ARTICLE IN PRESS

Int. J. Production Economics 113 (2008) 928–942

www.elsevier.com/locate/ijpe

Understanding ERP system adoption from the user’s perspective

Man-Kit Changa, Waiman Cheungb,, Chun-Hung Chengc, Jeff H.Y. Yeungb
a
Department of Finance & Decision Sciences, Hong Kong Baptist University, Kowloon Tong, Kowloon, Hong Kong
b
Department of Decision Sciences & Managerial Economics, The Chinese University of Hong Kong, Shatin, NT, Hong Kong
c
Department of Systems Engineering & Engineering Management, The Chinese University of Hong Kong, Shatin, NT, Hong Kong
Received 23 July 2004; accepted 24 August 2007
Available online 27 December 2007

Abstract

An enterprise resource planning (ERP) system is a new management technology that advocates an integrated
approach to conduct business. While organizations are hoping to apply this technology to improve overall
performance, they must understand what it takes for their employees to use it. Although the use of ERP systems
may not be voluntary, the understanding of system adoption from the user’s perspective is useful in helping
the organizations prepare their employees to face new challenges and learn how to make good use of the
technology. To analyze factors affecting the ERP system usage, we proposed a conceptual model derived from
the Triandis framework. The use of the Triandis framework is based on the previous research that documents
the importance of social factors on the adoption of a technology. An empirical study was conducted in Hong
Kong to understand the adoption process. Our research results show that social factors are the most significant
determinant affecting the ERP system usage. Other factors such as compatibility and near-term consequences are
also significant. Based on our findings, we also propose some important managerial implications in connection to
promoting the usage.
r 2008 Elsevier B.V. All rights reserved.

Keywords: Enterprise resource planning; Technology adoption; Triandis model; Survey

1. Introduction decisions affect the functions of other units. On the

It has been well recognized in the management
literature that it takes all business units or depart-
ments of an organization to work together to
achieve its overall objectives and goals. This
integrated view of management requires that each
unit not only function efficiently and effectively
within, but also understand how its activities and
Corresponding author. Tel.: +852 2609 7816;
fax: +852 2603 5104.
E-mail address: wcheung@cuhk.edu.hk (W. Cheung).
other hand, information systems have been devel-
oped in an ad-hoc manner focusing on their
corresponding business units. This creates islands
of information in the organization making informa-
tion sharing difficult. To support an integrated
management approach, enterprise resource plan-
ning (ERP) has been proposed as a solution
(Shanmugam et al., 2000).
ERP emphasizes resource planning from an
enterprise’s perspective. ERP systems implement
ERP concepts enterprise wide and cover all business
functions. Many benefits have been realized from

0925-5273/$ - see front matter r 2008 Elsevier B.V. All rights reserved.
doi:10.1016/j.ijpe.2007.08.011
 ARTICLE IN PRESS
M.-K. Chang et al. / Int. J. Production Economics 113 (2008) 928–942
the use of ERP systems. The advantages include
better information sharing within the organization,
improved planning and decision quality, smoother
coordination between business units resulting in
higher efficiency, and quicker response time to
customer demands and inquiries. Building on top
of these advantages, organizations may promote
customer relationship management that would
strengthen customer loyalty and satisfaction, and
achieve larger market share.
ERP systems, similar to other management
information systems, are often perceived as very
complex and difficult to be implemented (Liang
et al., 2007; Xue et al., 2005). For many organiza-
tions, ERP systems are the largest systems they have
worked with in terms of the financial resource
invested, the number of people involved and the
scale of implementation. Several recent cases of
ERP system implementation have experienced con-
siderable difficulties (Goldberg, 2000; Krasner,
2000; Wah, 2000; Xue et al., 2005). The failure rate
of ERP implementation is very high (Yeh et al.,
2007). Among other obstacles, technical problems
and people obstacles have been cited as the major
barriers (Botta-Genoulaz and Millet, 2006; Krasner,
2000). To further understand the ERP adoption
process, this research attempts to identify key
factors that determine the ERP system usage using
a well-established theoretical framework, the Trian-
dis model.
In the next section, we provide a brief description
of ERP function and characteristics and then we
discuss our research model based on the Triandis
framework in Section 3. The methodology of the
study is described in Section 4, followed by the
results of our analysis in Section 5. Discussions of
findings, implications and conclusions are presented
in Sections 6, 7 and 8, respectively.
2. ERP functions and characteristics
The integration of business functions through
ERP is achieved through a uniform software plat-
form and database. It is evolved from Material
Requirement Planning (MRP) systems and Manu-
facturing Resource Planning (MRPII) systems.
MRP systems are concerned with the integration
of manufacturing functions involving purchasing,
planning, materials and operations, whilst MRPII
systems ensure the coordination and interactions of
manufacturing with other functional areas such as
marketing, finance and engineering.
929
Although ERP systems have the foundation
originally developed for a manufacturing setting,
they have evolved into a platform to support almost
all aspects of business and industrial operations.
Typical ERP systems provide many modules. Some
have more advanced and powerful modules than the
other. A typical ERP system may consist of these
modules: Accounting and Financial Module;
Human Resources Management Module; Manufac-
turing Module; Procurement Management Module;
and Distribution and Supply Chain Module.
Unlike most home-growth legacy systems that
were designed to fit individual working convention,
ERP systems provide best practices, in other
words generic processes and functions at their
outset. Alignment of the standard ERP processes
with the company’s business process has been
considered as an important step in the ERP
implementation process (Botta-Genoulaz et al.,
2005). Implementing a packaged ERP system
inevitably changes the way people work. Compat-
ibility between the new system and the existing
business procedures and data format are the major
issues reported by the companies (Lucas et al., 1988;
Soh et al., 2000; Van Everdingen, 2000). The
mismatches created considerable adoption problems
(Kumar, 2000). Therefore, we believe the compat-
ibility of these systems with the existing business
working conventions is an important factor explain-
ing the usage.
As an enterprise system, the success of ERP
implementation requires a close cross-functional
cooperation (Motwani et al., 2005). The informa-
tion entered by one division will be used by other
divisions, even in real time. Thus, employees may be
expected by their peers to use the ERP in order to
make the ERP more useful. Since ERP is a major
investment of a firm and the implementation may
involve substantial organizational changes, top
management support has been found to be a key
factors of success, but more importantly top
management need to develop a shared vision and
to communicate it to the employees so that the
expectation is clear (Ehie and Madsen, 2005;
Motwani et al., 2005; Tchokogué et al., 2005).
Thus, the expectation of both peers and top
management may influence the behavior of the
ERP users. This social factor has been dropped
from the original Technology Acceptance Model
(TAM), a popular framework for study of technol-
ogy adoption, because the technology investigated,
a word processor, was of personal and individual
 ARTICLE IN PRESS
930 M.-K. Chang et al. / Int. J. Production Economics 113 (2008) 928–942
nature (Davis, 1989). Thus its use is not likely
driven by social influences (Cheung et al., 2000).
However, we have found that social pressure plays
an important role in explaining the use of the
Internet. Thus, we believe that a user’s attitude
toward using ERP systems will be strongly influ-
enced by his/her perception of the superiors’ and
colleagues’ expectation.
Facilitating conditions (i.e., the availability of the
necessary resources and supports to the ERP system
usage) is another factor we believe to be important.
In order to provide a smooth access to these
systems, a large number of elements must work
closely together. These elements include supports in
hardware, software, training and information pro-
vision.
Based on the discussion above, we adopt a more
comprehensive social psychological model—the
Triandis framework (Triandis, 1977, 1980) for this
study. The original Triandis framework includes
affect, social factors, facilitating conditions and
consequences as determinants of formation of
behavior. It has been successfully used in a number
of technology adoption researches (Bergeron et al.,
1995; Thompson et al., 1991, 1994). In our previous
study (Cheung et al., 2000), we adopted Thomp-
son’s idea to separate consequences into near-term
consequences, long-term consequences and com-
plexity (Thompson et al., 1991). Specific to ERP
system usage, we include compatibility as a new
extension to the basic Triandis framework.
3. Research model
Triandis (1980) proposed a theoretical network
that postulated the relationship attitude and beha-
vior with many constructs such as culture and
biological factors. Fig. 1 depicts the subset of
Triandis framework similar to the one adopted in
Thompson et al. (1991) that is relevant to this study.
According to the Triandis model, the probability of
performing a given behavior is determined by a
number of factors: (1) habit of performing the
behavior, (2) facilitating condition and (3) intention.
Habit is a behavior that is or has become automatic
in a given situation. The facilitating condition is
important because, as stated by Triandis (1980),
even if the intention is high and the habit is
well established, if some objective obstacles exist,
such as geographic and resource limitations, the
expected behavior will not occur. In his model,
behavioral intention is determined by social factors,
Habit
Hierarchies
Social Perceived
Affect
Factors Consequences
Intentions
Facilitating
Conditions
Behavior
Fig. 1. Subset of the Triandis model [adopted from Triandis
(1980)].
affect and perceived consequences of performing the
behavior.
In this study, we applied the Triandis model to
predict ERP system usage at work. Since we were
interested in the prediction of actual behavior,
‘‘intention’’ was dropped from the model. Instead,
we studied the direct effect of social factors, affect
and perceived consequences on the current beha-
vior. Habit was excluded from our investigation
because in a cross-sectional study the measurements
for habit and actual behavior are the same.
There has been a study in the social psychology
area comparing the application of the theory of
reasoned action (TRA), the theory of planned
behavior (TPB) and Triandis theory. The result
shows that the Triandis theory and the TPB
are better than the TRA. The authors further
suggested that some of the constructs in the Triandis
theory should be added to the TPB to improve its
predictive power (Godin et al., 1996). We use the
Triandis theory directly in this study because it
contains many more constructs, such as habit
hierarchies and culture. Future studies can further
extend its application.
Fig. 2 shows the modified model. According to
the adopted model, the probability of performing a
given behavior is determined by a number of
factors: (1) perceived consequences, which include
near-term consequences and long-term conse-
quences, (2) affect, (3) complexity, (4) compatibility,
(5) facilitating conditions and (6) social factors.
These factors fall into three categories: individual,
technological and organizational characteristics

