Omega 29 (2001) 143–156
www.elsevier.com/locate/dsw
Resource constraints and information systems implementation
in Singaporean small businesses
James Y.L. Thong ∗
Department of Information and Systems Management, School of Business and Management, Hong Kong University of Science
and Technology, Clear Water Bay, Kowloon, Hong Kong
Received 3 April 2000; accepted 5 July 2000
Abstract
While the information systems (IS) literature has identied potential factors of IS implementation success, none has
investigated the relative importance of these factors in the context of small businesses. Small businesses have very dierent
characteristics from large businesses; notably, small businesses suer from resource poverty. Without knowing the relative
importance of key factors, small businesses may be expending their limited resources and energy on less important factors which
have limited contribution to IS implementation success. This paper develops a resource-based model of IS implementation for
small businesses based on Welsh and White’s (Harv Bus Rev 59(4) (1981) 18–32) framework of resource constraints in small
businesses and Attewell’s (Organ Sci 3(1) (1992) 1–19) knowledge barrier theory. The model is then tested on a sample of
114 small businesses. The results show that small businesses with successful IS tend to have highly eective external experts,
adequate IS investment, high users’ IS knowledge, high user involvement, and high CEO support. External expertise is the
predominant key factor of IS implementation success in small businesses. ? 2001 Elsevier Science Ltd. All rights reserved.
Keywords: External expertise; Implementation; Information systems; Resource-based theory; Small business
1. Introduction
The computer-based information systems (IS) are a
major technological innovation during this century. The
benets of IS to large businesses are well-documented in
the popular press, trade magazines, and IS text books. At
the same time, the majority of IS research has concentrated
on IS implementation in large businesses. However, the literature has also argued that there is a relationship between
organizational size and IS implementation success (e.g.
[1– 4]). Due to the inherent dierences between small and
large businesses, research ndings based on large businesses
cannot be generalized to small businesses (e.g. [5 –8]). As
small businesses constitute over 90% of operating businesses in many countries, there is a great need for more rigorous research that is relevant to this important sector of the
economy [9].
∗
Tel.: +852-2358-7631; fax: +852-2358-2421.
E-mail address: jthong@ust.hk (J.Y.L. Thong).
Small businesses are under increasing pressure to employ
IS to maintain their competitive positions or simply to survive. At the same time, there are more barriers to IS implementation in small businesses than in large businesses due to
the high capital investment and skilled manpower involved
in implementing and operating IS. If the IS implementation is successful, potential benets to small businesses can
include increased sales, improved protability, increased
productivity, improved decision-making, and secured competitive positions (see [10 –14]). On the other hand, if the
IS implementation is unsuccessful, it will have severe repercussions on small businesses with their limited resources as
they can rarely rely on organizational slack to act as a buer
[15,16]. Hence, a key research question is identifying the
determinants of successful IS implementation in small businesses.
A review of the IS, small business, and management literature was conducted to assess the current state of research on
small businesses computerization. As we are examining IS
specically, it is highly relevant to review the IS literature.
0305-0483/01/$ - see front matter ? 2001 Elsevier Science Ltd. All rights reserved.
PII: S 0 3 0 5 - 0 4 8 3 ( 0 0 ) 0 0 0 3 5 - 9
144
J.Y.L. Thong / Omega 29 (2001) 143–156
It is also relevant to include the small business literature as
we are studying the context of small businesses. Finally, the
management literature is included as it also publishes studies on the management of IS in small business. It should be
noted that we are interested in studies of factors that lead to
successful IS implementation rather than studies of factors
that lead to the decision to adopt IS or further IS adoption.
Adoption, implementation, and post-implementation (including further adoption) are three dierent stages in the
technology innovation cycle. The adoption stage is where a
decision is made about whether to adopt a new technology.
If the decision is to go ahead with adoption, the implementation stage involves implementing the technology in the
business. Once the technology has been implemented successfully, the post-implementation stage is concerned with
how much organizational learning takes place within the
business so as to facilitate further technology adoption
[17–19]. As the current study is concerned with the implementation stage, literature involving the second stage of the
technology innovation cycle will be most relevant.
Based on case studies and surveys, a number of potential
factors have been identied in the literature as critical to IS
implementation success in small businesses (e.g. [20 –28]).
However, there are limitations with the prior research. None
of the prior literature has investigated the relative importance of the identied factors of IS implementation success.
Without knowing the relative importance of these factors,
small businesses may be expending their limited resources
and energy on less important factors which have limited contribution to IS implementation success. Thus, there is a need
to identify the more important factors so that implications
and guidelines can be drawn for eective IS implementation
in small businesses. In addition, most of the prior studies
were carried out in the 1980s. Since then, computer hardware have become more powerful, software applications are
more sophisticated, and prices are increasingly more aordable even for small businesses. Some of the previously identied factors are no longer applicable in today’s business
environment (e.g. type of computer, interactive versus batch
systems, use of custom-developed applications). Thus, there
is also a need to reexamine the factors of IS implementation
success in small businesses in today’s context.
The prior literature also tends to use bivariate analysis,
specically correlation analysis. Bivariate analysis which
considers two variables at a time allows only a limited testing
of the research question. In comparison, multivariate analysis allows us to examine the eects of multiple independent
variables on a dependent variable simultaneously. Some researchers have even claimed that unless a problem is treated
as a multivariate problem, it is treated supercially. Further,
developments in structural equation modeling techniques, or
second-generation multivariate analysis, allow even stronger
tests to be brought to bear on the research question [29 –
31]. Thus, statistical techniques other than simple bivariate
analysis need to be utilized to provide a stronger test of the
research question. Finally, most prior studies used only one
key informant as representative of user evaluation of IS implementation success and sometimes this key informant may
not even be a user of the IS. The research design would be
more rigorous if multiple respondents were included [32].
The objective of this paper is to develop an updated IS
implementation model in small businesses. The model is
then tested on a survey sample of small businesses with
multiple respondents from each small business. The relative
importance of the key factors can then be examined using a
structural equation modeling technique [33].
The rest of this paper is organized as follows. The following section describes the theoretical background for this
research. After that, we present the research model and develop the research propositions. We then describe the research methodology. This is followed by a presentation of
the data analysis results and discussion of the ndings. Finally, we conclude the paper with implications for practice
and research.
2. Theoretical background
Small businesses tend to have simple and highly centralized structures with the chief executive ocers (CEOs),
who are also the owners, making most of the critical decisions [34,35]. These CEOs have a great inuence on the
technology adoption decision [36,37]. Small businesses,
especially those with less than 100 employees, also tend to
employ generalists rather than specialists [38,39]. Operating
procedures are not written down or standardized. Other
distinctive characteristics include reliance on short-term
planning rather than long-term strategic plans, fewer bureaucratic procedures, less complex interpersonal and
political relations, and less organizational inertia [34,40].
