulness (PU) and ignores perceived ease of use (PEOU) construct
which is generalisable. The model is extended to TAM3 which integrates TAM2 with PU
determinants to explain PEOU (Venkatesh and Bala, 2008). The additional factors to
TAM3 are computer self-efficacy, perception of external control, computer anxiety and
computer playfulness. Computer self-efficacy attribute refers to as an individual
perception whether the user could complete the task. Perception of external control is

42

user perception about the existence of support for the system. Computer anxiety is
defined as the fear related to using a new technology while computer playfulness is
defined as internal motivation to use a new technology. Perceived enjoyment describes
the feeling when using a system becomes satisfactory. Objective usability is the effort
needed to perform a task (Venkatesh and Bala, 2008).

Although the TAM3 is more comprehensive in that it increases the PEOU and PU, the
effects are focused only on the user and not in a much wider perspective. Broader
organisational issues, e.g. the influence of peers and user involvement in decisionmaking may also influence the adoption of a new technology. However, many studies
find that TAM is not comprehensive enough in explaining users decisions to use the IS.
TAM is used as a base model and additional variables are added according to the
conditions and nature of IS being investigated.

For example, in a study by

Kamarulzaman (2007), TAM was used as base model and adds other variables such as,
personal cognitive influence. TAM was combined with other adoption models as well, for
example, DOI was integrated with TAM in a study on online banking by Hernandez and
Mazzon (2007).

In summary, TAM has been successfully employed in various studies to explain

individual acceptance and usage behavior in a working environment. However, there is
a need to extend the model by adding additional variables depending on the types of
technologies and environment.

43

2.5.2 Diffusion of Innovations (DOI)

Innovation Diffusion Theory (Rogers, 1995) is another user acceptance theory which
has been praised by many researchers as an effective model in explaining user
behavior in technology acceptance. DOI theory as shown in Figure 4, focuses on the
individual characteristics that relate to technology adoption behavior. The DOI is a
widely used model in behavioral sciences to investigate adoption of innovations by
individuals (Oliveira et al., 2011). The aim of the DOI theory is to comprehend how and
why users either embrace or reject innovations (Rogers, 1995). In 1995, he studied the
characteristics of individuals in terms of openness to innovations, and he developed
DOI, which proposes that individuals react differently to change based on a stable
predisposition.

DOI focuses on diffusion of innovation process when the new technology is transmitted
through the channels in the social system (Rogers, 2003). Diffusion of innovations
refers to how new ideas are employed and employed by the users within a specific
setting. According to DOI, an innovation will be diffused at an increased rate if it could
be tested before adoption (trialability). In addition, an innovation offers observable
results (observability) and new technology has an advantage relative to other
innovations (relative advantage). Another innovation attributes include complexity (the
technology is not overly complex) and whether the innovation is compatible with existing
practices and values (compatibility). Fichman, (2000) defines innovation diffusion as the
process when a technology permeates and is adopted across a population.

44

Figure 4.Perceived Attributes Influencing Individual Adoption of Innovation

Perceived Attributes of Innovations

Source: Rogers, (2003)

As depicted in Figure 4, these attributes are empirically inter-related, however each

attributes are different from each other and these attributes are based on previous
literature. Description of the attributes is given below:

a).

Relative advantage refers to how an innovation is believed to be superior to

previous technology. Kendall et al., (2001) define relative advantage as benefits
such as, lower business costs, wider market coverage, preference to upgrade
other business ventures and importance of doing business on the internet in the
future.

45

b).

Compatibility is defined as how an innovation is in conformance with the culture

that will adopt it. Compatibility refers to how the innovation is in conformance
with the users values and requirements (Agarwal and Prasad, 1998).

c).

Complexity is defined as the difficulty for the users to learn to use an innovation
and how much effort needed to use the new technology. Complexity also refers
to the users perception on the difficulty of using the innovation and how an
innovation negatively affects the adoption rate (Murphy, 2003). In all of the
studies reviewed, if an innovation has high complexity, it is hypothesized to
have a negative impact upon the individuals attitude toward use (Kendall et al.,
2001).

d).

Trialability is defined as how an innovation can be experimented on a smaller

scale to examine the usefulness prior to acceptance (Rogers, 2003). A new
technology that is on a short trial is more receptive to the users than any
technology which is pushed upon them. Trialability also refers to the
demonstrability of a new technology, which can be tested on a limited period
priority. The purpose of triability is to learn how well it performs within a specific
setting (Murphy, 2003). Innovations that can be experimented before adoption
are adopted more rapidly than those that have no prior experience especially
among those who have adopted the innovation earlier.

46

e).

Observability is defined as how a user believe the expected results will be from
using the new technology. An innovation will be more easily adopted if the
intended users can clearly see the benefits of employing the technology (Denis
et al., 2002; Ovretveit, 2002). Efforts to increase the awareness of the benefits
of a technology via demonstrations or deliberate testing will lead to higher
possibility of the technology being accepted. Observability is also highly related
to innovation acceptance (Rogers, 2003)

f).

Rate of Adoption describes the expected behavioural outcome and user

inclination towards adopting a technology (Rogers, 1995).

Researchers argue that DOI serves as a fundamental theoretical base of innovation

adoption research in many disciplines, including sociology, communications, marketing,
education, and other areas of research (Oliveira et al., 2011). While Frambach and
Schillewaert, (2002) contend that DOI has been employed in various empirical studies
to investigate the adoption of innovations at organizational level of analysis.

In summary, the DOI is one of the most employed theories in various technology or
innovation-related studies due to its ability to understand organizational behaviour in
terms of technology usage (Braak, 2001).

47

2.5.3 Personal Innovativeness in the Domain of Information Technology (PIIT)

Personal Innovativeness in the Domain of Information Technology (PIIT) was

conceptualized by Agarwal and Prasad, (1998) as the willingness of an individual to try
any new information technology. Users with greater intrinsic innovativeness are posited
to have positive opinions about an innovation, which in turn will have more positive
intentions toward use. According to Individual Innovativeness Theory, innovators are
people who are risk takers interested in taking the initiative and time to try out
something new. Rogers, (1995) contends that individual innovativeness role in
innovation adoption is confirmed in numerous diffusion researches.

In a study on the Internet, PIIT is confirmed to act as mediator on the relationships

between compatibility and intention to use, including PEOU and PU of other web-based
technologies (Lewis et al., 2003). There is also significant associations between PIIT,
social influences and perceptions (usefulness and ease of use) that has an impact on
user intention in IS adoption (Lu et al., 2005). PIIT model as depicted in Figure 5, is an
important trait that influences individual-level efforts in trying to innovate with technology
(Magni et al., 2011).

48

Figure 5. Moderator Model of Personal Innovativeness in Information Technology

(PIIT)

Source : Agarwal and Prasad, (1998)

PIIT constructs are as follows:

a.

Usefulness is described as how easy it is to use the information system, which is

the same as performance expectancy;

b.

Ease of use refers to how the users believe the system will help them to improve
performance and it is similar to effort expectancy;

c.

Compatibility is defined as how a new technology is in conformance with the

norms and the requirements of users; and

d.

Behavioural intention refers to as individual willingness to put an effort for a given

behaviour.

49

PIIT is proposed to have mediator role on individuals belief on an innovation (Agarwal

and Prasad, 1998). PIIT would be the primary moderator for the antecedents as well as
the effects of the beliefs (for example, ease of use and compatibility), especially for
compatibility (Agarwal and Prasad, 1998). They proposed that PIIT has moderator
effects between ease of use and intention, however their research confirmed the
significant moderation effect only for compatibility.

Previous studies have proven the moderating effect of PIIT on the adoption of new web
services (Lee et al., 2007; Crespo et al., 2008). PIIT is a confirmed mediator on the
relationships between switching costs and e-store loyalty (Wu and Qi, 2010). A study on
mobile-RFID acceptance shows the moderating effect of PIIT between perceived need
and intention to use (Jeong et al., 2008).

In conclusion, PIIT is hypothesized to have mediator roles on the users beliefs and
perception about an innovative information system.

2.5.4 Unified Theory of Acceptance and Use of Technology (UTAUT)

Unified Theory of Acceptance and Use of Technology (UTAUT) provides a foundation for
research in individual acceptance of technology in a working environment (Venkatesh
et. al, 2003). The Unified Theory of Acceptance and Use of Technology (UTAUT) model
combines established models associated with technology adoption and usage among
others, Theory of Reasoned Action (TRA), Technology Adoption Model (TAM), the

50

Motivational Model and the Theory of Planned Behavior (TPB). UTAUT is frequently
employed in identifying the rate of adoption of IT, especially within large sample
populations (Anderson and Schwager, 2004). In their empirical study, the authors
validate the UTAUT and identified performance expectancy as key motivating factor for
the adoption of tablet device in institutes of higher learning. In their empirical study, Li
and Kishore (2006) also find significant effects of the construct (performance
expectancy and effort expectancy) on user adoption of an online community web. Other
studies which have applied the UTAUT are in mobile commerce adoption (Mallat, 2007)
and communication system (Wu et al., 2007).

Figure 6. Unified Theory of Acceptance and Use of Technology (UTAUT)

Source: Venkatesh et. al, (2003)

51

UTAUT Model (depicted in Figure 6) is formulated with four antecedent factors

(performance expectancy, effort expectancy, social influence and facilitating conditions)
on intention and usage behavior, as follows:

a. Performance expectancy (PE) is referred to as the degree a user believes that

using the system will improve his/her job performance and is posited as a
predictor of intention and use behaviour;

b. Effort expectancy (EE) refers to the ease of use of the new technology;

c. Social influence (SI) is defined as the degree to which an individual perceives

that the important others believe he or she should use the new system;

d. Facilitating conditions (FC) refers to a users belief about the existence of

organizational support for the use of the information system. The construct is
contended to have significant effect on the use of the IS. Prior literature also
confirms the significant effects of facilitating conditions on behaviour intention of
a new technology (Wu et al., 2007);

e. User experience is the experience enjoyed by the user from first use of the
system. Venkatash and Davis (2000) contend the negative moderating influence
of experience on the relationship between subjective norm and performance

52

expectancy. They argue that an increase in inexperience will decrease the effects
of effort expectancy and social influence on behaviour intention;
f. Voluntariness is described as when users believe that they have a choice in the
technology adoption or use decision (Brown, et al., 2002). This is in contrast to a
mandatory use environment, which when users believe that it is compulsory to
use the IS (Venkatesh, et al., 2003);
g. The influence of gender on IS acceptance have been demonstrated in various
ways. Previous studies find that males are more experienced in computing areas
compared to females. Males also have more positive attitudes towards
computing than females (Durndell et al., 2000);

h. Age Studies show older users show the strongest propensity to employ an
innovation when he or she possesses positive attitude about the innovation. On
the other hand, younger users have more positive attitude about using a new
technology (Elias et al., 2012).

The four determinants of UTAUT are independent factors, which will affect behavioral
intention and usage behavior. Behavior intention is referred to as user intention to
perform a specified task while usage attitude is referred to as the user reactions when
using the IS (Venkatesh, et al., 2003).

53

Since its inception in 2003, researchers have been employing UTAUT in various fields
of studies. UTAUT consistently predicts 70% of the variance in intention to use. There is
evidence that UTAUT is superior to the eight models and their extensions, which have
lower per cent of variance (Venkatesh et al., 2003). Anderson & Schwager (2004)
transform UTAUT into an analysis process in which gender, age, experience and
voluntariness are employed as the defining variables. This helps the researchers to
identify patterns of use in twelve specific areas based on the four constructs and the
qualities of the sample population.

In summary, UTAUT is an effective model for assessing and presenting data on user
acceptance, especially when user demographic information is taken into account. The
UTAUT model has contributed significant understanding towards user acceptance in
providing a unified framework for overlapping user acceptance theories.

54

2.5.5 Model of Information System Success (ISSM)

In information systems (IS) literature, the Information System Success Model (ISSM) is
widely employed in various IS empirical studies. In 1992, the authors formulate the IS
Success Model based on their review on theoretical and empirical research of
numerous information systems (DeLone and McLean, 1992). The dimensions comprise
system quality (SQ), information quality (IQ), system use, user satisfaction, individual
impact and organizational impact. System and information quality are both posited to
affect use and user satisfaction. System quality and information quality are also
measures of individual impact; as such it will produce an effect at organizational level.
According to the authors, a successful information system (IS) is determined by three
dimensions, namely information quality (IQ), system quality (SQ) and service quality
(SVQ). These dimensions have significant influencing effects on intention to use which
contribute towards satisfaction. By using the information system, users will reap certain
benefits which will either positively or negatively influence satisfaction and ultimately
adoption of the system.

The ISSM was updated by the authors in 2003, by including two important modifications
and a clarification. As shown in Figure 7, the improved model incorporates the service
quality dimension and groups both impact measures (individual and organizational
impact) into net benefits. The authors also contend information systems usage should
precede user satisfaction. In addition, positive experience with the use of the system will

55

increase user satisfaction. As such, a higher level of satisfaction will improve the
intention to use which will also affect IS usage (DeLone and McLean, 2003).
Figure 7. Updated Information System Success Model

Source : Delone and McLean, (2003)

56

In the 10-Year Update of the Model, the authors further explain about every construct
and they argue that the constructs should be employed not only in current systems, but
also systems in development such as e-business, e-government, knowledge systems
and online technologies. The updated model comprises the following constructs:-

a. System quality (SQ) is defined as attributes that are required in an information

system (IS). The authors operationalise the attributes as easy to use, flexibility,
reliability, easy to learn, sophistication and speed of responses. The criteria for
SQ comprise of criteria such as flexible, reliable, responsibility, accessible and
integrate with other systems (Nelson et al., 2005).

b. Information quality (IQ) is referred to as system output, such as the relevance,

accurateness, completeness and usefulness of the product to the individual. It is
also referred to as the quality outcome produced by the system. Nelson et al.,
(2005) propose that other than output quality, information quality should consider
who uses the information, the application being used and the task being
completed. They posit that information quality consists of four dimensions,
namely accuracy, completeness, currency and format.

c. Service quality (SVQ) refers to the support rendered by technical support

personnel for example, empathy and technical know-how (DeLone and McLean,
2003). SVQ also refers to the service that the system support provides to the
consumers during the course of the online transactions. Nelson et al., (2005)

57

posit that service quality consists of five dimensions, namely tangibles, reliability,
responsiveness, assurance and empathy.

d. System use is defined as the measure of use of the system. It is measured by

volume, frequency, visits and intention of use. Usage also refers to any kind of
transactions that consumers or users have done with the website.

e. User satisfaction refers to how the consumer feels about using the system. It
starts from using the system itself proceeding to the outcome and the kind of
services provided by the support group. DeLone and McLean (2003) contend
satisfaction measures the users opinions during transactions with the system.

f. Net benefit is defined as system contribution towards the overall impact to the
organization, economy, society or countries (Petter and Mclean, 2009). Net
benefits is operationalizes as the effects of the IS on consumers, employees,
organizations, governments, industries, economy and the society. Net benefits
include the various impacts such as societal impact, individual impact, and
organizational impact (Venkatesh, et al., 2003).

They also argue that information system success factors comprise three types of quality
factors, namely the information quality, service quality and system quality. These factors
are regarded as the key factors of success in an information system. ISSM has been

58

widely employed by researchers in various information systems research (Oliviera et al.,

2011).

In summary, the information system and technology acceptance literature provides a

useful theoretical lens to study e-procurement auctions implementation, particularly
eBidding adoption. It allows us to examine the significance of different antecedents
impacting procuring officials behavior in eBidding use.

59

CHAPTER 3

THEORETICAL FRAMEWORK
3.1 Introduction
This chapter proposes an empirical model of an information systems (IS) adoption in a
public sector setting along with the hypotheses to be tested in the study. A brief review
of the theoretical framework is provided to guide the reader through the development of
the conceptual model to explain the factors influencing the behavioural intention to use
eBidding. The proposed model is developed based on the conceptual model and
hypotheses are specified for testing.

3.2 Conceptual development

Many information system (IS) researchers have studied new technology acceptance
over the past two decades. From an analysis of the literature, there are three most cited
theories in individual acceptance of technology mainly the Diffusion of Innovation
Theory (DOI), Technology Acceptance Model (TAM) and the Unified Theory of
Acceptance and Use of Technology (UTAUT). TAM, DOI and UTAUT have been widely
employed to investigate user behavior associated with adoption of innovations. Tobbin
(2011) argues that the most employed and tested model in previous studies is the TAM,
UTAUT and followed by DOI.

60

Technology Acceptance Model (TAM) is a behavior adoption theory that proposes

usefulness and ease of use as influencing factors for behavior intentions to use new
technology (Davis, 1989). These two beliefs create intentions about information system
and how it affects its usage. Perceived Usefulness (PU) refers to how a user believes
that using an IS will increase his/her job performance. Perceived Ease of Use (PEOU)
refers to how a user believes that employing the system will be free of effort (Davis,
1989).

Diffusion of Innovation (DOI) theory examines the adoption behaviors which includes
several innovation attributes that influence acceptance (Rogers, 1995). They are
relative advantage, complexity, trialability, compatibility and observability. Relative
advantage refers to how a user believes that using the system is better than the
previous practice. Compatibility refers to how the user believes that by using the
system, he/she will be conforming to the norms. Complexity refers to how a user
perceives the difficulty in using the system. Trialability is defined as how a new
technology can be tried before making a decision while observability refers to the
visibility and demonstrativeness of the output that are visible to others (Rogers, 1995).

Unified Theory of Acceptance and Use of Technology (UTAUT) is a unified theory that
explains user acceptance by integrating eight different models. The underlying premise
of the UTAUT model is that three factors will influence behavioral intentions, namely
performance expectancy (similar to perceived usefulness), effort expectancy (similar to
perceived ease-of-use), facilitating conditions and social influence. The UTAUT model

61

explains that these constructs have significant influence towards adoption decision. The
moderating variables in the UTAUT include gender, age, experience and voluntariness
of use.

Personal Innovativeness in the Domain of Information Technology (PIIT) is based on the

premise that users with high degree of personal innovativeness will have propensity of
trying out new technology. The key constructs of PIIT (similar to PU and PEOU in TAM)
comprise usefulness and ease of use. Usefulness refers to how easy it is to use the
system which is similar to performance expectancy. Similar to effort expectancy, ease
of use refers to how a user believes that by employing the technology it will improve job
performance. Compatibility refers to how a new technology is viewed in conformance
with the culture and social norms. Behavioural intention refers to the willingness of
individuals to exert effort to achieve the given behaviour. PIIT is argued to have a
mediator effect because it measures user's beliefs in the innovation (Agarwal and
Prasad, 1998).

The Information System Success Model (ISSM) is another model that is related to
behavioural intention-based theory (DeLone and McLean, 1995). In 2003, the authors
proposed an updated model of IS success to predict e-commerce applications (DeLone
and McLean, 2003). They extended their earlier model by adding additional dimensions,
namely the service quality, system quality and information quality. The authors also
propose that both intention and actual use behavior are measures of system success.

62

The Model is updated by the inclusion of net benefits derived from intention to use and
user satisfaction

3.3 Models Comparisons

TAM and DOI are considered to complement one another and similar in some
constructs. Wu and Wang (2005) contend that there are similarities between relative
advantage and perceived usefulness and perceived ease of use that it is argued that
the TAM constructs are subsets of the DOI.

The weakness of TAM is that although it includes user behavior intention and
information systems theory, the model ignores social factors and the model is often
criticized for focusing on the intention to use as opposed to actual usage. In addition,
the shortcoming of TAM only provides 40% of variance in usage intentions and there is
inconsistent relationship among constructs investigated (Zhang et al., 2006). The
weakness of DOI is that the model does not show the proofs how attitude is translated
into acceptance or rejection (Chen et al., 2002). As such, both models are considered
unsuitable to be employed in the study.

IS Success Model limitation is that it cannot provide explanation as to why the same
application system can be adopted in different ways, with different effects in various
settings (Tsiknakis and Kouroubali, 2008). Whereas UTAUT limitation is that it did not

63

take into account the interactions between individuals and organizations simultaneously
with interaction between organization and its environment (Ghobakhloo et al., 2010).

IS Success Model (ISSM) and the UTAUT model are both by itself is a comprehensive
model. Each is internally sound and based on well-tested behaviour intention models.
Both models define almost similar dependant constructs, where in IS Success Model it
is called intention to use or use, whereas in UTAUT, it is called behavioural intention.
However, for this dependant construct, each model defines different independent
constructs. In UTAUT, behavioural intentions are determined by performance
expectancy (PE), effort expectancy (EE), social influence (SI) and facilitating conditions
(FC) while in the Updated IS Success Model, usage is influenced by system quality
factors, namely information quality (IQ), system quality (SQ) and service quality (SVQ).

The UTAUT model incorporates moderators, namely gender, age, experience and
voluntariness that influence user adoption of technology. PIIT is posited to have
mediator roles on user's behavior on intention. On the other hand, the IS Success
Model, DOI and TAM has none of the moderating variables. Table 4 provides a
comparison of the models

64

Table 4. Models Comparisons

No

Authors

Independent
Variables
- Perceived
usefulness
- Perceived ease of
use

Moderating
Variables
-

Dependent
Variables
Behavioral
intention

1.

Technology
Acceptance
Model
(Davis et al., ,
1989)

2.

Diffusion of
Innovation (DOI)
(Rogers, 2003)

- Trialability
- Observability
- Relative
advantage
- Complexity
- Compatibility

Rate of adoption

3.

Personal
innovativeness
in the domain of
information
technology
(PIIT)
(Agarwal and
Prasad, 1999)

- Perceived
usefulness
- Perceived ease of
use
- Compatibility

- PIIT

Behavioral
intention

4.

Unified Theory
of Acceptance
and Use of
Technology
(UTAUT)
Venkatesh et al.,
(2003)
Updated IS
success Model
(ISSM)
(DeLone and
McLean, 2003)

- PE
- EE
- SI
- FC

5.

- Gender
- Age
- Experience
- Voluntary of
use

- SQ
- IQ
- SVQ

Usage behavior

Net benefits

65

3.4 Proposed research model

In the study, the UTAUT is adapted as a base while system quality, information quality
and service quality and satisfaction (dimensions drawn from the IS Success Model) are
added variables. The proposed framework incorporates attitude toward information
systems (IS), individual innovativeness and willingness to use new innovation. The
system quality factors are drawn from Updated IS Success Model while user behaviour
factors are based on UTAUT constructs that measure EE (similar to ease of use),
performance expectancy (similar to usefulness), SI and FC.

The UTAUT is a comprehensive model based on the combination of eight theories as

such it is adequate and suitable for the study. UTAUT has been employed by numerous
IS researchers to investigate the antecedents influencing the adoption and
implementation of new technology by individuals (Oliveira et. al, 2011). The theory
explains both user behavioral intentions to use the IS. By combining eight adoption
models, UTAUT is the most comprehensive to explain IS use behavior at individual level
and demonstrated up to 70% accuracy in predicting user acceptance (Venkatesh et al.,
2003). In short, UTAUT is adequate to be employed as the base model for its extensive
use by researchers in various fields and its high explanatory power of variance.

All constructs in UTAUT are posited to influence behavioral intention to adopt eBidding.
However, UTAUT does not address the issues relevant to technical and system quality

66

factors such as, system, service and information quality dimensions. The model
limitation is that it cannot provide explanation as to why the same application system
can be adopted in different ways in various settings (Tsiknakis and Kouroubali, 2008).
Hence, it is posited that, by combining all these independent factors it could better
represent factors that determine intention to use or behavioural intention of eBidding.

The effect of users personal innovativeness in the modified UTAUT framework will also
be investigated by including PIIT construct as a moderating variable. UTAUT model is a
comprehensive model, but it is unable to include the measure to examine individual
traits, such as innovativeness and risk taking that contributes towards technology
acceptance (Rosen, 2004). By including PIIT to the integrative UTAUT and IS Success
Model, it is expected to boost the amount of variance explained in the adoption behavior
that allows a much wider perspective of user adoption of an IS. The proposed
framework for this research is as shown in Figure 8.