M.-K. Chang et al. / Int. J. Production Economics 113 (2008) 928–942
Organizational Factors
Facilitating
Conditions
Individual Factors
Near-term +ve
Consequences
+ve
Long-term Usage
+ve
Consequences
+ve
Affect -ve
Complexity
Technological Factors
Fig. 2. Research model showing the hypotheses.
(Jeyaraj et al., 2006). Individual characteristics
consist of perceived near-term consequences, per-
ceived long-term consequences and affect. Techno-
logical characteristics include complexity and
compatibility while organizational characteristics
include facilitating conditions and social factors.
In a review of 99 studies on IT innovation adoption,
Jeyaraj et al. (2006) have concluded that top
management support and user support (organiza-
tional characteristics), as well as the perceived
usefulness (individual characteristic) are amongst
the best predictors of IT adoption by individual.
3.1. Perceived consequences
Among factors influencing behavior, is perceived
consequences of a behavior. This suggests that a
behavior has some consequences with value to the
performer (Triandis, 1980). This construct is con-
sistent with the expectancy theory of motivation.
The expectancy theory postulates that an individual
chooses a behavior based on the desirability of
the rewards (Thompson et al., 1991). Therefore, the
higher the perceived value of the consequences, the
higher is the likelihood that a person will perform
the behavior. Thompson et al. (1991) included three
dimensions of perceived consequences, i.e., near-
term consequences, long-term consequences and
ARTICLE IN PRESS
931
complexity in his research framework. However,
we would argue that the nature of perceived
Social Factors
complexity as a consequence is quite different from
perceived short-term and long-term consequences.
It can be viewed as a belief on whether using the
innovation is easy or hard. Thus, we treat perceived
+ve complexity as the perception of the characteristic of
the technology being used instead of the conse-
quences of using the technology.
3.1.1. Near-term consequences
Perceived near-term consequences in our research
context are defined as the extent to which an
+ve individual believes that using the ERP systems
can enhance the performance of his/her job. The
impacts are on the individual’s current job. This
concept closely resembles the Perceived Usefulness
compatibility concept in the TAM. Studies have consistently
found that this factor is an important determinant
of intention and behavior (Bergeron et al.,
1995; Morris and Dillon, 1997; Thompson et al.,
1991, 1994). Perceived usefulness has also been
found to have a strong positive impact on the
satisfaction of the ERP users (Calisir and Calisir,
2004). Thus, we predict that perceived near-term
consequences have a positive impact on the ERP
system usage. The first hypothesis is shown as
follows:
Hypothesis 1. Perceived near-term consequences are
positively related to the ERP system usage.
3.1.2. Long-term consequences
Apart from the near-term and the more direct
impact on the existing job, knowing how to use a
technology also has a longer-term impact on career
development. Therefore, perceived long-term con-
sequences of using information technology is
another dimension of perceived consequences in-
vestigated in this study. Long-term consequences
include the increased flexibility in changing job or
increase in opportunities of having a better job. Past
findings regarding this factor are mixed. Thompson
et al. (1991) found a positive relationship between
long-term consequence and usage but later, Thomp-
son et al. (1994) did not find such a relationship in
inexperienced users. Chau (1996), when predicting
behavioral intention of using MS Word and Excel,
found long-term consequences to have a significant
impact. Consistent with the Triandis model, we
hypothesize that there is a positive relationship
between perceived long-term consequences and the
 ARTICLE IN PRESS
932 M.-K. Chang et al. / Int. J. Production Economics 113 (2008) 928–942
ERP system usage. The second hypothesis is offered
as follows:
Hypothesis 2. Perceived long-term consequences are
positively related to the ERP system usage.
3.2. Affect
Affect, an individual characteristic, is the direct
emotional response to the thought of the behavior
and is referred to as the ‘‘feelings of joy, elation or
pleasure, or depression, disgust, displeasure, or hate
associated by an individual with a particular act’’
(Triandis, 1980). Many behaviors have positive
consequences, but are unpleasant to perform.
Therefore, another factor that influences the use of
the ERP systems is affect. Existing findings relating
affect to usage in information systems are mixed.
Paré and Elam (1995), studying the adoption of PC,
concluded that anxiety (an negative affect) has a
negative relationship with usage. Other studies did
not found any significant relationship between
affect and usage (Bergeron et al., 1995; Thompson
et al., 1991, 1994). Thompson et al. (1991) explained
their insignificant result by stating that the use of
PC may not evoke any strong emotional response.
On the other hand, enjoyment, which we consider
similar to affect, was found to have a positive
relationship with PC (Igbaria et al., 1995) and the
Internet usage (Cheung et al., 2000; Teo et al.,
1999). In addition, affect was also found significant
in predicting other behaviors (Valois et al., 1988).
Therefore, we hypothesize that affect is positively
related to ERP system usage (as shown in the
following hypothesis).
Hypothesis 3. Affect is positively related to the ERP
system usage.
3.3. Complexity
Complexity, is a characteristic of the technology,
ERP in this case, that will affect adoption decision.
Perceived complexity is defined as ‘‘the extent to
which an innovation is perceived as relatively
difficult to understand and use.’’ (Thompson
et al., 1991)
Perceived complexity is the opposite of perceived
ease of use (EOU) in the TAM. In general,
complexity is found to have a negative impact on
the adoption of information technologies (Igbaria
et al., 1995; Teo et al., 1999; Thompson et al., 1991).
For the ERP systems, Calisir and Calisir (2004)
found that perceived EOU affected the end-user
satisfaction positively and Bradford and Florin
(2003) found that the complexity of the ERP system
affected satisfaction negatively. In our model, we
expect that perceived complexity is negatively
related to the ERP system usage. Alternatively, we
restate our postulate in the following hypothesis:
Hypothesis 4. Perceived complexity is negatively
related to the ERP system usage.
3.4. Compatibility
It is well known that information systems that are
incompatible with the culture and convention of
the organization may fail (Lucas et al., 1988; Yusuf
et al., 2004). ERP systems, unlike legacy systems,
provide generic functions and processes. They
inevitably alter the culture and convention they
used to be familiar.
Due to limited resources and short deadlines,
many organizations are totally occupied by techni-
cal issues in ERP system implementations. Little
attention is paid to customization of ERP modules.
ERP systems are likely to create tensions, frustra-
tion, instability and conflict in the user group.
Soh et al. (2000) indicated that procedural and
data compatibility are crucial to the acceptance of
the system by the employees. Holsapple et al. (2005)
has also found that compatibility is positively
related to ERP satisfaction. Hence, ERP system
compatibility with the user group’s existing opera-
tions should have effect on the usage. We expect
the higher the compatibility, the higher the system
usage.
Hypothesis 5. Compatibility is positively related to
the ERP system usage.
3.5. Social factors
Social factors, an organizational characteristic
(Jeyaraj et al., 2006), are defined as ‘‘an individual’s
internalization of the reference group’s subjective
culture, and specific interpersonal agreements that
the individual has made with others, in specific
social situations’’ (Triandis, 1980). Subjective cul-
ture consists of ways of categorizing experiences,
beliefs, attitudes, ideals, roles, norms and values,
which can be understood as the characteristic way
that a human group views the human-made part of
its environment. This concept is similar to subjective
norm in the TRA (Ajzen and Fishbein, 1980). Using
 ARTICLE IN PRESS
M.-K. Chang et al. / Int. J. Production Economics 113 (2008) 928–942
the Triandis model as a guiding framework, a
number of studies found a positive relationship
between what the subject thought others want him/
her to do and the usage of various systems including
PC and EIS (Bergeron et al., 1995; Thompson et al.,
1991, 1994). The concept of social factors includes
not only senior management commitment, but also
the expectation and pressure from various parties
and colleagues to whom the ERP users interact.
Successful implementation of ERP systems requires
close coordination amongst various parties and
departments within the companies. Therefore, social
pressure is expected to play an important role in
determining employees’ ERP system usage. Thus,
consistent with previous empirical findings and the
Triandis model, we predict that social factors have a
positive impact on the ERP usage. The following
hypothesis is formulated:
Hypothesis 6. Social factors are positively related to
the ERP system usage.
3.6. Facilitating conditions
Facilitating conditions, another organizational
characteristics, refer to the ‘‘objective factors, ‘out
there’ in the environment, that several judges or
observers can agree make an act easy to do’’
(Triandis, 1980). People require necessary resources
and supports to perform a behavior. According to
Triandis (1980), a person may intend to do some-
thing, but is unable to perform it due to geographic
barrier that prevents the act to be realized. To deal
with this situation, Triandis includes facilitating
conditions to predict behavior. In the context of our
study, the facilitating conditions include hardware,
software, network connection, training, informa-
tion, etc. that allow individuals to access the ERP
systems when they want to. In addition, they include
the supports provided by the company to facilitate
the use of the system at work.
Past findings on the relationship between facil-
itating conditions and usage are mixed. Thompson
et al. (1991) found no significant relationship and in
a later study, Thompson et al. (1994) found a
significant negative relationship between PC usage
and facilitating conditions for both the inexper-
ienced and experienced users. On the other hand,
Paré and Elam (1995) found a positive relationship
between the two variables. Past studies on the ERP
systems have shown that user’s training is one of the
critical success factors for ERP implementation
933
(Amoako-Gyampah and Salam, 2004; Bradford and
Florin, 2003; Zhang et al., 2005). Thus, we predict a
positive relationship between the usage and facil-
itating conditions. The hypothesis is restated as
follows:
Hypothesis 7. Facilitating conditions are positively
related to the ERP system usage.
3.7. Summary of hypotheses
Based on the Triandis model, seven hypotheses
(shown in Fig. 2) are formed to analyze factors that
affect the ERP system usage. As shown in the
model, we predicted that factors relating to indivi-
dual, innovation and organization will together
contribute to the adoption decision of the ERP
users. Our analysis predicts that all factors except
complexity will have a positive effect on the usage.
4. Research methodology
The purpose of this research is accomplished by
conducting a survey using a questionnaire. The
questionnaire, the operationalization of research
construct and the subjects will be described in this
section.
4.1. Questionnaire and measures
The questionnaire (given in the Appendix A) is
designed based on Triandis (1977) and Triandis
(1980). It is applied to suite the specific context of
the ERP system utilization. We also refer to
previous empirical study results in the design of
the questionnaire (Bergeron et al., 1995; Paré and
Elam, 1995; Thompson et al., 1991, 1994).
Near-term and Long-term consequences: Five
items on near-term consequences and five items on
long-term consequences were adapted from Thomp-
son et al. (1991). Respondents were asked whether
they agreed or disagreed that certain consequences
would happen. A 5-point scale ranging from
strongly disagree (1) to strongly agree (5) was used.
Complexity: The measurement scale was devel-
oped also based on Thompson et al. (1991). Three
statements, such as ‘‘Working with the ERP is
complicated, it is difficult to understand what is
going on’’, were used. They were rated on a 5-point
scale ranging from strongly disagree (1) to strongly
agree (5).
 ARTICLE IN PRESS
934 M.-K. Chang et al. / Int. J. Production Economics 113 (2008) 928–942