Further, few small businesses use management techniques
such as nancial analysis, forecasting, and project management [6]. The decision-making process of small business
managers is more intuitive and less dependent on formal
decision models [41]. Due to their distinctive dierences
from large businesses, there is a need to examine small
businesses separately rather than as scale models of large
businesses (e.g. [5,8,42,43]).
The theoretical framework leading to the proposed IS implementation model for small businesses is based on the
resource-based theory of the rm [44 – 46]. This theory has
been hailed as a promising approach to the study of rms
[47– 49]. According to the resource-based theory, rms are
characterized as collections of resources or capabilities. A
rm’s resources may include both tangible and intangible
assets including capabilities, organizational processes, information, and knowledge, controlled by a rm that enable
the rm to conceive and implement strategies that improve
its eciency and eectiveness [50]. It emphasizes understanding the internal capabilities that enable rms to secure
competitive positions [47]. Further, the value of a resource
is likely to be partially contingent upon the presence of other
J.Y.L. Thong / Omega 29 (2001) 143–156
resources; i.e. a system of resources matter more than individual resources taken separately [51]. In this paper, the
resource-based theory of the rm is applied to the issue of
IS implementation success in small businesses.
Closely related to the concepts of resource-based theory is
Welsh and White’s [43] framework of resource constraints
in small businesses. According to them, the unique characteristics of small businesses are exemplied in the condition
known as resource poverty where small businesses operate under severe time constraints, nancial constraints, and
expertise constraints. Time constraints refer to the limited
amount of time available for activities beyond the normal
job responsibilities of individuals in the small businesses.
Due to their limited time, the CEOs and their employees
tend to have a short-range perspective with regard to IS
implementation and are not very involved in the IS implementation projects. If the CEOs and potential users do not
participate in the IS implementation, the quality of the IS will
suer. Financial constraints refer to the limited amount of
nance available for activities beyond the normal operations
of the small businesses. Due to their nancial constraints,
small businesses have to control their cash ows carefully
and do not have unlimited funds for their IS implementation
projects. They tend to choose the cheapest system which
may be inadequate for their purpose and underestimate the
amount of time and eort required for IS implementation
[52]. Expertise constraints refer to the limited amount of expertise within the small businesses to carry out activities beyond designated job responsibilities. They do not have the
necessary inhouse IS expertise or a formal IS department.
Small businesses tend to engage consultants and IT vendors
to develop and support their information systems. Due to
their lack of internal IS expertise, small businesses do not
have the capability to undertake their own IS implementation projects. In summary, resources such as time, nance,
and expertise, that are necessary for planning represent the
most critical diculties in small businesses [7]. Inadequate
resources spent on IS implementation increase the risk of IS
implementation failure.
While the resource-based theory emphasizes the importance of internal resources in a rm, external resources are
also important in the context of small businesses. Attewell’s
[53] technology diusion theory emphasizes the role of
external entities, such as consultants and IT vendors, as
knowledge providers in lowering the knowledge barrier or
knowledge deciency on the parts of potential IS adopters.
Small businesses tend to delay inhouse IS implementation
because they have insucient knowledge to implement IS
successfully. In response to this knowledge barrier, mediating entities come into existence which progressively lower
this barrier, and make it easier for small businesses to adopt
and implement IS without extensive inhouse expertise.
These mediating entities can capture economies of scale
in learning. After developing many systems, the IT vendor
would have learned from earlier attempts and developed a
relatively error-free system. Similarly, the consultant would
145
Fig. 1. Research model.
have acquired a wealth of experience in IS implementation.
Hence, external resources in the form of external experts
are also important to small businesses in implementing IS
successfully.
3. Research model
The research model is developed based on the prior theoretical background discussion (see Fig. 1). By adopting the
resource-based view of the rm and taking into account the
distinctive characteristics of small businesses, the IS implementation environment in a small business is conceptualized
in terms of three categories of resource constraints. The categories of resource constraints are time, nance, and expertise. While resources can also be viewed as facilitators, we
have adopted the view of constraints to be consistent with
Welsh and White’s [43] concept of resource poverty. Further, according to Attewell’s [53] knowledge barrier theory,
small businesses with their lack of internal IS expertise will
need to engage IS expertise from the external environment.
3.1. IS implementation success
Two popular constructs of IS implementation success
are user information satisfaction and organizational impact.
Based on a literature review of prior constructs of IS implementation success, DeLone and McLean [54] developed a
meta-model that included both user information satisfaction
and organizational impact as appropriate constructs of the
eectiveness of IS. These two constructs are elaborated on
below.
146
J.Y.L. Thong / Omega 29 (2001) 143–156
3.1.1. User information satisfaction
According to Seddon and Kiew [55], user information satisfaction may be considered the best “omnibus” construct of
IS implementation success. Other researchers [32,56] have
also argued that user information satisfaction provides the
most useful assessment of IS implementation success. User
information satisfaction may be dened as the extent to
which users believe the IS meet their information requirements [57]. If the IS meet the requirements of the users,
the users’ satisfaction with the IS will increase. Conversely,
if the IS do not provide the needed information, the users
will become dissatised. This denition of IS success is a
popular one in the IS literature. It can also be a meaningful surrogate for the critical but unquantiable result of the
IS, namely, changes in organizational eectiveness [57]. In
small business research, user information satisfaction has often been used as the dependent variable (e.g. [4,24 –26,58]).
Hence, user information satisfaction is used as a construct
of IS implementation success in this study.
3.1.2. Organizational impact
Organizational impact is dened as the impact of the IS
on the performance of the small business. IS implementation success only has meaning to the extent that the IS contribute to organizational eectiveness. In a small business,
the impact and value of the IS are likely to be achieved
through sta productivity, operations eciency, improvement in decision-making, increased sales revenue, increased
prot, and increased competitive advantage [59,60]. Further,
the eects of various factors on organizational impact of the
IS are mediated by user information satisfaction [54]. The
higher the level of user information satisfaction with the IS,
the greater will be the organizational impact on the business.
If the information needs of top management are met satisfactorily by the IS then their decision-making performance will
improve leading to increased positive organizational impact.
Hypothesis 1: Organizational impact is positively related
to the level of user information satisfaction.
3.2. Time constraint
3.2.1. CEO support
The CEO of a small business has limited time to spend
on IS implementation. But if the CEO can aord to be more
involved in the IS implementation project, the probability
of a successful implementation is much higher. The CEO
has the authority to inuence other members of the business and is more likely to succeed in overcoming organizational resistance to accept the IS [61]. Jarvenpaa and Ives
[62] noted that hands-on management in IS implementation
may be much more important in a small business where the
CEO commonly makes most key decisions and is perhaps
the only one who can harness IS to business objectives.
This is especially true in a small business where the CEO
is the person who understands the business best. A support-
ive CEO is also more likely to commit scarce resources and
adopt a longer-range perspective to the benets of IS implementation [63,64]. This is also believed to be the case in
small businesses [21,28,34]. Hence, it is hypothesized that
user information satisfaction is likely to be high when the
level of CEO support is high.
Hypothesis 2: User information satisfaction is positively
related to the level of CEO support.