67

Figure 8. Proposed Research Model

Personal
Innovativeness in
IT (PIIT)

Performance
Expectancy

Effort
Expectancy
Social Influence

Facilitating
conditions
Information
Quality
System Quality

Experience

H1a
H1

H2a

H2b

H4a
H3a

H2

H3

eBidding
Adoption

H4

H5

H6

H7

Service
Quality

H7a

H6a

Satisfaction

H5a

68

A research framework is proposed based on the combinations of construct drawn from

both models. The proposed integrative model would provide a more realistic picture of
individual level eBidding adoption by the procuring officials. In the proposed framework,
eBidding adoption is the dependent variable with performance expectancy (PE), effort
expectancy (EE), social influence (SI) and facilitating conditions (FC), information
quality (IQ), system quality (SQ) and service quality (SVQ) as independent variables.
User satisfaction is proposed as the mediator while PIIT is added as the moderator
between PE and EE with eBidding adoption.

The adapted UTAUT model will be tested without two moderator variables, age and
gender. Gender variable is excluded due to the environment in the government setting
that did not discriminate the work functions and responsibilities between male or female
officials. Both genders receive similar training, exposure and rulings on the use of any
e-government applications. Similarly, age variable is also eliminated from the study due
to the government setting that did not differentiate the work functions and
responsibilities with regards to officials age. Voluntariness variable is excluded since
eBidding usage is optional and totally voluntary. A variable for moderating role namely
experience, is drawn from UTAUT. Prior research suggests that increased inexperience
will lower the influence of effort expectancy (EE) and social influence (SI) on the
adoption of the system (Venkatesh et al., 2003). The PIIT is added to examine the
mediator effects between PE and EE with eBidding adoption. PIIT is posited to have a
moderator role on users perceptions about an innovation (Agarwal and Prasad, 1998).

69

Thus, this study assumes the possibility that experience and PIIT have moderating
effects on eBidding adoption by government users.

For the IS Success Model, the two IS success categories, namely the organizational
impact and net benefits variables are excluded because it is not the intention of the
study to investigate these two categories, but the behavioural intention category to
adopt eBidding without the various impacts such as societal, individual, and
organizational impact of the system. Variable satisfaction is posited as a mediating
factor between system factors, namely the information quality (IQ), system quality (SQ)
and service quality (SVQ) and eBidding adoption. The research model put forward in
this study would be tested through the following hypotheses.

3.5 Hypotheses Development

3.5.1 Dependent Variable

3.5.1.1 eBidding Adoption

In the proposed theoretical model in the study, adoption of eBidding is treated as

dependent variable. Delone and Mclean (2003) contend system quality usage has been
widely employed in literature to measure how successful is the system in terms of
implementation. In the study, the dependent variable is actual use instead of
behavioural intention to use. In the studys context, adoption refers to the willingness of
the procuring officials to put in the effort to use and adopt the eBidding adoption. The

70

adoption of the eBidding is operationalised based on the works of Venkatesh et al.,

(2003), namely planning, intention, and predicting the use and actual use. The level of
system use is also quantitatively assessed using criteria, for example planning,
intention, actual use, use frequency and amount of use of specific features (Nelson et
al., 2005).

The study will focus solely on actual adoption because it is more practical to measure
the use of technology instead of intention to use. There is a strong correlation between
intentions to use with actual adoption. TAM explains the relationship between intention
to use and actual usage (Davis, 199). Behavioural intention is found to be a valid
predictor of usage and adoption (Sun and Zhang, 2005). As such it is adequate to test
only one of the variables because of the positive significant relationship between these
variables to each other. It is posited that adoption is influenced by performance
expectancy (PE), effort expectancy (EE), facilitating conditions (FC), social influence
(SI) and system quality factors with three moderators, PIIT, voluntariness and
experience mediated by satisfaction.

71

3.5.2 Independent Variables

3.5.2.1 Performance expectancy

The construct refers to how the user believes that by employing the IS, this will facilitate
them to enhance the achievements in their jobs (Venkatesh et al., 2003). They also
contend that PE is similar to Perceived Usefulness (PU) as in the TAM. Carlsson et al.,
(2006) employed UTAUT in a mobile service study and confirm the significant effect of
performance expectancy on the behaviour intention of the services. Perceived
usefulness (similar to PE) was found to effect the adoption of on-line shopping system
(Gefen et al., 2003).

Performance expectancy found to have significant effect on mobile service users in Lu

et al., (2009) in a survey of 1320 respondents using the service. Park et al., (2007) find
that performance expectancy significantly affects users to employ mobile services.
Performance expectancy is shown to significantly influence behavioural acceptance
both in voluntary and mandatory situations. It has been moderated by gender and age
whereby men and younger adults are found to be affected by performance expectancy.
The

performance

expectancy measurement

factors

productivity, time and cost saving (Venkatesh et al., 2003).

72

include

job

performance,

In the context of this study, performance expectancy refers to the perception that
eBidding usage would assist the procuring officials to achieve certain outcomes, such
as improve job performance and productivity in their working environment. When an
official perceives that using the system would aid in achieving gains in the workplace,
he or she would be more likely to have the propensity to employ the system. The more
the official believes that using eBidding will help him/her improves job performance
there will be more likelihood that they will use the system.

It is posited that the officials will use eBidding if it will improve their job performance,
reduce time and cost and increase job productivity. Therefore, PE is predicted to have
positive significant relationship with the behaviour intention to use eBidding. Thus, it is
posited as follows:

H1: Performance expectancy is significantly related to officials adoption of eBidding

73

3.5.2.2 Effort expectancy

Effort expectancy refers to the degree of ease of system usage. The construct is similar
to TAMs perceived ease of use which is confirmed by UTAUT authors. In TAM and
TAM2, the effort expectancy is represented as perceived ease of use in DOI. The TAM
factors include ease of use, ease of achieving level of expertise and easily
understandable (Venkatesh et al., 2003).

Effort expectancy construct is posited to be an antecedent of behaviour adoption and

use. Park et al., (2007) find that effort expectancy significantly affects user adoption of
mobile service. In Carlsson et al., (2006) study, the authors support the relationship
between effort expectancy and adoption. The authors contend that effort expectancy
has significant influence on the behavioural acceptance of the service. Previous
research also supports the contention that lesser effort is required to learn and use the
system will ultimately influence its acceptance (Gefen and Straud, 2000).

In the study, effort expectancy is the belief that using eBidding would assist the
procuring officials to gain certain advantages for example, improve job performance and
increase productivity in their work environment. As officials increasingly become used to
a new technology, the effort needed to use eBidding will decline. The more the officials
believes that the system is easy to use and help them become very skillful in using the
system the more the inclination to use eBidding. It is posited that the officials will use
eBidding if the system is easy to use, easy to operate and learn and require less effort

74

to understand. As such in the research model, effort expectancy is posited to have

significant relationship with the adoption of eBidding. Hence, it is posited as the
following:

H2: Effort expectancy is significantly related to officials adoption of eBidding

3.5.2.3 Social influence

The construct refers to how a user believes that other significant individuals (senior
managers/peers) believe he or she should employ the innovation. The SI factors include
organizational support, acknowledgment from peers and department encouragement
(Venkatesh et al., 2003). Previous research indicates peers, superiors and other people
who are significant to the user often affect his/her behaviour to adopt the system (Wolin
and Korgaonkar, 2003). When individuals believe that their peers, supervisors and
others around them support their use of the technology, the propensity to use the IS is
higher. Previous research has shown the impact of system influence on behavioral
intention (Karahanna et al., 1999; Venkatesh and Davis, 2000). Singh et al., (2010)
argue that users decisions to use mobile commerce are significantly affected by peers
and family. In a study from 158 clients in a major bank in Malaysia, Amin et al., (2008)
find that user intention to employ banking service is significantly affected by their circle
of friends.

75

In the context of the study, system influence refers to the officials perception that their
colleagues and superiors believe he/she should use eBidding. It is posited the more
one believes that the head of departments, superiors and others around them support
their use of eBidding, the propensity to use eBidding will be higher. As such in the
proposed research model, SI is predicted to have positive significant relationship with
the adoption of eBidding. It is posited as the following:

H3: Social influence is significantly related to eBidding adoption

3.5.2.4 Facilitating conditions

Facilitating conditions refers to how a user perceives the technical support provided by
the organization to facilitate system usage (Venkatesh et al., 2003). The authors
contend facilitating conditions are directly related to actual usage of new technology.
The construct includes characteristics of the technology or organization conditions (for
example, regulations, incentives and training) that facilitate system use.

Facilitating conditions is posited to have significant influence on acceptance and use.

Prior literature indicates facilitating conditions significantly influences acceptance
behaviour (Wu et al., 2007). From the study, the authors explain that these factors
relates to financial, technological and training support provided to the users in assisting
the system usage. The facilitating conditions factors include availability of technical
assistance, resources, knowledge and compatibility of the new technology with other IS

76

(Venkatesh et al., 2003). When officials believe there is organizational support available
to the IS, they will be more willing to use the system. In a study on online banking,
Joshua and Koshy (2011) argue clients greater accessibility to the Internet will result in
higher number of proficient use which is translated to a higher adoption rate.

In the context of the study, facilitating conditions refers to the existence of supporting
infrastructure available to the users of the information system. The construct is
important because without these supporting infrastructures, the officials may believe
that it will be more difficult to conduct the procurement auction transactions. It is posited
that the officials will be inclined to use eBidding if the system is compatible with other
applications, there is technical assistance and resources and the knowledge to operate
the system readily available. In the proposed research model, facilitating conditions is
predicted to have positive significant relationship with behaviour intention to use
eBidding. As such, it is posited as the following:

H4: Facilitating conditions is significantly related to officials adoption of eBidding

77

3.5.2.5 Information Quality

Information quality is referred as individuals belief of the output quality generated by the
system (Wixom and Todd, 2005). They contend that output quality produced by the
system plays an important function in users satisfaction. The higher the user perceives
the quality of output produced by the system, the higher possibility he or she would be
satisfied with the information, which will influence behavioural intention. Nelson et al.,
(2005) define information quality as information that is accurate, up-to-date, complete
and in the right format based on the users perspectives. A user who is satisfied with the
quality of information produced from the system will have higher possibility he or she will
perceive that it will enhance the work performance. If the information generated by the
system is satisfactory to the individual, the satisfaction will be likely to influence their
beliefs of the usefulness and actual use (Wixom and Todd, 2005). Wang (2008)
contends that information quality significantly influences users perception of value
which increases users' intentions to use an IS. Lee et al., (2007) contend that
information quality is an important factor in increasing customers intentions to use an
online service. Lin (2006) finds information quality significantly affects users'
participation in a virtual community and confirms the significant influence of information
quality on actual use.

78

In the context of the study, information quality refers to the perception that the output
generated by eBidding will help the procuring officials to attain benefits for example,
improve job performance and productivity in their working environment. When an official
perceives that the information generated by the eBidding system is useful and accurate,
he or she would be more likely to have a positive attitude towards using the technology.
The more they believe that the information will improve job performance, the officials
will be more willing to use the system. User satisfaction from using the output produced
by the system is predicted to have positive relationship with adoption behaviour.

It is posited that the officials will use eBidding if the system output is accurate, in correct
format, up-to-date and complete, thus will improve their job performance, reduce time
and cost and increase job productivity. Therefore, information quality is predicted to
have positive significant relationship with user satisfaction and eBidding adoption.
Hence, it is posited as the following:

H5: Information quality is significantly related to officials adoption of eBidding

79

3.5.2.6 System Quality

System quality refers to performance quality of the system. Nelson et al., (2005) define
system quality as system attributes, for example flexibility, reliability, response time,
accessibility and integration. System quality is also associated with system satisfaction
(Wixom and Todd, 2005). A satisfied user would be more influenced, in terms of
perceptions of usefulness and actual use (Wixom and Todd, 2005). DeLone and
McLean, (2003) argue ease of use, reliability and flexibility are measures employed by
researchers to investigate system quality. For example, Lin (2006) contends that other
than information quality, system quality is another factor that affects users' participation
in a virtual community. System quality significantly affects respondents use of an online
e-learning via user satisfaction (Chiu et al., 2007).

If the officials perceive that the system is flexible, reliable, fast and accessible, the
satisfaction with the system will be higher which results in higher propensity to use
eBidding. Therefore, system quality is predicted to have positive significant relationship
with satisfaction and the eBidding adoption. Hence, it is posited as the following:

H6: System quality is significantly related to officials eBidding adoption

80

3.5.2.7 Service Quality

Service quality refers to the quality of support provided by the system provider (DeLone
and McLean, 2003). Service quality is defined as the users perception of the level of
support provided by the e-commerce service provider (Molla and Licker, 2001). SVQ is
measured by attributes such as response to queries, problem solving, providing details
about transaction processes or questions about the tender (Wu et al., 2007). Cai and
Jun (2003) identify four key dimensions of service quality as web site design, website
content, trustworthiness, prompt and reliable service and communication. Nelson et al.,
(2005) definition of service quality are the tangibles, reliability, assurance and empathy
provided by the system service providers.

In the case of the study, service quality refers to the responsiveness and assurance by
the service providers to the officials to help make their experience better when using
eBidding. The availability of support for the government users is imperative for the
success of the eBidding system. Users, who are satisfied with the support and services
provided by the system, will be more likely to influence others about their beliefs of
usefulness and actual use (Wixom and Todd, 2005). The service providers are the
concessionary company, CommerceDotCom Sdn. Bhd. supported by the help-desks
and in-house technical departments. If the official perceives that the service quality by
the service provider is high, then he or she is more likely to be satisfied with the system.
The officials will be more inclined to use eBidding if they are satisfied with the service

81

quality which is perceived as fast, with empathic and supportive service providers,
hence is inclined to use eBidding. Service quality is predicted to have positive significant
relationship with user satisfaction and the adoption behaviour of eBidding. Hence, it is
posited as the following:

H7: Service quality is significantly related to officials adoption of eBidding

3.5.3 Mediating Variable

3.5.3.1 Satisfaction

A mediator is the third variable between a predictor and an outcome variable. Any
construct has a mediator function if it explains the relationships between the
antecedents and the results (Creswell, 2009). Satisfaction is defined as to how the user
favours the system and the interaction dynamics with the system (Wixom and Todd,
2005). Satisfaction also refers to how a user feels throughout the transactions with the
system. Reichheld and Schefter (2000) defined the construct as the satisfactions during
their interactions with the system, repeated transactions or recommendation to friends
or peers. Wixom and Todd (2003) define user satisfaction as individuals attitude toward
information quality, system quality and service quality of the information system. If a
user is satisfied with the system, he or she will be influenced to continue using the
system. If the information produced by the system is satisfactory for the user, there is
more possibility that this will influence their beliefs of the IS usefulness.

82

Wixom and Todd (2005) argue that satisfaction has mediating effects on information
quality, system quality and service quality and subsequently influence individual
adoption of the system. Satisfaction is a confirmed mediator between perceptions of
quality and behavioral intentions (Cronin et al., 1992). Overall customer satisfaction has
been tested as a mediating variable in determining the influence of satisfaction
antecedents on behavioral consequences (Cheung and Lee, 2005; Kim et al., 2009). Liu
et al., (2000) find service quality has a significant influence on behavioral intentions and
the mediating role of satisfaction on influencing behavioral intentions in the Indian
service industry. In an empirical study on e-government portal, Cora K.L. (2009) find
user satisfactory use of an e-government website has partial mediating effect on the
relationships between success factors and intention to use.

In the studys context, it is predicted that satisfaction may play mediating role in the
adoption of eBidding, particularly when the system involves new technology and is quite
complex, which requires excellent web design, good services and technical support. A
reasonable hypothesis is that the more procuring officials are satisfied with the system
quality factors, the more likely eBidding will be adopted. The more the officials believe in
the output quality, the more likely they will adopt eBidding due to higher user
satisfaction. Similarly, if the officials have more positive perception of the services
provided by the system providers, they will be more inclined to use eBidding via the
mediating role of user satisfaction. On the other hand, if the officials have less positive

83

perception of the quality of eBidding performance, the officials will be less inclined to
adopt due to lower user satisfaction. Hence, it is posited as the following:
H5a: Satisfaction significantly mediates relationship between information quality and
eBiddingAdoption
H6a: Satisfaction significantly mediates relationship between system quality and
eBidding Adoption
H7a: Satisfaction significantly mediates relationship between service quality and
eBidding adoption

3.5.4 Moderating variables

3.5.4.1 Personal Innovativeness in the Domain of Information Technology

Moderating variables are qualitative or quantitative variables which affect the

relationships between an independent or dependent variable (Creswell, 2009). The
author contends that moderator variables are important because specific factors have
contingent effect on the relationship between the predictor variables with the outcome
variable.

PIIT refers to users willingness to risk and try an innovation (Agarwal and Prasad,
1998). PIIT refers to personal traits or the individuals willingness to change that is a
function of individuals tolerance of risk (Thatcher and Perrewe, 2002). PIIT is
operationalised by Agarwal and Prasad (1998) as a persons inclination to take the risk,
try out and experiment with an innovation. PIIT also measures general attributes of
technology innovativeness and may moderate the relationships of numerous other

84

constructs (Agarwal and Prasad, 1998). The authors argue that PIIT is a significant
determinant of usefulness, ease of use and compatibility. In other words, highly
innovative individuals may believe that the technology is more useful, easier to use, and
more compatible because of their intrinsic innovativeness. The trait is similar to the
individual characteristics of earlier adopters in the DOI (Rogers, 2003). However,
Agarwal and Prasad, (1998) also find that PIIT acts as a moderator in the relationship
between perceived ease of use (effort expectancy) and usefulness (performance
expectancy) with usage intention. PIIT is also a confirmed mediator between
compatibility and intention to use web technologies, as well as perceived ease of use
(PEOU) and perceived use (PU) of an innovation (Agarwal and Prasad, 1998; Lewis et
al., 2003).

Malaysias eBidding is a newer technology compared to conventional procurement

system. It is argued that PIIT might have a higher effect on eBidding in comparison to
the conventional manual system. Officials with higher innovativeness are expected to
feel comfortable around new innovations such as eBidding. It is believed that highly
innovative officials would have the natural inclination to learn the innovation easily and
they would have higher propensity to employ the eBidding. Therefore, PIIT might
positively moderate relationship between eBidding perceptions and behavioural
intention. It is posited that innovativeness trait in the official will moderate performance
expectancy and effort expectancy relationships with eBidding adoption. Officials with
higher personal innovativeness are more likely to appreciate the productivity and
performance of the system and will be more inclined to adopt eBidding. Similarly,

85

officials with higher PIIT will be more likely to appreciate the ease of the system (effort
expectancy) and will adopt eBidding. Hence, the following hypotheses are posited as
the following:

H1a: PIIT significantly moderates the relationship between performance expectancy

and
eBidding adoption
H2a: PIIT significantly moderates the relationship between effort expectancy and
eBidding adoption

3.5.4.2 Experience
Experience is defined as the user experience whether the individual is affected since
the last time the user initially employed the information system (Venkatesh, et al., 2003).
In TAM2, experience construct was a confirmed moderator between social influence and
acceptance behaviour of an IS (Venkatesh and Davis, 2000). The relationship between
effort expectancy and adoption is moderated by gender, age and experience whereas,
the relationship between facilitating conditions and intention is moderated by age and
experience (Venkatesh et al., 2003). They contend effort expectancy, facilitating
conditions and social influence have significant effects on the system usage and
moderated by experience.

Literature shows that people with a high level of experience are different from those with
lower experience especially in the knowledge structure (Sderlund, 2002). Users
evaluate their satisfaction with an information system differently by depending on their

86

level of experience and knowledge (Hawkins et al., 2009). Venkatesh et al. (2003) argue
that with increasing experience, the effect of effort expectancy and social influence
decreased whereas the effect of facilitating conditions increased. In the study,
experience is included as moderating variable to determine whether the officers are
affected since the last time he/she has initial use of similar system.

In the studys context, eBidding is a new application introduced under the ePerolehan
system, thus prior experience with similar system is expected to have an impact on the
usage of eBidding. Prior negative experience with application will moderate the officials
perception, in terms of perceived ease of use (PEOU) and social influence towards
system adoption. It is posited that officials prior experience with eBidding or similar
system will negatively moderate the relationships between effort expectancy and social
influence with officials adoption of eBidding, whereas it is posited that prior experience
will positively moderate the relationships between facilitating conditions and eBidding
adoption.

If the officials believe that they only need to spend less effort to learn and use eBidding
before adopting it, there is high possibility that they will use the system. If the officials
have previous negative experience in using similar systems and find those system
requires too much time and difficult to learn, it is likely that this experience would
negatively affect their belief on eBidding ease of use and their adoption of the system.
Similarly, if the officials and their superiors share prior bad experience about similar
systems, they will have negative perception of eBidding, thus it is likely that this

87

experience will negatively affect their inclination to adopt eBidding. However, if the
officials had positive experience with the facilities and support associated with the
system, it is likely that this experience will positively affect their inclination to adopt
eBidding. Hence, the following hypotheses are posited as the following:

H2b: Experience negatively moderates the relationship between effort expectancy

with eBidding adoption
H3a: Experience negatively moderates the relationship between social influence
with eBidding adoption
H4a: Experience positively moderates the relationship between facilitating
conditions with eBidding adoption

88

CHAPTER 4
METHODOLOGY
4.1 Introduction
This chapter explains the research design and describes the process to identify the
sampling population, the research method and data anaysis. Is also outlines the validity
and reliability of the questionnaire including a brief description of the preliminary pilot
study.

4.2 Population and Sampling

There are 2,558 Responsibility Centers (RCs) in Peninsular of Malaysia, in which 2,047
RCs are ePerolehan-enabled. There are 604 RCs in Putrajaya and Klang Valley
(comprised 1,507 procuring officials), which will be the focus of this study. The centres
are chosen because they are well-equipped with eBidding systems and trained
personnel. The RCs located in Putrajaya administrative capital conducts the main bulk
of procurement transaction activities for the public sector (ePerolehan PTJ, 2012). The
government procuring officials include personnel from grade 27 to 54 that perform
various tasks, including supervisory, requisition, goods receiving, payment and
supporting roles in the RCs.

89

4.3 Data Collection and Sampling procedure

4.3.1 Sampling procedure

Sampling procedure includes the gathering of a sizable number of items from the
population. The sample and its characteristics were then analysed and the results from
the analysis is generalized to the population (Sekaran, 2003). A representative or good
sample will ensure the ability of final results to be generalised to the population. Sample
bias must be avoided by ensuring adequate sample and sample randomness (Sekaran,
2003). The population of this study involves 1,507 officials in the 604 RCs in the Klang
Valley and Putrajaya. A simple random sampling is deemed appropriate since there is
adequate sampling frame and the population is geographically concentrated (Hair et al.,
2006). The respondent designations and mailing list of the RC agencies were sourced
from the ePerolehan Unit, Ministry of Finance (MoF) across different ministries in
Putrajaya, Kuala Lumpur and Shah Alam.

The main method of data collection for the study is by using mail survey in which the
questionnaires were administered by postal service. The advantages of the method are
the respondents can provide their responses to the questionnaire at their own leisure. At
the same time this method can draw a sample from wider and larger geographical area
(Sekaran, 2003). Although better response rate could be gained by distributing the
quesionnaire by hand, time and cost limitations were the major constraints since the
government agencies in the sample size are located all over Klang Valley and

90

Putrajaya.A random number generator called Research Randomizer (Urbaniak & Plous,
1997) was employed to randomly select respondents for the study. Before the data
collection exercise, permission was obtained from the Secretary Generals of the 28
Ministries located in Putrajaya.

The process of determining the right and conducive sample size was based on the
sample size table of adequate sample size for the size of a population. This was done
by using the guidelines by Bartlett, Kotrlik and Higgins (2001). According to Table 5, a
population of 1,507 require a minimum sample size of 112 responses. The sample size
plays an important role in calculation of efficiency of coefficients of internal consistency.
Ozdamar (1999) argued that for reliability coefficient calculation, the sample size should
be more than 50. Yi (2012) proposed that the sample size should be above 100 for a
study that employ Structural Equation Modelling (SEM). A sample size of 200 is
recommended but less than 400 because a higher number will render the SEM analysis
to be too sensitive (Hair, et al., 2006).

91

Table 5. Table for Determining Returned Sample Size for Given Population Size for
Continuous and Categorical Data by Bartlett, Kotrlik, & Higgins (2001)

Source: Bartlett, Kotrlik, and Higgins (2001)

92

4.3.2 Data Collection

The participation of RCs in the survey was solicited by mail survey. Their contact
information (email, postal addresses, telephone numbers) were sourced from
ePerolehan Unit, Ministry of Finance and invitations to participate were sent to the RCs
starting from the first week of February 2013. After approvals were obtained from the
Ministries, the questionnaires were sent to the Head of Departments/Undersecretary of
Procurement/Finance/Development Divisions which supervise the RCs and undertake
procurement transactions based on the list provided by the ePerolehan Unit. The
qquestionnaires were sent through postal services, emails and followed by phone calls
to the RCs. Further assistance and phone calls to clarify the questionnaires to the
respondents were done as and when requested by the respondents. The face-to-face
engagements will enable the respondents to clarify any doubts in the questionnaires
with the researcher (Sekaran, 2003).