Compatibility: ERP system compatibility is con- accessibility to the ERP systems and other items for
cerning the possible mismatches between the ERP general support were added. The 4-item scale was
system and the user’s working mode. We developed rated on a 5-point scale ranging from strongly
a new scale to measure the construct. The subjects disagree (1) to strongly agree (5).
were asked whether their ERP system is compatible Usage: Triandis (1980) suggested that acts differ
with their work practices and styles. Two statements in a number of ways such as duration, intensity and
were developed based on the discussion in Soh frequency. Three items were used to measure the
et al. (2000). They were rated on a 5-point scale intensity of use and the frequency of use. They were
ranging from strongly disagree (1) to strongly rated on a 5-point scale ranging from strongly
agree (5). disagree (1) to strongly agree (5).
Social factors: We developed a new scale accord-
ing to Triandis (1980) to measure the construct,
social factors or SOCIAL. This is because Thomp- 4.2. Subjects
son et al. (1991) discovered that their scale seriously
overlapped with the scale for measuring facilitating The questionnaires were distributed to 600 practi-
conditions. In our new scale, the subjects were tioners who had experience in using ERP systems.
asked whether they perceived their work group Two hundred and forty questionnaires were returned.
(i.e., top management, immediate supervisor and The demographic information is shown in Table 1.
coworker) thought that they should use the ERP Most respondents have been in their careers for
system (i.e., normative beliefs or NB), and whether a long time. The means and standard deviations
they would follow what these people thought of the number of years in their careers were about 6
they should do (i.e., motivation to comply or and 5.5 years, respectively. Out of all the respondents,
MC). A 5-point scale ranging from strongly disagree 55% were female. Over 45% of the respondents
(1) to strongly agree (5) was used. The construct was held an undergraduate or graduate degree. The
measured by multiplying the two scores in pairs respondents work for many companies from both
(i.e., SOCIAL ¼ NB x MC). This operationaliza- service and manufacturing industries. About 43%
tion is similar to the one used by Bergeron et al.
(1995).
Affect: We developed our own scale to measure
affect. Thompson et al. (1991) used three items to Table 1
operationalize this construct asking only whether Demographic characteristics
using PC is interesting and fun. The resulting No. of years in current position
reliability (a ¼ .61) was poor. Paré and Elam Mean 6.0 years
(1995) dropped the enjoyment scale from their Standard deviation 5.5 years
analysis because the items were loaded together Gender
with perceived consequences in their factor analysis. Male 98 (40.8)a
To measure different aspects of affect, we followed Female 133 (55.4)
Triandis’ suggestions to operationalize the affect Omitted 9 (3.8)
by using a 4-item semantic differential scale Educational level
(Triandis, 1980). On a 5-point fully anchored Master degree or above 24 (10.0)
scale, respondents were asked whether they felt Undergraduate 84 (35.0)
Associate degree 28 (11.7)
using the ERP systems was enjoyable–disgusting,
Others 104(43.3)
dull–exciting, pleasant–unpleasant or interesting–
boring. In this scale, the higher the final score, the Position
Departmental manager or above 51 (21.3)
better they felt when thinking of using the ERP
Assistant manager 8 (3.3)
systems. Professional staff 16 (6.7)
Facilitating conditions: In terms of the ERP Technical staff 48 (20.0)
system utilization, facilitating conditions are con- Front line operator 13 (5.4)
ceptualized as the support provided by the company Others 34 (14.2)
Omitted 70 (29.1)
and the accessibility to the ERP systems. Two
items were adapted from Thompson et al.’s (1991) a
Numbers of respondents and percentage of respondents (in
original scale. In addition, one item measuring the parentheses).
 ARTICLE IN PRESS
M.-K. Chang et al. / Int. J. Production Economics 113 (2008) 928–942 935