3.2.2. User involvement
Employees in a small business tend to be generalists rather
than specialists in a certain eld [38,39]. They have multiple
job functions to perform within a limited time. Hence, if the
small business encourages the ultimate users to be involved
in the IS implementation by allowing them time-o from
their normal responsibilities, then the IS implementation is
more likely to be successful. Benets can include a better
t of the IS with user requirements, ease of operating the IS
due to learning experience during the design phase, feeling
of ownership, and reduced resistance to change [65 – 67].
Hence, it is hypothesized that user information satisfaction
is likely to be high when the level of user involvement is
high.
Hypothesis 3: User information satisfaction is positively
related to the level of user involvement.
3.2.3. IS planning
Similarly, if the small business can aord to spend more
time on IS planning, the chances of IS implementation success will be higher. The importance of IS planning in terms
of requirements analysis, system analysis and design, and
resource controls has been stated in the literature [68,69].
Ginzberg [64] identied the extent of project denition and
planning as a key recurrent issue in IS implementation success of large businesses. More eort spent on IS planning
can lead to better t of the business requirements with the
nal system. There is also some evidence of a positive relationship between user information satisfaction and the level
of IS planning in small businesses [25]. Hence, user information satisfaction is expected to be high when the level of
IS planning is high.
Hypothesis 4: User information satisfaction is positively
related to the level of IS planning.
3.3. Financial constraint
3.3.1. IS investment
Research has shown that sucient nancial resources increases the likelihood of IS implementation success [70,71].
As small businesses often lack nancial resources, insucient funds may be allocated for IS implementation.
Insucient nancial resources place constraints on the IS
implementation eort and often lead to selection of less
eective IS. Selection of such IS, although cheaper, will
J.Y.L. Thong / Omega 29 (2001) 143–156
result in false economy because the IS do not meet the
requirements of the business and will be inadequate in the
long run [52]. IS implementation involves capital investments and often has organization-wide implications. The
future of the business may be jeopardized by unsuccessful
investments in IS because a technical failure in the IS can
have a major negative impact on the business that is heavily
dependent on them. The setback has even greater ramication for a small business as it may even result in business
failure [72]. Hence, it is hypothesized that increased allocation for IS investment will increase the likelihood of IS
implementation success.
Hypothesis 5: User information satisfaction is positively
related to the level of IS investment.
3.4. Expertise constraint
3.4.1. Users’ IS knowledge
There is generally a lack of internal IS expertise in a small
business [22,39]. Employees are usually employed to work
on the daily operations of the business and not for their
ability to program or use software packages. Further, it is
dicult to recruit and retain IS professionals in a small business due to the tight labor market for IS professionals and
the absence of a career ladder for them in a small business.
However, if the employees have adequate IS knowledge,
they can contribute more eectively to the IS implementation through their involvement in the requirements and design phases. They will also have more realistic expectations
from the IS and be more comfortable participating in the IS
implementation process [66,73]. Insucient IS knowledge
has been found to lead to IS selection failure and failure
to use the IS [74]. Hence, it is hypothesized that a higher
level of IS knowledge is likely to increase IS implementation
success.
Hypothesis 6: User information satisfaction is positively
related to the level of users’ IS knowledge.
3.5. External expertise
Due to the lack of inhouse IS expertise, small businesses
are likely to be much more dependent on external expertise such as consultants and vendors [20,72]. According to
Attewell’s [53] theory, these external experts act as mediators to compensate for the lack of IS knowledge in the
small business and lower the IS knowledge barrier to successful IS implementation. The responsibilities of a consultant are to provide consultancy service specically to help
businesses implement eective IS. Consultancy service can
include performing information requirements analysis, recommending suitable computer hardware and software, and
managing the IS implementation [38]. The responsibilities
of a vendor generally include providing the computer hardware, software packages, technical support, and training
of users. It is also important to maintain a good working
147
relationship among the various parties (i.e. the CEO, users,
consultant, and vendor) in the IS implementation. In IS implementation of the small businesses, the vendor may also
play the role of a consultant, and thus performs additional
duties besides the usual responsibilities [58]. In view of the
possibility of the consultant being the vendor, we will treat
the responsibilities of the external experts as a combination
of the duties of the consultant and the vendor. In the research
model, external expertise is a second-order latent variable
of both types of external expertise. There is some prior
evidence to suggest positive correlations between user information satisfaction and external expertise [24,39]. Hence, it
is hypothesized that user information satisfaction is likely
to be high when the level of external expertise is high.
Hypothesis 7: User information satisfaction is positively
related to the eectiveness of external expertise.
4. Research methodology
4.1. Measurement of variables
The measures used in this study were developed through
an extensive literature review followed by iterative reviews
by both practitioners and experienced IS faculty. Further, the
measures had been used in prior studies and were found to
demonstrate adequate reliability and validity. The questionnaires containing the measures were also pilot-tested before
the main survey. Except for IS investment, the remaining
variables were measured as perceptual items on seven-point
Likert scales. Pre-tax prot and sales revenue were measured as seven-points Likert scales as small businesses are
reluctant to reveal actual revenue or prot. In order to secure their assistance in completing the survey questionnaire,
perceptual rather than objective measures were used. According to Dess and Robinson [75], subjective measures can
be appropriate surrogates for organizational performance.
Table 1 presents the measurement of the variables.
4.2. The sample
There is no generally accepted denition of a small business. Three commonly used criteria for dening a small
business are number of employees, annual sales, and xed
assets [76]. In this study, the criteria for dening a small
business were adopted from the Association of Small and
Medium Enterprises (ASME) in Singapore. A small business must satisfy at least two of the following criteria: (1)
number of employees should not exceed 100; (2) xed assets should not exceed US$7.2 million; and (3) annual sales
should not exceed US$9 million.
The names and addresses of small businesses that have
computerized were obtained from a database maintained
by the Singapore National Computer Board (NCB). The
148
J.Y.L. Thong / Omega 29 (2001) 143–156
Table 1
Measurement of variables
Variable
Measure
Source
User information satisfaction
1.
2.
3.
4.
5.
6.
Currency of reports
Timeliness of reports
Reliability of reports
Relevancy of reports
Accuracy of reports
Completeness of reports
[57,85]
adapted
Organizational impact
1.
2.
3.
4.
5.
6.
Pre-tax prot
Sales revenues
Sta productivity
Competitive advantage
Operating cost
Quality of decision-making
[59,60]
CEO support
1.
2.
3.
4.
5.
CEO
CEO
CEO
CEO
CEO
attendance at project meetings
involvement in information requirements analysis
involvement in reviewing consultant’s recommendations
involvement in IS decision-making
involvement in monitoring project
[86]
User involvement
1.
2.
3.
4.
User
User
User
User
attendance at project meetings
involvement in information requirements analysis
involvement in reviewing consultant’s recommendations
involvement in IS decision-making
[86]
IS planning
1.
2.
3.
4.
5.
