Government Information Quarterly 23 (2006) 97 – 122

Determinants of user acceptance of the

e-Government services: The case of online
tax filing and payment system
Shin-Yuan Hung4, Chia-Ming Chang, Ting-Jing Yu
Department of Information Management, National Chung Cheng University, 160 San-Hsin Village,
Ming Hsiung, Chia-Yi, Taiwan, R.O.C.
Available online 27 January 2006

Abstract

This paper identifies the factors that determine the publics’ acceptance of e-Government services.
The online tax filing and payment system (OTFPS) is a well-known e-Government service in Taiwan.
Using a theoretical model based on the theory of planned behavior, this study aims to (1) identify the
determinants for acceptance of the OTFPS; (2) examine the causal relationships among the variables of
acceptance behavior for the OTFPS; and (3) explore the relative importance of each determinant for
both those who use the OTFPS and those who do not. The article reports on a survey that collected
1,099 usable responses. The results indicate that the proposed model explained up to 72 percent of the
variance in behavioral intention. In addition, the important determinants of user acceptance of the
OTFPS are perceived usefulness, ease of use, perceived risk, trust, compatibility, external influences,
interpersonal influence, self-efficacy, and facilitating condition. Finally, the academic and practical
implications of these findings are discussed.
D 2005 Elsevier Inc. All rights reserved.

Keywords: Electronic government; e-Government services; Information technology acceptance; Theory of

planned behavior; Online tax filing and payment system

4 Corresponding author. Fax: +886 5 2721501.

E-mail addresses: syhung@mis.ccu.edu.tw (S.-Y. Hung)8 chiaming@mis.ccu.edu.tw (C.-M. Chang)8
tingjing2001@yahoo.com.tw (T.-J. Yu).

0740-624X/$ - see front matter D 2005 Elsevier Inc. All rights reserved.
doi:10.1016/j.giq.2005.11.005
98 S.-Y. Hung et al. / Government Information Quarterly 23 (2006) 97–122

1. Introduction

Since the late 1990s, numerous governments have made huge investments in electronic
government (e-Government) services to link government networks and deploy a variety of
service infrastructure to provide extensive and proactive services. However, low levels of user
acceptance of these services are recognized as an endemic problem for government policy
makers, government agencies, and e-Government services providers. For example, the online
tax filing and payment system (OTFPS) is a well-known e-Government service in Taiwan.
The OTFPS system was launched in 1998. The Taiwanese government tax agency has
invested millions of dollars in promoting the OTFPS. However, the acceptance rates for
OTFPS were low (15.05 percent in 2003 and 21.06 percent in 2004).1 In addition, the
acceptance rates for the online tax filing were also low in United States (20.11 percent in
2002 and 22.61 percent in 2003) (http://www.irs.gov/taxstats/indtaxstats/index.html). Thus,
governments are in need of research that identifies the determinants of user acceptance to
apply to increase the user acceptance of e-Government services.
Understanding of user acceptance of e-Government services is insufficient owing to two
major problems. First, previous studies on information systems (IS) adoption and acceptance
focus on business and for-profit organizations, not on governmental and public organizations.
There are significant differences between private and public organizations.2 For example,
governments are less market exposure, more legal and formal constraints, and more complex
constraints for information systems. Second, although numerous studies on improving e-
Government services have been proposed, useful empirical research is deficient. E-
Government is more than online service delivery system. Behavioral issues of e-Government
research are markedly more important than technological ones. More empirical studies on
user acceptance of e-Government services are needed to assist governments in improving the
effectiveness and quality of e-Government services.
In recent years, Taiwan’s effort to provide e-Government services in government to
government (G2G), government to business (G2B), and government to citizens (G2C) has
been recognized. According to a report by the Center for Public Policy at Brown University,
Taiwan’s e-Government services were ranked first among 198 countries in 2004.3 Thus,
identifying and understanding the factors that determine user acceptance of e-Government
services in Taiwan should be a valuable reference for other countries. One of numerous well-
known G2C e-Government services in Taiwan is the online tax filing and payment system
(OTFPS).
A theory of planned behavior (TPB) model is developed in this study to evaluate user
acceptance. User acceptance of information technology is defined as bthe act of receiving
information technology use willingly.Q 4 User acceptance research studies the following
questions to explain individual acceptance of information technology: individual reactions to
using information technology, intentions to use information technology, and actual use of
information technology.5 The findings from user acceptance research suggest that when
users are presented with a new software package, a number of factors influence their
decision about how and when they will use it. Specifically, this study has three objectives:
(1) identify the determinants for acceptance of the OTFPS; (2) examine the causal
 S.-Y. Hung et al. / Government Information Quarterly 23 (2006) 97–122 99

relationships among the variables of acceptance behavior for the OTFPS; and (3) explore
the relative importance of each determinant for both those who use the OTFPS and those
who do not.
The contributions of this study are fourfold. First, it increases our understanding of the
determinants of e-Government services acceptance. Second, it provides an explanatory
framework of the external variables–core TPB variables relationship in a new IT context:
implementation of e-Government services. Third, it provides empirical support for the effects
of external factors on the attitude, subjective norms, and perceived behavioral control, which
further affect the user’s intention of accepting the e-Government services. Finally, it
investigates and tests an existing IS theory in a new IT context: implementation of e-
Government services. As well as the obvious academic value, the managerial implications of
this study will also benefit governmental policy makers, governmental agencies, and system
designers.
The remainder of this article is organized as follows. Section 2 presents a comprehensive
review of previous e-Government studies and e-Government applications in Taiwan and the
decomposed TPB model. Section 3 presents a description of the proposed research model and
hypotheses. Section 4 presents description of the research method. Analytical results are
reported and discussed in Section 5. Finally, Section 6 presents the conclusion and
implications of findings.

2. Literature review

2.1. Electronic government and related studies

Electronic government (e-Government) has attracted more and more research interest in
the worldwide. To bring fundamental transformation of the government services, numerous
regions’ e-Government development status have been reported. For example, Chen6
discussed e-Government outsourcing projects in the U.S. local government. Strejcek and
Theil7 discussed e-Government problems from the perspective of a lack of legislative
coordination in the European Union. Huang et al.8 explored the adoption factors for e-
Government in the Australian public sectors. In Asia, Hu et al.9 provided an overview of e-
Government development in China. Poon10 explored the e-Government application
bESDLifeQ in Hong Kong. Hung et al.11 briefly introduced the history, current status and
the architecture of Taiwan’s e-Government.
Furthermore, survey reports were released to provide government units useful information
about leading e-Government exemplars in the world. Since 2000, the Brown University
Center for Public Policy has annually investigated various countries’ efforts in e-Government
implementation.12 In its 2003 annual report, as categorized by its distinct features, Taiwan
was also listed as one of the top-ranked five best practices of e-Government sites over the 198
investigated nations.13 The distinct feature of e-Government implementation in Taiwan is that
100 percent of the sites evaluated offer services that can be fully executable online. In its
2004 annual report, Taiwan was ranked highest, with the largest number of services provided
100 S.-Y. Hung et al. / Government Information Quarterly 23 (2006) 97–122