The survey respondents were informed of the privacy of the data provided in the survey.
The respondents returned the completed questionnaire in one month from the date the
questionnaires were being delivered to them. By the end of the period, follow-ups were
conducted via phone calls, and email reminder. Officials from ePerolehan Unit also
provide assistance in terms of reminding the procuring officials in the list to complete the
questionnaire in their emails to the RCs. One hundred and ten responses (110) were
returned within the specified time. Queries on the low return of the survey
questionnaires revealed that eBidding module has not been employed at the time

93

considered as an early part of the year and the respondents were busy with the
scheduled annual budget planning work and did not have time to complete the
questionnaire. After the cut-off date, another forty (40) responses were returned after
follow-ups and personal visits to the RCs. This brought the accumulated returned
questionnaires to 150 valid survey forms, about 10% of the sampling population. This
response exceeded the required minimum responded sample size of 112 for a
population of 1,500 according to Bartlett, Kotrlik and Higgins (2001).

4.4 Questionnaire Design

The questionnaire in the study is divided into three parts as in Appendix 1. The first part
of the survey is to collect the data of factors influencing the procurement officials
adoption of the eBidding. This part is divided into seven sections, corresponding to the
number of independent variables. The measurements were adapted from the studies of
UTAUT and Information System Success Model. The second part of the instrument is to
acquire respondents profiles (position, department, years of service, age, and gender);
and his/her (years of service, department size and location). This section consists of
closed question formats and open-ended format, using nominal scale. The third part of
the questionnaire was to obtain the data of officials actual use of the eBidding.
Questions are on the actual use, level of usage in transactions, frequency of use in
years and period of utilization. This section is based on closed question format, using a
Likert scale.

94

4.5 Measurement of Variables

The proposed instruments are adapted from the works of DeLone and McLean (2003)
and Venkatesh et al., (2003) with modifications to the government settings. The
behavioral dimensions of the concept order are examined in order to operationalize a
construct (Sekaran, 2003). The concept was then translated into measureable elements
for the construct. Operationalisation for the constructs in the proposed framework was
drawn from the studies of UTAUT and IS Success Model. Table 6 summarizes the
research instruments based on the dimensions used, variables and number of items on
each variables and its source of reference.
Table 6. Description of Research Instruments

No

Secti
on

Variables

No. of
Stateme
nts
7

Scale of
Measurement

References

Interval

Interval

Interval

Facilitating conditions
(FC)
System quality (SQ)

Interval

Interval

Interval

Interval

Interval

Venkatesh et al.,
(2003)
Venkatesh et al.,
(2003)
Venkatesh et al.,
(2003)
Venkatesh et al.,
(2003)
Delone and Mclean,
(2003)
Delone and Mclean,
(2003)
Delone and Mclean,
(2003)
Wixom and Todd,
(2005)

Information quality
(IQ)
Service quality (SVQ)

Satisfaction

Performance
expectancy (PE)
Effort expectancy
(EE)
Social influence (SI)

95

No

Secti
on

Experience

10

11

12

Variables

No. of
Stateme
nts
5

Scale of
Measurement

References

Interval

PIIT

Interval

Use

Interval
Nominal

Venkatesh et al.,
(2003)
Agarwal and
Karahanna, (2000)
Venkatesh et al., (2003)

Respondents profile

Source: Developed for the study

Items for PE, EE, SI, FC and experience were drawn from user behavior studies by
Venkatesh et al., (2003), while items for SQ, IQ, SVQ and satisfaction were adapted
from DeLone and McLean (2003) with some adjustments to reflect the specific target
behavior of eBidding adoption/actual use. Measurement items of PIIT are adapted from
the works of Agarwal and Karahanna (2000). Measurement items of satisfaction were
adapted from the works of Wixom and Todd, (2005). A seven-point Likert scale was
employed for the questionnaire except the demographic section. Items were measured
on a seven-point Likert scale, where 1 = strongly disagree, 2 = moderately disagree, 3 =
disagree, 4 = neutral (neither disagree nor agree), 5 = agree, 6 = moderately agree, and
7 = strongly agree (Babbie, 2010).

96

4.5.1 Actual Use

This variable measures actual usage of the IS. The construct is measured from
planning, intention, a visit to the IS site, navigation, information query and actual
transaction (Venkatesh et al., 2003). Respondents are asked to choose their preference
based on the statements using a seven-point Likert scale from (1) extremely
infrequent to (7) extremely frequent; and one (1) statement using ordinal data scale
with a rating scale attached ranging from(1 ) less than 1 year ;( 2 ) 1-2 years;(3) 3-5
years; ( 4 ) more than 5 years .

4.5.2 Performance Expectancy (PE)

Performance expectancy measures how users believe that using the IS will increase
their job performance (Venkatesh et al., 2003). Seven (7) statements were used to
measure this construct that the respondents could choose using a seven-point Likert
scale.

4.5.3 Effort Expectancy (EE)

Effort expectancy refers to ease of use of the information systems (Venkatesh et al.,
2003). Seven (7) statements are used to measure this construct that the respondents
could choose using a seven-point Likert scale.

97

4.5.4 Social Influence (SI)

Social influence refers to how a user belief that important others believe he or she
should use the IS (Venkatesh et al., 2003). Seven (7) statements are used to measure
this construct that the respondents could choose based on the statements using a
seven-point Likert scale.

4.5.5 Facilitating Conditions (FC)

Facilitating conditions construct refers to how a user believes the existence of

supporting infrastructure exists to facilitate the use of the IS. Seven (7) statements are
used to measure this construct that the respondents could choose using a seven-point
Likert scale.

4.5.6 System Quality (SQ)

System quality measures the technical characteristics of the system, namely the
adaptability, availability, reliability, response time and usability (DeLone and McLean,
2003). Seven (7) statements are used to measure this construct that the respondents
could choose using a seven-point Likert scale.

98

4.5.7 Information Quality (IQ)

Information quality is defined as the attributes of the IS, for example, completeness,
personalisation, accuracy and relevance (DeLone and McLean, 2003). Seven (7)
statements are used to measure this construct that the respondents could choose using
a seven-point Likert scale.

4.5.8 Service Quality (SVQ)

Service quality refers to the support that a user receives from the system support
personnel, namely assurance, empathy and responsiveness (DeLone and McLean,
2003). Seven (7) statements are used to measure this construct that the respondents
could choose using a seven-point Likert scale.

4.5.9 Satisfaction

Satisfaction refers to the users opinions of the information system which covers the
users entire experience cycle. The cycle involves information query, actual transactions,
and surveys (Wixom and Todd, 2005). The construct is measured by number of repeat
visits, transactions and user surveys. Five (5) statements are used to measure this
construct that the respondents could choose using a seven-point Likert scale

99

4.5.10 Experience

Experience refers to individuals experience whether he or she is affected since the last
time he/she has used the system (Venkatesh, et al., 2003). Five (5) statements are
used to measure this construct that the respondents could choose using a seven-point
Likert scale.

4.5.11 Personal Innovativeness in the Domain of Information Technology (PIIT)

PIIT refers to a users propensity to take the risk and try an innovative technology
(Agarwal and Karahanna, 2000). Five (5) statements are used to measure this construct
that respondents could select using a seven-point Likert scale.

4.6 Pilot Study

Verbal and written feedback regarding the questionnaire construction was obtained from
fifty (50) participants and pilot-tested in UPM. The respondents were chosen based on
convenience sampling, however they were not included as respondents in this study.
Pilot testing is to improve the reliability and validity of the proposed research
instruments (Sekaran, 2003).The pilot test enables removal of vague and confusing
questions, determine the time taken to complete the survey and to check for proper
sequencing of the questions. The feedback is to be employed to finalize the content of
the questionnaire. Recommendations from the respondents were incorporated and

100

modifications were made. Several items were re-sorted in order to make them more
convenient for the respondents. Several sentences were rephrased and reworded to
make them more appropriate with the objectives of the study.

Analysis of data from the trial determines the reliability of the survey questions.
Cronbachs alpha reliability coefficient was employed to examine the reliability of the
research instruments. Coefficient (r) from 0 and 1 with coefficient closer to one indicates
higher reliability (George and Mallery, 2003). Reliability coefficients should be at least .
70 or higher to be considered reliable for an effective research instrument (Wallen and
Fraenkel, 2001).

Cronbach's alpha of each of the instrument was obtained. For n=50, all the items for
each variable posed a Cronbach's alpha value of 0.8 to 0.9 in terms of reliability (Hair et
al., 2006). All Cronbach's alpha coefficient of the scale from the pilot test is above the
acceptance value of 0.7. Hence, the survey questions are valid and reliable and can be
used at the RCs for the purpose of measuring the outcome of eBidding adoption.

101

4.7 Data Analysis Approaches

The data analysis approaches in the research are based on descriptive and inferential
statistics. The data will be tested for goodness of fit using factor analysis and reliability
of measures using Cronbach Alpha. In addition, proposed hypotheses will be examined
using relevant statistical tests. Data analysis as depicted in Figure 9, involves several
steps before hypothesis testing in ensuring good quality data for further analysis by data
preparation; feel for the data; test the goodness of data; examine the hypothesis using
structural equation modelling ; and examine the model fit (Sekaran, 2003).

Figure 9: Flow Diagram for Data Analysis Process

Data Analysis
Interpretation
of Results

Discussion

Data
Collection

Feel for data

-Normality
-Correlations

Goodness of
data
-Reliability
- Validity

Source: Sekaran, (2003)

102

Hypothesis
testing
-Appropriate
statistical
manipulations
(SEM,
Hierarchical,
Regression)

Answer for
research
questions

Testing model
fit
-RMSEA
-TLC etc

4.7.1 Descriptive Statistics

Descriptive statistics is employed in order to organise the results which covers the tests
of frequencies, mean and standard deviation. Frequency distribution enables
researchers to view the entire responses to the questionnaire. Mean is used to measure
average response by adding all the numbers and dividing by the number of cases.
Standard deviation shows the distribution scores from the mean value (Fink, 2008).

4.7.2 Inferential Statistics

In the study, inferential statistics is used to enable the results obtained from samples to
be generalised. Structural Equation Modelling (SEM) and AMOS software are used to
examine the measurement model adequacy and structural model goodness-of-fit (GoF)
including the hypotheses testings.
4.7.3 Structural Equation Modeling (SEM)

The study employs the Structural Equation Modeling (SEM) technique to test the
relationships among the variables in the model. The method involves multiple
regression analysis of factors among a single measured dependent variable and a
group of predictors (Ullman 2007). Hair et al., (2006) argued that SEM is able to
examine two types of models, i.e. the measurement model that represents the theory
and the model which represent the latent factors. A structural equation modeling (SEM)
employs multivariate analyses and will show relationships between constructs and the

103

causal dependencies between endogenous and exogenous variables (Hair et al., 2006).
The SEM also allows two groups comparisons which makes it suitable for testing the
hypotheses of the research. Therefore, SEM is selected in this study to maintain
parsimony in the proposed model while benefiting from its strength in testing the
research hypotheses.

The study uses AMOS version 6 to examine the relationships between the independent
variables (PE, EE, SI, FC, SQ, IQ, SVQ), moderator variables (PIIT and experience)
and one mediator (satisfaction) and behavioural actual use of eBidding service. There
are two main reasons behind the decision to adopt this software. First, the software is
available in the graduate school resource centre, as such it is accessible for the
analysis of data. Second, AMOS is rarely employed in previous empirical and
conceptual research of user acceptance (Tong, 2007).

Structural Equation Modeling (SEM) in AMOS involves drawing a circle and arrow path
diagram. AMOS is easy to learn and use and it is capable of analysing many goodnessof-fit measures. AMOS also offers flexibility and ability to analyse numerous linear
models (Garson, 2009). The data will be analysed with descriptive and inferential
statistics tests. The following statistical methods are used in this study to organize the
data:

104

a) Data are keyed-in and sorted in the AMOS 6 software ;

b) An exploratory data analysis was conducted to check central tendency and

data dispersion to calculate the mean, range, standard deviation and the
variance in the data.;
c) Two test procedures were undertaken to maximize the instruments validity
and reliability;
d) Cronbachs alpha was used to analyse the reliability of the measures and the
correlations strength between each items. The internal consistency of
reliability is higher when the value is closer to 1. Reliability analysis test was
performed prior to testing the proposed model. Cronbachs Alpha was
employed for evaluating the consistency between multiple research variables
measurements, where the generally acceptable value for reliability is 0.7
(Creswell and Miller, 2000);
e) Validity test ensures that the instruments are reliable, however an instrument
can be reliable without being valid. Confirmatory factor analysis (CFA) was
used to evaluate and establish evidence of sufficient construct validity
(Creswell, 2003). The CFA also tests whether measures of research
constructs are consistent or fit the validity of the research model. For the
model

to

have

goodness-of-fit,

assumptions

of

homoscedasticity and normality distribution must be met.

105

multi-collinearity,

4.7.4 Testing Process Using Structural Equation Modeling (SEM)

Structural Equation Modelling (SEM) was used to examine the hypotheses. SEM with
AMOS combines both regression analysis and factor analysis. SEM is a multi-variate
statistical method used to test direct and indirect associations between a predictor
variable with one or more outcome variables. SEM consists of seven major stages, from
developing the model, constructing path diagram causal relationship, building and
modifying measurement model and finally model fit (Hair et al., 2006).

a)

Stage 1 : Developing a Model

The present study aims to understand the factors influencing procuring officials
behaviour in eBidding adoption in government setting. The proposed model in
the study must fulfil the established criteria for making causal relationships. Hair
et al., (2006) argued there must be sufficient relationships between variables,
existence of cause versus effect and theoretical justifications for the
relationships. Appendix II shows the list of the indicators of the present study.

106

b) Stage 2: Path Diagram Relationship

A Structural Equation Modeling model consists of an observed variable (OV)

and latent variable (LV). The relationships of observed and latent variables in
SEM are shown using path diagrams (Gefen, 2003; Tong, 2007).

i.

Actual Use/Adoption This variable is an endogenous variable, which

consists of four (4) observed factor variables/indicators, namely Use1,
Use2, Use3 and Use4;

ii.

Performance Expectancy (PE) - This variable is an exogenous variable,

which consists of seven (7) observed factor variables, namely PE1, PE2,
PE3, PE4, PE5, PE6, PE7;

iii.

Effort Expectancy (EE) - This variable is an exogenous variable, which

consists of seven (7) observed factor variables, namely EE1,
EE2,EE3,EE4,EE5,EE6 and EE7;

iv.

Social Influence (SI) - This variable is an exogenous variable, which

consists of seven (7) observed factor variables, namely SI1, SI2,SI3,SI4,
SI5,SI6 and SI7;

v.

Facilitating Conditions (FC)- This variable is an exogenous variable

which consists of seven (7) observed factor variables, namely FC1,
FC2,FC3, FC4,FC5,FC6 and FC7;

107

vi.

System Quality (SQ) - This variable is an exogenous variable, which

consists of seven (7) observed factor variables, namely SQ1, SQ2, SQ3,
SQ4, SQ5, SQ6 and SQ7;

vii.

Information Quality (IQ)- This variable is an exogenous variable, which

consists of seven (7) observed factor variables, namely IQ1, IQ2, IQ3,
IQ4, IQ5, IQ6 and IQ7;

viii.

Service Quality (SVQ) - This variable is an exogenous variable, which

consists of seven (7) observed factor variables, namely SVQ1, SVQ2,
SVQ3, SVQ4, SVQ5, SVQ6, and SVQ7;

ix.

Satisfaction- This variable is an exogenous variable, which consists of

seven

(7)

observed

factor

variables,

namely

Satis1,

Satis2,Satis3,Satis4,Satis5, Satis6 and Satis7;

x.

Experience (Exp) - This variable is an exogenous variable, which

consists of five (5) observed factor variables, namely Exp1, Exp2, Exp3,
Exp4 and Exp5;

xi.

Personal Innovativeness in IT (PIIT)- This variable is an exogenous

variable, which consists of five (5) observed factor variables, namely
PIIT1, PIIT2, PIIT3, PIIT4 and PIIT5.

108

c) Stage 3: Structural and Measurement Models

SEM integrates both path and factor analyses including multiple observed
variables or indicators for every unobserved variable as in factor analysis and
specific paths connecting the unobserved variables in path analysis (Tong,
2007). Confirmatory factor analysis (CFA) establishes a model fit to the data.
Unobserved variable is tested to see if they are measured by the observed
constructs where each unobserved variable is assumed to be related with a set
of observed variables (Tong, 2007). The measurement model is then evaluated
using the Goodness-of-Fit (GoF) test. By analysing the results from
measurement and structural models testing, a more precise hypothesis testing
can be undertaken (Garson, 2009).

d) Stage 4: Matrix Type and Proposed Model Estimation

Data screening was done to check the raw data in order to identify any missing
data, which involve testing data normality and skewness. The key assumption of
SEM software is that endogenous and exogenous variables must be normally
distributed. A distribution is considered as normal if it has skewness indices of
less than 3 and kurtosis value being less than 10 (Tong, 2007). In addition,
multi-collinearity problem will occur if and when the observed variables are
closely associated. Pearson coefficient test detects multi-collinearity which is
not present if the coefficient value is less than 0.8 (Sekaran, 2000).

109

e) Stage 5: Assessing the Model Identification

The first type of model identification is a simple model, which has the same
number of variables as the co-variance elements. In this study, an overidentified model was obtained because it will significantly enhance the reliability
of the estimates. An over-identified model has more co-variance elements
compared to variables (Garson, 2009).

f)

Stage 6 : Evaluating Goodness-of-Fit Criteria (GoF)

Goodness-of-fit is employed to determine the model rejection or acceptance for

further analysis (Tong, 2007). Although AMOS offers 25 different goodness-of-fit
measures, however Garson (2009) advised researchers not to report all the 25
goodness-of-fit measures. He argued that by reporting all goodness-of-fit
measures, it implies that the researcher is on a fishing expedition. However,
Kline (2005) recommended the four most commonly reported goodness-of-fit
tests are chi-square; Goodness-of-Fit Index (GFI), Normed Fit Index (NFI),
Comparative Fit Index (CFI), and Root Mean Square Error of Approximation
(RMSEA).

Chi-square (2) is the most popular goodness-of-fit test used for SEM (Garson,
2009). AMOS outputs list chi-square as chi=\CMIN. The model chi-square
values tend to decrease (better fit) when more paths are inserted or created in a

110

model. On the other hand, significant 2 value shows poor model fit (Garson,
2009; Kassim, 2001; Tong, 2007). Relative chi-square (2/df) is used in SEM so
that the 2 value will be less influenced by the size of the sample (Ullman,
2007). Relative X2 is listed in AMOS as RATIO = Cmin/df, and a relative X2
value of 3 or less is considered as acceptable (Tong, 2007).

Comparative Fit Index (CFI) is another measurement for Goodness-of-Fit test.

CFI is used to measure the model fit improvement compared to a null model
(Garson, 2009; Tong, 2007). AMOS lists CFI as CFI=\cfi (Tong, 2007). CFI value
which is close to 1 indicates an excellent model fit (Garson, 2009; Tong, 2007).
Normally, CFI has to be .90 for a model to be accepted. CFI has to be .90 for
a model to be accepted (Tong, 2007).

Normed Fit Index (NFI) is an alternative index of CFI (Tong, 2007). NFI ranges
from 0 (bad fit) to 1 (good fit) (Ullman, 2007). NFI value of 0.50 means the SEM
model improves by fifty per cent in comparison to null model (Tong, 2007). NFI
values above .95 are considered outstanding, NFI values from 0.90 to 0.95 are
desirable, but the researcher has to re-specify the model if NFI value is below
0.90 (Garson, 2009; Tong, 2007).

Tucker Lewis Fit Index (TLI) is another incremental fit index similar to NFI
(Garson, 2009). TLI is relatively independent of sample size. TLI is expressed
as the fit per degree of freedom (Tong, 2007). AMOS lists TLI as TLI =\TLI

111

(Tong, 2007). TLI value close to 1 or TLI value > 0.90 shows a good model fit,
while TLI value below 0.90 shows the need to re-specify the model (Tong,
2007).

Another important indicator is Goodness of Fit Index (GFI) which analyses the
per cent in the model co-variances (Garson, 2009). AMOS lists GFI as GFI =\gfi
(Tong, 2007). GFI should be equal to 0.9 or higher for a parsimonious model
(Garson, 2009) whereas, Schumacker and Lomax (2004) recommended a GFI
value of 0.95 or higher.

Another criterion for Goodness-of-Fit test is Root Mean Square Error of

Approximation (RMSEA), which is the difference per degree of freedom (Tong,
2007). RMSEA is an effective measure of model fit as there is no need to
compare a SEM model with a null model (Tong, 2007). AMOS lists RMSEA as
RESEA =\rmsea. RMSEA value of 0.8 as good model fit (Tong, 2007).

g) Stage 7: Modifying the Model

The model was examined for potential model modifications after an initial SEM
model was established. A re-specification process was done by trimming the
model by adding path arrows and removing parameters to achieve a model with
good fit (Garson, 2009).

112

CHAPTER 5

DATA ANALYSIS AND DISCUSSION

5.0 Introduction

The chapter outlines the findings of the study with analysis of the data gathered in the
study. The hypotheses and the proposed framework are examined to analyse the
behavioural and technological factors associated with the adoption of an e-government
system within the government setting.

The chapter outlines the descriptive and inferential analysis of the data including
reliability and normality tests. The reliability and validity of the constructs were tested
using conformatory factor analysis. SEM was conducted to test associations between
the constructs in proposed framework. SEM was also used for the statistical analysis in
relation to the objectives of the study and the proposed hypotheses.

5.1 Feel for Data

In the study, feel for data was done by examining the central tendency and dispersion.
As suggested by Sekaran, (2003), the analysis is organized into the following:

113

1). the frequency distribution of demographic variables ;

2). normality, the mean, standard deviation, range, and variance of dependent and
independent variables; and
3). inter-correlation matrix of the variables.

5.2 Frequency Distribution of Demographic Variables

For this study, the procurement officials in the responsibility centres represent the unit of
analysis. Respondents of the study vary according to a number of demographic
variables such as agency size, agency location and respondent position.

5.2.1 Respondents Gender and Age

The results of the study show 54% of the respondents are female while 46% of the
respondents are male as in Table 7. In term of age, approximately 53.3% of the
respondents are in the age range of 31-40 years old, followed by 34.7% of the
respondents less than 30 years old, 6.7% of the respondents are above 50 years old
and 5.3% are in the range 41-50 years old.

114

Table 7 : Respondents Gender and Age

No.
1

Demographic Variable
Gender
Male
Female
Age
less 30 years old
31 to 40 years old
41 to 50 years old
more 50 years old

69
81

46
54

52
80
8
10

34.7
53.3
5.3
6.7

Source: Computed Data Analysis

5.2.2 Respondents Position dan Working Experience

The officials positions are categorised based on Head of Departments, Management

and Support staff (as Table 8). By looking at the professional designation, at least half of
the officials (53.3%) are in the managerial position, followed by 45.3% in the supporting
position and 1.3% are department heads. By studying in the aspects of working
experiences, 35.3% of the respondents have working experiences of 1-3 years, 28.7%
of the respondents with working experiences 4-6 years,18% less than 1 year and 18%
more than seven years.

115

Table 8: Respondents Professional Designation and Working Experiences

No.

Demographic Variable

Professional Designation

Department heads
Managerial
Supporting
Working Experience
Less 1 year
1 to 3 years
4 to 6 years
More 7 years

2
80
68

1.3
53.3
45.3

27
53
43
27

18
35.3
28.7
18

Source: Computed Data Analysis

5.2.3 Location and Department Size

In term of location, 92.7% of the respondents are in the Putrajaya, followed by 4.7% of
the respondents are in Kuala Lumpur, 2% are in the Seremban and 0.7% are in Shah
Alam (as Table 9). Approximately 74% of the respondents work in departments which
have more than 50 staff, followed by 12% work in the departments which have 31-50
staff, 6% has 10-20 staffs, 2.7% has less than 10 staff.