Financial resources planning
Human resources planning
Information requirements analysis
Implementation (software development, installation, and conversion)
Post-implementation (operation, maintenance, future needs)
IS investment
1. Investment in computer hardware and software
[4,70,87]
Users’ IS knowledge
1. Years of computer experience
2. Understanding of computers in comparison to persons in
your position at other small companies
3. Number of computer courses taken
[59]
Consultant eectiveness
1.
2.
3.
4.
Eectiveness in performing information requirements analysis
Eectiveness in recommending suitable computer solution
Eectiveness in managing implementation
Relationship with other parties in the project (CEO, Users, Vendor)
[58]
Vendor support
1.
2.
3.
4.
5.
6.
Adequacy of technical support during IS implementation
Adequacy of technical support after IS implementation
Quality of technical support
Adequacy of training provided
Quality of training provided
Relationship with other parties in the project (CEO, Users, Consultant)
[58]
NCB conducts a national IT survey on a large cross-section
of business organizations every two years. Stratied random sampling was used to ensure that the sample was
representative of the national prole. Hence, our sample
was not a convenient sample per se. Nonprot businesses,
[86]
public-listed businesses, and wholly owned subsidiaries of
large businesses were excluded from the survey sample.
Three hundred and four small businesses fulll the ASME
criteria and were included in the survey sample. Two weeks
after the questionnaires were mailed, follow-up telephone
J.Y.L. Thong / Omega 29 (2001) 143–156
calls were made to nonresponding businesses to encourage
a higher response rate. One hundred and thirty small businesses responded, giving a response rate of 43%. Responses
from 16 businesses were excluded from the nal sample
because of incomplete data resulting in 114 usable sets of
questionnaires. In order to assess the possibility of nonresponse bias, we compared the responses of the early returns
with the late returns [77,78]. The rationale for this test is that
late respondents are likely to have similar characteristics
to nonrespondents. The MANOVA test did not detect any
signicant dierences in the research variables (Wilks’
= 0:958; F-sig: = 0:919), and hence nonresponse bias was
not a major concern.
4.3. Data collection
The study was conducted in two phases: a pilot study
and a questionnaire survey. Two questionnaires, the Project
Manager Questionnaire and the Computer User-Manager
Questionnaire, were designed for data collection. The rst
questionnaire was to be completed by the inhouse person
who was administratively responsible for IS implementation. It solicited data on the research variables and IS
characteristics. The second questionnaire was to be completed by senior managers who were users of the IS and the
computer-generated reports. It requested data on IS eectiveness and level of IS knowledge. Views from multiple
respondents were solicited to provide a more representative picture of the IS implementation success in the small
businesses.
In the pilot study phase, ve small businesses were randomly chosen from the small business database to pretest the
questionnaires. Five project managers and fteen managers
who used the businesses’ IS completed the questionnaires.
Interviews were conducted with these respondents in order
to determine whether there were problems with the questionnaires. Based on their feedback, some statements were
reworded and explanations were provided where necessary
to clarify the questions. Responses from these ve small
businesses were not included in the nal sample.
In the questionnaire survey phase, a package containing
four items was mailed to the CEO of each of the small business in the survey sample. The four items were: a cover
letter, one Project Manager Questionnaire, three Computer
User-Manager Questionnaires, and a prepaid envelope. The
cover letter requested the CEO to direct the relevant questionnaires to the manager in charge of the IS implementation project and three senior managers who used the IS.
The respondents were assured of the condentiality of their
responses. They could also return the completed questionnaires in individually sealed envelopes.
Interviews were also conducted with the project managers
and computer user-managers in 67 of the small businesses.
As far as possible, all managers who used the reports generated by the IS were included in the study. Responses from
the managers were not revealed to each other or the CEOs
149
of the small businesses. During the interviews, the respondents were asked to explain in greater detail their responses
to the questionnaires and to qualitatively relate their experience with the IS implementation projects. The interviews
helped us to interpret the questionnaire data through deeper
insights into IS implementation issues faced by these small
businesses. To check whether the CEOs in the noninterviewed businesses were biased in selecting managers who
were more satised with the IS, a MANOVA test was conducted on the variables between the 67 interviewed businesses and the 47 noninterviewed businesses. No signicant
dierence was found (Wilks’ = 0:905; F-sig: = 0:440).
Thus, there is no evidence of selection bias.
As the unit of analysis is at the organization level rather
than at the individual user level, the managers’ responses
were aggregated within each small business for purpose of
statistical analysis. In this study, respondents were members
of top management in the small businesses and were in a suitable position to view the success of the IS implementation.
Aggregation of individual responses into an organizational
response was done by averaging the scores by multiple respondents in the same business. The ANOVA tests revealed
signicantly greater variance for user information satisfaction and users’ IS knowledge between the small businesses
than within them, suggesting close agreement between respondents from the same business [4]. Hence, there was statistical support for aggregating the individual responses to
the organization level.
5. Findings
5.1. Sample characteristics
Table 2 presents the characteristics of the survey sample.
The responding small businesses were from the manufacturing, commerce, and service sectors. They all satised the
criteria of a small business as dened earlier. They employed
an average of 50 employees and the mean annual sales was
US$6 million. They had a mean of four years of computer
experience, and the majority had spent more than US$30,000
on their IS implementation. Their hardware platforms were
distributed evenly between microcomputers, microcomputers with local area networks, and minicomputers. Most of
the small businesses had implemented operational and management information systems applications such as accounting systems, inventory control, sales analysis, sales order
processing, and payroll. Finally, all of them had engaged
external experts to implement their information systems.
The eects of ve sample characteristics (number of
employees, annual sales, computer experience, type of
hardware, and business sector) on IS implementation success were examined. Correlation analysis showed that there
was no evidence of signicant correlations at the 10% level
between the IS eectiveness measures and the rst three
variables. The eects of type of hardware conguration and
150
J.Y.L. Thong / Omega 29 (2001) 143–156
Table 2
Characteristics of sample
5.2. Statistical analysis
Frequency (n = 114)a
Sector
Service
Commerce
Manufacturing
31
25
55
Number of employees
1–24
25 – 49
50 –74
75 –100
¿100
48
24
15
21
5
Annual sales (US$ million)
¡ $1:499
1.5 –$2.999
3.0 –$5.999
6.0 –$9.0
¿$9:0
28
27
14
14
25
Computer experience (yr)
0 –1
2–3
4 –5
6 –10
¿10
18
34
22
32
8
Computer expenditure (US$’000)
0 –30
31– 60
61–120
¿120
42
25
21
26
Hardware
Minicomputers and microcomputers
Microcomputers and LAN
Microcomputers only
43
34
33
Top 10 software applications
Accounts receivable
General ledger
Accounts payable
Inventory control
Sales analysis
Sales order processing
Payroll
Purchasing
Budgeting
Job Costing
96
90
87
74
50
47
40
29
24
23
a Figures
may not add up due to missing data.
business sector on the IS eectiveness measures were tested
using one-way ANOVAs. Similarly, there was no evidence
of signicant relationships (p ¿ 0:10). In summary, these
sample characteristics had no eect on IS implementation
success.