by e-Government implementation in 2004.14 Hence, investigation of the implementation

experience of Taiwan e-Government should provide worthwhile lessons.
A number of research papers in e-Government were also published in recent years to help
practitioners improve government service responsiveness, convenience, and quality. Their
effort can be grouped into the five issues: (1) the concept, history, structure, initiatives,
policies, key principles, challenges, development success factors of e-Government;15 (2) the
technological applications, tools, or approach to facilitating e-Government services develop-
ment, implementation, and evaluation;16 (3) the assessment, measurement, or investment
model of e-Government services to clarify economic, social, or public benefits of
implementing e-Government services;17 (4) management support or implementation
strategies, including knowledge management, process-oriented management, service lifecycle
management, customer relationship management, and reputation recovery management, to
facilitating e-Government services management;18 and (5) key factors affecting acceptance,
expectation, and usage intention of e-Government services.19
While important issues from conceptual framework, technological innovation, services
measurement, and management support of e-Government services have been studied
extensively, determinants of user acceptance have not been well understood. Therefore,
this study intends to further our understanding of user acceptance of e-Government
services.

2.2. e-Government applications in Taiwan and online tax filing and payment system

Taiwanese e-Government plan (http://www.rdec.gov.tw/) was formulated by the Research,

Development, and Evaluation Commission (RDEC). The plan outlines e-Government
concept, vision, goals, effectiveness indicators, implementation strategies, specific measures,
current development progress, anticipated benefits, and current development in Taiwan. In
recent years, some of the development milestones of e-Government services have been
achieved. One well-known case in the e-Taxation plan is the implementation of the online tax
filing system and tax tracking system.20
The online tax filing and payment system (OTFPS) allows taxpayers to complete all their
tax filing and payment obligations online using online ID and electronic transaction
authentication mechanisms. The services provided by the OTFPS include the online
reporting of various kinds of income tax withholding (exemption) data, online filing and
payment of general income tax, online filing and payment of business income tax, online
payment of business tax, and online payment of taxes owed following a tax audit (http://
www.dot.gov.tw/).

2.3. Theory of planned behavior

The theory of planned behavior (TPB) is an extension of the theory of reasoned action
(TRA).21 Numerous studies have applied TPB to investigate behavior prediction using
attitudinal variables. According to TPB, people’s actions are determined by their intentions
and perceptions of control, where as their intentions are influenced by attitudes towards
 S.-Y. Hung et al. / Government Information Quarterly 23 (2006) 97–122 101

behavior, subjective norms, and perceptions of behavioral control. Ajzen reviewed numerous
studies that demonstrated the applicability of TPB to various content domains.22
Ajzen23 also demonstrated the ability of TPB in providing a very useful theoretical
framework for understanding and predicting the acceptance of new information technology.
Abundant empirical evidence suggests that TPB effectively explains individual intentions and
behavior in adopting new information technologies. Such evidence includes the acceptance of
telemedicine technology by physicians,24 the widespread adoption of virtual banking,25
computer resource center adoption and usage,26 IT adoption in work settings,27 acceptance of
electronic brokerage services,28 and others.
This study is to examine e-Government services acceptance, a comprehensive theoretical
model is proposed based on the TPB. The rationale for choosing TPB as theoretical foundation
is as follows: (1) the acceptance of e-Government services is not entirely under citizens’ control.
The condition satisfies core assumption of the TPB that the presence of constraints can inhibit
both the intent to perform a behavior and the behavior itself; (2) in accepting e-Government
services, citizens tend to be affected by external and interpersonal influences. Social influence is
an important determinant of behavior; (3) individual’s self-efficacy and facilitating resources
are possible barriers to user acceptance of e-Government services. Individual knowledge and
facilitating resources are possible adoption barriers; and (4) a decomposed TPB model is an
appropriate model for providing concrete managerial implications for practitioners. This model
can effectively elicit e-Government services users’ salient belief structure and acquire stable,
easily understood, and managerially relevant factors.

3. Research model and hypotheses

Fig. 1 illustrates a research model for e-Government services acceptance, as constructed

based on TPB. Thirteen hypotheses and their supporting studies are summarized in Table 1.

4. Research method

The data were collected using an e-mail questionnaire survey. First, this study created a
sampling frame with the assistance of the National Tax Administration of the Taiwan’s
Ministry of Finance. Then, 1,099 usable responses of 8,500 registered taxpayers were
collected using a systematic sampling method and analyzed using structural equation
modeling (SEM). To verify the construction of the research model and measurement
instruments, several professors and users were also interviewed to modify the research model
and the construction of the questionnaire.

4.1. Instrument development and pre-test

Regarding instrument construction, the items used to operationalize the constructs of each
investigated variables were mostly adopted from relevant previous studies, with necessary
102 S.-Y. Hung et al. / Government Information Quarterly 23 (2006) 97–122

Fig. 1. The research model.

validation and wording changes (see Table 2). Specifically, items measuring perceived
usefulness and ease of use were adapted from Davis,29 whereas items measuring subjective
norms, perceived behavioral control, and attitudes were taken from Taylor and Todd.30 Items
assessing perceived risk were adapted from Stone and Gronhaug,31 personal innovativeness
was adapted from Agarwal and Prasad,32 and trust was adapted from Gefen et al.33 Moreover,
measurements of behavioral intention were derived from Taylor and Todd.34 All items were
measured using a seven-point Likert-type scale with anchors ranging from bstrongly agreeQ to
bstrongly disagreeQ. The items used to measure each variable are listed in the Appendix A. To
ensure the desired balance and randomness in the questionnaire, half of the items were
worded with proper negation and all items in the questionnaire were randomly sequenced to
reduce the potential ceiling (or floor) effect, which induces monotonous responses to the
items for measuring a particular construct. Moreover, the final questionnaire was validated by
two professional translators to ensure that no syntax or semantic biases occurred during the
translation from English to Chinese.
Furthermore, to ensure validity and reliability, this study first pre-tested the questionnaire
by having three professors and users review it. Once the final survey was administered,
 S.-Y. Hung et al. / Government Information Quarterly 23 (2006) 97–122 103