Table 9: Respondents Location Size and Department Size

No.
5

Demographic Variable
Location
Putrajaya
Kuala Lumpur

116

139
7

92.7
4.7

Shah Alam
Seremban
Department Size
Less 10 employees
10 to 20 employees
21 to 30 employees
31 to 50 employees
More 50 employees

1
3

0.7
2

4
9
8
18
111

2.7
6
5.3
12
74

Source: Computed Data Analysis

5.2.4 Normality Test

The data was examined using a normality test to ensure that data was from a
population with normal distribution. The tests regarding feel for data will furnish
information, such as how fine the scales are and is gives an overview of the data coding
and data entry (Sekaran, 2003). In this study, testing feel for data was done by
examining the central tendency and dispersion as shown in Table 10.

117

Table 10: Normality Test

No.

Constructs

1.

Performance Expectancy (PE)

PE1
PE2
PE3
PE4
PE5
PE6
PE7
Effort Expectancy (EE)
EE1
EE2
EE3
EE4
EE5
EE6
EE7
Social Influence (SI)
SI1
SI2
SI3
SI4
SI5
SI6
SI7
Facilitating Conditions (FC)
FC1
FC2
FC3
FC4
FC5
FC6
FC7
System Quality (SYQ)
SYQ1
SYQ2
SYQ3
SYQ4
SYQ5
SYQ6

2.

3.

4.

5.

Mean

Standard
Deviation

Skewnes
s

Kurtosis

4.75
4.89
4.69
4.68
4.66
4.59
4.77

1.28
1.43
1.32
1.29
1.35
1.28
1.40

-.326
-.635
-.394
-.386
-.453
-.409
-.701

-.108
.338
.128
.103
-.094
.020
.390

4.73
4.78
4.85
4.81
4.51
4.67
4.53

1.16
1.07
1.05
1.07
1.25
1.06
1.20

-.425
-.321
-.246
-.400
-.532
-.125
-.516

.462
.518
.607
.542
.118
-.334
.231

4.59
4.57
4.84
5.07
4.61
4.80
4.96

1.33
1.34
1.33
1.29
1.29
1.25
1.34

-.515
-.681
-.497
-.523
-.712
-.693
-.485

.317
.375
.186
.253
.638
.540
-.392

5.23
4.57
4.35
4.68
4.75
5.18
4.99

1.15
1.22
1.27
1.25
1.21
1.19
1.24

-.388
-.451
-.302
-.062
-.344
-.403
-.309

-.302
.423
.072
-.801
.424
-.260
-.226

4.55
4.69
4.57
4.64
4.53
4.69

1.30
1.32
1.23
1.30
1.24
1.26

-.804
-.571
-.464
-.500
-.419
-.573

.818
.065
.336
.106
.154
.394

118

No.

6.

7.

8.

9.

10.

11.

Constructs

Mean

SYQ7
Information Quality (IQ)
IQ1
IQ2
IQ3
IQ4
IQ5
IQ6
IQ7
Service Quality (SVQ)
SVQ1
SVQ2
SVQ3
SVQ4
SVQ5
SVQ6
SVQ7
Satisfaction
Satisfaction1
Satisfaction2
Satisfaction3
Satisfaction4
Satisfaction5
Satisfaction6
Satisfaction7
Experience (Exp)
Exp1
Exp2
Exp3
Exp4
Exp5
Personal Innovativeness (PIIT)
PIIT1
PIIT2
PIIT3
PIIT4
PIIT5

Actual use (Use)

USE1

Skewnes
s
-.294

Kurtosis

4.39

Standard
Deviation
1.21

4.81
4.83
4.81
4.85
4.66
4.57
4.78

1.17
1.11
1.10
1.14
1.19
1.21
1.19

-.751
-.427
-.535
-.363
-.559
-.454
-.553

1.316
.474
.487
.201
.330
.370
.177

4.72
4.77
4.79
4.77
4.89
4.84
4.81

1.15
1.17
1.19
1.18
1.14
1.11
1.16

-.364
-.192
-.320
-.019
-.211
-.332
-.131

.206
.027
.049
-.261
.334
.217
.174

4.67
4.72
4.61
4.59
4.69
4.55
4.59

1.23
1.22
1.32
1.30
1.27
1.41
1.32

-.200
-.287
-.217
-.433
-.363
-.415
-.432

.008
-.277
-.232
.004
.259
.009
.180

4.22
4.07
3.65
4.45
3.91

1.32
1.38
1.70
1.38
1.75

-.077
-.090
.077
-.301
-.019

-.169
-.376
-.790
-.237
-.660

4.45
4.86
4.48
4.79
4.27

1.36
1.20
1.41
1.29
1.42

.041
-.031
-.161
-.335
-.252

-.398
-.718
-.317
-.044
-.093

4.21

1.56

-.091

-.478

119

-.013

No.

Constructs
USE2
USE3
USE4

Mean
3.26
3.36
2.91

Standard
Deviation
1.92
1.84
1.73

Skewnes
s
.272
.266
.845

Kurtosis
-1.219
-1.067
-.326

Source: Computed Data Analysis

Table 5.4 shows the mean and standard deviation, skewness and kurtosis of each item
in data testing. Various opinions can be found concerning the acceptable level of
skewness (distributions symmetry) and kurtosis (the clustering of scores toward the
centre of a distribution) for a particular variable. The underlying assumption of SEM is
that it needs data to be multivariate normality. Any violation of this assumption will lead
to a biased result (Sekaran, 2003). The data are considered normal if the value of
skewness and kurtosis is around 1. Departures from normality cause the chi-square
test to be higher and standard errors to be lower. By following the rule of thumb that
both skewness and kurtosis should fall in the acceptable range of 1.0 (Hisham, 2008),
hence the result demonstrates that all constructs in the study is normally distributed.

120

5.3 Testing Goodness of Data

5.3.1 Reliability and Validity

A reliability test examined the consistency of respondents in answering the

questionnaire items. Construct reliability was used to measure the degree to which the
items were free from random error to produce consistent results (Sekaran, 2003). If the
items are independent measures of the similar concept, they are correlated with one
another. Cronbachs alpha is a widely used method in testing consistency reliability
between items that is used for multipoint-scaled items such as Likert scale. It is believed
that higher coefficient will lead to better measurement. As such, Cronbach alpha value
of 0.5 and higher is considered sufficient in determining reliability of the item (Sekaran,
2003).

Each constructs were tested for validity using convergent and discriminant validity tests.
Convergent and discriminant validity tests will establish construct validity. At this stage,
confirmatory factor analysis (CFA), average variance extracted results (AVE), composite
reliability tests and Cronbachs alpha have been performed. Average variance extracted
(AVE) results have been established when the variance has extracted an estimate to
measure an amount of variance gained by a construct associated to the variance from
the random measurement error. An AVE value higher than 0.5 indicates the presence of
convergent validity and AVE variance from 0 to 1 refers to the ratio of the total variance.

121

Composite reliability is the overall reliability of the whole scale. Composite reliability
value is recommended to be above 0.7 and AVE above 0.50 (Hair et al., 2006).

Table 11. Reliability and Validity

Construct

No

1.

2.

Performance
Expectancy (PE)
PE1
PE2
PE3
PE4
Effort Expectancy
(EE)
EE4
EE5
EE6
EE7

3.

4.

5.

6.

Social Influence
(SI)
SI 1
SI 3
SI 5
SI 7
Facilitating
Conditions (FC)
FC2
FC 4
FC 5
FC 7
Information Quality
(IQ)
IQ2
IQ4
IQ5
IQ6
System Quality
(SYQ)

Factor
loading

Average
Variance
Extracted
(AVE)

Composite
Reliability

Cronbach's
Alpha

.75

.92

.929

.79

.94

.932

.7

.9

.908

.888
.949
.831
.798
.948
.974
.897
.703

.873
.879
.686
.888

.886
.787
.763
.733
.95

.66

.88

.891
.822
.803
.735
.957

122

.7

.9

.72

.91

.891

No

7.

8.

9.

Construct

SYQ3
SYQ5
SYQ6
SYQ7
Service Quality
(SVQ)
SVQ1
SVQ3
SVQ5
SVQ6
Actual Use (USE)
USE1
USE2
USE3
USE 4
Satisfaction
Satisfaction1
Satisfaction2
Satisfaction3
Satisfaction4

Average
Variance
Extracted
(AVE)

Factor
loading

Composite
Reliability

Cronbach's
Alpha

.568
.906
.899
.969
.878
.65

.88

.67

.89

.892

.53

.82

.818

.803
.883
.78
.754
.756
.934
.911
.638
.76
.66
.69
.80

Source: Computed Data Analysis

Table 11 indicates all factor loadings for the study constructs are found significant and
surpassed the 0.5 guideline recommended by Hair et al., (2006). All constructs variance
extracted estimate are also found to surpass the 50 per cent. The composite reliability
values are higher than 0.6 ranging from 0.82 to 0.94. In short, these constructs are
proven to have adequate convergent reliability.

123

5.3.2 Discriminant Validity

Discriminant validity measures whether one variable is internally correlated, unique and
distinct from other variables (Tong, 2007). Correlation tests are used to examine the
discriminant validity and examine the directions in the correlation relationships (Hair et
al., 2006). A correlation value of 0.5 shows distinct concept, whereas a correlation
value of 0.8 and higher shows a lower distinct concept.
Table 12. Discriminant Validity
Variable
PE (X1)
Adoption/Use (X2)
EE (X3)
SI (X4)
FC (X5)

X1

X2

1
.660**

X3

X4

X5

X6

.643

.771**

.462

**

**

.605**

.406

**

.242

**

.622**

.542

**

.425

**

.379**

.039

**

X8

.809

1
1

SQ (X6)

.683

IQ (X7)

.423**

.698**

.513**

.798**

.785**

.328**

**

**

**

**

**

**

.675**

.557**

.482**

.522**

SVQ (X8)

.437

Satisfaction (X9)

.710**

X9

**

.129

X7

.721

.815**

.543

.696**

.723

.570**

.561

.191*

.358

1
1

**. Correlation is significant at the 0.01 level (2-tailed).

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

Source : Computed Data Analysis

The results in Table 12 shows all constructs are less than 0.8 indicating the presence of
discriminant validity. There are two correlations which show the value of Pearsons

124

Correlation greater than 0.8 which are 0.809 (Social influence and Adoption of eBidding) and 0.815 (Satisfaction and Adoption). However the values are considered low
and acceptable. Therefore the analyses indicate the scales developed for the study
have good discriminant validity.

5.4 Measurement Model

The measurements for predictor and dependent variables are examined using
confirmatory factor analysis (CFA). CFA is also used to test the convergent and
discriminant validity of the variables. The validity of the measurement model must be
confirmed before testing the structural model. Hair et al., (2006) contended that the
SEM involves specifying the model; constructing model identification; estimating the
parameter; measuring the goodness-of-fit (GOF) and re-specifying the model.

The measurement model can be re-specified if there is no reasonable fit. If the model fit
is within recommended guidelines, the structural model will be tested based on the
measurement indices (MI) (Kline, 2005). The measurement model was tested against
five indicators recommended by Hair et al., (2006). The goodness-of-fit tests explains
how the data will fit the model. Incremental fit and absolute fit indices are used to test
the study models. Multiple indices are used to test the GOF measures, namely the chisquare (X2), normed chi-square (NC), GOF index (GFI) and the Root Mean Square
Error of Approximation (RMSEA). Validity is tested using RMSEA and GFI indices.
Incremental fit indices include Normed Fit Index (NFI), Comparative Fit Index (CFI) and

125

Tucker Lewis Index (TLI). Table 13 shows the criteria for the GoF indices and the
desirable range.

Table 13. Goodness-of Fit (GOF) Indices

Determiner
Absolute fit indices
Cmin

Symbol

Specification
Desirable Criteria

Cmin

p > .05 means significant

1.0<cmin/df<5.0

Normed Cmin

Cmin/df

Root Mean Square Error of

Approximation

RMSEA

RMSEA<.08

Comparative Fit Index

CFI

NFI > or equal to .9 indicates

satisfactory fit
.8<NFI< .9 indicates
acceptable fit

Tucker-Lewis index

TLI

TLI > .9 indicates acceptable fit

Normed Fit Index

NFI

NFI >= .9 indicates acceptable fit

0.8<NFI< .9 indicates acceptable fit
GFI >= .9 indicates acceptable fit

Goodness-of-Fit Index

GFI
0.8<GFI< .9 indicates acceptable fit

Source: Hair et al. (2006)

The maximum likelihood (ML) estimates to measure parameters of the measurement

models of the uni-dimensional construct in the study are shown in Figure 10 to Figure
17. The values of standard errors related to the estimates and the goodness-of-fit
indices for each measurement models are shown in Table 12 to Table 21.

Figure 10. Measurement Model for Performance Expectancy

126

Source: Generated from study

Table 14. Goodness-of-fit for Performance Expectancy

GOF
indices

NC
GFI
AGFI
RMSEA
NFI
CFI
TLI

Value
60.165
4.297
.881
.762
.149
.897
.918
.878

Figure 11. Measurement Model for Effort Expectancy

127

Source: Generated from study

Table 15. Goodness-of-fit for Effort Expectancy

GOF
indices

NC
GFI
AGFI
RMSEA
NFI
CFI
TLI

Value
53.80
3.84
.90
.80
.14
.93
.94
.92

Figure 12. Measurement Model for Social Influence (SI)

128

Source: Generated from study

Table 16. Goodness-of-fit for Social Influence

GOF
indices

NC
GFI
AGFI
RMSEA
NFI
CFI
TLI

Value
36.41
2.6
.94
.87
.10
.93
.95
.93

Figure 13. Measurement model for Facilitating Conditions (FC)

129

Source: Generated from study

Table 17. Goodness-of-fit for Facilitating Conditions

GOF indices

NC
GFI
AGFI
RMSEA
NFI
CFI
TLI

Value
57.652
4.118
.885
.77
.145
.868
.895
.842

Figure 14. Measurement model for Information Quality

130

Source : Developed for the study

Table 18. Goodness-of-fit for Information Quality
GOF indices

NC
GFI
AGFI
RMSEA
NFI
CFI
TLI

Value
40.614
2.901
.925
.851
.113
.908
.937
.905

Figure 15. Measurement Model for System Quality

131

Source: Generated from study

Table 19. Goodness-of-fit for System Quality

GOF indices

Value
53.13

NC

3.795

GFI
AGFI
RMSEA
NFI
CFI
TLI

.903
.806
.137
.905
.928
.891

Figure 16. Measurement Mode for Service Quality

132

Source : Developed for study

Table 20. Goodness-of-fit for Service Quality

GOF indices

NC
GFI
AGFI
RMSEA
NFI
CFI
TLI

Value
26.644
1.903
.956
.913
.078
.926
.963
.944

Figure 17. Measurement Model for Satisfaction

133

Source : Developed for Study

Table 21. Goodness-of-fit for Satisfaction

GOF indices

NC
GFI
AGFI
RMSEA
NFI
CFI
TLI

Value
246.632
17.617
.784
.568
.334
.481
.488
.231

134

5.5 Data Refinement and Validation

Following from the previous section, if the fit indices show a poor fit, the model could be
re-specified by modifying the models. Hair et.al, (2006) outlined four methods to respecify the model indicators by relating the indicator to another factor; deleting the
indicator; relating the indicator to multiple factors or employing correlated measurement
errors.

For model modification, modification indices were used (MI) as guidelines to detect
specification errors so that better specified models may be tested. Model respecification was undertaken in order to enhance the goodness-of-fit (GOF) indices and
validate the constructs (see Table 22 to 29). The AMOS programme also provides
modification indices (MI) that shows the values whether there is an expected decrease
or increase in Chi-square value. The MI values indicate whether the addition or deletion
of the linkages deteriorated or improved the fitness of the model.

135

i. Performance Expectancy

There are seven (7) items under the PE. As shown in Figure 18, after re-specification,
three (3) items have to be removed, as the loading is less than 0.5, which is considered
to be weak. This leaves four items for the next stage of analysis..

The goodness-of-fit indices from Table 22 show the X2 value produces non-significant
results which show that the model fits the data after modification. RMSEA value is less
than 0.06, which suggests a good fit. GFI, CFI, NFI, and TLI values are 1, indicating a
perfect fit. PE2 produces the highest factor loading of 0.95, followed by PE1 (0.89),
PE3 (0.83) and PE4 (0.80).

136

Figure 18. Revised Measurement Model for Performance Expectancy

Source : Developed for the study

Table 22. Goodness-of-fit for revised Performance Expectancy
GOF Indices
CMIN
CMIN/DF
GFI
CFI
NFI
TLI
RMSEA

Value
.001 *
.0001
1.00
1.00
1.00
1.00
.00

137

ii. Effort Expectancy

There are seven (7) items under the EE. After re-specification, three (3) items have to
be removed as the loading is less than 0.5 which indicates weak relationship (Figure
19). This leaves four (4) items remaining for the next stage of analysis. Item ee5
produces the highest factor loading (0.97), followed by ee4 (0.95), ee6 (0.90) and ee7
(0.70). Results from Table 23 illustrate the X 2 value produced a non-significant result,
which indicates that the model fits well.

RMSEA value is lower than 0.08 which

indicates a good fit. GFI, CFI, NFI and TLI values are 1, indicating a perfect model fit.

138

Figure 19. Revised Measurement Model for Effort Expectancy

Source : Developed for the Study

Table 23. Goodness-of-fit revised Effort Effort Expectancy

GOF Indices
CMIN
CMIN/DF
GFI
CFI
NFI
TLI
RMSEA

Value
.117 (p>.05)
.117
1.00
1.00
1.00
1.00
.00

139

iii. Social Influence

There are seven (7) items under the construct SI. After re-specification, three items (3)
are omitted as factor loadings are less than 0.5 (Figure 20). As shown in Table 24, test
of the fitness model yields X2 value of 5.795, degree of freedom of 1 with CMIN/DF is
reported to be 5.795. This value produces significant results (p<0.05) indicating the data
does not fit the model well.

However, other indices CFI (0.989), GFI (0.989), NFI (0.987) and TLI (0.932) are higher
than 0.9 indicating the data fits the model well.

140

Figure 20. Revised Measurement Model for Social Influence (SI)

Source: Developed for the Study

Table 24. Goodness-of-fit for revised Social Influence

GOF Indices
CMIN
CMIN/DF(1)
CFI
GFI
NFI
TLI
RMSEA

Value
5.795 (p<0.05)
5.795
.989
.989
.987
.932
.072

141

iv. Facilitating Conditions (FC)

There are seven (7) items under the FC construct. After re-specification, three (3) have
to be omitted with four (4) items remained for the next stage of analysis (Figure 21). As
shown in Table 25, the X 2 value generates non-significant results, indicating the data fit
the model very well. RMSEA value was 0.00 indicating a perfect fit. Other indices GFI
(1.00), CFI (1.00), NFI (1.00) and TLI (1.00) indicate perfect model fit.

142

Figure 21. Revised Measurement Model for Facilitating Conditions (FC)

Source : Developed for the Study

Table 25. Goodness-of-fit for revised Facilitating Conditions
GOF Indices
CMIN
CMIN/DF(1)
GFI
CFI
NFI
TLI
RMSEA

Value
0.000 (p>0.05)
.993
1.000
1.00
1.00
1.00
.00

143

v. Information Quality (IQ)

There are seven (7) items placed under IQ. After re-specification, three (3) items are
removed and four (4) items remain for the next stage of analysis (Figure 22). As in
Table 26, the analysis reveals that X 2 value of 2.631; degree of freedom 2 with CMIN/DF
value of 1.315 (p>0.05), indicating model fit. RMSEA is 0.046 ; confidence interval (0.00
to 0.176).

Other indices, namely GFI (0.991), CFI (0.998), NFI (0.993) and TLI (0.995) indicating a
good model fit. Overall factor loadings for each item are higher than the 0.5 limit
indicating model fit.

144

Figure 22. Revised Measurement for Information Quality (IQ)

Source: Developed from study

Table 26. Goodness-of-fit for revised Information Quality
GOF Indices
CMIN
CMIN/DF(2)
GFI
CFI
NFI
TLI
RMSEA

Value
2.631(p>0.05)
1.315
.991
.998
.993
.995
.046

145

vi. System Quality (SQ)

There are seven (7) items under the SQ construct. Three (3) items are removed after
re-specification with four (4) items remaining for the next stage (Figure 23). The results
in Table 27, indicate the X 2 value is 0.949 ; degree of freedom 2, CMIN/DF value of
0.475. RMSEA value of 0.000 is lower than 0.06 that indicate it is a model. The
confidence interval for RMSEA is 0.000-0.130.

Other indices such as, GFI (0.997) and NFI (0.998) values are close to 1 indicating that
the data is compatible with the model. Similarly, CFI and TLI produced the value of 1
indicating perfect fit. In addition, all factor loadings are above 0.5 as recommended by
Hair et al, (2006).

146

Figure 23. Revised Measurement Model for System Quality

Source: Developed from Study

Table 27. Goodness-of-fit for revised System Quality
GOF Indices
CMIN
CMIN/DF(2)
GFI
CFI
NFI
TLI
RMSEA

Value
0.949 (p>0.05)
.475
.997
1.00
.998
1.00
.00

147

vii. Service Quality

There are seven (7) items placed under the SVQ construct. After re-specification, three
(3) items are removed with four (4) items remaining for the next stage (Figure 24). As
shown in Table 28, the X2 value of was 5.465 ; degree of freedom of 2; CMIN/DF is
2.733. The non-significant results showed that the model was well fitting.

On the other hand, the RMSEA value is 0.079 with confidence interval 0.000-0.134.
Other indices such as, TLI (0.967), NFI (0.983) and GFI (0.982) further highlight
satisfactory values which are close to 1. The findings show that the data fit well with the
model.

148

Figure 24. Revised Measurement Model for Service Quality

Source: Developed from Study

Table 28. Goodness-of-fit for revised Service Quality
GOF Indices
CMIN
CMIN/DF(2)
GFI
CFI
NFI
TLI
RMSEA

Value
5.465 (p>0.05)
2.733
.982
.989
.983
.967
.079

149

ix. Satisfaction

There are seven (7) items placed under the satisfaction construct. After re-specification
process, three (3) items are removed with four (4) items remaining for the next steps of
analysis (Figure 25). As shown in Table 29, X 2 value is 1.542 with degree of freedom of
2 and CMIN/DF value of 0.771. The non-significant results show the model fits with the
data.

Meanwhile, RMSEA value is 0.00 with confidence interval 0.000-0.150. Results from
GFI (0.995) and NFI (0.992) further prove satisfactory values which are close to 1. CFI
and TLI produce the value of 1 indicating good fit. These findings indicate the fitness of
data with the model.

150

Figure 25. Revised Measurement Model for Satisfaction

Source : Developed for Study

Table 29. Goodness-of-fit for revised satisfaction
GOF Indices
CMIN
CMIN/DF(2)
GFI
CFI
NFI
TLI
RMSEA

Value
1.542
.771
.995
1.00
.992
1.00
.00

151

5.6 Structural Equation Modelling (SEM)

5.6.1 Hypothesized First Order CFA Model

The SEM statistical technique was used to examine the relationships between
behavioural factors with eBidding adoption, the role of satisfaction as mediating well as
the moderating effects of experience and PIIT between behaviour factors and eBidding
adoption by government procurement officials. CFA was undertaken to measure
parameters of the measurement models. A Goodness-of-Fit test was used to assess the
proposed structural model to decide either to accept or reject the model. The overall fit
statistics results obtained from testing the model is presented in Table 30.

Table 30. Goodness-of-fit for the study Model

GOF Indices

Value

CMIN
CMIN/DF(1)
GFI
CFI
NFI
TLI
RMSEA

3013.3
2.07
.63
.60
.61
.74
.09

152

A summary of the results indicates the X2 statistics are 3013.3, GFI at 0.63 and CFI at
0.60, RMSEA (0.09), TLI (0.79) and NFI (0.61). Based on the guidelines given by Hair
et.al, (2006), the proposed model does not have a good overall fit and any parameter
estimates in SEM with a poor fit are not generalizable. Model re-specification is
undertaken to improve and develop a model that is generalizable.

5.6.2 Revised Measurement Model

Confirmatory factor analysis (CFA) was used to improve the GOF indices of the model.
After re-specification, the overall fit for the revised model were examined based on the
output obtained as Figure 26. A summary of results is presented in Table 31. The test of
fitness of the model used on the whole sample produces a value of 788.38 while the
CMIN/DF is reported to be 1.919. Referring to the X 2 value, the model does not seem
to be compatible. Other indices are also used as indicators to determine the goodness
of fit of the study model.

