Abstract
Although there are many studies in project management literature on factors that affect the success or failure of projects, few have focused specifically on Information System (IS) projects, especially in China. This research sets out to study the critical success factors (CSFs) of contemporary information system projects in China. This paper presents the results of a large-scale questionnaire survey conducted in the Mainland of China to collect empirical support for the set of CSFs postulated based on extensive literature review. The survey reported here has collected responses of 247 IS project practitioners from three different sample sources. The first is randomly selected from the telephone directory’s yellow pages; the second from class lists of part-time students in Master of Business Administration (MBA) and professional training courses in Shanghai; and the third from a web-based survey. The data collected were analyzed using the computer programs SPSS® and SAS®. The result largely supports the findings of similar previous studies done in the West. This China survey has identified a newly added factor, relationship management, which we believe is the manifestation of the unique characteristics of Chinese culture and management practice. Furthermore, differences are also identified and discussed in the CSFs of this study when compared with those from the USA and Canada.
1. Introduction
There have been many studies reported in project management literature on factors that contribute to the success or failure of projects. However, many of these critical success factors (CSFs) identified are rather too general as they tend to be derived from a wide variety of project types and not specific to a certain class of projects; in particular, the information system (IS) projects. The IS projects are still growing fast and continuing to be invested with huge amounts of money all over the world today. They have, however, also suffered greatly from a high rate of failure. Unfortunately, extensive literature review finds little effort focused specifically on the CSFs of IS projects. Two such studies are found in literature. Jiang, Klein, & Balloun (1996) reported their empirical findings of system implementation success factors in the United States of America (USA) based on Pinto and Slevin (1987) success-factors research in 1987. Recently, Hartman and Ashrafi presented their empirical result of the 10 most important CSFs from their interviews of 12 Information Systems /Information Technology projects in Canada (Hartman & Ashrafi, 2002). China’s IS project failure rate is even worse than those in the developed countries. Our recent literature review finds no published work about CSF studies in the IS projects of China. The business and industrial environment of China is still very different from the West. We believe that the CSFs identified in the developed countries from the West may not be totally applicable in the same manner to the Chinese IS project management practice and environment.
To bridge this gap, this research sets out to: 1) identify and study the major CSFs of IS projects in China; and 2) compare the difference of these CSFs with those found in the Western developed countries, namely the USA and Canada. The research findings will help the contemporary Chinese IS project practitioners to manage their project more effectively.
To achieve this research objective, the paper is organized as follows. First, in the second section, extensive literature of CSF in project management and IS are reviewed and a set of CSFs are postulated. Then the questionnaire is designed, the data collection in the Mainland of China is introduced, and the analysis of the survey data is described. In the third section, the survey results and comparison of the findings with previous research results are discussed. The fourth section is devoted to conclusions and future research directions.
2. Research Design
2.1 Review of checklist of success factors
The study of critical factors in project management has been conducted for many years. Notable contributions include Sayles and Chandler (1971); Martin (1976); Cleland and King (1983); Baker, Murphy, & Fisher (1983); Locke (1984); Morris and Hough (1987); Pinto and Slevin (1987); etc. Many of these studies can be regarded as having covered the majority of the important factors generic to most projects. However, factors affecting success of the project or the extent of their influence vary from one type of project to another, industry to industry, or even country to country (Pinto & Covin, 1989, Odedra-straub, 1993). What is universally agreed upon is that “managing projects is difficult” and “the world often interferes with our plans” (Cicmil, 1997).
To generate a hypothetical CSF list generic enough to cover most of the concerns of Chinese IS project managers, extensive literature has first been reviewed. The most commonly cited set of CSFs is compiled and used as the basis of the questionnaire designed. The postulated CSFs are:
a) Effective communication
Numerous authors have stressed that good communication is the key to a successful project (Cleland & King, 1983; Locke, 1984; Keider, 1984; Pinto & Slevin, 1988; Barki, Rivard, & Talbot, 1993). IS projects are probably no exception. In fact, Martinez’s study has revealed that the greatest threat to the success of any IS project is a failure to communicate effectively (Martinez, 1996).