of the respondents declared that they used ERP Table 2

throughout the day, and 22% of them indicated Results of the final factor analysis
that they used the system more than once every day. Factors Cronbach’s a Factor loading
Only 8% of the respondents have never used ERP
system. Social factor .86
The most-frequently used module by the respon- SF1 .86
SF2 .87
dents is Finance, which is in turn followed by
SF3 .76
Distribution, and Manufacturing. The least-used
module is Human Resources. Complexity .83
Complex1 .81
Complex2 .85
4.3. Data analysis Complex3 .86
Near-term consequences .84
The first step in our analysis was to assess
NT1 .57
the psychometric properties of the measurement NT3 .79
scales. Factor analysis was used to demonstrate NT4 .84
the discriminant validity of the measurement NT5 .78
scales. Items should load higher on its associated Long-term consequences .85
construct than on any other construct. The internal LT1 .75
consistency and reliability of the scale were LT2 .75
assessed using Cronbach a. Ordinary least-square LT3 .74
LT4 .76
multiple regression was used to test the proposed
LT5 .64
model.
Facilitating condition .71
FC2 .80
FC3 .70
5. Research findings FC4 .80

Compatibility .79
5.1. Assessment of the measurement scales
Compat1 .87
Compat2 .84
In order to assess the discriminant validity of the Affect .88
measurement scales, principal component factor
Affect1 .87
analysis with varimax rotation was used. Scree Affect2 .78
plot and eigenvalues were used to determine the Affect3 .86
number of factors to be extracted. According to the Affect4 .78
criteria of extracting factors that have eigenvalue Usage .91
greater than 1, eight factors were extracted. The Usage1 .90
number of factors extracted corresponds to the Usage2 .89
number of factors measured. The result of the initial Usage3 .87
factor analysis was examined to identify the
presence of items failing to load highly upon any
factor. One item from near-term consequences
and one item from facilitating conditions were higher than .35 is considered statistically significant
excluded from the final analysis because they did at an a level of .05 for a sample size of 250; thus
not load highly upon any factor. The remaining all the items loaded significantly on its own
items were again factor analyzed and an eight-factor factor. Moreover, each item loaded higher on its
structure was suggested by the eigenvalue criteria. associated construct than on any other construct;
The resulting factor loadings are shown in Table 2 this supported the discriminant validity of the
with all factor loadings less than .4 suppressed. measurement.
The extracted factors account for 75.2% of the Cronbach’s a’s were computed for each measure-
total variance. All items loaded onto the expected ment scale. As shown in the second column of
factors as they were originally designed. Factor Table 2, the reliability coefficients ranged from .71
loadings were all higher than .5 on its own factors. to .91, which was higher than the acceptable level of
As suggested in Hair et al. (2006), a factor loading .7 for this kind of study (Nunnally and Bernstein,