The TLI (0.92) and GFI (0.8) values are within desirable range, which suggest the
model can fit the data. CFI and NFI show reasonable values which 0.93, and 0.87
(close to 1), which suggest that the model and the data are harmonious with one
another. Additionally, the RMSEA value is 0.08 within the desirable range for models fit.
In other words, the re-specification process had therefore improved the models fit.

153

Figure 26. Revised Measurement Model of the Study

Source: Developed from study

Table 31. Goodness-of-fit for Revised Study Model
GOF Indices
CMIN
CMIN/DF(1)
GFI
CFI
NFI
TLI
RMSEA

Value
788.38
1.91
.80
.93
.87
.92
.08

154

The revised measurement model shows a good fit, as such the model can be used to
examine the hypotheses proposed in this research. The relationship between the
variables examined and eBidding adoption are analysed by examining the significance
of the path coefficients in the model. In the next section, the data will be analysed in
terms of mediating and moderating effects including the forwarded hypotheses in the
model. The reports of the findings will be compared with published studies reported in
the literature including possible explanations for the results.

5.7 Tests for Mediating Effects (Satisfaction)

A mediating effect is present when the variable explains the relationships between the
predictor and the outcome variable of the study. Hair et al., (2006) argued that
mediating effect is not supported if the relationships between the predictor and the
outcome variable is unchanged when the mediating varibale is added to the model. As
depicted by Figure 27, MacKinnon (2000) contended a mediating variable is an
asymmetric relations among the variables in the model.

155

Figure 27. Path with Mediator

Source : (Hair et al, 2006).

MacKinnon (2000) proposed that these criterias must be fulfilled for mediation:

a. The predictor variable must be significantly related with the mediator;

b. The predictor variable must affect the dependent variable in the second equation; and
c. The mediating variable must affect the outcome (dependent) variable in the third
equation.

Full mediation effect is achieved if the relationship between the predictor and the
outcome variable controlling the mediator is zero. Partial mediation is achieved if the
relationship between the predictor variable and the outcome variable is small but
greater than zero when the mediating variable is in the equation (Path c) compared to
when the mediating variable is not in the equation (Path c).

156

i. Test for Mediating Effects of Satisfaction on (System Quality Adoption)

The measurement model is shown as Figure 28. As shown in Table 32, chi-square value
produces significant results which show that the data do not fit the model. RMSEA value
(0.08) suggests not a good fit. However, other indices such as CFI (0.98), NFI (0.95),
GFI (0.92) and TLI (0.96) indicate a model fit.
Figure 28. Measurement Model Mediating Effect of Satisfaction (System Quality
Adoption)

Source : Developed for study

157

Table 32.Test of Mediating Effects of Satisfaction on System QualityAdoption

Relationship
GOF Indices

Value
76.98
1.83
.92
.98
.95
.96
.08

CMIN
CMIN/DF(51)
GFI
CFI
NFI
TLI
RMSEA

Table 33. Hypotheses Testing of the Mediating Effects (Satisfaction on System

Quality-Adoption Relationship)
Step

Estimate

SC

C.R./t

Adoption

<---

SQ

.069

.039

.642

2
3

Satisfaction
Adoption
Direct effect
Indirect
effect

<--<---

SQ
Satisfaction

.638
1.464

.57
0.918

6.044
9.771
.039

Results
Not
Significant
Significant
Significant

.52326

The findings in Table 33 show the relationship between system quality and adoption
(direct effect) produces non-significant results (=0.069, C.R.=0.642, p>0.05). However,
the

relationship

between

SQ

and

satisfaction

show

significant

results

(=0.638,C.R.=6.044, p<0.05). Furthermore, satisfaction has a significant effect on

adoption (=1.464, C.R.=9.771 p<0.05). Direct effect is 0.039 versus indirect effect
0.52. In conclusion, satisfaction is considered as full mediator between system quality
and adoption.

158

ii. Test for Mediating Effects of Satisfaction on (Information Quality Adoption

Relationship)
The measurement model is shown as in Figure 29. As shown in Table 34, the X 2 value
produces significant results (being less than 5), which indicate that the data do not fit
the model well. However, RMSEA being less than 0.08, suggests a good fit supported
by other indices such as GFI (0.953), CFI (0.995), NFI(0.969) and TLI (0.99) indicating a
reasonable model fit.

Figure 29. Measurement Model of Mediating Effect of Satisfaction on (Information

Quality Adoption)

Source: Developed for study

159

Table 34. Tests of Mediating Effects of Satisfaction on the Information QualityAdoption Relationship
GOF Indices
CMIN
CMIN/DF
GFI
CFI
NFI
TLI
RMSEA

Value
44.422
1.201
.953
.995
.969
.99
.037

160

Table 35. Hypotheses Testing (Mediating Effects of Satisfaction on the Information

Quality-Adoption Relationship)
Step
1
2
3

Adoption
Satisfaction
Adoption
Direct effect
Indirect
effect

<--<--<---

Estimate
IQ
IQ
Satisfaction

.73
.666
.494

SC

C.R./t

Results

7.658 Significant
7.099 Significant
4.62 Significant

.553
.664
.375
.553
.249

The findings (as in Table 35) show a significant relationship between information quality
and adoption (=0.73, C.R=7.658, p<0.05). It is confirmed that information quality
significantly influence satisfaction (=0.666, C.R=7.099,p<0.05). There is also
significant influence of satisfaction on adoption (=0.494, C.R=4.62, p<0.05). Direct
effect is 0.553 versus indirect effect 0.249. Hence, satisfaction is partial mediator
between information quality and adoption.

iii. Test for Mediating effects of Satisfaction on (Service Quality Adoption

relationship)
The measurement model is shown as in Figure 30. Results as Table 36 shows X 2 value
with significant results indicating poor model fit. However, RMSEA value (0.056) less
<.08 which indicate good fit. Other indices such as GFI (0.94), CFI (0.988), NFI (0.964),
TLI (0.976) also show a reasonable model fit.

161

Figure 30. Measurement Model to Test Mediating Effect of Satisfaction on (Service

Quality and Adoption)

Source : Developed for study

Table 36.Tests of Mediating Effect of Satisfaction on the Service Quality-Adoption

Relation
GOF Indices
CMIN
CMIN/DF
GFI
CFI
NFI
TLI
RMSEA

Value
48.659
1.475
.948
.988
.964
.976
.056

162

Table 37.Hypotheses Testing (Mediating Effects ofSatisfaction on the Service

Quality-Adoption Relation)
Step
1
2
3

Adoption
Satisfaction
Adoption
Direct effect
Indirect
effect

<--<--<---

Estimate
SVQ
SVQ
Satisfaction

.462
.943
.952

SC

C.R./t

.272
.735
.719

3.02
7.935
6.266

Results
Significant
Significant
Significant

.272
.528

The findings as in Table 37 show a significant relationship between SVQ and adoption
(=0.462, C.R=3.02, p<0.05). The relationship between SVQ and satisfaction also
produced significant findings (=0.943, C.R.= 7.935, p<0.05). There is a significant
relationship between satisfaction and adoption (=0.952, C.R=6.266, p<0.05). Based on
the findings, satisfaction becomes partial mediator between SVQ and adoption with
direct effect is 0.272 while indirect effect is 0.528. From the findings, satisfaction show
partial mediating effect between SVQ and eBidding adoption.
.

163

5.8 Tests for Moderating Effects (PIIT and Experience)

A moderating variable is the construct which has a strong contingent effect on the
relationships between the predictor and the outcome variables. A contingent effect
indicates an existence of another variable that influences the relationships between the
predictor and the outcome (Sekaran 2003). The moderating effect can be tested using
multiple-group analysis in AMOS which is able to estimate two or more groups
simultaneously (Arbuckle, 2005).

According to Frazier et al., (2004) moderating variable is present between predictor and
criterion variable. It modifies the correlation in two ways by influencing the changes in
the correlation strength or the changes in the causality direction (either negative or
positive). The moderating hypothesis will examine the paths between the predictor and
the outcome variables, in terms of magnitude and/or directions. For example, any
difference across the groups indicates that the predictor influence toward outcome
variable is moderated by that construct.

In the case of the study, two moderators (PIIT and experience) were tested to see
whether they will affect the influence of predictor variables toward the outcome variable
(eBidding adoption). As first moderator, PIIT was tested on the relationship between
performance expectancy (PE) and effort expectancy (EE) with eBidding adoption.

164

Experience (second moderator) was tested on the relationships between effort

expectancy (EE), social influence (SI) and facilitating conditions (FC).

i. Moderating Variable PIIT (Performance ExpectancyAdoption)

From Table 38, the results of the study show PE has a positive influence on adoption
(DF

2,147

=56.968,p<0.05). However, there is an insignificant relationship between PIIT

and adoption (p>0.05). In the second model it is proven that PIIT does not become
significant predictor between PE and adoption (p>0.05).
Table 38. Test for Moderating effect of PIIT (Performance Expectancy--Adoption)

Model

Unstandardized
Coefficients
B

(Constant)
PE
PIIT
(Constant)
PE
PIIT
PE* PIIT
a. Dependent Variable: Adoption
R2=0.443

1.671
.565
.039
2.676
.309
-.178
.054

165

Std.
Error
.388
.056
.081
.880
.209
.189
.042

Standardize
d
Coefficients

Sig.

4.310
10.145
.477
3.041
1.473
-.944
1.272

0.000
0.000
0.634
0.003
0.143
0.347
0.205

Beta
.652
.031
.356
-.141
.388

ii. Moderating variable PIIT (Effort ExpectancyAdoption)

The results of the study in Table 39 show EE does not influence adoption (DF
2,147

=107.514, p<0.05). The interaction between EE and PITT do not influence adoption

of eBidding adoption. This indicates PIIT does not become moderator variable of
relationship between EE and adoption of eBidding (p>0.05).

Table 39. Test for Moderating Effect of PIIT (Effort Expectancy- Adoption)

Model

Unstandardized
Coefficients
B

(Constant)
EE
PIIT
(Constant)
EE
PIIT
EE* PIIT
a. Dependent Variable: Adoption
R2=0.595 ,

1.325
.729
.012
1.794
.599
-.084
.026

166

Std.
Error
.332
.052
.069
.841
.221
.172
.043

Standardize
d
Coefficients

Sig.

3.993
14.134
.168
2.132
2.712
-.489
.606

0.000
0.000
0.866
0.035
0.007
0.626
0.545

Beta
.768
.009
.631
-.067
.176

iii. Moderating variable Experience (Effort ExpectancyAdoption)

The results in Table 40 show (EE and experience) have significant relationship with
eBidding adoption (DF

2,147

=113.824,p<0.05). Under the second model only EE has

significant relationship with adoption. The interaction between EE and experience also
produces an insignificant result. The result shows that experience is an insignificant
moderator between EE and eBidding adoption (p>0.05).

Table 40. Test for Moderating Effect of Experience (Effort Expectancy-adoption)

Model

Unstandardized
Coefficients

B
(Constant)
EE
Experience
(Constant)
EE
Experience
EE*Experience
a. Dependent Variable: Adoption
R2=0.609 .

.887
.696
.153
.378
.827
.273
-.030

167

Std.
Error
.290
.051
.067
.718
.177
.169
.038

Standardized
Coefficients

Sig.

3.061
13.516
2.270
.527
4.671
1.614
-.774

0.003
0.000
0.025
0.599
0.000
0.109
0.440

Beta
.733
.123
.871
.220
-.195

iv. Moderating variable Experience (Social InfluenceAdoption)

The results in Table 41 disclose (SI and experience) have significant association with
adoption of using eBidding (DF

2,147

=154.18, p<0.05). Under the second model, only SI

remains significant. The interaction between SI and experience does not yield
significant results. The results indicate experience is not a significant moderator
between SI and adoption.

Table 41.Test for Moderating Effect of Experience (Social Influence-Experience)

Model

Unstandardized
Coefficients

B
(Constant)
SI
Experience
(Constant)
SI
Experience
SI*Experience
a. Dependent Variable: Adoption
R2=0.679

-.128
.866
.192
.466
.719
.041
.037

Std.
Error
.289
.054
.060
.793
.191
.198
.045

168

Standardized
Coefficients

Sig.

-.443
15.928
3.202
.587
3.769
.208
.804

.659
0.000
0.002
0.558
0.000
0.836
0.423

Beta
.771
.155
.640
.033
.203

v. Moderating variable Experience (Facilitating Conditions > Adoption)

The findings in Table 42 show significant relationships between FC and experience and
eBidding adoption (DF

2,147

=28.098, p<0.05). Under the second model FC and

experience still yield significant results. However, the interaction between FC and
experience does not influence adoption. This result indicates experience is not a
significant moderator between FC and eBidding adoption.

Table 42. Test for Moderating Effect of Experience (FCExperience)

Model
(Constant)
FC
Experience
(Constant)
FC
Experience
FC*Experience
a. Dependent Variable:
Adoption
R2=0.281,

Unstandardized
Coefficients
Std.
B
Error
.380
.504
.496
.088
.416
.087
-.928
1.441
.807
.333
.727
.333
-.074
.076

169

Standardized
Coefficients

Sig.

.753
5.640
4.773
-.644
2.426
2.184
-.969

0.452
0.000
0.000
0.521
0.017
0.031
0.334

Beta
.396
.335
.644
.586
-.364

Table 43. SEM Output for Hypotheses Testing

Estimat
e

Parameter
PE

EE

SI

FC

IQ

SQ

SVQ

eBidding
adoption
eBidding
adoption
eBidding
adoption
eBidding
adoption
eBidding
adoption
eBidding
adoption
eBidding
adoption

S.E.

C.R.

.106

.04

2.64

.008

.407

.097

4.212

***

.307

.086

3.57

***

-.606

.163

-3.715

***

1.069

.332

3.217

.001

-.044

.037

-1.176

.239

.106

.06

1.768

.077

Source : Developed for the Study

Table 44. Summary of Results of Hypotheses Testing

Hypotheses

Results

H1

Performance expectancy is significantly related to officials

adoption of eBidding

Supported

H2

Effort expectancy is significantly related to officials adoption of

eBidding

Supported

H3

Social influence is significantly related to officials adoption of

eBidding

Supported

H4

Facilitating conditions is significantly related to officials adoption

of eBidding

Supported

H5
H6
H7
H5a
H6a

Information quality is significantly related to eBidding adoption

System quality is significantly related to eBidding adoption
.
Service quality is significantly related to eBidding adoption

Supported
Not Supported
Not Supported

Satisfaction significantly mediates relationship between

information quality and eBidding Adoption

Supported

Satisfaction significantly mediates relationship between system

quality and eBidding Adoption

Supported

170

Hypotheses

Results

Satisfaction significantly mediates relationship between service

quality and eBidding Adoption

Supported

PIIT positively moderates the relationship between performance

expectancy and eBidding adoption

Not Supported

H2a

PIIT positively moderates the relationships between effort

expectancy and eBidding adoption

Not Supported

H2b

Experience negatively moderates the relationship between effort

expectancy and eBidding adoption

Not Supported

H3a

Experience negatively moderates the relationship between

social influences and eBidding adoption

Not Supported

H4a

Experience positively moderates the relationship between

facilitating conditions and eBidding adoption

H7a
H1a

Source : Developed for the Study

171

Not Supported

5.9 Hypotheses Testing

5.9.1 Hypothesis Testing of Performance Expectancy

H1: Performance expectancy is significantly related to officials adoption of

eBidding

As in Table 44, the research findings in this study indicate that performance expectancy
positively influences eBidding adoption (=0.106, C.R.=2.64, p=0.008 <0.05). Hence,
null hypothesis has been rejected at =0.05. In support of hypothesis H1, performance
expectancy is found to have a significant and positive relationship with eBidding
adoption.

This result is in conformance with Venkatesh et al., (2003) works which confirmed that
performance expectancy has a significant and positive effect on system acceptance.
The implication of the result indicates that the greater the procuring officials perceives
that by employing eBidding they enhance the work performance, there will be more
willingness to use eBidding. The result is also consistent with recent studies related to
performance expectancy on e-government systems (Louho et al., 2006 ; Al-Qeisi 2009).

172

5.9.2 Hypothesis Testing of Effort Expectancy

H2 : Effort expectancy is significantly related to officials adoption of eBidding

The results in Table 44 shows that the hypothesis on effort expectancy is supported,
hypothesis null has been rejected at =0.05. Based on the coefficient (=0.407,
C.R=4.212, p=0.00<0.05). IT is determined that effort expectancy significantly
correlated with the dependent variable. Cody-Allen and Kishore (2006) argued that
effort expectancy is defined as how a user believes the level of difficulty to use the
system. Such perception will influence the decision whether to adopt or not to adopt the
IS. Other studies have found similar findings that effort expectancy significantly affects
the use intention. Therefore in the case of government procuring officials, the result is in
conformance with other findings that effort expectancy was significantly related to IS
usage (Helaiel, 2009; Rosen, 2004; Venkatesh et al., 2003).

173

5.9.3 Hypothesis Testing of Social Influence

H3 : Social influence is significantly related to officials eBidding adoption

The results in Table 44 show that the hypotheses on social influence are supported.
Based on the coefficient (=0.307, C.R.=3.57, p=0.000<0.05), social influence is
determined to have a significant correlation with the dependent variable. The findings
reveal that social influence positively influenced adoption of eBidding Thus, the null
hypothesis has been rejected at =0.05.

From the results, social influence is confirmed to be related to the eBidding adoption.
The result shows the higher social influence, the higher the officials propensity to adopt
eBidding. Therefore in the case of government procuring officials, it can be concluded
that SI has significant bearing on the adoption of eBidding. The result is also consistent
with several studies that shows e-government systems usage is affected by social
influence (Karahanna and Straub, 1999; Rosen, 2004; Venkatesh et al., 2003).

174

5.9.4 Hypothesis Testing of Facilitating Conditions

H4: Facilitating conditions is significantly related to officials adoption of

eBidding

The results in Table 44 show that the hypothesis on facilitating conditions is supported.
Based on the coefficient (=-0.606, CR=-3.715, p=0.000<0.05), facilitating conditions is
proven to be significantly correlated to the eBidding adoption. Thus, the hypothesis null
is rejected at =0.05.

In the case of government procuring officials, it can be concluded that there is

significant relationship between facilitating conditions and eBidding use. The finding is
consistent with the results of Hung et al., (2006) work on e-government systems usage
that found the influence of supporting conditions on e-government adoption.

175

5.9.5 Hypothesis Testing of Information Quality

H5: Information quality is significantly related to eBidding adoption

The results in Table 44 show that the hypothesis on information quality is supported.
Based on the coefficient (=1.609, C.R=3.217, p0.001<0.05), information quality is
determined to have a significant correlation with the dependent variable hence, the null
hypothesis is rejected at =0.05.

In the case of government procuring officials, we were able to conclude that information
quality has an influence on the adoption of eBidding. The result is consistent with the
findings of Nelson et al., (2005) empirical work that found the influence of information
quality on the adoption of e-government systems.

176

5.9.6 Hypothesis Testing of System Quality

H6: System quality is significantly related to eBidding adoption

The results in Table 44 show that the hypothesis on system quality is not supported.
Based on the (=-0.044, C.R=-1.176, p=0.239>0.05), system quality is determined to
have an insignificant correlation with the dependent variable thus, the null hypothesis is
rejected at =0.05. It suggests that system quality is not a significant predictor of
eBidding adoption.

Therefore, in the case of government procuring officials, it can be concluded that

system quality has no effect on eBidding use. The result is inconsistent with the findings
of Wixom and Todd, (2005) study which found system quality influences satisfaction on
an IS usage (Nelson et al., 2005).

177

5.9.7 Hypothesis Testing of Service Quality

H7: Service quality is significantly related to eBidding adoption

The results in Table 44 show that the hypothesis on service quality is not supported.
Based on the ( =0.106, CR=1.768, p=0.077>0.05), service quality is determined to have
an insignificant correlation with the dependent variable. The study finding shows that
service quality insignificantly influences eBidding adoption. It suggests that the service
quality is not a significant predictor of eBidding adoption.

In the case of government procuring officials, we are able to conclude that service
quality is not an influencing factor of eBidding adoption. The result is not consistent with
the findings of Nelson et al., (2005) work which found that service quality influences
user satisfaction and e-government adoption.

178

5.9.8 Hypotheses Testing Mediating Effects of Satisfaction

H5a: Satisfaction significantly mediates relationship between information quality

and eBidding Adoption
H6a: Satisfaction significantly mediates relationship between system quality and
eBidding Adoption
H7a: Satisfaction significantly mediates relationship between service quality and
eBidding Adoption

The findings in Table 35 show information quality influenced eBidding adoption by the
officials (=0.73, C.R=7.658, p<0.05), information quality also influenced satisfaction
(=0.66, C.R=7.09, p<0.05). There is a significant influence adoption on satisfaction
(=0.494, C.R.=4.62,p<0.05). Direct effect is reported to be 0.553 versus indirect effect
0.249. It can be concluded that satisfaction becomes partial mediator between
information quality and adoption of eBidding.

Based on the results in Table 33, system quality insignificantly influence eBidding
adoption (=0.069, CR=0.642, p>0.05) while system quality significantly influence
satisfaction (=0.638,C.R.=6.044, p<0.05) and there is significant relationship between
satisfaction and eBidding adoption (=1.464, C.R.=9.771, p<0.05). Direct effect is 0.039
while indirect effect is 0.523. Hence, it can be concluded that Satisfaction fully mediates
the relationship between system quality and adoption of eBidding.

179

The results in Table 37 show there is significant relationship between service quality and
adoption (=0.462, C.R=3.02, p<0.05). Meanwhile service quality is found to be
significantly related to satisfaction (=0.943, C.R.=7.935, p<0.05).

Satisfaction is

proven to be significantly related to eBidding adoption (=0.952, C.R=6.266, p<0.05).

Direct effect is reported to be 0.272 while indirect effect is recorded to be 0.528. Based
on the findings, satisfaction is proven to be partial mediator between service quality and
adoption. Therefore in the case of government procuring officials, we are able to confirm
the mediator role of satisfaction on the relationships with technology factors and
adoption of eBidding.

5.9.9 Hypotheses Testing on Moderating Effects of Personal Innovativeness in IT

(PIIT)
H1a: PIIT positively moderates the relationship between performance expectancy
and eBidding adoption
H2a: PIIT positively moderates the relationship between effort expectancy and
eBidding adoption

Based on the statistical results, PIIT is found to be not a moderator variable for
relationship between performance expectancy and eBidding adoption. The results of the
study (as in Table 38) show performance expectancy has a positive influence on
adoption (DF

2,147

=56.968,p<0.05) and there is an insignificant relationship between PIIT

and adoption (p>0.05). In the second model it is proven that PIIT does not become
significant predictor between performance expectancy and adoption (p>0.05).

180

From the results, there is also no interaction between effort expectancy and PIIT. This is
an indication that PIIT does not have moderating effects on the relationships between
effort expectancy with adoption of eBidding. The results of the study (as in Table 39)
show effort expectancy does influence adoption (DF

2,147

=107.514, p<0.05) but PIIT has

no significant relationship with adoption. In the second model effort expectancy did not
influence adoption (p>0.05). However, the interactions between effort expectancy and
PITT do not influence adoption of eBidding. This indicates that PIIT not a moderator
variable between relationship between effort expectancy and adoption of using eBidding
(p>0.05).

In the case of government procuring officials, we are able to conclude that PIIT has no
moderating influence on the relationship between performance expectancy and effort
expectancy with adoption of eBidding. These findings are not consistent with the
findings of other studies which find the moderating effects of PIIT on relationships
between user factors with decision to accept IS (Agarwal and Prasad, 1998 ; Agarwal
and Karahanna, 2000).

181

5.9.10 Hypotheses Testing on Moderating Effects of Experience

H2b: Experience will negatively moderate the relationships between effort
expectancy with eBidding adoption
H3a: Experience will negatively moderate the relationships between social
influences with eBidding adoption
H4a: Experience will positively moderate the relationships between facilitating
conditions with eBidding adoption

Based on the statistical results, experience is found not to have moderating effect on
the relationship between effort expectancy and eBidding adoption. The findings in Table
40 indicate effort expectancy is significantly related to experience and eBidding
adoption (DF

2,147

=113.824,p<0.05). In the second model, effort expectancy has

significant relationship with adoption however, relationship experience-adoption show

non-significant results. The interaction between effort expectancy and experience also
produced non-significant results which indicates that experience is not a significant
moderator between effort expectancy and eBidding adoption (p>0.05).