IS projects are often complex and cross-disciplinary in nature. Good communication among project team members should lead to expectations that are more realistic, better relationships, and more effective teamwork (Milis & Mercken, 2002). Outside the project team, the most important stakeholder to an IS project is its users. IS literature has repeatedly emphasized the importance of communication between the IS developers and its users (Edstrom, 1977; Cronan & Means, 1984; Martin & Fuerst, 1984; Bostrom, 1989). Furthermore, the communication should also be extended to all major stakeholders, such as top management and so on, as verified by empirical evidence (Keider, 1984; Effy & John, 2000). In the Chinese industrial environment of today, is this multi- dimensional communication such an important factor too?
b) Top management support
Literature echoes the general consensus view that top management support is one of the most critical factors to project success (Anderson & Narasimhan, 1979; Pinto & Slevin, 1987; The Standish Group, 1994; Belassi & Tukel, 1996; Fowler & Walsh, 1999). There is also reported empirical evidence that shows the crucial role of top management involvement in IS development projects (Ewusi-mensah & Parzasnyski, 1991, 1994).
Top management support for the project must be real as well as perceived. It is obvious that top management must be willing to provide sufficient resources (financial, manpower, time, etc) for the project (Martin, 1976; Cleland & King, 1983; Baker et al., 1983; Belassi & Tukel, 1996). Moreover, senior management has to realize the complexity of changes resulting from IS projects (Ewusi-mensah & Parzasnyski, 1994). Empirical surveys have found that the fear of unemployment and the fear of the unknown impact computers would cause were significant obstacles to IT adoption in China in the early 90‘s (Ishman, Pegels, & Sanders, 1994). It is natural to extend this to the implementation of IS projects in China.
c) User involvement
Many project management and IS studies have found that user involvement is crucial to the successful completion of IS development projects (Robey & Farrow, 1982; Ives & Olson, 1984; Barki & Hartwick, 1989; Ewusi-mensah & Parzasnyski, 1994; Jiang et al., 1996; Jiang, Chen, & Klein, 2002; Dvir, Raz, & Shenhar, 2003).
IS development projects are described as conceptually intensive (Ewusi-mensah, 1997) and quasi-occasional (Wateridge, 1999), as it is often difficult to completely and explicitly determine the system requirements at the start of the project. It is suggested that requirements must be allowed to be refined as a project progresses and then be fixed at some appropriate point in time (Wateridge, 1999). An effective way to refine system requirements is for the project manager to involve the users from the start of the project and have them continue for as long as they are needed (Dvir et al., 2003).
d) Project manager and team members
Because the project manager and the members of the project team are changed with the responsibility of managing and carrying out the project work, general agreement between them is critical to project success (Turner, 1993; Belassi & Tukel, 1996; Milis & Mercken, 2002). In the study, this factor will be further divided into project manager and project team members.
First, the project manager should be competent. Although some projects fail because of technology or design problems, most of the factors that lead to the failure of system development projects are within the control of the project manager (Keider, 1984). Meanwhile, your project team should be staffed with skilled and effective team players who possess the expertise that the project requires (Mcleod & Smith, 1996). Obviously, the project manager has to try to ensure that the team members can work cohesively, have a common goal, and be committed to the project.
e) Project definition
The right choice of project accounts for half of the project's success. The effort of the project team will not redeem a project that is doomed to fail because it has been ill-conceived or because of poor early decision-making (Munns & Bjeirmi, 1996).
Furthermore, the statement of the project objective must be clear and concrete (Pinto & Slevin, 1987; The Standish Group, 1994). Of course, it must be realistic (Milis & Mercken, 2002) and measurable (Wateridge, 1998) as well. Other researchers also suggest that the project objective should be congruent and agreed upon by the project stakeholders and participants (Clarke, 1999). Aiming at a clear and agreed-upon project objective, the project team’s effort must be within a definite and defined project scope. Scope changes are said to be the most common source of project overruns (Jolyon, 1998; Clarke, 1999). Inadequate definition of project scope was thought to be the second most frequent cause of most IS project failures (Keider, 1984).
f) Project planning
Literature provides much evidence that a well-set project plan plays a vital role in project success (Cleland & King, 1983; Baker et al., 1983; Milis & Mercken, 2002; Dvir et al., 2003).