Table 1
Summary of research hypotheses
Hypothesis Supporting studies
H1 Perceived Davis (1989); Taylor and Todd (1995)
usefulness Y attitude
H2 Perceived ease Davis (1989); Taylor and Todd (1995)
of use Y attitude
H3 Perceived risk Y attitude Gefen et al. (2002)
H4 Trust Y attitude Gefen et al. (2002)
H5 Personal Agarwal and Prasad (1998)
innovativeness Y attitude
H6 Compatibility Y attitude Taylor and Todd (1995); Mathieson (1991)
H7 External influence Y subjective Fishbein and Ajzen (1975); Ajzen (1991);
norms Taylor and Todd (1995); Bhattacherjee (2000)
H8 Interpersonal influence Y subjective Fishbein and Ajzen (1975); Ajzen (1991);
norms Taylor and Todd (1995); Bhattacherjee (2000)
H9 Self-efficacy Y perceived Bandura (1986); Ajzen (1991); Taylor and
behavioral control Todd (1995); Bhattacherjee (2000)
H10 Facilitating conditions Y perceived Bandura (1986); Ajzen (1991); Taylor and
behavioral control Todd (1995); Bhattacherjee (2000)
H11 Attitudes Y intention Ajzen (1991); Ajzen (2001); Liao et al. (1999);
Chau and Hu (2001); Chau and Hu (2002);
Taylor and Todd (1995); Bhattacherjee (2000);
Mathieson (1991)
H12 Subjective norms Y intention Ajzen (1991); Ajzen (2001); Liao et al. (1999);
Chau and Hu (2001); Chau and Hu (2002);
Taylor and Todd (1995); Bhattacherjee (2000);
Mathieson (1991)
H13 Perceived behavior Ajzen (1991); Ajzen (2001); Liao et al. (1999);
control Y intention Chau and Hu (2001); Chau and Hu (2002);
Taylor and Todd (1995); Bhattacherjee (2000);
Mathieson (1991)

analysis of the responses of thirty random respondents (twenty adopters and ten non-adopters)
revealed no problems with the survey design. Regarding reliability, the survey had strong
internal consistency with all multiple-item constructs achieving Cronbach’s alpha of 0.71 or
higher. Moreover, regarding validity, previously validated measurements were used to ensure
the validity of the measurements.

4.2. Data collection and sample representativeness

A sample frame was built with the assistance of the National Tax Administration of the
Taiwan Ministry of Finance. Using a systematic sampling method, 8,500 e-mail
questionnaires were delivered to individuals having registered with the OTFPS. Some
valuable coupons and digital cameras were used as incentive goods to motivate their
participation. The reason for using e-mail is that a complete e-mail list was provided by the
104 S.-Y. Hung et al. / Government Information Quarterly 23 (2006) 97–122

Table 2
Research variables and measurements
Construct Source
Perceived usefulness Davis (1989)
Perceived ease of use Davis (1989)
Perceived risk Stone and Gronhaug (1993)
Trust Gefen et al. (2002)
Personal innovativeness Agarwal and Prasad (1998)
Compatibility Taylor and Todd26
External influence Taylor and Todd
Interpersonal influence Taylor and Todd (1995)
Self-efficacy Taylor and Todd (1995)
Facilitating condition Taylor and Todd (1995)
Attitude Taylor and Todd (1995)
Subjective norms Taylor and Todd (1995)
Perceived behavior control Taylor and Todd (1995)
Intention Taylor and Todd (1995)

tax authorities, with data integrity validated by system’s in-built e-mail checks, as well as time
cost saving and data entry errors. Initially, all questionnaires were e-mailed to respondents
and a fifteen-day period was allowed for responses. After fifteen days, identical e-mail
questionnaires were resent to the non-respondents. After a further ten-day waiting period,
responses were solicited from the remaining non-respondents via e-mail again.
Of the 1,180 returned questionnaires, eighty-one were excluded because of incomplete
answers, leaving 1,099 usable responses (1,008 adopters and 91 non-adopters). The response
rate thus reached almost 13 percent. The chi-square goodness-of-fit test was used to test
whether the sample data ratio, including OTFPS adopters and non-adopters, comes from the
population with a specific distribution. The results indicated that this sample is representative
of the population having registered with the OTFPS. Moreover, this study also tested for
response bias between the responses and non-responses using the independent sample t test.
The results revealed no significant differences in terms of gender, age, education level, and
marriage status. Thus, no response bias existed in this study.

4.3. Statistical analysis

This study used the structural equation modeling (SEM) for hypotheses testing. Following
the two-stage approach suggested by Anderson and Gerbing,35 analysis of data from 1,099
samples was conducted through two stages. First, the measurement model was estimated
using confirmatory factor analysis to test whether the constructs possessed sufficient
validation and reliability. To ensure data validity and reliability, internal consistency,
convergent validity, and discriminant validity were demonstrated. Second, the structural
model that best fitted the data was identified, and the hypotheses were tested between
constructs using this model. Hair et al.36 specified the two main reasons for using SEM as
including: (1) to provide a straightforward method for dealing with multiple relationships
 S.-Y. Hung et al. / Government Information Quarterly 23 (2006) 97–122 105

simultaneously, while also providing statistical efficiency; and (2) to assess relationships
comprehensively, while providing a transition from exploratory to confirmatory analysis.
SEM is especially suited for testing a series of relationships constituting a large-scale model
or an entire theory. In fact, SEM has been identified as an appropriate covariance-based
approach in studies with a strong basis on a priori theory.37 This study is well suited for
confirmatory testing of the fit of the proposed theoretical model to observed data using SEM.
This study selected AMOS for Windows (version 4.0) as the SEM approach for estimating
the model. AMOS is a covariance-based approach similar to LISREL, in which the
covariance structure derived from the observed data is used to simultaneously fit measure-
ment and structural equations specified in the model. AMOS estimated both the measurement
and structural models using the full information maximum likelihood estimator.

5. Results and discussions

This section reports the analysis results. First, we explore the determinants of the OTFPS
acceptance. Second, the effects of external variables and TPB-related variables are examined
using the structural equation modeling approach. Finally, a simultaneous equation model to
evaluate adopters and non-adopters data is to examine whether there are the same dynamics in
both data sets.

5.1. Users’ profile of the OTFPS

Among 1,099 usable respondents (profiled in Table 3), 1,008 (91.7 percent) respondents
have adopted the OTFPS, whereas only 91 (8.3 percent) of respondents had registered with
and never adopted these services. Furthermore, the profile of OTFPS adopters is presented
in Table 4. The results show that adopters of these services in Taiwan are predominantly
male (83.5 percent male versus 16.5 percent female). Regarding age distribution, people
ages between thirty-one and fifty dominate among OTFPS adopters (75 percent). Moreover,
regarding educational level, adopters who have bachelors or higher degrees are dominant
(91 percent). Finally, married OTFPS adopters outnumber unmarried users (74 percent
compared to 26 percent, respectively). Therefore, we could conclude that married middle-
aged males, with bachelor degree or above, were more likely to adopt OTFPS than others in
Taiwan.

5.2. Relationships among the external variables and TPB-related variables

To examine the relationships among external variables and TPB-related variables, this
study uses structural equation modeling to test the hypothesized relationships between the
constructs. Following the two-stage approach suggested by Anderson and Gerbing,38 an
analysis of the 1,099 data samples was conducted in two stages. The measurement model was
first estimated using confirmatory factor analysis to test whether the constructs displayed
adequate validation and reliability. The measurement model specified in AMOS provided a
106 S.-Y. Hung et al. / Government Information Quarterly 23 (2006) 97–122

Table 3
Demographic profile of all respondents
Variable Count Percentage
Gender Male 913 83.1
Female 186 16.9
Age b20 years old 4 0.4
21–30 years old 122 11.1
31–40 years old 490 44.6
41–50 years old 332 30.2
N50 years old 151 13.7
Education Junior high school graduate 6 0.5
Senior high school graduate 91 8.3
Bachelor 664 60.4
Master or above 338 30.8
Marriage status Single 287 26.1
Married 812 73.9
Online tax filing and payment system use Non-users 91 8.3
Users 1008 91.7

convenient mean of validating the research instrument through confirmatory factor analysis
(CFA). CFA was considered more appropriate than exploratory factor analysis, since already-
validated scales were employed. In this study, to ensure data validity and reliability, we
demonstrated internal consistency, convergent validity, and discriminant validity. The second
stage then identified the structural model which best fitted the data. The two-stage processes
are detailed below.