Experience is also found not to have moderating effects on the relationships between
social influence and eBidding adoption. The results of simple regression in Table 41
shows both SI and experience have significant relationship with adoption (DF
2,147

=154.18, p<0.05). In the second model, only social influence has significant

relationship with adoption. The interactions between social influence and experience
does not yield significant results (p>.0.05). It is an indication that experience is not a
significant moderator between social influence and adoption.

182

Experience is also found not to have a moderating effect on the relationship between
facilitating conditions and eBidding adoption. The findings in Table 42 show both
facilitating conditions and experience have significant relationship with eBidding
adoption (DF

2,147

=28.098, p<0.05). In the second model, both facilitating conditions and

experience yield significant results. However, the interactions between facilitating

conditions and experience failed to produce significant relationship with adoption
(p>0.05). This is an indication that experience is not a significant moderator between
facilitating conditions and eBidding adoption.

Therefore, in the case of government procuring officials, we are able to conclude that
experience has no moderating influence on the relationships between effort expectancy,
facilitating conditions and social influence with eBidding adoption.The findings are
inconsistent with the findings of other research that found the moderating effects of
experience on the relationships between user behaviour factors with decision to accept
and use an information system (Venkatesh and Davis, 2000).

183

CHAPTER 6

CONCLUSIONS

6.1 Introduction

This chapter will discuss the study findings, recommendations for future research and
the conclusion. The research findings will be discussed in Section 6.2 while study
limitations will be explained in Section 6.7. Implications of research findings will be
discussed in Section 6.8. Recommendations for future research will be provided in
Section 6.9. The conclusion of the study will be explained in section 6.10.

6.2 Findings

The study results associated with users demographic characteristics, respondents

position and working experience, locations and department size and eBidding usage will
be briefly summarised and discussed. Other important key findings according to the
research objectives will be briefly summarised and discussed later in this chapter.

184

6.2.1 Demographic Characteristics

a. Gender and Age

Demographic characteristics of the respondents sample (as in Table 7) indicate that

female users are almost equal with the number of male users (female 46% : male
54%). The middle age group comprises of almost half of the respondents (age 31-40
years : 53.3%); young users (age less than 30 years); and a small group of older users
(age above 50 years : 6.7%). The proportion of younger users to older users is roughly
two to one. This indicates that young and middle-aged officials played important roles in
undertaking procurement related tasks in the Malaysian public sector.

b. Respondents Position dan Working Experience

By looking at the professional designation (as in Table 8), more than half of the
respondents (53.3%) are at the managerial position, followed by 45.3% in the
supporting position and 1.3% are the department heads. In terms of working
experiences, 35.3% of the respondents have 1-3 years working experiences, 28.7% of
the respondents have 4-6 years working experiences,18% less than 1 year and 18%
more than seven years. The experience and knowledge of users in the managerial
positions (53.3%) will benefit the tasks in procurement transaction by guiding and
managing the support group (45.3%) in completing the tasks. Only 1.3 % of the users
are in the higher positions (department heads). In a government department, the
support group will be tasked to complete the transaction under the supervision of the
185

middle managers. The department heads will act as advisors and provide strategic
directions to the overall procurement decisions. The majority of the officials have
adequate working experience (more than 5 years), which indicate familiarity with the
eBidding system. Since the eBidding has been operational for 7 years, it is assumed
that majority of the users have adequate knowledge and expertise to operate the
system.

c. Locations and Department Size

From the descriptive analysis of the locations and department size (as in Table 9), there
is evidence that 93% of the respondents are based in Putrajaya while the rest are in
other states (Shah Alam, Seremban and Kuala Lumpur). The main bulk of the
procurement transactions for the public sector are undertaken in Putrajaya, the nations
administrative capital. The majority of the Ministry and key departments are located in
Putrajaya, where the procurement and other purchases are tasked to the departments
of development and planning.

The main aim of this study is to examine the proposed research framework to evaluate
the procurement officials adoption of eBidding. The research incorporated an empirical
study through the development of a questionnaire to measure procuring officials
adoption of the electronic reverse auctions. The research problem that is addressed in
this study is to examine the key behavioural and system quality factors that affect a
government sourcing officials adoption of eBidding. The objectives of the research are

186

to establish and test a research model in a multi-dimensional, mediating (satisfaction)

and moderating (PIIT and experience) relationships between behavioural and system
factors on adoption of eBidding and also to test the proposed hypotheses by employing
Structural Equation Modeling.

6.3 Results from Hypotheses Testing

The proposed framework in the study is drawn from an overview of theoretical models
particularly, from UTAUT, IS Success Model and Personal Innovativeness in Information
Technology (PIIT) research stream. The proposed framework is based on the models as
discussed in chapter 2 which covers published literature on theory related to
behavioural factors of IS adoption. The chapter also discussed e-government evolution,
reverse auctions adoption in Malaysia, Malaysias government procurement auctions,
namely eBidding. Chapter 3 provides the theories for the research constructs and
provides the evidence linking the independent factors (i.e. behavioural factors and
system factors) and dependent factors (adoption of eBidding) with mediating role of
satisfaction and moderating variables of experience and PIIT. The process helps to
develop the research hypotheses. Chapter 4 operationalise the constructs and explains
the methods employed. Chapter 5 explains the descriptive analysis, findings from
hypotheses testing from the analysis of CFA and structural equation modelling.

To address the research objectives (see section 1.12), the study proposed a research
framework in an integrative model comprise of behavioural and system factors

187

constructs. For the second objective, hypotheses tests are performed using CFA which
shows that the measures fit the data. As a result, four dimensions of behavioural factors
(i.e., performance expectancy, effort expectancy, social influence, facilitating conditions)
are established as factors influencing eBidding adoption. For the third objective,
mediation and moderating analysis were performed and as the result, satisfaction is
found to be significant in mediating the relationships between system factors and
eBidding while both PIIT and experience are each found to be not a moderating factor
between behavioural factors and adoption. As for the fourth objective, measurement
development was performed using CFA and structural equation modelling (model respecification) which shows that the instruments chosen for the study fits the data. The
results from the tests support the following:
1) Performance expectancy is significantly related to eBidding adoption by
procuring officials (=0.106, C.R.=2.64, p=0.008 <0.05);
2) There is significant relationship between effort expectancy and eBidding
adoption by procuring officials (=0.407, C.R=4.212, p=0.00<0.05);
3) There is a significant relationship between social influence and eBidding
adoption by procuring officials (=0.307, C.R.=3.57, p=0.000<0.05);
4) There is a significant relationship between facilitating conditions and eBidding
adoption by procuring officials (=-0.606, CR=-3.715, p=0.000<0.05);
5) There is significant relationship between information quality and eBidding
adoption by procuring officials (=1.609, C.R=3.217, p=0.001<0.05);
6) There is an insignificant relationship between system quality and eBidding
adoption by procuring officials (=-0.044, C.R=-1.176, p=0.239>0.05); and

188

7) There is an insignificant relationship between service quality and eBidding

adoption by procuring officials ( =0.106, CR=1.768, p=0.077>0.05);
8) The role of satisfaction results is to show that it acts as mediator between
system quality, service quality and information quality with eBidding adoption
by procuring officials :

i.

There is significant relationship between

information quality and

satisfaction (=0.73, C.R=7.658, p<0.05). Satisfaction is significantly

related to eBidding adoption (=0.66, C.R=7.09, p<0.05), and information
quality is significantly related to adoption (=0.494, C.R.=4.62,p<0.05).
Direct effect is reported to be 0.553 versus indirect effect 0.249. It can be
concluded that satisfaction becomes partial mediator between information
quality and adoption of eBidding;

ii.

There is an insignificant relationship between system quality and adoption

(=0.069, CR=0.642, p>0.05). Satisfaction is significantly related to
system quality (=0.638,C.R.=6.044, p<0.05) while there is a significant
relationship between satisfaction and adoption (=1.464, C.R.=9.771,
p<0.05).

Direct effect is reported to be 0.039 while indirect effect is

recorded to be SC=0.523. Hence, in can be concluded that Satisfaction

becomes full mediator between system quality and adoption of eBidding
by procuring officials;

189

iii.

There is significant relationship between service quality and satisfaction

(=0.462, C.R=3.02, p<0.05). Satisfaction is significantly related to
eBidding adoption (=0.943, C.R.=7.935, p<0.05) and service quality is
significantly related to adoption =0.952, C.R=6.266, p<0.05). Direct effect
is reported to be 0.272 versus indirect effect is recorded to be 0.528. It can
be concluded that satisfaction becomes partial mediator between service
quality and adoption of eBidding

9).The moderating roles of PIIT and experience were tested and the results
indicate that both PIIT and experience do not have moderating influence on the
relationships between behavioural factors and eBidding adoption by procuring
officials as stated in the following findings:

i.

Based on the statistical results, PIIT is found not a moderator variable for

ii.

relationship on performance expectancy and eBidding adoption;

PIIT is also confirmed not a moderator on effort expectancy with eBidding

iii.

adoption by the officials;

Based on the statistical results, experience is found not to have a

iv.

moderating effect on effort expectancy and eBidding adoption;

Experience is proven not to have moderator effects on social influence

v.

and eBidding use; and

There is no moderating interaction between facilitating conditions and
experience indicating experience does not have a moderating effect on
facilitating conditions and adoption in using eBidding.

190

6.4 Determining Factors in UTAUT

The first determining factor in the UTAUT model, performance expectancy refers to how
the users believe that by using the system will facilitate them in achieving the outcome
in their work (Venkatesh et al., 2003). The finding indicates that the more the procuring
officials believe that the IS will enhance their job achievements, the more he or she is
willing to use eBidding. The second determining factor is effort expectancy which refers
to the degree of ease to use the IS (Venkatesh et al., 2003). The study also finds that
the effort expectancy has significant relationships with the dependent variable.
Therefore in the case of government procuring officials, it can be concluded that effort
expectancy has significant bearing on the adoption of eBidding. Social influence is the
third determining factor in UTAUT. Social influence refers to how a user believes his/her
peers or supervisors should use the system (Venkatesh et al., 2003). The study also
finds that the hypothesis on social influence is supported. In the case of government
procuring officials, it can be concluded that social influence has significant bearing on
the adoption of eBidding. Facilitating conditions refers to how the user believes the
infrastructure facilitates the IS usage (Venkatesh et al., 2003). Facilitating conditions is
also found to have a significant correlation with the dependent variable. Therefore in the
case of government procuring officials, it can be concluded that facilitating conditions is
significantly related to eBidding adoption.

These results are consistent with the findings of previous works which found that
performance expectancy, effort expectancy, social influence, facilitating conditions have

191

significant influence on the use of information systems (Al-Qeisi, (2009) ; Louho et al.,
2006 ; Helaiel, 2009 ; Rosen, 2004; Venkatesh et al., 2003). The findings demonstrate
the validation of UTAUT model in the examination of user factors in an information
system adoption in e-government setting.

6.5 The Determining Factors in ISSM

Service quality refers to the services provided by the system provider to the users of the
system (Delone and Mclean, 2003). From the study, it is shown that there is an
insignificant relationship between service quality and eBidding adoption. The second
determining factor, information quality is referred to the output quality of the IS Nelson et
al., (2005). Information quality is found to have significant influence on eBidding
adoption. As the third determining factor, system quality is referred to as performance
quality of the system. In the study, system quality is found not to have significant
correlation with the dependent variable. Study results show that procuring officials will
be more interested in using eBidding if the system is able to provide them with useful
information that will increase employee satisfaction in their work.

It is imperative that overall eBidding training needs to be aligned with updated

information about the daily operations of the officials and a good user interface to
improve the officials motivation to use the system, As such, the finding is consistent
with the findings of Nelson et al., (2005). However, service quality and system quality
are proven not to significantly influenced eBidding adoption. The findings are

192

inconsistent with the findings of Nelson et al., (2005) study that found SQ have an effect
on intention, usage and user satisfaction in e-government adoption.

The explanations for these results might be due to the officials perception that the
eBidding is a well-designed and high performance system. Since eBidding use is not
mandatory, the users might consider the system performance is not that important,
since the system is always operational and reliable with good technical support as
expected by the officials. In addition, eBidding was installed in the officials work stations
to enable its use. As the system is not a complex and difficult application, it needs no
extra support from technical support in the vicinity of the RCs. This might explain the
reason why the officials do not perceive the significance of support services of eBidding.
Therefore, the officials interpreted the factor as not significant in the use of eBidding.
Halonen and Martikainen (2011) also found that service quality of the system is not
significant in the use of an IS system as long as the application is easy and effortless to
use in a working environment.

193

6.6 Additional Variables

6.6.1 Mediating role of Satisfaction

In the study, satisfaction was examined as mediator in the relationship between system
factors with eBidding adoption. Satisfaction refers to users feeling about their
experience with the system. The experience starts from the interactions with the system
until reaching the overall output, including the support provided by the system (DeLone
and McLean, 2003). From the results, the study found that satisfaction becomes full
mediator between system quality and adoption (p<0.05). Satisfaction is found to
mediate (partial) relationship between information quality with eBidding adoption.
Similarly, satisfaction is also found to mediate (partial) relationships between service
quality with eBidding adoption. Hence, it can be concluded that satisfaction becomes full
mediator in the relationship between system quality and adoption of eBidding but
satisfaction becomes partial mediator between information quality and adoption of
eBidding; and between service quality and adoption of eBidding.

The implications of the findings, among others are eBidding implementers need to
ensure the satisfactory needs of procuring officials on availability of quality, relevant and
complete information. Firstly, the quality of information generated by eBidding seems
satisfactory for the officials because if the outputs are incomplete and unusable, this
may influence the officials perceived quality of eBidding. The beliefs will subsequently
affect officials satisfaction and influence the decision whether to use or not to use the

194

system. Secondly, if the reliability, quality and service of the systems are low, adoption
will be low because of low user satisfaction. The system quality of the eBidding system
has a significant relationship with satisfaction and adoption. The informational
dimension refers to providing all the necessary information for transactions and it has a
positive influence on the evaluation of eBidding. Thirdly, a high-quality and reliable
design of eBidding positively influences the user satisfaction to use the system. When
officials perceive a reliable, complete information and availability of service in eBidding,
their adoption behaviour is influenced in a favourable way mediated by satisfaction.

6.6.2 Moderating Role of Personal Innovativeness in the Domain of Information

Technology (PIIT)

In the study, the mediator role of Personal innovativeness in IT (PIIT) on the

relationships between performance expectancy and effort expectancy with eBidding
adoption was examined. From the analysis, PIIT was found not a moderator variable for
relationship between performance expectancy and eBidding adoption. There are also
no interactions between effort expectancy and PIIT which indicates that PIIT does not
have a moderating effect on the relationships between effort expectancy and adoption
of eBidding.

In the case of government procuring officials, we are able to conclude that PIIT has no
moderating influence on the relationships between performance expectancy and effort
expectancy with adoption of eBidding. However, the results are not consistent with the
findings of other works that found there is significant moderating effects of PIIT on

195

relationships between behaviour factors with decision to accept an IS (Agarwal and

Karahanna, 2000). In addition, Yi et al., (2006) investigated the moderating effect of
personal innovativeness on technology use and they posited that PIIT would moderate
the relationship between users perception and technology use.

In general, users with higher PIIT are more willing to try new innovation and are
expected to be more confident to use them. In contrast, users with lower PIIT are more
likely to encounter anxiety related to new technology (Agarwal and Karahanna, 2000;
Thatcherand Perrewe, 2002). Personal innovativeness (PI) is one of the variables that
potentially affect how people respond to innovations. PIIT also suggests that users with
higher PIIT have stronger intention to use regardless of other peoples opinion or
experience. However, these factors do not fully reflect user personality traits of
technology adoption because users could have favorable and unfavorable views of an
IS adoption (Parasuraman, 2000). In the case of the government procuring officials,
some officials might have the propensity for risk taking to try the new IS. On the other
hand, other officials might be more suspicious and hesitant to try the new system.

Perhaps, the PIIT needs to be re-defined whether it is a viable

indicator for

innovativeness since the scope specific of innovativeness is subjective and broad.

Personal innovativeness in IT (PIIT) is defined as an individual innovativenes within the
scope of IT. The question is whether the definition also equally applies to e-business
and e-government systems. It could be argued that a user who is innovative in mobile
applications may not also be innovative in e-commerce, even though they are both in

196

the IT domain. As an intrinsic trait, every user has different levels of innovativeness
which influence their acceptance behavior (Rogers, 2003). This finding was supported
by a study by Rosen A. (2004) that found PIIT did not have a moderator role between
performance expectancy and effort expectancy and behaviour intentions.

6.6.3 Moderating Role of Experience

Experience was examined as having moderating role in the relationship between effort
expectancy, facilitating conditions and social influence with eBidding adoption.
Experience is users experience whether they are affected since the last time they
employ the IS (Venkatesh, et al., 2003). From statistical results, experience was found
not to have a moderating effect between effort expectancy and eBidding adoption.
Experience is confirmed not a moderator between SI and eBidding usage. Experience is
also found not to be a moderator between social influence and eBidding usage.
Similarly, there is no moderating interaction between facilitating conditions and
experience indicating experience does not have a moderating effect on facilitating
conditions and adoption in using eBidding.
The results are not consistent with the findings of other studies that found the
moderating effects of experience on the relationships between behaviour factors with
decision to accept IS (Venkatesh and Davis, 2000 ; Venkatesh et al., 2003).
Technology Adoption Model 2 (TAM2) explains that experience is theorised as a
moderator and experience significantly moderates the influence of subjective norm
(social influence) on behaviour intention (Venkatesh and Davis, 2000). The results can
be understood in the context of officials situation. Given the countrys e-commerce
197

diffusion and high availability of the internet, the officials have been exposed to ebusiness and online auctions trading such as Lelong.com. As such, it is possible that
they may not be affected by the perception of performance, peer pressure and
availability of facilities related to eBidding use due to their familiarity with similar
applications in the Internet.In addition, It should be noted that this study did not measure
the affect of voluntariness as a moderator because the the eBidding use is not a
mandatory for officials in the procurement transactions. In previous research on UTAUT,
voluntariness and experience constructs were both tested and found as key moderators
in the relationships between behaviour factors and IS acceptance (Venkatesh et al.,
2003).

198

6.7 Study Limitation

This study was conducted in Klang Valley and Seremban with well-equipped and highly
trained officials that implemented the main bulk of the procurement transactions for the
government. Consequently, it is difficult to generalize the findings of this study to the
whole civil service in Peninsular and Sabah Sarawak. However, the findings have
provided new perspective with regards to user behaviour of sourcing officials on
technology acceptance within the context of government settings.

The next concern is about the survey accuracy. Officials had a period of one month to
work on the questionnaires during the busiest period finalizing financial and project
planning for the year. Since it is conducted in early February 2013, some departments
have not started eBidding transactions, as such there is a possibility that they hastily
complete the survey questions. With the short time frame, caution should be taken in
generalizing the study findings.

Another study limitation is the sample size. If similar research is to be replicated in the
future, the available sample size should be increased. With bigger sample size,
hypotheses could be better analysed and more generalisable. The limitation concerns
the sample size in the SEM analysis as she study sample size is 150 (less than 200
cases). The most common SEM estimation procedure is MLE and the recommended
minimum sample size to ensure stable MLE solution is 150 - 200 cases (Hair et al,

199

2006). In this regard, caution is required before generalising the findings to the
population. It can be noted that few analyses have been conducted (more than 10 times
for each testing for the impact of moderators by using AMOS version 6.0) and the
results were found consistently the same, with no unstable results found. Therefore the
results were deemed to be valid.

6.8 Implications of Research Findings

6.8.1 Theoretical Implications

The research employed in a model of technology acceptance and use that provides new
understanding of the effects of system factors and usage behaviour of a procurement
auctions system in public sector e-government environment. The findings provide
several theoretical implications:

a) The study successfully introduced and subsequently validated four constructs in

the UTAUT model (performance expectancy, effort expectancy, social influence,
facilitating conditions) and one construct (information quality) in the updated
DeLone and McLean (2003) ISSM model mediated by satisfaction in a public
sector reverse auctions environment. This would be the first attempt to apply
such a research model in exploring usage behaviour and system factors in
government auctions setting. Also, the study took place in an area that has not
been studied in this way before (government procurement auctions). In addition,

200

users satisfactions with system quality, service quality and information quality
perfectly explain eBidding usage behaviour.
b) The results of study indicate the higher the perceived system performance, effort
expectancy, social influence, facilitating conditions lead to higher officials
willingness to adopt the eBidding. The role of satisfaction as mediator, between
system quality, service quality and information quality factors, increases officials
propensity to adopt eBidding. The overall goodness-of-fit results provided the
statistical evidence to be generalised to the responsibility factors (RCs) of the
public sector as a whole.

c) Apart from the moderating variables (age and gender) the study validated the
UTAUT constructs in a public sector procurement auctions environment. The
results also contribute new findings on G2B auctions system. Due to the
instruments by Vankatesh et al., (2003) employed in the research, the results
enhance the validity of this research in an e-government setting.
d) In terms of methodological implications, SEM is recommended to be employed
for model testings. There are various benefits of SEM over other multivariate
techniques. SEM can provide estimates of error variance parameters, while
multivariate techniques are not able to correct measurement error. SEM can
analyse both latent variables and observed variables, whereas multivariate
techniques can only analyze observed measurements (Tong, 2007). In the case
of small sample size (less than 200 cases), bootstrapping procedure calculates

201

a X2 value (adjusted chi-square) against the originally obtained chi-square and an

adjusted p-value is then computed (Arbuckle, 2005).
6.8.2 Practical Implications

The results from the study find behavioural factors, such as performance expectancy,
effort expectancy, social influence, facilitating conditions and satisfaction with system
factors affect the use of eBidding. The implications of the results provide significant
benefits for public sector stakeholders i.e. e-government planners and administrators.

a) The findings indicate that officials perception of system performance

expectancy and effort expectancy, supporting infrastructure, peers influence
and satisfaction with the system factors of the information system appear to
impact positively on the use of eBidding. The findings indicate that the egovernment system administrators can devise procurement auctions
implementation plan/program with consideration for variables such as
satisfaction with system factors in using the technology. The findings also
indicate that the vast majority of respondents felt satisfied and at ease using
the system. It is believed that the officials were affected by their positive
beliefs of the performance, their peers and co-workers perceptions and the
infrastructure support for eBidding. Past experience with similar system did
not influence the willingness of officials to use the IS. The innate
innovativeness and risk taking attributes also did not moderate the willingness
to try or use the system. The training and communication programmes by

202

MAMPU and Ministry of Finance (MoF) might have successfully overcome the
officials hesitation in using eBidding with a well-designed, reliable and
satisfactory system.
b) The study confirmed Venkatesh et al., (2003) argument that SI significantly
influenced user intentions. Program managers and the early adopters of the
egovernment applications who supported the system would be likely to tell
their peers about the benefits of using the egovernment system. Other
officials would be influenced to use the e-government system and the
organisation would have a higher success rate of implementation. The
outcome of this study can be used to evaluate end-user satisfaction and to
aid in design, evaluation, and implementation of similar procurement auctions
systems.
6.8.3 Policy Implications

This research has several policy implications:

a) The findings indirectly assert that a majority of procuring officials considered

eBidding as reliable and productive system. However the study found that
manual system i.e. quotation system is more preferable due to the other
attributes such as (i.e. product specifiability, competitiveness and price
transaction value and human interactions with the suppliers. This information can
be an input in improving the eBidding process system.

203

b) The findings have implications at a national level. For example, this research
benefits policy makers in devising meaningful e-government procurement IT
policy in line with the national accountancy and transparency polices.

6.9 Recommendations for Future Research

This study discovers interesting facts regarding adoption and usage behaviour of a
procurement auctions in the Malaysian government. Subsequent studies could have
several different focuses as the following:

a) Further studies could be undertaken with similar constructs in other areas in the
Peninsular and Sabah Sarawak. A study could be attempted to validate the
proposed research model in an e-government procurement auctions system use
settings with the inclusion of the suppliers or in mandatory vs. voluntary settings
if and when the eBidding becomes mandatory by the government. Using the
proposed research model as framework, it would benefit stakeholders and
information system researchers to conduct a longitudinal (change of attitude over
time) study involving a cross-section of Responsibility Centers (RCs) in the
Malaysia government.
b) A longitudinal study is proposed to further examine the causal status of the
constructs observed in the study. Cresswell (2005) argued that a longitudinal
study could strengthen the validity of the results by examining the user
responses and the changes in the decisions over a longer period.