936 M.-K. Chang et al. / Int. J. Production Economics 113 (2008) 928–942
Organizational Factors
Facilitating
Conditions
Individual Factors
Near-term
Consequences
.175
Usage
Long-term
Consequences R2= .262
Affect
Complexity
Innovation Factors
Fig. 3. Model showing the results of the regression analysis.
1994, p.265). These results confirm that the scales
used are both valid and reliable.
5.2. Effect of the antecedents
Least-square multiple regressions were performed
to test the hypotheses and the results are summar-
ized in Fig. 3. To test the research framework, all
independent variables were entered into the regres-
sion simultaneously. The presence of multi-colli-
nearity for the regression equations was assessed
using the variance inflation factor (VIF). The VIF
values for the independent variables ranged from
1.2 to 1.7. As suggested by Neter et al. (1996),
‘‘a maximum VIF value in excess of 10 often
indicated that multi-collinearity may be unduly
influencing the least-square estimates.’’ The VIF
values we obtained were much smaller and very
close to 1, which indicated that there was no serious
multi-collinearity problem among the independent
variables. Moreover, the Durbin–Watson test was
used to detect the existence of autocorrelation
among the residuals. The results indicated that
there was no autocorrelation in any of the regres-
sion equations. Method variance may exist and
contribute to part of the correlation between two
constructs. However, its effect may not be very
substantial and it is hard to estimate. According to
ARTICLE IN PRESS
Table 3
Prediction of ERP system usage
Social Factors Variables Usage
Social factors .346a
Compatibility 186a
Affect .032
.346
Long-term consequence .071
Near-term consequences .175b
Complexity .063
Facilitating conditions .081
R2 .262a
a
Regression coefficient is significant at the .01 level.
b
Regression coefficient is significant at the .05 level
.186
compatibility Bagozzi and Yi (1990), they have used confirmatory
factor analysis via structural equation modeling to
analyze 11 data sets using Multi-traits Multi-
methods. They have found that individual method
loading in 6 data sets are not significant. As Fiske
and Campbell (1992) suggested, ‘‘method and trait
or content are highly interactive and interdepen-
dent. We may have to settle for the practice of
studying ‘trait-method units.’ ’’
Table 3 shows the regression results from testing
the determinants of ERP usage. The regression
equation is significant at the .001 level and the
independent variables account for 26% of the
variance of usage. As shown in the table and in
the order of significance, social factors (b ¼ .346),
compatibility (b ¼ .186) and near-term conse-
quences (b ¼ .175) are all positively related to the
current ERP usage; while affect, long-term con-
sequences, complexity and facilitating conditions
are not significant (at the a level of .05).
6. Discussion
In our study, three factors are found to have
significant effect on the ERP system usage: social
factors, compatibility and near-term consequence.
Among the significant factors, social factors, an
organizational characteristic, have the strongest
effect. This finding suggests that our respondents
are susceptible to social pressure exerted by their
colleagues and superiors. As mentioned above,
successful implementation of an ERP system
requires cooperation amongst different parties and
departments. Also top management may exert
 ARTICLE IN PRESS
M.-K. Chang et al. / Int. J. Production Economics 113 (2008) 928–942
pressure on their employee to use the system. These
may explain why the construct social factors
become the most important determinant of the
ERP system usage in this study.
Confirming our anticipation, compatibility is the
second significant factor determining the usage.
Although our respondents are expected by their
colleagues and superiors to use the ERP systems in
performing their jobs, they are more likely to use
them if the systems are compatible with their
working mode. This finding further confirms the
assertions made by other researchers (Holsapple
et al., 2005; Marble and Lu, 2007; Soh et al., 2000;
Yusuf et al., 2004).
Perceived near-term consequences, which mea-
sure the belief of enhancements in terms of quality
and efficiency to the users’ job after using the
ERP system, are found to have a positive effect
on the ERP system usage. This result is consistent
with previous studies (Thompson et al., 1994).
However, its relative importance among other
factors in determining usage is not the same. It
is ranked third in our study while previous studies
have shown that near-term consequence is the most
important determinant for PC usage (Thompson
et al., 1991, 1994). Triandis (1980) maintains that
there is evidence that the relative significance of
different factors on predicting intention and actual
behavior varies depending on the type of behavior,
social situation and person. Our results support that
such differences do exist. In this case, the effect of
near-term consequences has been superseded by the
social influences.
Other factors are found insignificant. These
include perceived long-term consequences, complex-
ity, facilitating conditions and affect. Perceived
long-term consequences do not have a significant
effect on the ERP usage. Our finding is different
from that of Thompson et al. (1991), but is
the same as Thompson et al. (1994) for inexper-
ienced PC users. Long-term consequences mainly
measure career development effect. In our study,
most of the respondents are not in the IS/IT
line of work. The ERP systems they are using
generally play a supportive role in their job.
Consequently, the factor of long-term consequences
is insignificant.
The complexity of the ERP systems does not
discourage users from using the systems. ERP
systems intend to integrate information and busi-
ness processes within and across functional areas,
therefore they are often very complicated and
937
difficult to implement. However, the complicated
and difficult to implement systems do not result in
hard to use systems. In our survey, the reported
perceived complexity of using ERP systems is
relatively low (i.e., 2.74 on a 5-point fully anchored
scale). Most ERP system users are front line
operators and junior managers. They only use a
few modules that are relevant to their work, usually
for data manipulation and simple reporting. As the
overall perceived complexity of using ERP systems
is low, its effect on the system usage becomes
insignificant. Another possibility is that complexity
does not affect usage directly but indirectly through
its positive impact on perceived usefulness (Dishaw
and Strong, 1999; Gefen and Keil, 1998). Calisir and
Calisir (2004) also found an indirect effect of
perceived EOU on end-user satisfaction through
perceived usefulness.
The insignificance effect of facilitating conditions
on ERP system usage is also evidenced in previous
study explaining PC usage (Thompson et al., 1991).
On the other hand, the same factor is found
significant on Internet usage (Cheung et al., 2000).
The importance of facilitating conditions may
vary depending on the technology being adopted.
It must be noted that in our study the overall
perceived complexity of using ERP systems is low
and the general perceived facilitating conditions is
adequate (3.28 on a 5-point fully anchored scale). It
is still important to provide sufficient hardware
installment for accessibility and software supports
for usability.
The effect of affect on ERP system usage is found
insignificant, which is consistent with past studies
investigating relationship between affect and cur-
rent behavior (Bergeron et al., 1995; Igbaria et al.,
1995; Thompson et al., 1991, 1994). However, Davis
has found affect has a significant impact on
formation of intention (Davis et al., 1989). The
dynamics of how factors affect the formation of
behavioral intention seems to be different from how
the factors affect current actual behavior. Future
study may be conducted to reveal the formation of
intention to use ERP system.
Overall, although some of the factors in all three
categories (individual, organizational and innova-
tion) affect the usage of ERP systems, the social
factors, an organizational factor, exert the strongest
influence. This may be due to the fact that ERP, as
an enterprise system, requires a lot of coordination
and corporation among the members of the
organizations to make it work. That may be the
 ARTICLE IN PRESS
938 M.-K. Chang et al. / Int. J. Production Economics 113 (2008) 928–942
reason why organizational environment is more
important than both the individual and the innova-
tion characteristics in determining the usage of the
systems.
7. Implications and limitations
ERP has been proposed as an approach to
achieve an organization’s objectives and goal
through an integrated approach of management.