For an IS project, Keider’s research indicated that the lack of good project planning is ranked as the most likely single cause of project failure (Keider, 1984). Effective planning is more than just setting up an elaborate plan at the start of a project. A good project manager monitors the progress closely and regularly updates the project plan throughout the project implementation.
g) Project control and change management
Regardless of the degree of project complexity, most project specifications change to some degree. During project implementation, change control and management are critical to the success of a project (Sayles & Chandler, 1971; Baker et al, 1983; Locke, 1984; Pinto & Slevin, 1987; Wateridge, 1999). The problem of changes is particularly severe in the case of software development and IS projects (Ewusi-mensah & Parzasnyski, 1994; Meredith & Mantel, 2000).
An IS project manager must concern himself/herself with keeping track of all changes and having a disciplined approach to cope with changes. A formal change management system is a useful tool and an integral part of a configuration management system (Mcleod & Smith, 1996). With it, a project manager can manage effectively not only user requirement change requests, one of the biggest problems of an IS project, but also any other changes in implementation (analysis, design, code, etc) (Wateridge, 1999).
h) Technology support
Although technology factors are said to be not as important a consideration as management and organization issues, they still play a contributory role to some IS projects’ abandonment, cancellation, or failure (Ewusi-mensah & Parzasnyski, 1991, 1994; Ewusi-mensah, 1997; Yeo, 2002; Hartman & Ashrafi, 2002). There must exist appropriate technical capability and infrastructure in the organization to effectively support the IS development and implementation.
i) Other three factors in the research
Apart from the above factors, the research also identifies project organization (Martin, 1976; Cleland & King, 1983; Baker et al., 1983), project management software, and techniques as CSF of an IS project. Finally, relationship management is also included as a factor in our postulated 12 CSFs to take care of the effect on the so-called unique Chinese characteristics. In this study, “relationship” is taken to be synonymous with the often-used “guanxi” or “connection.”
2.2 Questionnaire and Measurement
Although Rockart stated in his prominent seminal paper that interviewing is a useful way to acquire critical success factors (Rockart, 1979), others prefer to use questionnaire surveys (Martin, 1982; Wetherbe & Leitheisier, 1985). In this research, considering the immense geographical spread of China, a structured questionnaire survey is adopted as the main data-collection method.
A total of 12 CSFs and 65 associated CSF statements were defined in the questionnaire. Multivariate measurement was employed to assess CSF concepts. Between three and eight different statements were elicited from each of the 12 CSFs to represent differing aspects of the CSF. All these CSFs and most of the associated statements are generally based on previous CSF studies discussed. Only a few of them are adapted or newly proposed to better suit this study of Chinese practices.
A numerical scale of one to seven ranging from “very unimportant” to “very important” was used to measure a survey respondent’s perceived importance of the 65 CSF statements. A sample of the questionnaire is given in Appendix 1. Referable discussions about the design of the questionnaire are given elsewhere (Slevin & Pinto, 1986; The Standish Group, 1995).
2.3 Sample and Data Collection
There are three different sources of respondents in this survey. The first is a randomly compiled list of firms from websites and the yellow pages of a telephone directory. The second is compiled from the lists of MBA programs of Shanghai’s Fudan University and IS project professional training courses sponsored by Shanghai City government. In order to increase the generalizability of the research findings, the questionnaire is also posted on web pages of two project management websites in the Mainland of China to acquire more respondents of diverse backgrounds. The detailed distribution of questionnaires and usable return rate is presented in Table 1. After data screening and cleaning, a total of 247 usable questionnaires were reserved for data analysis.