204

c) Contrary to previous studies, the study findings show the non-moderating effects
of experience and personal innovativeness in IT. If experience and personal
innovativeness in IT are removed as moderating factors, future researches could
focus

on

reverse

auctions

attributes

(i.e.

value,

product

specifiability,

competitiveness) and tacit knowledge and the effects of technology adoption in

the context of e-government environment.

205

6.10 Conclusions

The study has attempted to fulfil the gap in a G2B procurement reverse auctions
literature. The study has formulated and examined a research model to investigate the
behavioural factors, system factors, the role of satisfaction, the role of experience and
the role of personal innovativeness in IT as experienced by the procuring officials in an
e-government setting. The hypotheses were tested in a cross-sectional survey of
officials operating in the Responsibility Centers (RCs) or Pusat Tanggung Jawab in
Klang Valley of Malaysia.

The results of this study indicate that performance expectancy, effort expectancy, social
influence and facilitating conditions significantly influenced the adoption behaviour of an
information system. System quality and service quality are proven to be not significantly
associated with eBidding adoption. Experience and personal innovativeness in IT are
confirmed not to exhibit mediating effects on relationships between user factors and
eBidding adoption. The mediation results confirmed that satisfaction is found to have a
full mediating effect on system quality and adoption and partial mediating effect on
information quality with adoption and service quality with adoption of eBidding. The
moderating analysis affirms that the role of experience is found not to be a moderator
between effort expectancy with adoption; social influence with adoption and facilitating
conditions with adoption. The moderating role of personal innovativeness in IT was also
found not to be a moderator between performance expectancy with adoption and effort
expectancy with adoption.

206

Given the significant impact of behavioural factors of officials in eBidding adoption and
role satisfaction, the organisational administrators and managers could introduce key
changes in the workplace to improve the satisfaction with the e-government system. It is
hoped that these changes will enhance the efficiency of the procuring officials in the
working environment. In conclusion, an efficient and effective e-procurement reverse
auctions system is important for an increasingly transparent and efficient public sector.

207

REFERENCES
Abdullah, R., and Zahadi E., D.,(2011).A framework of intelligent bidding system (iBS) of
registration number for road Transport Department, published in Research and Innovation in
Information Systems (ICRIIS), 2011 International Conference
Abdulhadi H.A., (2013), An Empirical Study of Online Shopping Customer Satisfaction in
Malaysia Evaluating eBay Malaysia, Journal Of Computing and Organisational Dynamics,
Vol.1, No.2, February 2013 pp.36-62
Accenture, (2011). Leading Governments Shifting Focus from e-government to e-Governance
Using New Technologies and Modified Service Delivery Models, According to Accenture
Report. Retrieved from http://newsroom.accenture.com/article_display.cfm?article_id=4880 (2
May 2011)
Agarwal, R., and Prasad, J. (1998). A Conceptual and Operational Definition of Personal
Innovativeness in the Domain of Information Technology. Information Systems Research, 9(2).
204-215
Agarwal, R. and Karahanna, E. (2000). Time flies when youre having fun: Cognitive absorption
and beliefs about information technology use. MIS Quarterly, 24(4). 665-694.
Ahmad, M. and Othman, R. (2007). Implementation of electronic government in Malaysia: The
status and potential for better service to the public. Public Sector ICT Management Review.
1(1). 2-10.
Akmaliah K. A. (2007), Online marketplace service innovation: the case of Lelong.com.my ,
International Journal of Technoentrepreneurship, Vol. 1, No. 2, 2007
Al-Qeisi, K. I. 2009. Analyzing the Use of UTAUT model in Explaining an Online Behaviour:
Internet Banking Adoption. PhD Thesis, Brunel University: London, UK, March 2009.
Amin, H., M. R. A. Hamid, S. Lada, and Z. Anis, (2008). The adoption of mobile banking in
Malaysia: The case of Bank Islam Malaysia Berhad, International Journal of Business and
Society, Vol.9, No. 2:43-53, 2008
Anthony P., and Law E. (2012). Reserve price strategy for seller agent in multiple simultaneous
auctions. International Journal of Knowledge-based and Intelligent Engineering Systems 16
(2012) 163175 163.
Anthony P., (2005), Overview of Electronic Auctions, Encyclopaedia of Information Science
and Technology, Second Edition pg 2953 - 2957
Anthony P.,(2009), Seller Agent for Online Auctions, The Second International Conference on
Innovations in Information Technology (IIT05).
Anderson, J. E., and Scwager, P. H. (2004). SME adoption of wireless LAN technology:
Applying the UTAUT model. Paper presented at the 7th Annual Conference of the Southern
Association for Information Systems. Retrieved from http://sais.aisnet.org/sais2004/Anderson
%20&%20Schwager.pdf. (November 14, 2010)

208

Arbuckle, J. L. (2005). Amos 6.0 Users Guide. United States of America: Amos
DevelopmentCorporation.
Babbie, Earl. (2010). The Practice of Social Research, 10th edition, Wadsworth,
Thomson Learning Inc.
Bandura, A. (1986). Social Foundations of Thought and Action: A Social Cognitive Theory.
Englewood Cliffs, N.J.: Prentice-Hall
Bartlett, J.E., Kotrlik, J.W., Higgins, C.C. (2001). Determining appropriate sample size in survey
research. Information Technology, Learning, and Performance Journal, 19 (1), pp. 43-50.
Beall, S., Carter, C., Carter, P. L., Germer, T. and Jap, S, (2003). The Role of Reverse Auctions
in
Strategic
Sourcing.
CAPS
Research,
Retrieved
from
http://www.capsresearch.org/publications/pdfs-protected/beall2003.pdf (April 12, 2010)
Belanger, F., and Hiller, J. (2006). A framework for e-government: Privacy implications.
Business Process Management Journal, 12(1), pg. 48
Braak, J.V. (2001). Factors influencing the use of computer mediated communication by
teachers in secondary schools. , Computers and Education, vol. 36, 2001, pp. 41-57.
Brown, M., Montoya-Weiss, M. M., & Burkman, J. R. (2002).Do I really have to? User
acceptance of mandated technology. European Journal of Information Systems, 11, 283-295
Cai, S. and Jun, M. (2003). "Internet users' perception of online service quality: a comparison of
online buyers and information searchers", Managing service quality, Vol. 13 No. 6, pp. 504-19.
Carter, C. R., Kaufmann, L., Beall, S., Carter, P. L., Hendrick, T. E., and Petersen, K. J. (2004).
Reverse auctions-grounded theory from the buyer and supplier perspective. Transportation
Research Part E, 40(3). 229254.
Carlsson, C., Hyvonen, K., Puhakainen, and Walden, P. (2006). Adoption of Mobile
Devices/Services-Searching for Answers with the UTAUT. Proceedings of the 39th Hawaii
International Conference on System Science-http://ieeexplore.ieee.org/stamp/stamp.jsp?
arnumber=01579556.(Retrieved: 29 Nov 2011).
Chadwick, A., and May, C. (2003). Interaction between states and citizens in the age of the
Internet: E-government in the United States, Britain, and the European Union.
Governance, 16(2). 271-300.
Chen, L., Gillenson, M. L., &Sherrell, D. L. (2002). Enticing online consumers: An extended
technology acceptance perspective. Information & Management, 39(8). 705719.
Cheung M.K., and Lee K.O. (2005). The Asymmetric Effect of Web Site Attribute Performance
on web satisfaction : An Empirical Study., e-Service Journal, Vol 3. No. 3 (Summer 200402005)
pp. 65-86

209

Chiu, C.-M., Chiu, C.-S. and Chang, H.-C. (2007). "Examining the integrated influence
of fairness and quality on learners' satisfaction and web-based learning continuance
intention", Information Systems Journal, Vol. 17 No. 3, pp. 271-87.
Cronin, J.J. Jr and Taylor, S.A. (1992). Measuring service quality: a re-examination and
extension, Journal of Marketing, Vol. 56 No. 3, pp. 55-68.
Cody-Allen, E. and Kishore, R. (2005). An Extension of the UTAUT Model with
E-Quality, Trust & Satisfaction Constructs. Retrieved on 28 September 2010,
from
http://portal.acm.org/citation.cfm?id=1125196.
Cora S. K. L. (2009), Towards a Model of Macaos E-government Portal Adoption,
Communications of the IBIMA, Volume 10, 2009 ISSN: 1943-7765
Creswell, J. W. & Miller, D. L. (2000). Determining validity in qualitative inquiry. Theory
into Practice, 39(3), 124-131.
Creswell J.W. (2009).Research Design, 3rd edition, Sage.
Crespo A. H. and Bosque I.R.,, (2008). The effect of innovativeness on the adoption of B2C ecommerce: A model based on the Theory of Planned Behaviour, Computers in Human
Behavior, vol. 24, pp. 2830-2847, 2008.
Davis, F. D. (1989). Perceived usefulness, perceived ease of use, and user acceptance of
information technology. MIS Quarterly, 13:319339.
Davis, F. D. (1992). Extrinsic and intrinsic motivation to use computers in the workplace.
Journal of Applied Social Psychology 22(14) 11111132.
Delone, W. H. and McLean, E. R. (1995). Information systems success: The quest for the
dependent variable. Information Systems Research, 3:6095.
DeLone, W. H. and McLean, E. R. (2003). The Delone and McLean model of information
systems success: A ten year update. Management Information Systems, 19(4):930.
Denis, J.L., Herbert, Y., Lozeau, D., and Trottier, L.H. 2002. Explaining Diffusion Patterns for
Complex Health Care Innovations. Health Care Management Review. Vol. 27, No. 3, pp. 60-73.
Durndell A., Haag Z. and Laithwaite H. (2000). Computer self efficacy and gender: a cross
cultural study of Scotland and Romania, Personality and Individual Differences, 28 (6). 10371044.
e-Bidding melalui system ePerolehan (2010) from ePerolehan Malaysia, retrieved from
http://home.ePerolehan.gov.my/home/images/stories/download/Surat%20Pekeliling
%20Perbendaharaan%20Bil.%203%20Tahun%202009.pdf (November 12, 2010)
eBidding
Transactions
Perolehan
Malaysia,
retrieved
from
http://home.eperolehan.com.my/v2/index.php/bm/mengenai-ep/statistik-sistem-ep (Oct,1, 2012)

210

e-Bidding is to expands ePerolehan procurement mode, says CommerceDotComSdn Bhd.

BERNAMA. Kuala Lumpur: (Dec 10, 2009)
e-Perolehan
Enabled
Pusat
Tanggung
Jawab.,
retrieved
from
http://home.eperolehan.com.my/v2/index.php/bm/senarai-ptj-ep-enabled (October 9, 2012)
Elias S.M., Smith W., L., and Barney C.,E., (2012). Age as a moderator of attitude towards
technology in the workplace: work motivation and overall job satisfaction., Behaviour &
Information Technology., Volume 31, Issue 5, 2012, pages 453-46
e-Perolehan
Enabled
Pusat
Tanggung
Jawab.,
retrieved
from
http://home.eperolehan.com.my/v2/index.php/bm/senarai-ptj-ep-enabled (October 9, 2012)
Emiliani, M. L., and Stec, D. J. (2004). Aerospace parts suppliers' reaction to online reverse
auctions.Supply Chain Management, 9(2). 139-153.
Eu-Gene, S. (2010). Malaysian C2C Auction e-Commerce Sites:A Case Study on eBay and
Lelong.com.my, MIS Quarterly
Frambach, R., and Schillewaert, N., (2002). Organisation Innovation Adoption: A Multi-Level
Framework of determinants and opportunities for future research, Journal of Business
Research, 55 (2)
Frazier, P., Tix, A.P. and Barron, K.E. (2004). Testing moderator and mediator effects in
counseling psychology research. Journal of Counseling Psychology, 51(1), 115134.
Fichman, R. G. (2000). The Diffusion and Assimilation of Information Technology Innovations,
in Framing the Domains of IT Management: Projecting the Future Through the Past, R. W.
Zmud (ed.). Cincinnati, OH: Pinnaflex Educational Resources, pp. 105-127.
Fink, A. (2008). How to conduct surveys: A step-by-step guide (4th ed.). Thousand Oaks, CA:
SAGE.
Gan K. S., Patricia, A., Jason, T, Kim O. (2009) Comparing the Performance of Deterministic
Dynamic Adaptation GA and Self Adaptive GA in Online Auctions Environment, Proceedings of
the 2009 IEEE International Conference on Systems, Man, and Cybernetics San Antonio, TX,
USA - October 2009
Garson,
G.
D.
(2009).
Structural
equation
modelling
http://faculty.chass.ncsu.edu/garson/PA765/structur.htm [2009, August 11].

[Online].

Geyskens, I., Gielens, K., and Dekimpe, M.G. (2002). "The Market Valuation of Internet Channel
Additions," Journal of Marketing (66:2). April 2002, pp 102-119
Gershon P. (2004). Independent Review of Public Sector Efficiency: Releasing Resources to
the Front Line, HM Treasury, HMSO, UK.
George, D., and Mallery, P. (2003). SPSS for Windows step by step: A simple guide and
reference. 11.0 update (4th ed.). Boston: Allyn & Bacon

211

George, P. (2007) User acceptance of the ePerolehan system among government users in
Malaysia, Unpublished Thesis, Universiti Putra Malaysia
Gefen, D. and Straub, D. (2000). The relative importance of perceived ease of use in IS
adoption: a study of e-commerce adoption. Journal of the Association for Information Systems,
1(8) 1-28.
Gefen D., Karahanna, E. and Straub, D., W., (2003). Trust and TAM in online shopping an
integrated model. MIS Quartely, 27, 51-90.
Ghobakhloo, Aziz and Zulkifli (2010). The Interactive Model of User Information Technology
Acceptance and Satisfaction in Small and Medium-sized Enterprises European, Journal of
Economics, Finance and Administrative Sciences, Issue 19
Gupta, M.P., Jana, D., (2003). E-government evaluation: a framework and case study.
Government Information Quarterly 20, 365387
Gunasekaran, A. , and Ngai, E.W.T., (2008). Adoption of E-procurement in Hong Kong: an
empirical research, International Journal of Production Economics 113, 159-175
Haque A., Alhasmi S.M., Parthiban R. (2010).Continuous Double Auction in Grid Computing: An
Agent Based Approach to Maximize Profit for Providers. Presented in 2010 IEEE/WIC/ACM
International Conference on Web Intelligence and Intelligent Agent Technology
Han, S. (2003). Individual adoption of information systems in organizations : A literature review
of technology acceptance model. Turku Centre for Computer Science (TUCS). Technical
Report No 540.
Hung, S. Y., C. M. Chang, and T. J. Yu. 2006. Determinants of User Acceptance of the EGovernment Services: the case of online tax filing and payment system, Government
Information Quarterly (23:1), pp 97-122
Hair, J.F., Black, W.C., Babin, B.J., Anderson, R.E. and Tatham, R.L. (2006). Multivariate Data
Analysis, 6th ed., Prentice-Hall, Englewood Cliffs, NJ.
Halonen, R., and Martikainen, O., (2011). "Model for the Benefit Analysis of ICT" . AMCIS 2011
Proceedings - All Submissions.Paper 19. http://aisel.aisnet.org/amcis2011_submissions/19
Hawkins, T.G., Randall, W.S., and Wittmann, C.M. (2009). An empirical examination of reverse
auction appropriateness in B2B source selection, Journal of Supply Chain Management, 45(4).
55-71.
Hazman, S.A Maniam, K. and Halimah, A. (2006). Citizens Expectations for Electronic
Government Services: Malaysian Perspectives. Paper presented at the e-Government Asia
Conference in Bangkok,Thailand, 26-28 April 2006.
Helaiel, A. 2009. Validating the Unified Theory of Acceptance and Use of Technology in Kuwait
Ministries: A Structural Equation Modeling Approach, International Journal of Information
Systems and Change Management (3:3), pp 191-219.

212

Hernandez, J.M.C. and Mazzon, J.A. (2007). "Adoption of internet banking: proposition and
implementation of an integrated methodology approach", International Journal of Bank
Marketing, Vol. 25 No. 2, pp. 72-88.
Hisham, (2008). Statistical Analysis [Online]. Retrieved from: www.hishammb.net/
doresearch/fanotes.pdf (October 12, 2012)
IBM Business Consulting Services (2003). How e-government are you? e-government in
France:
State
of
play
and
perspectives.
Retrieved
from
http://www03.ibm.com/industries/government/doc/content/bin/g510-3552-00-esr-egovernment.pdf (April 17, 2008).
Jamalludin, S., Jailani, N.,Ahmad, S., Abdullah, S., Mukhtar, M.,AbuBakar, M., Majid, M., A.,
Mokhtar, M., R., Abdullah, Z. (2011). "A Syariah compliant e-auction framework", Proceedings of
the 2011 International Conference on Electrical Engineering and Informatics, ICEEI 2011.
Jailani N., Yatim N.F.M, Yahya Y., Patel A., Othman M., (2008). Secure and auditable agentbased e-marketplace framework for mobile users, Computer Standards & Interfaces 30 (2008)
237252
Jap, S.D. (2007). The impact of online reverse auction design on buyer-supplier relationships,
Journal of Marketing, Vol. 71 No. 1, pp. 146-59.
Jeong C. H. (2007). Fundamental of Development Administration. Selangor: Scholar Press.
ISBN 978-967-5-04508-0
Jeong N., Y., Yoo Y., and Ku, J.Y., (2008). A study on moderating effect of personal
innovativeness in mobile-RFID services, In proceeding of: Advanced Communication
Technology, 2008. ICACT 2008. Volume: 1
Jeyaraj, A., Rottman, J. W., Lacity, M. C., (2006). A review of the predictors, linkages, and
biases and IT innovation adoption research, Journal of Information Technology, Vol. 21, No. 1,
pp. 123
Joshua, A. J. and M. P. Koshy, (2011).Usage patterns of electronic banking services by urban
educated customers: Glimpses from India, Journal of Internet Banking and Commerce, Vol. 16,
No. 1: 1-12, 2011.
Kaliannan, M., Halimah, A., and Raman, M. (2007). Technology adoption in the public sector:
An exploratory study of e-government in Malaysia, Proceedings of the 1st International
Conference on Theory and Practice of Electronic Governance ICEGOV '07 Macao. December
2007.
Kamarulzaman, N., H., and Rahman, A.,A. (2009). E-Auctions: A Potential Tool for the
Malaysian Agriculture Sector. Public Sector ICT Management Review, 3(2):19-24.
Kamarulzaman, Y. (2007). "Adoption of travel e-shopping in the UK", International Journal of
Retail and Distribution Management, Vol. 35 No. 9, pp. 703-19.

213

Kassim E.,S. and Hussin H., (2010). An Integrative Approach to Inter-Organizational System
Implementation and Evaluation: the Case of Government-to Business, IBIMA Publishing
Communications of the IBIMA., Vol. 2010 (2010). Article ID 694811, 12 pages
Karahanna, E., Straub, D.W. and Chervany, N.L.(1999). Information technology adoption
across time: A cross-sectional comparison of pre-adoption and post-adoption beliefs, MIS
Quarterly, 23(2). 183-213.
Kendall, J., Tung, L., Chua, K., Ng, C., and Tan, M. (2001). Electronic Commerce Adoption by
SMEs in Singapore. Hawaii International Conference on System Sciences. Hawaii.
Kim B., Choi M.and Han I., (2009). User Behaviour toward mobile services: The role of
perceived fee and prior experience. Experts Systems with Applications 36 (2009) 8528-8536
Kline, R. B. (2005). Principles and practice of structural equation modelling. United States: The
Guilford Press.
Layne, K., and Lee, J. (2001). Developing fully functional E-government: A four stage model.
Government Information Quarterly, 18, 122 -136.
Laporan Ketua Audit Negara (2010).Aktiviti Kementerian/jabatan dan Pengurusan Syarikat
Kerajaan
Persekutuan
retrieved
from
http://www.audit.gov.my/index.php?
option=com_content&view=article&id=458%3Alkan2010&catid=6%3Aberitapengumuman&lang=ms# (20 May 2012)
Laporan
Ekonomi
2010/2011
retrieved
http://www.treasury.gov.my/bajet2011/data/le/le11.html (October 10, 2012)

from

Lee, H.Y., Qu H., and Kim Y.S., (2007). A study of the impact of personal innovativeness on
online travel shopping behaviorA case study of Korean travellers, Tourism Management, vol.
28, pp. 886-897, 2007
Lee, H., Kim, J. and Kim, J. (2007). "Determinants of success for application service provider:
an empirical test in small businesses", International Journal of Human-computer Studies, Vol.
65 No. 9, pp. 796-815.
Lee, J. (2010), 10 year retrospect on stage models of e-Government: A qualitative
meta-synthesis Government Information Quarterly, 27(3), 220230.
Layne, K., and Lee, J. (2001). Developing fully functional E-government: A four stage model.
Government Information Quarterly, 18, 122 -136.
Lewis W., and Agarwal R., Sambamurthy V., (2003). Sources of influence on beliefs about
information technology use: an empirical study of knowledge workers. MISQuarterly 27(4):657
678
Li, J.P., and Kishore, R.,( 2006). How robust is the UTAUT instrument? A multigroup invariance
analysis in the context of acceptance and use of online community weblog systems. In
Proceedings of the 2006 ACM SIGMIS CPR Conference on Computer Personnel Research,
Claremont, CA, USA, pp. 183189.

214

Lin, H.F. (2006). "Understanding behavioral intention to participate in virtual communities",

CyberPsychology and Behavior, Vol. 9 No. 5, pp. 540-7.
Liu, Ben S., D. Sudharshan, and Lawrence O. H., (2000), After-Service Response in Service
Quality Assessment: A Real-Time Updating Model Approach, Journal of Service Marketing, 14
(2), 160-77
Louho, R., Kallioja, M., and Oittinen, P. 2006. Factors Affecting the Use of Hybrid Media
Applications, Graphic Arts in Finland (35:3), pp 11-21
Lu, J., Yao, J., and Yu, C. S. (2005). Personal innovativeness, social influences and adoption of
wireless internet services via mobile technology. Journal of Strategic Information Systems, 14
(3). 245-268.
Lu, J., C. S. Yu, and C. Liu, (2009). Mobile data service demographics in urban China, The
Journal of Computer Information Systems, Vol. 50, No. 2: 117-126, 2009.
Magni M., Maruping L., Caporarello L., and Basaglia S., (2011). "Innovating with technology in
team contexts: A trait activation theory perspective". ICIS 2011 Proceedings, retrieved from
http://aisel.aisnet.org/icis2011/proceedings/humanbehavior/1(December 5, 2011)
Malaysia
eBidding
module
wins
eIndia
Award.
2011,
retrieved
http://tender2u.blogspot.com/2011/03/malaysias-ebidding-module-wins-eindia.html(12
November 2011)

from

Mackinnon, D. P., Krull, J. L., & Lockwood, C. M. (2000). Equivalence of the Mediation,
Confounding, and Supporession Effect. Prevention Science, 1(4), 173-181.
Mallat N., (2007). The impact of use situation and mobility on the acceptance of mobile
ticketing services. Proceedings of the 39th Annual Hawaii International Conference on System
Sciences.
MAMPU (2009).The Role of ePerolehan in Improving Efficiency and Transparency in
Government Procurement from MAMPU, retrieved from http://arcadia.mampu.gov.my/1egov/papers/Plenary7/P7P3.pdf (January 2011)
MAMPU
(2010).
Electronic
Government.,
retrieved
http://www.mampu.gov.my/pdf/flipbook/eg2012/files/eg-book2012.pdf. (10 October 2012).

from

Maniam, K. (2008). Technology Acceptance for Government Procurement: A Study on ePerolehan in Malaysia. University of Malaya, Kuala Lumpur.
Maniam,K., A. Halimah, and S.A. Hazman, S.A. (2006). Citizens Expectations for Electronic
Government Services: Malaysian Perspectives,. The egovernment Asia Conference in
Bangkok, Thailand, 2006.
Matthew J. W., Omar F. E., and DorineB., (2008). Examining Healthcare Professionals
Acceptance of Electronic Medical Records Using UTAUT, Issues in Information Systems.,VOL
IX, No. 2, 2008 400

215

Molla, A., and Licker, P. (2001). "E-Commerce Systems Success: An Attempt to Extend and
Respecify the Delone and Mclean Model of Is Success," Journal of Electronic Commerce
Research (2:4). pp 131-141.
Moon, J. (2002). The Evolution of E-Government Among Municipalities: Rhetoric or reality?,
Public Administration Review 62(4): 424-433
Moon, J. (2005). E-procurement management in state governments: diffusion of e-procurement
practices and its determinants, Journal of Public Procurement, 5(1). 54-73.
Mukhtar M., Kurniasary, E., Jailani, N., Abdullah S., Yahya Y., and Abdullah Z. (2009).
Designing an e-Marketplace prototype for wood-based product industry in Malaysia, IEEE
2008
Murali, R., Maniam, K., and Cheng, M.Y. (2007). e-Business and E-government Issues and
Challenges in Malaysia, Information Technology Journal, 6(3). 428-434
Murphy, M. J. (2003). Computer technology for office-based psychological practice:
Applications and factors affecting adoption. Psychotherapy: Theory, Research, Practice,
Training, 40, 1019
Nelson, R., Todd, P, and Wixom, B. (2005). Antecedents of Information and System Quality : An
Empirical Examination within the Context of Data Warehousing. Journal of Management
Information Systems, 21 (4) 199-235.
OECD: The E-government imperative (2003). page 23
Oliveira, T and Martins, M, F. (2011). Literature Review of Information Technology Adoption
Models at Firm Level, The Electronic Journal Information Systems EvaluationVolume 14 Issue
1 2011, (pp. 110- 121)
Ovretveit, J. (2002). Quality Collaboratives: Lessons from Research. Quality and Safety in
Health Care. Vol. 11, pp. 345-351.
Ozdamar, K. (1999). Statistical data analysis by custom softwares-1 Eskisehir: KaanKitabev.
Pekeliling Perbendaharaan Bil. 3/2009 (2009). Tatacara Pengurusan Perolehan Kerajaan
Secara eBidding Melalui Sisteme Perolehan. Unit ePerolehan. Kementerian Kewangan. 2009
Park, J. K., S. J. Yang, and X. Lehto, (2007). Adoption of mobile technologies for Chinese
consumers, Journal of Electronic Commerce Research, Vol. 8, No. 3: 196-206, 2007
Parasuraman, A. (2000), "Technology Readiness Index (TRI): a multiple-item scale to measure
readiness to embrace new technologies", Journal of Service Research, Vol. 2 No.4, pp.307-20. Petter, S. and McLean, E. R. (2009). A meta-analytic assessment of the delone and mclean is
success model: An examination of is success at the individual level. Information and
Management, 46:159166.