While an organization applies this technology in
hoping for improved results, they must understand
what it takes for their employees to use the ERP
systems. This understanding will help organizations
prepare their employees to face new challenges
and learn how to make good use of the technology.
To analyze factors affecting the ERP system
usage, we propose a conceptual model derived
from the famous Triandis framework considering
many factors including compatibility, social factors,
etc. An empirical study was conducted in Hong
Kong to understand the adoption process. Factor
analysis and regression were used to analyze
the data collected. Our research results show that
users do not perceived using ERP systems as
complex and they believe that adequate level of
supports is currently provided. Social factors
are the most significant determinant affecting
the ERP system usage. Other factors such as
compatibility and near-term consequences are also
significant.
7.1. Theoretical implications
In this study, we have identified the factors and
their relative importance in terms of influencing the
usage of an ERP system in an organization based on
the Triandis model. IS researchers have been asked
to pay more attention to the IT artifacts that are
under investigation, instead of taking them for
granted (Orlikowski and Iacono, 2001), but few
studies has actually done so at the individual level
(Venkatesh, 2006). It is suggested that when
designing our studies we should take the character-
istics of the IT artifacts and its operating context
into account as ‘‘IT artifact are always embedded in
some time, place, discourse and community’’
(Orlikowski and Iacono, 2001, p.131) and the values
and assumption of the developers, investors, as well
as the users shape the IT artifacts. The findings of
the current study echo this point of view. ERP
systems are the focus of the current study and most
ERP systems, such as SAP, embed the best practices
of the industry, i.e. values and assumptions, into
their designs. Unlike more generic software, such as
word processor, compatibility becomes an impor-
tant factor in determining the usage of such systems.
Social factor is another example. In the context of
word processor usage, Davis et al. (1989) did not
find significant relationship between social norm
and usage and thus dropped this construct from the
TAM. However, social factor has become an
important factor for IT applications that require
cooperation among different parties to be success-
ful. ERP system is one such system that connecting
colleagues across functional areas to achieve
better efficiency and it is embedded into the social
environment of the companies more deeply than
those standalone applications. Another example
would be the use of the Internet which has gradually
become part of our life and topic of our social
conversation. These findings not only support
that the IT artifact is important in understanding
technology use and adoption, they also support
Triandis’ (1980) claim that the relative importance
of different factors in predicting behavior
varies depending on different behavioral contexts.
As new information systems relentlessly emerge, this
study calls for systematic investigation of the
relative contributions of different factors in deter-
mining usage by considering the characteristics of
various types of information systems and the
situations under which the technologies are being
used. A meta-analysis on the existing adoption
results will improve our understanding of the
adoption process.
The Triandis model that the current study is
based upon provides an additional theoretical
perspective for the investigation of ERP adoption
and adoption studies in general, in addition to
the oft-used model such as TAM. Only a small
subset of the Triandis model was used in the
current study to inform our research design. The
complete model contains other causal structures,
such as how subjective culture affects the social
factors, affect and perceived consequence, which
will be useful to inform future technology adoption
studies.
7.2. Managerial implications
Our results have a number of implications for
organizations that are trying to promote the use of
 ARTICLE IN PRESS
M.-K. Chang et al. / Int. J. Production Economics 113 (2008) 928–942
the ERP systems for job-related activities. First of
all, creating a supportive social atmosphere is very
important. Senior management should show sup-
port and encouragement to their subordinates in
using the systems. A strong and committed leader-
ship and support from senior management is a
necessary condition for successful ERP implemen-
tation (Sarker and Lee, 2003; Tchokogué et al.,
2005). The concept of social factors also includes
the expectation and pressure from various parties
and colleagues to whom the ERP users interact.
A successful implementation of ERP systems
requires close coordination amongst various
parties and departments within the companies.
Use a phase approach to implement the ERP
system may allow the experience of the existing
users to influence the adoption decision of the
potential users. Clearly define and communicate the
expected behavior of employees will also enhance
the chance of a successful ERP project (Tchokogué
et al., 2005).
Besides social support, organizations must ensure
that the ERP systems are compatible with the
working mode of the users. Modifying ERP
modules to suit individual use is almost necessary.
The significant effect of near-term consequences
further indicates that besides improving overall
management the ERP system must also benefit
individual productivity. The capacity of the ERP
system in enhancing quality as well as efficiency of
an individual’s job should be conveyed. Social
factor is found significantly influencing the usage
of ERP system therefore; experience sharing should
be arranged for employees so that the nonuser can
see the benefits of using the ERP system as told by
the users. This may motivate them to use the ERP
system (Roger, 1995).
System implementation success depends on many
factors. ERP system evaluation, vendor selection,
the ERP consultant, implementation plan and
execution are all critical to the success of imple-
menting an ERP system. Nevertheless, the intro-
duction of the system undoubtedly changes the
way people work. The platform is new, data
entry is changed and report formats are different.
Users often find these changes unnecessary and
therefore refuse to accept them. An adoption
study helps us to understand the determinants of
the usage. In the case of implementing ERP system
we should put more effort in customizing ERP
modules to compile with the existing work flow,
report formats and data needs. While involving
939
users as early as possible in system implementation
is generally a good strategy (Tchokogué et al.,
2005), we should involve them specifically for
finding ways to improve their work quality and
efficiency (i.e., creating positive short-term conse-
quence). It will also help to involve users from
different functional areas as a group in design,
briefing and training sections. Understanding
how a user’s participation in using the system can
improve work quality of another user can create
social tight among the users therefore improve the
adoption.
7.3. Limitations
There are limitations of the current study that
need to be mentioned. One limitation is that while
the respondents have provided the frequency of
ERP usage, a number of them did not dispose their
corporate positions. Hence, the overall profile of the
subjects was not completely clear. Another one, as
mentioned before, is that method variance may exist
and contribute to part of the correlation between
two constructs because self-reported rating was used
to measure all the constructs.
8. Conclusion
In this paper, we use the Triandis model as a
theoretical foundation to understand the determi-
nants of ERP system adoption. Although an
adoption study cannot completely imply system
implementation success, it helps understand the
determining factors for system success. Our study
found that Social factors, Compatibility and Near-
term consequences are the important factors affect-
ing ERP system usage. Based on our finding, it is
clear that it is important to create a supportive social
atmosphere to encourage system use. End-user
involvement in the implementation is also critical.
In addition, the benefits of ERP on individual
productivity must be conveyed to end-users.
Acknowledgements
This research has been partially supported by
Research Grant Council of Hong Kong Special
Administration Region (Grant No. CUHK4213/
01E) and Center of Cyber Logistics, Li & Fung
Institute of Supply Chain Management and Logis-
tics, The Chinese University of Hong Kong.
 ARTICLE IN PRESS
940 M.-K. Chang et al. / Int. J. Production Economics 113 (2008) 928–942