Partial least squares (PLS), a powerful second-generation
multivariate analysis technique or structural equation modeling technique, was used for hypotheses testing. PLS is
an approach to assess a causal model involving multiple
variables with multiple observed items by simultaneously
assessing both the structural model and the corresponding
measurement model in an optimal fashion [30,31]. It addresses both models at the same time; compared to factor
analysis which assesses the measurement model only and
path analysis which addresses the structural model alone.
Hence, PLS is superior to traditional regression and factor
analysis [31,79,80].
Fig. 1 presents the structural model in this study. The
structural model describes the relationships among the variables. Further, for each variable in the structural model, there
is a related measurement model (not shown in the gure),
which links each variable in the diagram with its respective
set of items. The items are listed in Table 1.
5.2.1. Testing the measurement model
Testing the measurement model involves examining internal consistency, convergent validity, and discriminant
validity. In PLS, internal consistency of the variables are
assessed with composite reliability. The advantage of
composite reliability over Cronbach’s alpha is that it is
not inuenced by the number of items in the scale [30].
Nunnally’s [81] guideline of 0.80 for assessing internal
consistency can be used to assess composite reliability.
Convergent validity is the degree to which two or more
items measuring the same variable agree [82]. In PLS, two
tests may be used to assess convergent validity. The rst test
is item reliability, which is measured by the factor loading
of the item on the variable. There is no generally accepted
level of what constitutes an acceptable factor loading in PLS
analysis. Fornell [79] recommended a minimum loading of
0.70 which suggests that the item explains almost 50 percent of the variance in the variable; while Falk and Miller
[83] recommended a loading should be at least 0.55 which
explains at least 30 percent of the variance in the variable.
Traditionally, researchers have used the 0.50 level [81]. The
second test is average variance extracted by each variable.
Fornell and Larcker’s [30] criterion that the average extracted variance should be 0.50 or more is usually used to
assess the average variance extracted coecients.
Table 3 presents the results of tests for internal consistency and convergent validity. The composite reliabilities
of all the variables were at least 0.80. All the factor loadings were greater than 0.55. While there were some loadings below the stricter limit of 0.70, the average variance
extracted of the variables were high. Using the average
variance extracted, which was a more conservative test of
convergent validity, all the variables had gures that exceeded 0.50. Hence, the variables in the measurement model
J.Y.L. Thong / Omega 29 (2001) 143–156
151
Table 3
Assessment of the measurement model in PLS
Variable
Mean
Standard
deviation
Factor
loading
User information satisfaction (1–7 scale)
Satisf1
Satisf2
Satisf3
Satisf4
Satisf5
Satisf6
5.55
5.55
5.52
5.73
5.73
5.38
0.95
0.87
0.99
0.83
0.93
0.82
0.82
0.81
0.86
0.85
0.86
0.82
Organizational impact (1–7 scale)
OrgImp1
OrgImp2
OrgImp3
OrgImp4
OrgImp5
OrgImp6
4.47
4.47
5.33
4.93
4.18
5.16
0.96
0.93
1.04
1.01
1.20
0.88
0.79
0.83
0.79
0.82
0.65
0.69
CEO support (1–7 scale)
CEO1
CEO2
CEO3
CEO4
CEO5
4.89
4.89
5.17
5.60
4.99
1.66
1.48
1.57
1.42
1.55
0.79
0.83
0.91
0.79
0.86
User involvement (1–7 scale)
Involve1
Involve2
Involve3
Involve4
5.33
5.34
4.71
5.13
1.43
1.36
1.49
1.39
0.96
0.92
0.58
0.66
IS planning (1–7 scale)
Plan1
Plan2
Plan3
Plan4
Plan5
4.67
4.51
5.09
4.97
4.84
1.36
1.25
1.24
1.16
1.12
0.65
0.72
0.83
0.76
0.68
IS investment (US$’000)
114
37
1.00
Users’ IS knowledge
CKnow1 (yr)
CKnow2 (1–7 scale)
CKnow3 (number)
5.19
4.48
2.99
3.42
1.14
3.99
0.71
0.96
0.56
Consultant eectiveness (1–7 scale)
Consult1
Consult2
Consult3
Consult4
5.03
4.53
4.76
5.13
1.13
1.28
1.24
1.28
0.81
0.78
0.88
0.88
Vendor support (1–7 scale)
Vendor1
Vendor2
Vendor3
Vendor4
Vendor5
Vendor6
4.76
4.48
4.67
4.29
4.28
4.79
1.46
1.71
1.51
1.60
1.60
1.31
0.87
0.83
0.86
0.90
0.90
0.84
Composite
reliability
Average variance
extracted
0.93
0.65
0.89
0.58
0.92
0.70
0.87
0.64
0.85
0.53
1.00
1.00
0.80
0.58
0.90
0.70
0.95
0.75
152
J.Y.L. Thong / Omega 29 (2001) 143–156
Table 4
Discriminant validity of measurement model in PLS
Variable
1
2
3
4
5
6
7
8
9
1.
2.
3.
4.
5.
6.
7.
8.
9.
0.650a
0.200
0.037
0.063
0.056
0.020
0.066
0.062
0.191
0.580
0.031
0.108
0.122
0.002
0.029
0.042
0.020
0.700
0.178
0.147
0.000
0.008
0.061
0.015
0.640
0.073
0.000
0.025
0.072
0.022
0.530
0.000
0.050
0.154
0.070
1.000
0.068
0.044
0.002
0.580
0.056
0.011
0.700
0.304
0.750
User information satisfaction
Organizational impact
CEO support
User involvement
IS planning
IS investment
Users’ IS knowledge
Consultant eectiveness
Vendor support
a Diagonals represent the average variance extracted; other entries represent the shared variances. Shared variances = 0:000 means below
0.001.
demonstrated adequate internal consistency and convergent
validity.
Discriminant validity is the degree to which items differentiate between variables or measure dierent variables
[82]. Discriminant validity can be assessed by examining the
correlations between variables. Each item should correlate
more highly with other items of the same variable than with
items of other variables. To assess this, the squared correlation (shared variance) between two variables should be less
than the average variances extracted by the items measuring
the variables [30]. Table 4 presents the results of the test
for discriminant validity. In all cases, the shared variance
between two variables was less than the average variances
extracted by the items measuring the variables. Hence, the
requirement for the test of discriminant validity was satised, indicating that the measurement model discriminated
adequately between the variables.