| Respondents source | Distributed questionnaires | Returned usable questionnaires | Usable return rate |
| 1. Randomly selected firms | 488 | 96 | 19.67% |
| 2. MBA students/IT Training practitioners | 292 | 81 | 27.74% |
| 3. PM Websites | / | 70 | / |
Table 1: Distributed and Returned Usable Questionnaires
2.4 Data analysis
To begin data analysis, the non-response bias was first examined using SPSS. A one-way multivariate analysis of variance was performed to investigate sample source difference in CSF statements. The result of MANOVA analysis obtained a Wilks' Lambda value of 0.51, with a significance value of 0.22. The result shows that there is no statistically significant difference among three different sample sources (Pallant, 2001), which means that there is no significant non-response bias. Therefore, it is acceptable to regard these three sample sources as one in the subsequent data analysis.
The scale reliability and validity of the combined data set was than analyzed. We assessed the internal consistency for multi-item scales by computing Cronbach’s alpha value. The 12 CSFs with different numbers of items have a composite scale reliability value ranging from 0.75 to 0.91, thus meeting the requirements for exploratory research.
In addition, we examined the construct validity of the measures by using confirmatory factor analysis (CFA). Because of the sample size restrictions, four separate tests were performed using SAS CALIS procedure, each having 3 CSF constructs. The result shows that 60 of 65 CSF statements have a standardized factor loading of more than 0.60 corresponding to their respective CSF. The average factor loading coefficient for all 65 items is 0.70. Meanwhile, all questionnaire items have significant t values with a confidence interval at p< 0.001.
The mean value of each CSF was also computed. The result is shown in Table 2.
| Rank | CSF | Mean | Cronbach's Alpha |
| 1 | Effective communication | 5.92 | 0.88 |
| 2 | Clear project definition | 5.87 | 0.80 |
| 3 | Top management support | 5.86 | 0.77 |
| 4 | Competent project manager | 5.76 | 0.84 |
| 5 | User involvement | 5.68 | 0.81 |
| 6 | Relationship management | 5.66 | 0.91 |
| 7 | Project control and change management | 5.66 | 0.91 |
| 8 | Project planning | 5.61 | 0.84 |
| 9 | Competent project team members | 5.56 | 0.85 |
| 10 | Project organization | 5.46 | 0.75 |
| 11 | Technology and expertise support | 5.37 | 0.83 |
| 12 | Project management software and techniques | 4.88 | 0.87 |
Table 2: CSF Mean Ranking and Reliabilities of Measures
3. Findings and discussion
Among the 12 CSFs identified for the present study of Chinese IS projects, certain factors, such as relationship management, have rarely been studied as a CSF. The present study result has ranked it in the middle, higher than a few other CSFs that have been studied more frequently. This is perhaps symptomatic of the unique characteristics of Chinese culture and management practice. Apart from this, most of the other CSFs’ rankings are similar to the research findings of empirical studies conducted in developed countries.
Two previous CSF studies in the USA (Jiang et al, 1996) and Canada (Hartman & Ashrafi, 2002) are cited in the introductory portion of the article, and the results from the present survey are compared with these two sets of findings. The comparison is summarized in Table 3.
| CSF | Rank | ||
| The Mainland of China | USA | Canada | |
| Effective communication | 1 | 7 | 3 |
| Clear project definition | 2 | 1 | 4 |
| Top management support | 3 | 2 | 5 |
| Competent project manager | 4 | 3 | / |
| User involvement | 5 | 6 | 1(2)* |
| Relationship management | 6 | / | / |
| Project control and change management | 7 | 12 | 9 |
| Project planning | 8 | / | 7 |
| Competent project team members | 9 | 4 | / |
| Project organization | 10 | / | / |
| Technology and expertise support | 11 | / | 8 |
| Project management software and techniques | 12 | / | / |
* Factors ranked 1 and 2 in the Canada study are both about user and its involvement.