5.2.1. Measurement model results

The findings indicate that the Cronbach’s alpha values of all items exceed 0.71, exceeding
the recommended value 0.70 for the social science research. Moreover, the average variances

Table 4
Demographic profile of online tax filing and payment system adopters
Variable Count Percentage
Gender Male 842 83.5
Female 166 16.5
Age b20 years old 4 0.4
21–30 years old 105 10.4
31–40 years old 450 44.6
41–50 years old 308 30.6
N50 years old 141 14.0
Education Junior high school graduate 5 0.5
Senior high school graduate 85 8.4
Bachelor 606 60.1
Master or above 312 31.0
Marriage status Single 259 25.7
Married 749 74.3
 S.-Y. Hung et al. / Government Information Quarterly 23 (2006) 97–122 107

extracted (AVE) value for each construct is between 0.52 and 0.81, exceeding the minimum
value of 0.50 proposed by Fornell and Larcker,39 which also demonstrates good internal
consistency.
Fornell and Larcker40 suggested that the criteria are determined by the AVE value, namely,
whether or not it exceeds the squared correlation between the constructs. Here, it
demonstrated good discriminant validity. Finally, Anderson and Gerbing41 suggested that
the assessment of convergent validity requires assessing the loading of each observed
indicator on its latent construct. The results of the CFA found that all loadings were
significant ( P value b0.01). Thus, strong evidence demonstrates satisfactory convergent
validity. Given that the required assessment of reliability and validity in the measurement
model are satisfactory, the subsequent process of identifying the structural model that best fits
the data can be illustrated in detail as follows.

5.2.2. Structural model results

The first step in model estimation involved examining the model fit results of the
hypothesized model. Some common fit indices reported in structural equation modeling are
designed to identify model goodness-of-fit. Common criteria for SEM have been previously
suggested and the results are presented in Table 5.42 In these results, the structural model
presented here indicates adequate fit with the observed data, compared with the suggested fit
criteria.
The second step in model estimation was to examine the significance of each hypothesized
path in the research model, as well as the variance explained (R 2 value) by each path. AMOS
reports raw and standardized estimates for all specified paths, along with standard errors and
test statistics for each path. The results are shown in Fig. 2 (the measurement model and belief
correlations are omitted for clarity). Twelve of the hypothesized thirteen paths are significant
at the 0.01 or 0.05 level.
Fig. 2 displays all structural relationships among the studied constructs. The figure
indicates that variance in OTFPS acceptance intention explained by the model is 72 percent,
which is very high given that numerous factors may affect acceptance intention for this
system. Variance in individual intention towards OTFPS adoption was totally explained 72
percent by attitude, subjective norms, and perceived behavior control. As noted in the TPB,
intentions to accept the OTFPS can be effectively explained by attitude, subjective norms,

Table 5
Results of the model goodness-of-fit
Fit index Recommended criteria Results in this study
Chi-square/Degree of freedom b3 2.93
P value N0.05 0.12
GFI (goodness-of-fit index) N0.9 0.93
AGFI (adjusted goodness-of-fit index) N0.9 0.91
CFI (comparative fit index) N0.9 0.97
RMR (root mean squared residual) b0.05 0.05
RMSEA (root mean squared error of approximation) b0.05 0.04
108 S.-Y. Hung et al. / Government Information Quarterly 23 (2006) 97–122

Fig. 2. Results of the model of the OTFPS acceptance.

and perceived behavior control. Furthermore, the results showed that variance in individual
intentions towards accepting the OTFPS can be explained by the large proportion of attitudes
in the sample, along with a relatively small proportion of perceived behavior control and
subjective norms.
In this context of the OTFPS acceptance, attitude is mainly determined by perceived
usefulness, perceived ease of use, perceived risk, trust, and compatibility. Personal
innovativeness is not significant in explaining variance of attitude toward OTFPS
acceptance. Regarding total variance of attitude, perceived usefulness can explain 31
percent of the variance, along with compatibility (26 percent), trust (24 percent), and
perceived ease of use (6 percent). Regarding total variance of subjective norms,
interpersonal influence can explain 13 percent of the variance, along with only 6 percent
of the variance explained by external influence. Furthermore, total variance of self-efficacy
 S.-Y. Hung et al. / Government Information Quarterly 23 (2006) 97–122 109

to accept the OTFPS can explain 64 percent of the variance. Only 21 percent of the variance
can be explained by facilitating conditions. Therefore, perceived usefulness, compatibility,
trust, interpersonal influence, and self-efficacy are more important than other determinants
for the OTFPS acceptance.
Some preliminary lessons can be drawn from the above analytical results. Clearly, much of
the path significance and variance explained are consistent with prior empirical findings.43
That is, e-Government services acceptance intention can be explained in terms of attitude,
subjective norms, and perceived behavior control.
Table 6 illustrates that twelve of the thirteen hypotheses were significantly supported, and
only hypothesis H5 was not empirically supported in this study. The results indicate that
personal innovativeness did not significantly affect attitudes towards the OTFPS. A possible
explanation is that the online tax filing task is always associated with negative image for
taxpayers. Hence, taxpayers with high personal innovativeness to accepting new ideas may
not easily be willing to challenge this system acceptance problem.

5.3. Relative importance of each influence on e-Government services acceptance for users of
e-Government

To examine the question of whether OTFPS acceptance intention follows the same
dynamics in adopters and non-adopters, we tested the above same model simultaneously with
two sets of data. The results show that the model fits the data from both groups quite well.
Furthermore, we focused on how the adopters’ estimates of conditions compare to the non-
adopters’ estimates (see Table 7).