5.2.2. Testing the structural model
Following conrmation of adequate psychometric properties in the measurement model, we proceeded to examine the
structural model. This evaluation consisted of an assessment
of the explanatory power of the independent variables, and
an examination of the size and signicance of the path coecients. Jackkning, a nonparametric technique, was used
to test the signicance of the paths. Fig. 2 presents the test
of the structural model. The model accounted for 26% of
the variance in user information satisfaction and 20% of the
variance in organizational impact. The percentage of variance explained was greater than 10%, implying a satisfactory and substantive model [83]. All the standardized path
coecients were signicant at the 5% level. However, a path
coecient may be statistically signicant but not meaningful. Pedhazur [84] recommended using the 0.05 level as the
threshold point. Following this guideline, hypotheses 1, 2, 3,
5, 6, and 7 were supported; only hypothesis 4 was not supported. The results show that user information satisfaction is
a key mediator between the antecedent factors and eventual
organizational impact on the small businesses. Among the
Fig. 2. Assessment of structural model.
key factors, external expertise is the variable most closely
related to user information satisfaction.
6. Discussion
The data analysis shows that external expertise is more
important than the other key factors in IS implementation
success of small businesses. In general, small businesses do
not have the resources to hire internal IS expertise and face
diculties in recruiting and retaining IS professionals. But
this lack of internal IS expertise may be compensated for
by engaging experienced external expertise, in the forms of
consultants and IT vendors, when undertaking IS implementation. Attewell [53] has argued that external institutions
J.Y.L. Thong / Omega 29 (2001) 143–156
play critical roles in lowering the knowledge barriers to IS
diusion, making it easier for businesses to adopt and use
the IS without extensive inhouse expertise. In the case of
small businesses, consultants and IT vendors perform the
role of external institutions that lower the knowledge barriers and make it easier for small businesses to implement IS
successfully. In a small business with its simple organizational structure and limited inter-personal and departmental
politics, IS implementation is essentially a technical matter. Under such circumstances, it is imperative to engage
external IS experts who are experienced, understand the requirements of small businesses, and able to maintain good
working relationships with all concerned parties. This was
borne in upon us in the interviews where the quality of the
external expertise was heavily stressed.
The level of users’ IS knowledge is another important
factor of IS implementation success. If the managers in the
small businesses have high levels of IS knowledge, the resulting IS are more likely to be successful. Potential IS users
should be sponsored by their companies to attend computer
courses to increase their appreciation of the IS implementation process and the potential benets of IS implementation.
With increased IS knowledge, these potential users will be
able to contribute more eectively to the IS implementation process and develop more realistic expectations of the
IS. Hence, taken together with the importance of external
IS expertise, a major nding of this study is that the lack
of IS expertise, whether internal or external, is the primary
barrier to successful IS implementation and is consistent
with Attewell’s [53] notion of “knowledge barriers” to successful IS implementation.
IS investment is the second most important determinant
of IS implementation success. This nding provides empirical support for Ein-Dor and Segev’s [2] hypothesis that
budgeting adequate nancial resources will increase the likelihood of IS implementation success. DeLone [1] has also
found that small businesses tend to spend proportionately
less of revenue on IS implementation than large businesses.
If small businesses could allocate sucient resources for
IS investment, not withstanding their tight cash ow, they
will be able to engage more experienced external experts
and contract for better IS that meet their objectives. If small
businesses decide to choose the lowest-cost external experts
and IS solutions, they may end up with IS solutions that do
not meet their business requirements. These scaled-down IS
solutions could ultimately even end up as white elephants,
as observed in some of the small businesses interviewed.
Thus, small businesses that are able to acquire the needed
capital for IS investment are more likely to secure successful IS. The interviews also revealed that small businesses
that lack nancial resources need to be proactive in securing
low-interest bank loans to nance their IS investment.
After technical expertise and nancial resource variables,
time-constrained variables are the next most important factors for IS implementation success. Due to time constraints
in small businesses, the CEOs and potential IS users could
153
not spend adequate time on IS planning and IS implementation. Among the three a priori time-constrained variables,
user involvement in IS implementation is the most important
for successful IS implementation. If the potential IS users
participate actively in the process of IS implementation, they
will be able to ensure that their suggestions and requirements are incorporated into the IS, feel a sense of ownership over the nal IS, and lower their resistance to adapt to
new work procedures. Adequate involvement of users can
compensate for the lower level of CEO involvement. While
the CEOs should be involved in key decisions aecting the
business and business processes, they need not be actively
involved throughout the IS implementation process. In fact,
given the heavy demand on the CEOs’ time and attention, it
is impractical to advise CEOs of small businesses to devote
a signicant amount of attention to the IS implementation
project. Surprisingly, the level of IS planning has no eect
on IS implementation success. A possible explanation that
emerged from the interviews is that most small businesses
do not consciously conduct IS planning. Beyond deciding
on how much to spend on IS, they are likely to depend on
the external experts in formulating other details of IS planning. In this study, IS planning is positively correlated with
the level of consultant eectiveness. Due to the recognized
importance of IS planning in prior IS implementation research in large businesses, further studies need to be conducted to determine the specic attributes of IS planning that
may have an eect on IS implementation success in small
businesses.
There are three limitations that should be noted in interpreting the ndings of this study. First, as this is a
cross-sectional study, causality of relationships cannot be
demonstrated completely. Further, feedback eects cannot
be investigated. Longitudinal studies are needed to conrm the direction of causality and test for feedback eects.
Second, signicant percentages of the IS implementation
success variances remain unexplained. More research on
this important topic is needed. Finally, in making generalization from the research sample, one has to take into consideration the context of Singapore, a newly industrialized
Asian country. The ndings may not be universally true,
but they are likely to be applicable to IS implementation
success in small businesses with similar cultural contexts
[39]. Findings from this study may also be applicable to
small businesses in developing countries that are interested
in adopting IS.
7. Conclusion
This paper has developed and tested a resource-based
model of IS implementation success in small businesses.
Based on Welsh and White’s [43] framework of resource
constraints in small businesses and Attewell’s [53] knowledge barrier theory, three types of resource constraints were
conceptualized: time constraint, nancial constraint, and
154
J.Y.L. Thong / Omega 29 (2001) 143–156
expertise constraint. Various key factors of IS implementation success were identied as technical expertise resource,
nancial resource, and time resource. The relative importance of these key factors of IS implementation success was
examined using a second generation multivariate technique
or structural equation modeling. The PLS analysis showed
good support for the IS implementation model with the
external technical expertise factor being the most important
followed by the nancial and time-constrained factors. User
information satisfaction was found to be a signicant mediator between these factors and the eventual organizational
impact of the IS.
The implication for small business management is that
to achieve a high level of IS implementation success,
they should direct their eorts at lowering three types
of resource barriers. The rst resource barrier is technical expertise constraints. Due to the lack of internal IS
expertise, small businesses need to engage experienced
consultants and IT vendors to undertake their IS implementation. They should also increase the level of IS knowledge
among potential IS users by sending employees for computer courses and training. The second resource barrier
is nancial constraints. Small businesses need to allocate sucient funds for their IS investment. While they
should not throw money at IS investment, which has not
been shown to be eective, they should also not opt for
the lowest-cost solutions which do not fulll their business requirements. Rather than purchase more expensive
custom-developed software, well-tested dedicated packages may well suit their needs. The third resource barrier
is time constraints. The busy CEO should ensure that
potential IS users are given time-o from their normal
duties to participate in the IS implementation process.