216

Radhakrishnan A., and Davis S. (2004). Factors affecting supplier adoption of governmental
reverse auctions : an exploratory case study, SEAMS Association, Clemson University August
31, 2004
Ramlah H.,, Karim N.,S. A., and Selamat M., H., (2007). The impact of technological factors on
information systems success in the electronic-government context. Business Process
Management Journal Vol. 13 No. 5, 2007 pp. 613-627.
Reichheld, F., and Schefter, P. (2000). E-Loyalty, Harvard Business Review (78:4).pp105-113
Rogers, E. M. Diffusion of Innovations, The Free Press, New York. (1995).
Rogers, E., M. (2003). Diffusion of innovations (5th ed.). New York: Free Press
Rosen , A., P. (2004). The effect of personal innovativeness in the domain of Information
technology on the acceptance and use of technology: a working paper., retrieved from
http://cba.okstate.edu/~rosenp/dsi2004.pdf (15 January 2010)
Schumacker, R. E., and Lomax, R. G. (2004). A beginner's guide to structural equation
modelling (2nd ed.). Mahwah, NJ: Lawrence Erlbaum.
Siau, K., and Long, Y., (2009). Factors impacting e-government development, Journal of
Computer Information Systems, Fall 2009, 50 (1). pg. 98-99
Siddiquee A. (2008). E-government and Innovations in Service Delivery: The Malaysian
Experience. Intl Journal of Public Administration, 31: 797815, 2008
Sun, H., and Zhang, P. (2005). The role of moderating factors in user technology acceptance.
International Journal of Human-Computer Studies (IJHCS). 64(2) 53-78.
Sderlund, M. (2002). Customer familiarity and its effects on satisfaction and
intentions. Psychology & Marketing, 19(10). 861879

Behavioral

Star, 2010 Online,. Bidding of govt contracts to be made public via MyProcurement, retrieved
from http://thestar.com.my/news/story.asp?file=/2010/4/2/nation/5984619&sec=nation
(9 July 2010)
Sekaran, U. (2003). Research Methods for Business: A Skill Building Approach. USA: John
Wiley and Sons
Settoon, R and Wyld, D C (2003). The Ski Slope to Prosperity: An Analysis of the Potential
Impact of Reverse Auctions in Government Procurement in Five Southeast Asian Nations, Asia
Pacific Journal of Marketing and Logistics, 15(3). 3-19.
Singh, S., Srivastava, V., and R. K. Srivastava, (2010). Customer acceptance of mobile
banking: A conceptual framework, SIES Journal of Management, Vol. 7, No. 1: 55-64, 2010.
Thatcher, J. B., and Perrewe, P.L. (2002). An empirical examination of individual traits as
antecedents to computer anxiety and computer self-efficacy, MIS Quarterly26(4). 381-396

217

Tan, M., H. (2001) Design Of E-Procurement Online Auction System For Small Medium
Enterprise. Masters thesis, Universiti Utara Malaysia.
Tobbin, P. (2011). Adoption of Mobile Money Transfer Technology: Structural Equation Modeling
Approach European, Journal of Business and Management, ISSN 2222-1905 (Paper) ISSN
2222-2839 (Online) Vol 3, No.7, 2011 p.59
Tolbert, C. J. and McNeal R. S.. (2003). Unraveling the Effects of the Internet on Political
Participation. Political Research Quarterly 56(2): 175-185
Tsiknakis, M., and Kouroubali, A. (2008). Organizational factors affecting successful Adoption of
innovative e-health services: A case study employing the framework. International Journal of
Medical Informatics.
Tong, D. Y. K. (2007). An Empirical Study of e-Recruitment Technology Adoption in Malaysia:
Assessment of Modified Technology Acceptance Model. Unpublished doctoral thesis,
Multimedia University, Malaysia.
Tully, S. (2000). The B2B tool that really is changing the world, Fortune, Vol. 141 No. 6, pp.
132-45.
Ucapan Penggulungan Bajet 2009 (2009) from Treasury Malaysia, retrieved from
http://www.treasury.gov.my/index.php?option=com_content&view=article&id=1012%3Aucapanpenggulungan-bajet-2009&catid=53%3Aucapan&Itemid=251&lang=my(12 August 2010)
Ullman, J. (2007). Structural Equation Modelling in: B. G. Tabachnick and L.S. Fidell, Using
Multivariate Statistics. US: Pearson Education Inc, 676-780.
Urbaniak, G. C., and
http://222.randomizer.org

Plous,

S.

(1997).

Research

Randomizer,

2004,

from

UN. (2005). UN Global e-Government Readiness Report 2005. New York: United Nations
Vaidya, A., S., M., Sajeev, and Guy C. (2006). Critical factors that influence e-procurement
implementation success in the Journal of Public Procurement, 2006; 6, 1/2; Academic
Research Library pg. 70
Venkatesh, V. and F. Davis, (2000). Theoretical Extension of the Technology Acceptance
Model: Four Longitudinal Field Studies, Management Science, Vol. 46, No. 2: 186204, 2000.
Venkatesh, V., Morris, M. G., Davis, G. B., and Davis, F. D. (2003). User acceptance of
information technology: toward a unified views. MIS Quarterly, 27:425478
Venkatesh, V, and Bala H. (2008). Technology Acceptance Model 3 and Research Agenda on
Interventions , Decision Sciences 39(2):273-315
Wallen, N.E. and Fraenkel, J.R. (2001). Educational research: A guide to the process (2nd
ed.), Mahwah, NJ: Lawrence Erlbaum Associates, Publishers.

218

Wang, Y.,S,. (2008). "Assessing e-commerce systems success: a respecification and validation
of the DeLone and McLean model of IS success", Information Systems Journal, Vol. 18 No. 5,
pp. 529-57
West, D. (2004). Urban e-government, 2004. Providence, RI: Center for Public Policy, Brown
University.
Wixom, B. H., and Todd, P. A. (2005). A Theoretical Integration of User Satisfaction and
Technology Acceptance. Information Systems Research, 16(1). 85-102
Wolin, L., D., and Korgaonkar, P.(2003). Web advertising: gender differences in beliefs,
attitudes and behaviour. Internet Research: Electronic Networking Applications and
Policy.13(5). 375-385.
Wu, J. and S.C. Wang (2005). What drives mobile commerce? An empirical evaluation of the
revised TAM, Information and Management, (42) 5, pp. 719-729
Wu, Y. Tao, Y. and Yang, P., (2007). Using UTAUT to explore behaviour of 3G mobile
communications users. IEEE International Conference on Industrial Engineering
Wu J., Qi D.,(2010). Moderating Effect of Personal Innovativeness in the Model for E-store
Loyalty. 2010 International Conference on E-Business and E-Government
Wyld, D. (2011). Current Research on Reverse Auctions Part 1. Understanding the Nature of
Reverse Auctions and the Price and Process Savings Associated With Competitive Bidding,
International Journal of Managing Value and Supply Chains (IJMVSC) Vol. 2, No. 3, September
2011
Wyld, D. (2012). Reverse Auctioning Saving Money and Increasing Transparency, Retrieved
from
http://www.businessofgovernment.org/sites/default/files/Reverse%20Auctioning.pdf
(October, 10, 2012)
Yi, Y., Wu, Z., & Tung, L. (2006). How individual differences influence technology usage
behavior? Toward an integrated framework. The Journal of Computer Information Systems,
46(2), 52-63.
Yi, Y. (2012) "Specification, evaluation, and interpretation of structural equation models". Journal
of the Academy of Marketing Science, 40 (1), 834.
Zhang, P. N. Li, and H. Sun, (2006). Affective Quality and Cognitive Absorption: Extending
Technology Acceptance Research., Series Affective Quality and Cognitive Absorption:
Extending Technology Acceptance Research., ed., Editor ed., 2006, pp121.

219

Appendix 1
The Results for Pilot Study Reliability Test
GET
FILE='D:\eBidding actual and modified\eBidding pilot6.sav'.
DATASET NAME DataSet1 WINDOW=FRONT.
RELIABILITY
/VARIABLES=PE1 PE2 PE3 PE4 use1 use2 use3 use4 ee4 ee5 ee6 ee7 si1 si3 si5
si7 fc2 fc4 fc5 fc7 sys3 sys5 sys6 sys7 info2 info4 info5 info6 svq1 svq3 svq5 svq6
satis1 satis2 satis3 satis4 PIIT001 PIIT002 PIIT003 PIIT004 PIIT005 EXP002 EXP003
EXP004 EXP005
/SCALE('ALL VARIABLES') ALL
/MODEL=ALPHA
/SUMMARY=TOTAL.
[DataSet1] D:\eBidding actual and modified\eBidding pilot6.sav
Scale: ALL VARIABLES
Case Processing Summary
N
%
Valid
50
100.0
Excluded
0
.0
Cases a
Total

50

100.0

a. Listwise deletion based on all

variables in the procedure.
Reliability Statistics
Cronbach's
N of
Alpha
Items
.958
45

220

PE1
PE2
PE3
PE4
use1
use2
use3
use4
ee4
ee5
ee6
ee7
si1
si3
si5
si7
fc2
fc4
fc5
fc7
sys3
sys5
sys6
sys7
info2
info4
info5
info6
svq1
svq3
svq5
svq6
satis1
satis2

Item-Total Statistics
Scale Mean
Scale
Corrected
Cronbach's
if Item
Variance if
Item-Total Alpha if Item
Deleted
Item
Correlation
Deleted
Deleted
175.3000
1589.520
.580
.957
175.0800
1592.565
.636
.956
174.9800
1582.020
.679
.956
175.0800
1563.544
.719
.956
174.8400
1565.770
.795
.955
175.2200
1569.889
.867
.955
175.4400
1570.904
.825
.955
174.4800
1576.051
.691
.956
175.3400
1578.270
.662
.956
174.8800
1598.557
.598
.957
174.6400
1608.807
.641
.956
175.0800
1561.381
.889
.955
175.2400
1598.758
.676
.956
175.1400
1595.388
.777
.956
174.7400
1608.768
.596
.957
175.0800
1595.626
.686
.956
174.7000
1647.439
.418
.957
174.6600
1650.515
.291
.958
174.7400
1638.564
.406
.957
175.2000
1613.633
.466
.957
175.2200
1584.828
.682
.956
175.0000
1630.408
.441
.957
174.9200
1628.279
.472
.957
174.5600
1615.272
.450
.957
175.1000
1625.439
.542
.957
175.1400
1619.633
.603
.957
175.1200
1625.863
.592
.957
175.2600
1562.196
.820
.955
174.8200
1619.783
.662
.956
174.8600
1615.143
.736
.956
174.7800
1629.522
.586
.957
174.6600
1616.392
.648
.956
174.5800
1605.269
.647
.956
174.7200
1597.593
.722
.956
221

satis3
satis4
PIIT00
1
PIIT00
2
PIIT00
3
PIIT00
4
PIIT00
5
EXP00
2
EXP00
3
EXP00
4
EXP00
5

174.3800
174.8200
174.3400

1617.302
1603.538
1655.902

.603
.651
.242

.957
.956
.958

174.0200

1656.632

.281

.958

174.4000

1648.041

.317

.958

174.0400

1659.100

.215

.958

174.7000

1646.908

.346

.958

174.9600

1612.937

.581

.957

175.3600

1647.051

.219

.959

174.4200

1618.453

.480

.957

174.9200

1649.544

.179

.959

Cronbach's alpha of each of the instrument was obtained. All the items for each variable
posed a Cronbach's alpha value of 0.95 to 0.97 in terms of reliability. All Cronbach's
alpha coefficient of the scale from the pilot test was above the acceptance value of 0.7
as recommended by Hair et al (2006).

222

Appendix 11 Questionnaire

PUTRA BUSINESS SCHOOL UNIVERSITY PUTRA MALAYSIA

Analyzing Factors That Influence eBidding Utilization in Malaysian Public

Sector
This study focuses on an empirical research on the adoption of the procurement reverse auctions
system (eBidding) implemented by the Government of Malaysia. The aim of this study is to
examine the factors that influence the adoption of eBidding by the government procuring and
sourcing officials and to understand the users behavior in adopting the eBidding. Information
concerning Performance Expectancy, Effort Expectancy, Social Influence, Facilitating
Conditions, System Quality, Service Quality and Information Quality, eBidding Adoption, User
Satisfaction, Experience and Personal Innovativeness in Information Technology and the user
profile will be gathered. This study will provide a model that best explains the eBidding adoption
behavior among the government users.
There is no right or wrong answers. Please think carefully before answering all questions as
honestly as possible. Your response will be kept confidential and will be aggregated with other
responses so that individual respondents cannot be identified. Information provided in this
survey will only be meant for the researcher.
This questionnaire has twelve (12) sections. Section A covers the Performance Expectancy,
Section B covers the Effort Expectancy, Section C covers the Social Influence, Section D covers
the Facilitating Conditions, Section E covers the System Quality, Section F covers the
Information Quality, Section G covers the Service Quality, Section H covers the User
Satisfaction, Section I covers the Experience and Section J covers the Personal Innovativeness in

223

Information Technology, Section K covers the Actual Use and Section L will cover

the

Respondents Profile.
If you have any queries, please contact Megat Shariffudin B. Hj Zulkifli at 0193800118 or
email : megatsz1970@gmail.com.
Thank you.
Please use the following scale to indicate the degree of your agreement or disagreement for each
statement with each item below for each question that best matches your opinion. Using a rating
scale of 1 to 7, please circle the number that indicates your level of disagreement/agreement with
the following statements:

Strongly disagree

Strongly agree

SECTION A : PERFORMANCE EXPECTANCY

Statements
Scale
Performance expectancy is concerned which how an individual believe that using the system will
help attain gains in performing procurement tasks. (Venkatesh et al., 2003)
1

Using eBidding proves to be useful in doing my

job
Using eBidding enables me to accomplish
procurement tasks more quickly

Using eBidding increases the productivity of my

tasks

Using eBidding increases the job performance

Using eBidding makes it easier to procure goods

and services

Using eBidding enhances the effectiveness of my

work

224

Statements
7

Scale

Using eBidding gives me control over the

procurement auctions transactions.

SECTION B : EFFORT EXPECTANCY

Statements
Scale
Effort expectancy is concerned with the ease of use of the system (Venkatesh et al., 2003)
8

I find interactions with eBidding is straightforward

and understandable

I find that it is easy to become skilful at using the

eBidding

10

I find that it is easy to learn to use eBidding system

11

I find it easy to get the eBidding system to do

what I want it to do

12

I find the system is flexible to interact with

13

I find it is easy to navigate around the eBidding

system
I find that interactions with eBidding does not
require a lot of effort

14

225

SECTION C : SOCIAL INFLUENCE

Statements
Scale
Social influence is concerned with how an individual perceives others believe he/she should use
the information system. (Venkatesh et al., 2003)
People who are important to me think that I should
use eBidding

16

People who influence my behaviour think I should

use eBidding

17

In general, the department has been helpful in the

use of eBidding

18

In general, the department encourages the use of

eBidding

19

My colleagues think that I should use eBidding

20

My superiors think the system is useful and easy to

use
The helpdesk and technical support staff are
helpful in the use of the system

15

21

SECTION D : FACILITATING CONDITIONS

Statements
Scale
Facilitating conditions is concerned with organizational and technical infrastructure exists to
support the use of the information system (Venkatesh et al., 2003)
2 I have the technical resources necessary to use
2
eBidding (e.g. PCs, internet access)

23

I have the knowledge necessary to use eBidding

24

The eBidding is compatible with other systems

25

A specific technical support group is available to

assist

26

The legal infrastructure, circulars and laws are

available to enable the effective use of eBidding

226

Statements

Scale

27

The Government endorses the use of eBidding

system for procurement auctions transactions

28

The department promotes and ensures the effective

use of eBidding by providing training, internet
access and technical support group

SECTION E : SYSTEM QUALITY

Statements
Scale
System quality is concerned with the consistency of the system interface, availability, reliability,
usability and quality of features provided. (Delone and Mclean, 2003)
2 In terms of reliability, the system is rated high.
9
30 Overall, the system loads quickly

31
32

The system is user friendly and easy to use

The availability is very high

1
1

2
2

3
3

4
4

5
5

6
6

7
7

33

The system adaptability is very high

34
35

The system is capable of high speed of accessing

The system is easy to correct, in terms of system
error, such as runtime error, system down etc.

1
1

2
2

3
3

4
4

5
5

6
6

7
7

SECTION F : INFORMATION QUALITY

Statements
Scale
Information quality is concerned with the accuracy, timeliness, currency, understandability and
completeness of information provided to the users by the system.(Delone and Mclean, 2003)
36

The information generated by eBidding

complete and in correct format

is

37

The information generated by the application is

current

38

The information content generated

application is understandable

by

227

the

39

Statements
The information generated is relevant to the tasks

Scale
3
4

40

The system is able to be personalized according to

tasks

41

The information presentation is able to be

customized for different requirements

42

The information generated is timely and accurate

SECTION G : SERVICE QUALITY

Statements
Scale
Service quality is concerned with the support that users receive from the systems service
provider, namely assurance, empathy and responsiveness. (Delone and Mclean, 2003)
43

The helpdesk response is quick and prompt

44

The helpdesk personnel empathise with issues

faced by users

45

The helpdesk is helpful and provides assurances to

solve problems faced by users

46
47

The helpdesk provide follow-up service to users

The service providers (MAMPU, CDCSB and
GITN) have the necessary knowledge and means
to help the users

1
1

2
2

3
3

4
4

5
5

6
6

7
7

48

The service providers ensure a provision of secure

and safe environment for the transactions

49

The
service
providers
provide
troubleshooting services to users

on-site

228

SECTION H : USER SATISFACTION

Statements
Scale
User satisfaction is concerned with ones overall feelings of using the system. (Wixom and Todd,
2005)
The eBidding system is efficient and effective

The eBidding fulfils the needs and requirements of

my department

52

The outcome of the eBidding system meet my

expectations

53

My feeling after using the system was satisfactory

and contented

54

My decision to use the system was a good one

55

I would always visit the system for the

procurement transactions

56

Overall, I am very satisfied with the eBidding

system

50
51

SECTION I : EXPERIENCE
Statements
Scale
Experience is concerned an individuals experience, whether he or she is affected since the last
time he/she have used the system (Venkatesh et al., 2003)
7

5 I frequently visited the eBidding website for

information gathering

58
59

I frequently used eBidding for my job

I have used similar websites before this one to
perform the task

1
1

2
2

3
3

4
4

5
5

6
6

7
7

60

I am pleased with experience spent with the

eBidding

61

My experience with other auction systems was a

delightful one

229

SECTION J : PERSONAL INNOVATIVENESS IN INFORMATION TECHNOLOGY

Statements
Scale
Personal innovativeness is concerned with the users willingness to take the risk, to experiment
and try out the new system (Agarwal and Karahanna, 2000).
62

If I hear a new IT application, I would be the first

to experiment with it

63

In general, I am not hesitant to try out new

information technologies

64

I will take the risk to try out every new Internet

technologies

65

I like to experiment with new information

technologies

66

Among my peers, I am usually the first to try out

new information technologies

SECTION K : ACTUAL USE

Statements

Scale

Actual use measures everything from a visit to a Web site, to navigation within the site, to
information retrieval, to execution of a transaction by the users (Delone and Mclean, 2003).
Please circle your response on the appropriate space provided below.
67. On average, I use the eBidding system.
Extremely Infrequent

Extremely
Frequent

230

68. On average in a week, I use the eBidding system..

Not al all

More than
once each day

69. On average in a month, I execute at least 80 per cent of my transaction using the eBidding
system.
Strongly disagree
1

Please tick [ / ] your response on the appropriate space provided below

70. How long have you been using the eBidding system?
(
(
(
(
(

) under one year

) 1-2 years
) 2-3 years
) 3-4 years
) More than 4 years

231

Strongly
agree
7

SECTION L: RESPONDENTS PROFILE

Name of Ministry/Department :
71. Gender
1Male
2Female
72. Age
1.< than 30 years old
231 40 years old
341 - 50 years old
4> 50 years old

73. Users professional designations

1.Department Heads
2.. Managerial
3Supporting
74. Duration in present department
1.< 1 year
2..3 years
3..4 6 years
4..>7 years

75. Where is your department located?

232

1.Putrajaya
2.Kuala Lumpur
3.Shah Alam
4 Seremban
76. Department size. Please indicate the appropriate size of your Agency.
1< 10 employees
2..10 20 employees
3..21 30 employees
4..31 50 employees
5> 50 employees
-THANK YOU FOR YOUR PARTICIPATION & CONTRIBUTION

233

Biodata of Student

The candidate was born on November 4 th, 1970 in Kangar, Perlis. He completed his
primary education in Sekolah Tunku Mahmud, Kluang Johor (1982) and his secondary
schooling in Sekolah Dato Jaafar, Johor Bahru, Johor (1987) before furthering his study
at USM Matriculation at Sekolah Tunku Abdul Rahman, Ipoh Perak in 1989. After
completing the matriculation courses, the candidate attended University Science of
Malaysia (USM) in 1990 majoring in Applied Science and graduating in 1994.

The

candidate

was

employed

at

Mutiara

Telecom

Berhad

(now

DIGI

Telecommunications Bhd) before joining the government service in 1996 as

administrative diplomatic officer as programme manager for IRPA program in Ministry of
Science, Technology and Environment (MOSTE). The candidate completed the
compulsory Diploma in Public Administration in INTAN in 1997 and continue working
with MOSTE until 2002. He left for post graduate degree course majoring in Science
and Technology Studies in University of Wollongong in Australia and graduating in 2003.
Back in Malaysia, the candidate worked with the Ministry of Energy, Communication and
Water (MECW) in the IT Infrastructure Division. In 2005, he was promoted to senior
assistant director in the Economic Planning Unit, Prime Minister Department (EPU,
JPM) managing the IT infrastructure and industry development desk.

In 2009, he pursued a Ph.D degree in Management at the Graduate School of

Management (now Putra Business School), University Putra Malaysia.

234