Table 6
Summary of the research findings
Hypothesis Hypothesized Critical Finding
direction ratio
H1: Perceived usefulness Y Attitude + 10.59444 Supported
H2: Perceived ease of use Y Attitude + 2.40444 Supported
H3: Perceived risk Y Attitude 6.24444 Supported
H4: Trust Y Attitude + 12.76444 Supported
H5: Personal innovativeness Y Attitude + 0.91 Not supported
H6: Compatibility Y Attitude + 8.04444 Supported
H7: External influence Y Subjective norms + 5.40444 Supported
H8: Interpersonal influence Y Subjective norms + 10.63444 Supported
H9: Self-efficacy Y Perceived behavior control + 20.77444 Supported
H10: Facilitating condition Y Perceived behavior control + 7.24444 Supported
H11: Attitude Y Intention + 18.22444 Supported
H12: Subjective norms Y Intention + 2.41444 Supported
H13: Perceived behavior control Y Intention 5.92444 Supported
Note. The critical ratio (CR) is defined as bcritical ratio = estimate / standard errorsQ. The CR values greater than
1.96 and 2.32 are statistically significant at 0.05 and 0.01 levels, respectively. The results of structural equation
modeling are standardized maximum likelihood path coefficients for the hypothesized model.
444 Denotes significance at the P b 0.01 level.
110 S.-Y. Hung et al. / Government Information Quarterly 23 (2006) 97–122

First, for both adopters and non-adopters, perceived usefulness, trust, and compatibility
are all significantly important determinants of the attitude toward OTFPS. Second, personal
innovativeness does not significantly affect adopters’ or non-adopters’ attitudes. According
to the definition of personal innovativeness provided by Agarwal and Prasad,44 individual
traits in the adoption of information technology can help identify users who are likely to
adopt information technology innovations earlier than others. This empirical finding
demonstrates that users who adopt e-Government services earlier than others do not have
significantly higher levels of personal innovativeness. Possibly, the acceptance or use of
OTFPS is not a problem but rather an interesting task for adopters with high personal
innovativeness.
Third, perceived ease of use and perceived risk significantly influence adopters’ attitudes
toward these e-Government services. Gefen et al.45 have noted that perceived risk is
individual subjective expectation of suffering a loss in pursuit of a desired outcome. The
empirical results show that the increased ease of use and reduced perceived risk both
significantly influence adopters’ attitudes. However, this is not significant for non-adopters.
Hence, it is clear that greater perceived ease of use and lower risk level should be
beneficial determinants for e-Government services adopters. As for non-adopters, possible
reasons are that these two determinants not easily perceived or that little weight is assigned
to them.
Fourth, interpersonal influence is a significant determinant for adopters or non-adopters’
perceived subjective norms. This means that peers or friends’ opinions about e-Government
services clearly influence both adopters and non-adopters. Relatively speaking, external

Table 7
Comparison between adopters’ and non-adopters’ estimates
Hypothesis Hypothesized Adopters Non-adopters
direction CR CR
H1: Perceived usefulness Y Attitude + 10.53444 2.0744
H2: Perceived ease of use Y Attitude + 2.1544 1.31
H3: Perceived risk Y Attitude 6.14444 0.55
H4: Trust Y Attitude + 12.05444 2.40444
H5: Personal innovativeness Y Attitude + 0.79 0.67
H6: Compatibility Y Attitude + 6.51444 4.50444
H7: External influence Y Subjective norms + 5.01444 1.12
H8: Interpersonal influence Y Subjective norms + 10.22444 2.33444
H9: Self-efficacy Y Perceived behavior control + 19.30444 8.28444
H10: Facilitating condition Y Perceived behavior control + 8.06444 0.26
H11: Attitude Y Intention + 15.94444 7.12444
H12: Subjective norms Y Intention + 1.04 3.14444
H13: Perceived behavior control Y Intention 7.92444 0.72
Note. The critical ratio (CR) is defined as bcritical ratio = estimate / standard errorsQ. The CR values greater than
1.96 and 2.32 are statistically significant at 0.05 and 0.01 levels, respectively. The results of structural equation
modeling are standardized maximum likelihood path coefficients for the hypothesized model.
44 Denotes significance at the P b 0.05 level.
444 Denotes significance at the P b 0.01 level.
 S.-Y. Hung et al. / Government Information Quarterly 23 (2006) 97–122 111

influences such as news reports or mass media significantly influence adopters rather than
non-adopters. Apparently, adopters have paid some attention to external reports on e-
Government services, whereas non-adopters neglect such mass media reports or pay less
attention on them.
Fifth, self-efficacy is another significant determinant for adopters and non-adopters’
perceived behavior control. Self-efficacy is defined as people’s judgments of their
capabilities to organize and execute courses of action required to attain designated types
of performance.46 In the context of e-Government services, it is also consistent with other
studies exploring the adoption of technological innovation.47 This study also finds that the
increase of facilitating condition towards e-Government services in Taiwan will result in a
positive increase of perceived behavioral control for adopters, whereas this causal
relationship is not significant for non-adopters. Possibly, resources required to use these
e-Government services are sufficient for adopters as perceived entirely within their control.
In contrast, non-adopters may not perceive the resources or fail to have their interests
aroused.
Sixth, in the context of e-Government services in Taiwan, the theory of planned behavior
postulates that individual intention to adopt OTFPS is determined by attitude, subjective
norms, and perceived behavioral control. To examine the causal relationship between the
two group samples, the results highlighted different causal dynamics in adopters and non-
adopters. Attitude and perceived behavior control have significant effect on adopters’
intentions, whereas attitude and subjective norms are significant for non-adopters. This
highlights the fact that subjective norms have little effect on adopters’ adoption intention,
whereas perceived behavior control has little effect on non-adopters’ intent to adopt.
Furthermore, as adopters’ attitude is heavily influenced by additional factors (perceived
ease of use and risk), comparative results indicate that despite the influence of other
determinants of attitude, adopters may perceive and be determined by greater ease of use
and reduced risk than non-adopters. In addition, adopters may perceive subjective norms as
a trivial matter. In contrast, non-adopters’ intentions are only determined by attitude and
subjective norms rather perceived behavior control. To examine the association between
subjective norms, perceived behavior control, and its determinants, non-adopters may
heavily focus on only friends or peers’ opinions rather than mass media reports and
perceive the adoption of e-Government services as a trivial matter which also entirely
within control.

6. Conclusion and implications

An empirical study was conducted to identify determinants of user acceptance for e-

Government services. The critical factors included perceived usefulness, ease of use,
perceived risk, trust, compatibility, external influence, interpersonal influence, self-efficacy,
and facilitating condition. Furthermore, the causal relationships among the variables that
determine e-Government services acceptance were also examined. The results demonstrated
that e-Government services acceptance can be explained in terms of attitude, subjective norm,
112 S.-Y. Hung et al. / Government Information Quarterly 23 (2006) 97–122

and perceived behavior control. Finally, the relative importance of each determinant for users
and non-users was explored.
The academic value of this study is fourfold. First, this study proposed a comprehensive
theoretical framework to identify determinants of e-Government services acceptance. Second,
the sample group was real taxpayers. Thus, the validity of the findings is not limited. Third,
Taiwan’s efforts at e-Government services should be instructive for other countries,
particularly Chinese-speaking countries with cultures and languages in common with Taiwan.
Finally, by comparing three intention-based theoretical models (TAM, TPB, and Decomposed
TPB), this study confirmed that the proposed model has a high explanatory power for intention
(R 2 = 72 percent). Compared with previous studies, the total variance in intention to accept
OTFPS was higher or comparable to prior studies (e.g., R 2 = 70 percent in Venkatesh et al.,48
R 2 = 60 percent in Taylor and Todd,49 and R 2 = 62 percent in Mathieson50).
In addition, managerial implications for government policy makers, government agencies,
and system developers are also discussed. The findings can provide useful recommendation
to development of practice and policy making, which are customer oriented and evidence
based. For governmental policy makers responsible for future strategic planning of e-
Government services or effective OTFPS in particular, this study provides the following
recommendations:

(1) Several studies pointed out lacked appropriate performance measurements for e-
Government services.51 The nine important antecedents (i.e., perceived usefulness,
perceived ease of use, trust, compatibility, external influence, interpersonal influence,
self-efficacy, and facilitating condition) of user acceptance of e-Government services
were identified. Accordingly, to effectively evaluate the performance of e-Government
services, we suggest that policy makers can improve strategic planning for e-
Government services investments through monitoring these nine factors as performance
indicators.
(2) Four factors (i.e., perceived ease of use, perceived risk, external influence, and
facilitating condition) are significant determinants for adopters but not for non-adopters.
Bhattacherjee52 indicated that user continuance intentions are determined by satisfaction
with past use experience. Accordingly, to retain e-Government service adopters, we
suggest policy makers that marketing strategies for e-Government services seeking to
increase user loyalty and retention can focus on these points. Alternatives include
improving user interface of e-Government services, enhancing services security
mechanisms, employing mass media marketing, and increasing the availability of
necessary hardware and software for e-Government service use. Furthermore, for
supporting and marketing e-Government service adoption, we suggest that policy
makers should emphasize the advertising of adopters’ successful experience to attract
non-adopters.
(3) Adopters ranked higher scores in perceived usefulness, trust, compatibility,
interpersonal influence, and self-efficacy than non-adopters did. Accordingly, we
suggest that policy makers should invest more on user training and incentives to
attract non-adopters.
 S.-Y. Hung et al. / Government Information Quarterly 23 (2006) 97–122 113

(4) Subjective norm significantly affects non-adopters’ intention to not use, while
perceived behavioral control significantly affects adopters’ intention to use. From
this differentiating key motivational force behind e-Government service use, we
suggest that policy makers should make plans of action on enhancing external
and internal influences for non-adopters and continuously increasing users’ self-
efficacy and providing resources required to use e-Government services for
adopters.

For governmental agencies responsible for developing implementation strategies for

e-Government services, this study provides the following recommendations:

(1) Important determinants of e-Government services use in sequence are self-efficacy,

trust, interpersonal influence, perceived usefulness, compatibility, facilitating condition,
perceived risk, external influence, and perceived ease of use. To successfully implement
e-Government services given constraints in resources, we suggest that governmental
agencies can set priorities based on the relative importance of the factors.
(2) To increase positive attitude towards e-Government services, we suggest that
governmental agencies should develop implementation strategies that emphasize the
usefulness of e-Government services, workstyle compatibility, and user trust.
(3) To manipulate subjective norms, we suggest governmental agencies that marketing
strategy should seek to produce peer impact and external impact as well. For example,
encouraging e-Government service adopters to enhance their peer influence through
various channels; endorsing e-Government service by well-known stars.
(4) To reduce perceived behavioral control in using e-Government services, we suggest
governmental agencies that a market segment focus on low self-efficacy user will
clearly be the most efficient means of achieving a reduction in perceived behavior
control. Low self-efficacy users have lower perceived behavior control than other
users. Relatively, their perceived behavior control has a weak suppressive effect on
e-Government service adoption. We thus can train and educate citizens to increase
users’ self-efficacy.

For system developers responsible for e-Government services design, this study provides
the following recommendations:

(1) Perceived ease of use, trust, and compatibility are key factors influencing user
acceptance of the e-Government services. Therefore, we suggest that system developers
should provide a user-friendly interface, reinforce security mechanisms for e-Govern-
ment services, and design a suitable information system flow more compatible with
user’s workstyle.
(2) To support e-Government service adoption, we suggest that system developers could
focus on developing effective user guidance, continuously improving security
mechanisms, and using community of practice on Internet for promoting the e-
Government services and sharing use experience.
114 S.-Y. Hung et al. / Government Information Quarterly 23 (2006) 97–122

This study has several limitations. First, threats to external validity of this
investigation in Taiwan cannot be avoided. Future research could expand the subject
group to include e-Government services users of the Greater China economic region.
Second, the sample size of non-users is small compared to that of users. Internal validity
of this study may be reduced due to this unbalanced design problem. Third, due to the
existing limitations of the TPB model, additional factors (e.g., moral norms, self-identity,
individual past behaviors and habits, and affective belief) were overlooked in the study.
These factors should be considered in the future research. Finally, longitudinal studies
are also required to observe the users continuous or discontinuous use of e-Government
services.

Appendix A. The Questionnaire

Gender (1) Male (2) Female

Age (1) 20 or less (2) 21–30 (3) 31–40
(4) 41–50 (5) 51 or above
Marriage status (1) Single (2) Married
Education (1) Junior high school (2) Senior high school
(3) Bachelor (4) Master or above
Average hours of computer use in a week (1) 1 or less (2) 1–3 (3) 3–5 (4) 5–7 (5) 8 or above
You use online tax filing and payment system (1) Yes (2) No

Please indicate your agreement with the next set of statements using the following rating
scale.

Item Rating scale

Extremely Quite Slightly Neither Slightly Quite Extremely
likely likely likely unlikely unlikely unlikely
Usefulness
Using online tax filing and
payment system would
improve my performance in
preparing income tax filling.
Using online tax filing and
payment system would
improve my productivity in
preparing income tax filling.
Using online tax filing and
payment system would
enhance my effectiveness in
preparing income tax filling.
 S.-Y. Hung et al. / Government Information Quarterly 23 (2006) 97–122 115

Appendix A (continued)
Item Rating scale
Extremely Quite Slightly Neither Slightly Quite Extremely
likely likely likely unlikely unlikely unlikely
Usefulness
I would find online tax filing
and payment system useful
in preparing income tax
filling.

Ease of use
Learning to use online tax
filing and payment system
would be easy for me.
I would find it easy to prepare
income tax filing using
online tax filing and payment
system.
It would be easy for me to
become skillful at using
online tax filing and payment
system.
I would find online tax filing and
payment system easy to use.

Personal innovativeness
I am generally cautious about
accepting new ideas.
I find it stimulating to be
original in my thinking and
behavior.
I am challenged by ambiguities
and un-solved problems.
I must see other people using
innovations before I will con-
sider them.

Compatibility
Using the online tax filing and
payment system will fit well
with the way I work.
Using the online tax filing and
payment system will fit into
my workstyle.
The setup of the online tax
filing and payment system
will be compatible with the
way I work.
(continued on next page)
116 S.-Y. Hung et al. / Government Information Quarterly 23 (2006) 97–122

Appendix A (continued)
Item Rating scale
Extremely Quite Slightly Neither Slightly Quite Extremely
likely likely likely unlikely unlikely unlikely
Trust
Even if not monitored, I’d trust
the online tax filing and
payment system to do the job
right.
I trust the online tax filing and
payment system.
I believe that the online tax
filing and payment system is
trustworthy.