Potential IS users can provide useful inputs that will
lead to IS that better meet the requirements of the small
business.
The implication for research is that the resource-based
view of the rm and the resulting framework of resource
constraints are useful theories to ground future research on
IS implementation in small businesses. In this study, the
resource-based view was applied in conjunction with the
unique characteristic of small businesses and Attewell’s [53]
knowledge barrier theory to conceptualize the key factors of
IS implementation success in small businesses in terms of
three types of resource constraints. The proposed research
model for IS implementation in small businesses is one of
the few models that are guided by a strong theory. Future
research can extent the proposed model by examining other
key factors due to the three types of resource constraints
that may lead to successful IS implementation in small businesses. Further, the proposed model can be tested on samples of small businesses in other countries to determine its
generalizability. While small businesses everywhere suer
from resource constraints, do their contextual environments
have any eects? If found to be important, these factors will
need to be incorporated into the proposed model.
References
[1] DeLone WH. Firm size and the characteristics of computer
use. MIS Quarterly 1981;5(4):65–77.
[2] Ein-Dor P, Segev E. Organizational context and the success
of management information systems. Management Science
1978;24(10):1064–77.
[3] Iacovou CL, Benbasat I, Dexter AS. Electronic data
interchange and small organizations: adoption and impact of
technology. MIS Quarterly 1995;19(4):465–85.
[4] Raymond L. Organizational context and information systems
success: a contingency approach. Journal of Management
Information Systems 1990;6(4):5–20.
[5] Blau PM, Heydebrand WV, Stauer RW. The structure
of small bureaucracies. American Sociological Review
1966;31(2):179–91.
[6] Blili S, Raymond L. Information technology: threats and
opportunities for small and medium-sized enterprises.
International
Journal
of
Information
Management
1993;13(6):439–48.
[7] Cohn T, Lindberg RA. How management is dierent in small
companies. New York: American Management Association,
1972.
[8] Dandridge TC. Children are not little ‘grown-ups’: small
business needs its own organizational theory. Journal of Small
Business Management 1979;17(2):53–7.
[9] Bannock G, Daly M. Small business statistics. London:
Chapman, 1994.
[10] Dwyer E. The personal computer in the small business:
ecient oce practice. Oxford: NCC Blackwell, 1990.
[11] Lincoln DJ, Warberg WB. The role of microcomputers
in small business marketing. Journal of Small Business
Management 1987;25(2):8–17.
[12] Massey Jr. TK. Computers in small business: a case
of under-utilization. American Journal of Small Business
1986;11(2):51–60.
[13] Meredith J. The strategic advantages of new manufacturing technologies for small rms. Strategic Management
Journal 1987;8(3):249–58.
[14] Poutsma E, Walravens A. Technology and small enterprises:
technology autonomy and industrial organization. Delft,
Holland: Delft University Press, 1989.
[15] Carter NM. Small rm adaptation: responses of physicians’
organizations to regulatory and competitive uncertainty.
Academy of Management Journal 1990;33(2):307–33.
[16] Gerwin D. A theory of innovation processes for
computer-aided manufacturing technology. IEEE Transactions
on Engineering Management 1988;35(2):90–100.
[17] Carayannis EG. The strategic management of technological
learning in project=program management. Technovation
1998;18(11):697–703.
[18] Lefebvre E, Lefebvre LA, Roy M-J. Technological penetration
and organizational learning in SMEs: the cumulative eect.
Technovation 1995;15(8):511–22.
[19] Lefebvre LA, Lefebvre E, Harvey J. Intangible assets as
determinants of advanced manufacturing technology adoption
in SMEs: toward an evolutionary model. IEEE Transactions
on Engineering Management 1996;43(3):307–22.
[20] Cragg PB, King M. Small-rm computing: motivators and
inhibitors. MIS Quarterly 1993;17(1):47–60.
[21] DeLone WH. Determinants of success for computer usage in
small business. MIS Quarterly 1988;12(1):51–61.
J.Y.L. Thong / Omega 29 (2001) 143–156
[22] Gable GG. Consultant engagement for computer system
selection: a pro-active client role in small businesses.
Information Management 1991;20(2):83–93.
[23] Gunasekaran A, Cecille P. Implementation of productivity
improvement strategies in a small company. Technovation
1998;18(5):311–20.
[24] Lees JD. Successful development of small business
information systems. Journal of Systems Management
1987;25(3):32–9.
[25] Montazemi AR. Factors aecting information satisfaction in
the context of the small business environment. MIS Quarterly
1988;12(2):239–56.
[26] Raymond L. Organizational characteristics and MIS success
in the context of small business. MIS Quarterly 1985;9(1):
37–52.
[27] Sohal AS. Introducing new technology into a small business:
a case study. Technovation 1999;19:187–93.
[28] Yap CS, Soh CPP, Raman KS. Information system success
factors in small business. Omega 1992;20(5=6):597–609.
[29] Bagozzi RP. Causal models in marketing. New York: Wiley,
1980.
[30] Fornell C, Larcker DF. Evaluating structural equation models
with unobservable variables and measurement error. Journal
of Marketing Research 1981;18(1):39–50.
[31] Wold HA. Soft modeling: the basic design and some
extensions. In: Joreskog KG, Wold H, editors. Systems under
indirect observation. Amsterdam: North-Holland, 1982.
[32] Hamilton S, Chervany NL. Evaluating information system
eectiveness part II: comparing evaluator viewpoints. MIS
Quarterly 1981;5(4):79–86.
[33] Chin WW, Gopal A. An examination of the relative
importance of four beliefs constructs on the GSS adoption
decision: a comparison of four methods. Proceedings of the
26th Hawaii International Conference on System Sciences,
vol. IV, 1993. p. 548–57.
[34] Lefebvre E, Lefebvre LA. Firm innovativeness and CEO
characteristics in small manufacturing rms. Journal of
Engineering Technology Management 1992;9(3):243–77.
[35] Mintzberg H. The structuring of organizations. Englewood
Clis, NJ: Prentice-Hall, 1979.
[36] Lefebvre LA, Harvey J, Lefebvre E. Technological experience
and the technology adoption decisions in small manufacturing
rms. R&D Management 1991;21(3):241–9.
[37] Lefebvre LA, Mason R, Lefebvre E. The inuence prism
in SMEs: the power of CEOs’ perceptions on technology
policy and its organizational impacts. Management Science
1997;43(6):856–78.
[38] Gable GG. A multidimensional model of client success
when engaging external consultants. Management Science
1996;42(8):1175–98.
[39] Thong JYL, Yap CS, Raman KS. Top management support,
external expertise, and information systems implementation
in small businesses. Information Systems Research 1996;
7(2):248–67.
[40] Harvey J, Lefebvre LA, Lefebvre E. Exploring the relationship
between productivity problems and technology adoption in
small manufacturing rms. IEEE Transactions on Engineering
Management 1992;39(4):352–8.