Appendix A. Questionnaire items

Strongly Strongly
disagree agree

(A) Social factors

1. Top management thinks that I should use ERP system 1 2 3 4 5
in my job
2. My immediate supervisor thinks that I should use ERP 1 2 3 4 5
system in my job
3. My coworkers think that I should use ERP system in 1 2 3 4 5
my job
4. Generally speaking, I would do what top management 1 2 3 4 5
thinks I should do
5. Generally speaking, I would do what my immediate 1 2 3 4 5
supervisor thinks I should do
6. Generally speaking, I would do what my coworkers 1 2 3 4 5
think I should do
(B) Complexity of ERP system
1. Working with the ERP system is complicated, it is 1 2 3 4 5
difficult to understand what is going on
2. It takes too long to learn how to use the ERP system to 1 2 3 4 5
make it worth the effort
3. In general, the ERP system is very complex to use 1 2 3 4 5
(C) Near-term consequences
1. Use of the ERP system will have positive effects on the 1 2 3 4 5
performance of my job
2. Use of the ERP system can decrease the time needed for 1 2 3 4 5
my important job responsibilities
3. Use of the ERP system can significantly increase the 1 2 3 4 5
quality of output of my job
4. Use of the ERP system can increase the effectiveness of 1 2 3 4 5
performing job tasks (e.g., communication)
5. The ERP system can increase the quantity of output for 1 2 3 4 5
the same amount of effort
(D) Long-term consequences
1. Use of the ERP system will increase the opportunity for 1 2 3 4 5
preferred future job assignments
2. Use of the ERP system will increase the amount of 1 2 3 4 5
variety on my job
3. Use of the ERP system will increase the opportunity for 1 2 3 4 5
more meaningful work
4. Use of the ERP system will increase the flexibility of 1 2 3 4 5
changing jobs
5. Use of the ERP system will increase the opportunity to 1 2 3 4 5
gain job security
(E) Facilitating conditions
1. The ERP system is assessable to me when 1 2 3 4 5
I need it
 ARTICLE IN PRESS
M.-K. Chang et al. / Int. J. Production Economics 113 (2008) 928–942 941

2. A person (or group) is available for assistance with 1 2 3 4 5

ERP system difficulties
3. Specialized instruction concerning the ERP system is 1 2 3 4 5
available to me
4. Formal training sessions are offered periodically by my 1 2 3 4 5
organization to improve the use of ERP system
(F) Compatibility
1. Data captured in the ERP system and their format do 1 2 3 4 5
not match my current data needs
2. The ERP system does not match my current processing 1 2 3 4 5
procedure
(G) Usage
1. I use the ERP system very intensively (many hours per 1 2 3 4 5
day, at work)
2. I use the ERP system very frequently (many times per 1 2 3 4 5
day, at work)
3. Overall, I use the ERP system a lot 1 2 3 4 5
(H) Feeling of using ERP system
Using ERP system to do my job would be: (Please put an ‘‘X’’ over the line)
Enjoyable – – – – – Disgusting
Dull – – – – – Exciting
Pleasant – – – – – Unpleasant
Beneficial – – – – – Harmful