Perceived risk
Overall, the thought of using the
online tax filing and payment
system causes me to be
concerned with experiencing
some kind of loss if I went
ahead with the tax filling
action.
All things considered, I think I
would be making a mistake if
I used the online tax filing and
payment system for my tax
filling action.
When all is said and done, I
really feel that using the
online tax filing and payment
system poses problems for me
that I just don’t need.

Peer influence
My peers/colleagues/friends
thought that I should use
online tax filing and payment
system for tax filling action.
People I knew thought that
using online tax filing and
payment system was a good
idea.
People I knew influenced me to
try out online tax filing and
payment system for tax filling
action.
 S.-Y. Hung et al. / Government Information Quarterly 23 (2006) 97–122 117

Appendix A (continued)
Item Rating scale
Extremely Quite Slightly Neither Slightly Quite Extremely
likely likely likely unlikely unlikely unlikely
External influence
I read/saw news reports that
using online tax filling and
payment system was a good
way of tax-filling work.
The popular press depicted a
positive sentiment for using
online tax filing and payment
system.
Mass media reports influenced
me to try out online tax filing
and payment system.

Self-efficacy
I would feel comfortable using
online tax filing and payment
system on my own.
I would be able to use online tax
filing and payment system
reasonably well on my own.
I would be able to use online tax
filing and payment system
even if there was no one
around to help me.

Facilitating conditions
Resources required to use online
tax filing and payment system
were available to me.
I had access to hardware,
software, and services needed
to use online tax filing and
payment system.
I was constrained by the lack of
resources needed to use
online tax filing and payment
system.

Attitude
Using online tax filing and
payment system would be a
good idea.
(continued on next page)
118 S.-Y. Hung et al. / Government Information Quarterly 23 (2006) 97–122

Appendix A (continued)
Item Rating scale
Extremely Quite Slightly Neither Slightly Quite Extremely
likely likely likely unlikely unlikely unlikely
Attitude
Using online tax filing and
payment system would be a
foolish idea.
I like the idea of using online
tax filing and payment system
for tax-filling action.
Using online tax filing and
payment system would
be a pleasant experience.

Subjective norms
People (peers and experts)
important to me supported my
use of online tax filing and
payment system.
People who influenced my
behavior wanted me to use
online tax filing and payment
system instead of any
alternative means.
People whose opinions I valued
preferred that I use online tax
filing and payment system.

Perceived behavioral control

I would be able to use online tax
filing and payment system
well.
Using online tax filing and
payment system was entirely
within my control.
I had the resources, knowledge,
and ability to use the tax
filling systems

Intention
I intend to use online tax filing
and payment system.
It is likely that I will use online
tax filing and payment
system.
I expect to use online tax filing
and payment system.
 S.-Y. Hung et al. / Government Information Quarterly 23 (2006) 97–122 119

Notes and References

1. Research, Development, and Evaluation Commission (RDEC). (2005). E-government de-

velopment in Taiwan. Retrieved from http://www.rdec.gov.tw/lp.asp?ctNode=3702&CtUnit=
605&BaseDSD=7
2. Kraemer, K. L., Danziger, J. N., Dunkle, D. E., & King, J. L. (1993). The usefulness of
computer-based information to public managers. MIS Quarterly, 17(2), 129–148;
Caudle, S. L., Gorr, W. L., & Newcomer, K. E. (1991). Key information systems issues
for the public sector. MIS Quarterly, 15(2), 171–188;
Mohan, L., Holstein, W. K., & Adams, R. B. (1990). EIS: It can work in the public sector.
MIS Quarterly, 14(4), 435–448;
Henderson, J. C., & Schilling, D. A. (1985). Design and implementation of decision
support systems in the public sector. MIS Quarterly, 9(2), 157–169.
3. West, D. M. (2002). Global e-Government survey. World Markets Research Centre,
Brown University. Retrieved from http://www.insidepolitics.org/egovt05int.pdf;
West, D. M. (2003). Global e-Government, 2003. Center for Public Policy, Brown
University. Retrieved from http://www.insidepolitics.org/egovt03int.pdf;
West, D. M. (2004). Global e-Government, 2004. Center for Public Policy, Brown
University. Retrieved from http://www.insidepolitics.org/egovt04int.pdf
4. Saga, V. L., & Zmud, R. W. (1994). The nature and determinants of IT acceptance,
routinization, and Infusion. In L. Levine (Ed.), Diffusion, transfer, and implementation of
information technology. Pittsburgh, PA7 Carnegie Mellon University, Software Enginee-
ring Institute.
5. Venkatesh, V., Morris, M. G., Davis, G. B., & Davis, F. D. (2003). User acceptance of
information technology: Toward a unified view. MIS Quarterly, 27(3), 278– 425
(Minneapolis, MN).
6. Chen, Y. C. (2002). Managing e-Government outsourcing projects: Lessons from U.S.
local government. Proceedings of the Eighth Americas Conference on Information
Systems, 558–563.
7. Strejcek, G., & Theil, M. (2003). Technology push, legislation pull? E-government in the
European union. Decision Support Systems, 34(3), 305–313.
8. Huang, W., D’Ambra, J., & Bhalla, V. (2002). Key factors influencing the adoption of e-
Government in Australian public sectors. Proceedings of the Eighth Americas Conference
on Information Systems, 577–579.
9. Hu, H., Li, Q., & Lin, Z. (2003). From concept towards implementation: E-
Government in China. Proceedings of the Ninth Americas Conference on Information
Systems, 1104–1108.
10. Poon, S. (2002). ESDLife of Hong Kong e-Government application with an e-business
spirit. Proceedings of the Eighth Americas Conference on Information Systems, 585–591.
11. Hung, S. Y., Ku, C. Y., & Chang, C. M. (2005). E-Government applications and user
acceptance in Taiwan. In W. Huang, K. Siau, & K. Kee Wei (Eds.), Electronic
government strategies and implementation (pp. 421–440). Idea Group Publishing.
12. Ibid., p. 3.
120 S.-Y. Hung et al. / Government Information Quarterly 23 (2006) 97–122