Table 3: Ranking Comparison of Corresponding CSFs in the Mainland of China, USA (Jiang et al, 1996) and Canada (Hartman & Ashrafi, 2002)
This study and the two previous studies in the USA and Canada differ greatly in research design and survey sample, and the definition and scope of CSFs proposed for the researches may also be slightly different. But they generally produce very similar insights and reveal an analogous underlying structure of CSFs for an IS project.
As mentioned earlier, one noticeable difference in our China survey results is the ranking of a newly added factor, relationship management. It is ranked 6th among the 12 factors, implying that it plays an important role in Chinese IS project implementation. A reasonable interpretation is that it is a manifestation of the significant influence of Chinese traditional belief and culture on Chinese organizational behavior and management practice today. Good personal relationships are often critical to the dealings among the Chinese, whether it is in a social or job-related context. The IS project environment is no different. Thus the relatively high ranking of the relationship management factor in this survey result is not surprising.
The top-ranked CSF in this study is “effective communication,” which ranked in the middle in the USA study and third in the Canada study. This reason for this difference is perhaps because, to Chinese project personnel, “effective communication” is not only needed to resolve problems during the project implementation, but it is also essential to maintain the previously-mentioned good personal relationships.
The next top-ranked factor is “clear project definition,” which is ranked at the top in the USA study and fourth in the Canada study. Apparently, there is a common acceptance of this factor's importance. It is easy to understand that a project must match with an organization's goals, and that its objective has to be clearly defined. In fact, the expectation of a clearly defined objective for the Chinese is not surprising because of the long history of central control in China (Andersen, Dyrhaug, & Jessen, 2002).
The CSFs that are related to the stakeholders follow next, with the exception of “competent project team members,” These CSFs include “top management support,” “competent project manager,” and “user involvement.” There is general agreement that “top management support” is always vital to project success. The results show that “competent project team members” in China play a less critical role than in the USA, while “user involvement” is judged to be more important than in the USA but less than in Canada. Although “user involvement” is deemed as important, it may not be quite as critical as many purport (Jiang et al, 1996).
“Project planning” and “project control and change management” are factors associated with project life cycle, which appears to take up the middle rankings of China’s CSFs. “Project control and change management” has relatively higher importance in China than in the USA and Canada. This may be the result of poor project control and management of change in China’s IS project.
Lowest ranked of the China’s CSFs are the “project management software and techniques” and “technology and expertise support”--which are factors associated with management techniques and technical issues. This result apparently agrees with previous research findings, and confirms that administrative, human, and organization factors are more important than technical factors. But this is not to say that technical factors are unimportant to IS projects, just that they are comparatively less important than the others.
4. Conclusion
The literature has reported few CSF studies on IS projects, especially in China. The survey study sets out to collect empirical support for the set of CSFs relevant to contemporary China’s IS projects. The preliminary result is also compared with those from two previous studies in the USA and Canada.
The result of this China survey about 12 CSFs of Chinese IS projects has for the most part supported the research findings of similar studies done in the West. One of the key findings in this preliminary analysis is that “relationship management” is viewed as an important CSF for Chinese IS projects. Also, there are noticeable differences are identified and discussed in the CSFs ranking from this China survey compared with those from USA and Canada studies.
Although there are not many surprises in the result, this empirical research using a large-scale survey still represents the first effort to study the contemporary practice of Chinese IS project management. The research findings about the CSFs and their ranking order give a better insight to contemporary Chinese IS project practitioners and help them to manage their projects more effectively. The detailed sub-factors of each CSF provide IS project managers various implications to help them cope with different project environments. Furthermore, detailed analysis of the CSFs will be carried out, which will help to lay the foundation for the development of a CSF management model for Chinese IS projects. It is envisaged that this CSF model will further lead to an IS project diagnosis and evaluation system that can provide pragmatic guides for China’s IS project managers.
5. Acknowledgement
Dong Chao gratefully acknowledges the generous PhD studentship he receives from City University of Hong Kong in support of his PhD study. This research is carried out in collaboration with the research team of Fudan University, Shanghai, under supervision of Dr. Zhai Li, who has received funding support from China’s Nation Natural Science Foundation.