[41] Rice GH, Hamilton RE. Decision theory and the
small businessman. American Journal of Small Business
1979;4(1):1–9.
155
[42] D’Amboise G, Muldowney M. Management theory for
small business: attempts and requirements. Academy of
Management Review 1986;13(2):226–40.
[43] Welsh JA, White JF. A small business is not a little big
business. Harvard Business Review 1981;59(4):18–32.
[44] Penrose ET. The theory of the growth of the rm. Oxford:
Oxford University Press, 1959.
[45] Prahalad CK, Hamel G. The core competence of the
corporation. Harvard Business Review 1990;68(3):79–91.
[46] Wernerfelt B. A resource-based view of the rm. Strategic
Management Journal 1984;5(2):171–80.
[47] Conner KR. A historical comparison of resource-based theory
and ve schools of thought within industrial organization
economics: do we have a new theory of the rm? Journal of
Management 1991;17(1):121–54.
[48] Conner KR, Prahalad CK. A resource-based theory of the
rm: knowledge versus opportunism. Organization Science
1996;7(5):477–501.
[49] Montgomery CA. Of diamonds and rust: a new look at
resources. In: Montgomery CA, editor. Resource-based and
evolutionary theories of the rm: towards a synthesis. Norwell,
MA: Kluwer Academic Publishers, 1995.
[50] Barney J. Firm resources and sustained competitive advantage.
Journal of Management 1991;17(1):99–120.
[51] Foss NJ, Knudsen C, Montgomery CA. An exploration
of common ground: integrating evolutionary and strategic
theories of the rm. In: Montgomery CA, editor.
Resource-based and evolutionary theories of the rm: towards
a synthesis. Norwell, MA: Kluwer Academic Publishers, 1995.
[52] Yap CS. Issues in managing information technology. Journal
of the Operational Research Society 1989;40(7):649–58.
[53] Attewell P. Technology diusion and organizational learning:
the case of business computing. Organisation Science
1992;3(1):1–19.
[54] DeLone WH, McLean ER. Information system success:
the quest for the dependent variable. Information Systems
Research 1992;3(1):60–95.
[55] Seddon P, Kiew MY. A partial test and development of the
DeLone and McLean model of IS success. Proceedings of
the 15th International Conference on Information Systems,
Vancouver, BC, 1994. p. 99 –110.
[56] Powers RF, Dickson GW. MIS project management:
myths, opinions, and reality. California Management Review
1973;15(3):147–56.
[57] Ives B, Olson MH, Baroudi JJ. The measurement of
user information satisfaction. Communications of the ACM
1983;26(10):785–93.
[58] Thong JYL, Yap CS, Raman KS. Engagement of external
expertise in information systems implementation. Journal of
Management Information Systems 1994;11(2):209–31.
[59] DeLone WH. Determinants of success for small business
computer systems. Unpublished PhD thesis, University of
California, Los Angeles, 1983.
[60] Heikkila J, Saarinen T, Saaksjarvi M. Success of software
packages in small businesses: an exploratory study. European
Journal of Information Systems 1991;1(3):159–69.
[61] Markus ML. Power, politics, and MIS implementation.
Communications of the ACM 1983;26(6):430–44.
[62] Jarvenpaa SL, Ives B. Executive involvement and participation
in the management of information technology. MIS Quarterly
1991;15(2):205–27.
156
J.Y.L. Thong / Omega 29 (2001) 143–156
[63] Cerveny RP, Sanders GL. Implementation and structural
variables. Information Management 1986;11(4):191–8.
[64] Ginzberg MJ. Key recurrent issues in the MIS implementation
process. MIS Quarterly 1981;5(2):47–59.
[65] Barki H, Hartwick J. Rethinking the concept of user
involvement. MIS Quarterly 1989;13(1):53–63.
[66] Hirschheim RA. User experience with and assessment
of participative systems design. MIS Quarterly 1985;9(4):
295–304.
[67] Robey D, Farrow D. User involvement in information system
development: a conict model and empirical test. Management
Science 1982;28(1):73–85.
[68] Bowman B, Davis G, Wetherbe J. Three stages model of MIS
planning. Information Management 1983;6(1):11–25.
[69] King WR, Zmud RW. Managing information systems:
policy planning, strategic planning and operational planning.
Proceedings of the Second International Conference
on Information Systems, Cambridge, MA, 1981. pp.
299 –308.
[70] Ein-Dor P, Segev E, Blumenthal D, Millet I. Perceived
importance, investment and success of MIS, or the MIS ZOO.
Systems Objectives Solutions 1984;4:61–7.
[71] Tait P, Vessey I. The eect of user involvement on
system success: a contingency approach. MIS Quarterly
1988;12(1):91–108.
[72] Senn JA, Gibson VR. Risks of investment in microcomputers
for small business management. Journal of Small Business
Management 1981;19(3):24–32.
[73] Nelson RR, Cheney PH. Educating the CBIS user: a case
analysis. Data Base 1987;18:11–6.
[74] Neidleman LD. Computer usage by small and medium
sized European rms: an empirical study. Information &
Management 1979;2(2):67–77.
[75] Dess GG, Robinson RB. Measuring organizational
performance in the absence of objective measures: the case
[76]
[77]
[78]
[79]
[80]
[81]
[82]
[83]
[84]
[85]
[86]
[87]
of the privately-held rm and conglomerate business unit.
Strategic Management Journal 1984;5:265–73.
Ibrahim AB, Goodwin JR. Perceived causes of success
in small business. American Journal of Small Business
1986;11(2):41–50.
Armstrong JS, Overton TS. Estimating non-response bias in
mail surveys. Journal of Marketing Research 1977;14(3):
396–402.
Gatignon H, Robertson TS. Technology diusion: an empirical
test of competitive eects. Journal of Marketing Research
1989;53(1):35–49.
Fornell C. A second generation of multivariate analysis,
methods: vol. 1. New York: Praeger, 1982.
Fornell C, Bookstein FL. Two structural equation models:
LISREL and PLS applied to consumer exit-voice theory.
Journal of Marketing Research 1982;19(4):440–52.
Nunnally JC. Psychometric theory. New York: McGraw-Hill,
1978.
Cook TD, Campbell DT. Quasi-experimentation: design and
analysis issues for eld settings. Boston: Houghton Miin
Co., 1979.
Falk RF, Miller NB. A primer for soft modeling. Akron, OH:
The University of Akron Press, 1992.
Pedhazur EJ. Multiple regression in behavioral research:
explanation and prediction, 2nd ed. New York: Holt, Rinehart
and Winston, 1982.
Raymond L. Validating and applying user satisfaction as a
measure of MIS success in small organizations. Information
Management 1987;12(4):173–9.
Yap CS, Thong JYL, Raman KS. Eect of government
incentives on computerization in small business. European
Journal of Information Systems 1994;3(3):191–206.
Srinivasan A, Kaiser KM. Relationships between selected
organizational factors and systems development. Communications of the ACM 1987;30(6):556–62.