13. Ibid., p. 3.
14. Ibid., p. 3.
15. Devadoss, P. R., Pan, S. L., & Huang, J. C. (2003). Structurational analysis of e-
Government initiatives: A case study of SCO. Decision Support Systems, 34(3), 253–269;
Jaeger, P. T. (2002). Constitutional principles and e-Government: An opinion about
possible effects of federalism and the separation of powers on e-Government policies.
Government Information Quarterly, 19(4), 357 –368;
Ibid., p. 7;
Aldrich, D., Bertot, J. C., & McClure, C. R. (2002). E-government: Initiatives,
developments, and issues. Government Information Quarterly, 19(4), 349–355;
Pons, A. (2004). E-Government for Arab countries. Journal of Global Information
Technology Management, 7(1), 30–46;
Teicher, J., Hughes, O., & Dow, N. (2002). E-government: A new route to public sector
quality. Managing Service Quality, 12(6), 384–393.
16. Adkins, M., Burgoon, M., & Nunamaker, Jr. J. F. (2003). Using group support systems
for strategic planning with the United States air force. Decision Support Systems, 34(3),
315–337;
Chen, Y. C., & Gant, J. (2001). Transforming local e-Government services: The use of
application service providers. Government Information Quarterly, 18(4), 343–355;
Lowry, P. B., Albrecht, C. C., Nunamaker, Jr. J. F., & Lee, J. D. (2003). Evolutionary
development and research on Internet-based collaborative writing tools and processes
to enhance e-writing in an e-Government setting. Decision Support Systems, 34(3),
229–252.
17. Chircu, A. M., & Lee, D. H. D. (2003). Understanding IT investments in the public
sector: The case of e-Government. Proceedings of the Ninth Americas Conference on
Information Systems, 792–800;
Kaylor, C., Deshazo, R., & Van Eck, D. (2001). Gauging e-Government: A report on
implementing services among American cities. Government Information Quarterly,
18(4), 293–307;
Yang, L., & Huang, J. (2004). An exploratory study on the assessment model of e-
Government in China. Proceedings of the Tenth Americas Conference on Information
Systems, 585–593.
18. Vassilakis, C., Laskaridis, G., Lepouras, G., Rouvas, S., & Georgiadis, P. (2003). A
framework for managing the lifecycle of transactional e-Government services. Telematics
and Informatics, 20(4), 315–329;
McHenry, W. K. (2003). Using knowledge management to reform the Russian criminal
procedural codex. Decision Support Systems, 34(3), 339–357;
Stuart, J. B., & Vidgen, R. (2004). Interactive e-Government services: Modeling user
perceptions with eQual. Electronic Government, 1(2), 213–228;
Weber, R. O., & Aha, D. W. (2003). Intelligent delivery of military lessons learned.
Decision Support Systems, 34(3), 287–304.
19. Gefen, D., Pavlou, P. A., Warkentin, M., & Rose, G. M. (2002). E-government adoption.
Proceedings of the Eighth Americas Conference on Information Systems, 569–576;
 S.-Y. Hung et al. / Government Information Quarterly 23 (2006) 97–122 121

Ibid., p. 8;
Lee, J. K., Braynov, S., & Rao, H. R. (2003). Effects of a public emergency on citizens’
usage intention toward e-Government: A study in the context of war in Iraq. Proceedings
of the Twenty-Fourth International Conference on Information Systems, 896–902;
Zhang, J., Cresswell, A. M., & Thompson, F. (2002). Participants’ expectations and the
success of knowledge networking in the public sector. Proceedings of the Eighth
Americas Conference on Information Systems, 597–604.
20. Research, Development, and Evaluation Commission (RDEC). (2005). E-government
in Taiwan-policy, strategy, and status. Retrieved from http://www.rdec.gov.tw/lp.asp?
ctNode=3702&CtUnit=605&BaseDSD=7;
Ibid., p. 1.
21. Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human
Decision Process, 50, 179–211;
Fishbein, M., & Ajzen, I. (1975). Belief, attitude, intention and behavior: An introduction
to theory and research. Addison-Wesley.
22. Ajzen, I. (2001). Nature and operation of attitudes. Annual Review of Psychology, 52,
27–58.
23. Ibid., p. 22.
24. Chau, P. Y. K., & Hu, P. J. H. (2001). Information technology acceptance by individual
professionals: A model comparison approach. Decision Sciences, 32(4), 699–719;
Chau, P. Y. K., & Hu, P. J. H. (2002). Investigating healthcare professionals’ decisions to
accept telemedicine technology: An empirical test of competing theories. Information and
Management, 39(4), 297–311.
25. Liao, S., Shao, Y. P., Wang, H., & Chen, A. (1999). The adoption of virtual banking: An
empirical study. International Journal of Information Management, 19(1), 63–74.
26. Taylor, S., & Todd, P. A. (1995). Understanding information technology usage: A test of
competing models. Information Systems Research, 6(2), 144–176.
27. Venkatesh, V., Morris, M. G., & Ackerman, P. L. (2000). A longitudinal field
investigation of gender differences in individual technology adoption decision-making
processes. Organizational Behavior and Human Decision Processes, 83(1), 33–60.
28. Bhattacherjee, A. (2000). Acceptance of e-commerce services: The case of electronic
brokerages. IEEE Transactions on Systems, Man, and Cybernetics: Part A. Systems and
Humans, 30(4), 411–420.
29. Davis, F. D. (1989). Perceived usefulness, perceived ease of use, and user acceptance of
information technology. MIS Quarterly, 13(3), 319–340.
30. Ibid., p. 26.
31. Stone, R. N., & Gronhaug, K. (1993). Perceived risk: Further considerations for the
marketing discipline. European Journal of Marketing, 27(3), 39–50.
32. Agarwal, R., & Prasad, J. (1998). A conceptual and operational definition of personal
innovativeness in the domain of information technology. Information Systems
Research, 9(2), 204–215.
33. Ibid., p. 19. Gefen et al.
34. Ibid., p. 26.
122 S.-Y. Hung et al. / Government Information Quarterly 23 (2006) 97–122

35. Anderson, J. C., & Gerbing, D. W. (1988). Structural equation modeling in practice: A
review and recommended two step approach. Psychological Bulletin, 103, 411–423.
36. Hair, J. F., Tatham, R. L., Anderson, R. E., & Black, W. (1998). Multivariate data
analysis. (5th ed.). New York, NY7 Macmillan Publishing Company.
37. Ibid., p. 36.
38. Ibid., p. 35.
39. Fornell, C., & Larcker, V. F. (1981). Evaluating structural equation models with
unobservable variables and measurement error. Journal of Marketing Research, 18, 39–50.
40. Ibid., p. 39.
41. Ibid., p. 35.
42. Joreskog, K. G., & Sorbom, D. (1993). LISREL 8: Structural equation modeling with
SIMPLIS command language. Chicago, IL7 Chicago, IL: Scientific Software International;
Ibid., p. 36.
43. Ibid., pp. 25, 26, 28.
44. Ibid., p. 32.
45. Ibid., p. 19. Gefen et al.
46. Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory.
New York, NY7 Macmillan Publishing Company.
47. Teo, T. S. H., & Pok, S. H. (2003). Adoption of WAP-enabled mobile phones among
Internet users. Omega, 31(6), 483–498;
Ibid., p. 25;
Ibid., p. 28.
48. Ibid., p. 5.
49. Ibid., p. 26.
50. Mathieson, K. (1991). Predicting user intention: Comparing the technology acceptance
model with the theory of planned behavior. Information Systems Research, 2(3), 173–191.
51. Steyaert, J. C. (2004). Measuring the performance of electronic government services.
Information and Management, 41(3), 369–375;
Ibid., p. 2. Caudle et al.
52. Bhattacherjee, A. (2001). Understanding information systems continuance: An
expectation-confirmation model. MIS Quarterly, 25(3), 351–370.

Further reading

1. Steyaert, J. C. (2004). Measuring the performance of electronic government services.

Information and Management, 41(3), 369–375.