References
Ajzen, I., Fishbein, M., 1980. Understanding Attitudes and
Predicting Social Behavior. Prentice-Hall, Englewood Cliffs, NJ.
Amoako-Gyampah, K., Salam, A.F., 2004. An extension of the
technology acceptance model in an ERP implementation
environment. Information & Management 41 (6), 731–745.
Bagozzi, R.P., Yi, Y., 1990. Assessing method variance in
multitrait-multimethod matrices: The case of self-reported
affect and perception at work. Journal of Applied Psychology
75 (5), 547–560.
Bergeron, F., Raymond, L., Rivard, S., Gara, M.F., 1995.
Determinants of EIS use: Testing a behavioral model.
Decision Support Systems 14, 131–146.
Botta-Genoulaz, V., Millet, P., 2006. An investigation into the
use of ERP systems in the service sector. International
Journal of Production Economics 99 (1–2), 202–221.
Botta-Genoulaz, V., Millet, P., Grobot, B., 2005. A survey on the
recent research literature on ERP systems. Computers in
Industry 95 (2), 510–522.
Bradford, M., Florin, J., 2003. Examining the role of innovation
diffusion factors on the implementation success of enterprise
resource planning systems. International Journal of Account-
ing Information Systems 4 (3), 205–225.
Calisir, F., Calisir, F., 2004. The relation of interface usability
characteristics, perceived usefulness, and perceived ease of use
to end-user satisfaction with enterprise resource planning (ERP)
systems. Computers in Human Behavior 20 (4), 505–515.
Chau, P.Y.K., 1996. An empirical assessment of a modified
technology acceptance model. Journal of Management
Information Systems 13 (2), 185–204.
Cheung, W., Chang, M.K., Lai, V., 2000. Prediction of internet
and world wide web usage at work: A test of an extended
Triandis Model. Decision Support Systems 30 (1), 83–100.
Davis, F.D., 1989. Perceived usefulness, perceived ease of use,
and user acceptance of information technology. MIS Quar-
terly 13 (3), 319–340.
Davis, F.D., Bagozzi, R.P., Warshaw, P.R., 1989. User
acceptance of computer technology: A comparison of two
theoretical models. Management Science 35 (8), 982–1003.
Dishaw, M.T., Strong, D.M., 1999. Extending the technology
acceptance model with task-technology fit constructs. Infor-
mation and Management 36, 9–21.
Ehie, I.C., Madsen, M., 2005. Identifying critical issues in
enterprise resource planning (ERP) implementation. Compu-
ters in Industry 56 (6), 545–557.
Fiske, D.W., Campbell, D.T., 1992. Citations do not solve
problems. Psychological Bulletin 112 (3), 393–395.
Gefen, D., Keil, M., 1998. The impact of developer responsiveness
on perceptions of usefulness and ease of use: An extension
of the Technology Acceptance Model. DATA BASE 29 (2),
35–49.
Godin, G., Maticka-Tyndale, E., Adrien, A., Manson-Singer, S.,
Willms, D., Cappon, P., 1996. Cross-cultural testing of three
social cognitive theories: An application to condom use.
Journal of Applied Social Psychology 26 (17), 1556–1586.
Goldberg, A., 2000. The ERP trap. Upside 12 (11), 32.
Hair, J.F., Black, B., Babin, B., Anderson, R.E., Tatham, R.L.,
2006. Multivariate Data Analysis. Pearson Prentice Hall,
Upper Saddle River, NJ.
Holsapple, C.W., Wang, Y.-M., Wu, J.-H., 2005. Empirically
testing user characteristics and fitness factors in enterprise
 ARTICLE IN PRESS
942 M.-K. Chang et al. / Int. J. Production Economics 113 (2008) 928–942
resource planning success. International Journal of Human–
Computer Interaction 19 (3), 323–342.
Igbaria, M., Iivari, J., Maragahh, H., 1995. Why do individuals
use computer technology? A Finish case study. Information
and Management 5 (2), 7–38.
Jeyaraj, A., Rottman, J.W., Lacity, M.C., 2006. A review of the
predictors, linkages, and biases in it innovation adoption
research. Journal of Information Technology 21, 1–23.
Krasner, H., 2000. Ensuring e-business success by learning from
ERP failures. IT Pro 2 (1), 22–27.
Kumar, K., 2000. ERP experiences and evolution. Communica-
tions of the ACM 43 (4), 22–26.
Liang, H., Saraf, N., Hu, Q., Xue, Y., 2007. Assimilation of
enterprise systems: The effect of institutional pressures and the
mediating role of top management. MIS Quarterly 31 (1), 59–87.
Lucas, H.C., Walton, E.J., Ginzberg, M.J., 1988. Implementing
packaged software. MIS Quarterly 12, 537–549.
Marble, R.P., Lu, Y., 2007. Culturalizing enterprise software for
the Chinese context: An argument for accommodating
guanxi-based business practices. International Journal of
Production Economics 107 (2), 364–379.
Morris, M.G., Dillon, A., 1997. How user perceptions influence
software use. IEEE Software 14 (July/August), 58–64.
Motwani, J., Subramanian, R., Gopalakrishna, P., 2005. Critical
factors for successful ERP implementation: Exploratory findings
from four case studies. Computers in Industry 56 (6), 524–544.
Neter, J., Kutner, M.H., Nachtsheim, C.J., Wasserman, W.,
1996. Applied Linear Statistical Models. McGraw-Hill,
New York.
Nunnally, J.C., Bernstein, I.H., 1994. Psychometric Theory.
McGraw-Hill, New York.
Orlikowski, W.J., Iacono, C.S., 2001. Desperately seeking the
‘‘IT’’ in IT research—A call to theorizing the IT artifact.
Information System Research 12 (2), 121–134.
Paré, G., Elam, J., 1995. Discretionary use of personal computers by
knowledge workers: Testing of a social psychology theoretical
model. Behavior & Information Technology 14 (4), 215–228.
Roger, E.M., 1995. Diffusion of Innovations. Free Press,
New York.
Sarker, S., Lee, A.S., 2003. Using a case study to test the role of
three key social enablers in ERP implementation. Information
and Management 40 (8), 813–829.
Shanmugam, R., Forcht, K., Busing, M.E., 2000. SAP R/3: A
reengineering tool at Tenneco, Inc. Journal of Computer
Information Systems 41 (1), 18–24.
Soh, C., Kien, S.S., Tay-Yap, J., 2000. Cultural fits and misfits: Is
ERP a universal solution? Communications of the ACM 43
(4), 47–51.
Tchokogué, A., Bareil, C., Duguay, C.R., 2005. Key lessons
from the implementation of an ERP at Pratt & Whitney
Canada. International Journal of Production Economics 95 (2),
151–163.
Teo, T.S.H., Lim, V.K.G., Lai, R.Y.C., 1999. Intrinsic and
extrinsic motivation in Internet usage. Omega, International
Journal of Management Sciences 27, 25–37.
Thompson, R.L., Higgin, C.A., Howell, J.M., 1991. Personal
computing: Toward a conceptual model of utilization. MIS
Quarterly 15 (1), 125–143.
Thompson, R.L., Higgin, C.A., Howell, J.M., 1994. Influence of
experience on personal computer utilization: Testing a
conceptual model. Journal of Management Information
Systems 11 (1), 167–187.
Triandis, H.C., 1977. Interpersonal Behavior. Brooks/Cole,
Monterey.
Triandis, H.C., 1980. Values, attitudes, and interpersonal
behavior. In: Page, M.M. (Ed.), Nebraska Symposium on
Motivation, 1979: Beliefs, Attitudes, and Values. University
of Nebraska Press, Lincoln, pp. 195–259.
Valois, P., Desharnais, R., Godin, G., 1988. A comparison of the
Fishbein and Ajzen and the Triandis attitudinal models for
the prediction of exercise intention and behavior. Journal of
Behavioral Medicine 11 (5), 459–472.
Van Everdingen, Y., 2000. ERP adoption by European midsize
companies. Communications of the ACM 43 (4), 27–31.
Venkatesh, V., 2006. Where to go from here? Thoughts on future
directions for research on individual-level technology adop-
tion with a focus on decision making. Decision Sciences 37
(4), 497–518.
Wah, L., 2000. Give ERP a change. Management Review 89 (3),
20–24.
Xue, Y., Liang, H., Boulton, W.R., Snyder, C.A., 2005. ERP
implementation failures in China: Case studies with implica-
tions for ERP vendors. International Journal of Production
Economics 97 (3), 279–295.
Yeh, T.M., Yang, C.C., Lin, W.T., 2007. Service quality and ERP
implementation: A conceptual and empirical study of
semiconductor-related industries in Taiwan. Computers in
Industry 58 (8-9), 844–854.
Yusuf, Y., Gunasekaran, A., Abthorpe, M.S., 2004. Enterprise
information systems project implementation: A case study of
ERP in Rolls-Royce. International Journal of Production
Economics 87 (3), 251–266.
Zhang, Z., Lee, M.K.O., Huang, P., Zhang, L., Huang, X., 2005.
A framework of ERP systems implementation success in
China: An empirical study. International Journal of Produc-
tion Economics 98 (1), 56–80.