Empirical evaluation of the revised end user computing acceptance model

Transcription

1 Computers in Human Behavior Computers in Human Behavior 23 (2007) Empirical evaluation of the revised end user computing acceptance model Jen-Her Wu a,b, *, Yung-Cheng Chen a, Li-Min Lin c a Department of Information Management, National Sun Yat-sen University, 70 Lien-Hai Road, Kaohsiung 80424, Taiwan b Institute of Health Care Management, National Sun Yat-sen University, 70 Lien-Hai Road, Kaohsiung 80424, Taiwan c Mei-Ho Institute of Technology, Pingtung 912, Taiwan Available online 10 March 2006 Abstract This paper proposed a revised technology acceptance model for measuring end user computing (EUC) acceptance. An empirical study was conducted to collect data. This data was empirically used to test the proposed research model. The structural equation modeling technique was used to evaluate the causal model and confirmatory factor analysis was performed to examine the reliability and validity of the measurement model. The results demonstrate that the model explains 56% of the variance. This finding contributes to an expanded understanding of the factors that promote EUC acceptance. The implication of this work to both researchers and practitioners is discussed. Ó 2004 Elsevier Ltd. All rights reserved. Keywords: End user computing; Technology acceptance model; Computer self-efficiency; Network externalities; Computer enjoyment; Task-technology fit 1. Introduction With the recent growth of practical information technology in such areas as engineering and business, the topics of end user computing (EUC) deserve careful attention. Today, knowledge workers are increasingly using sophisticated tools to develop their own * Corresponding author. Tel.: x4722; fax: addresses: jhwu@mis.nsysu.edu.tw (J.-H. Wu), m @student.nsysu.edu.tw (Y.-C. Chen), x3213@mail.meiho.edu.tw (L.-M. Lin) /$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved. doi: /j.chb

2 J.-H. Wu et al. / Computers in Human Behavior 23 (2007) information systems to help them efficiently manage work. EUC acceptance has been established as one of the critical success factors in achieving business success. It is becoming a fundamental part of the organizational plan. End user computing acceptance is one of the most widely researched topics in the information field. The definition of the EUC is not consistent in the literature. Here, the EUC is defined as the adoption and use of information technology by personnel outside the information systems department to develop software applications in support of organizational tasks (Brancheau & Brown, 1993). The reasoned action theory (TRA) is a well-established model and has been broadly used to predict and explain human behavior in various domains. Davis proposed the technology acceptance model (TAM) derived from TRA that has been tested and extended by numerous empirical researches (Davis, 1989; Henderson & Divett, 2003; Igbaria, Zinatelli, Cragg, & Cavaye, 1997; Legris, Ingham, & Collerette, 2003; Venkatesh & Davis, 2000). As Davis (1989) pointed out, the original TAM model consists of perceived ease of use (PEOU), perceived usefulness (PU), attitude toward using (AT), behavioral intention to use (BI), and actual system use (AU). PU and PEOU are the primary determinates of system use while prior researches have indicated that attitude towards the technology is not a significant mediating variable. TAM has been proven for its validity and ability to satisfactorily explain end user system usage (SU). Igbaria et al. (1997) assumed that the antecedents of the end user s perception are intraorganizational and extra-organizational factors. However, Igbaria et al. pointed out that the model variables in their study only explained 25% of the variance in system usage and suggested that further research should incorporate other variables into the model. In addition, some other EUC acceptance researches using TAM are summarized in Table 1. Table 1 shows that none of the explained variance for the model is above 30%. Comprehending the essentials of what determines EUC acceptance can provide great management insights for promoting EUC success. Therefore, this research adopts the TAM, from Igbaria et al. (1997), and integrates it with the task-technology fit theory (TTF), network externality, subject norm, computer self-efficacy and computer enjoyment variables to investigate what determines EUC acceptance. The proposed model is then evaluated. Table 1 Prior TAM for EUC Reference Model The explained variance of the model (%) Adams et al. (1992) Perceived ease of use! Usage 30 Perceived usefulness! Usage Igbaria et al. (1996) Organizational support! Usage 28 Complexity! Usage Usefulness! Usage Enjoyment! Usage Social pressure! Usage Igbaria et al. (1997) Internal computing support! Usage 28 Internal computing training! Usage Management support! Usage Internal computing support! Usage External computing support! Usage Perceived ease of use! Usage Perceived usefulness! Usage

3 164 J.-H. Wu et al. / Computers in Human Behavior 23 (2007) The rest of the paper is organized as follows. Section 2 reviews the related works and describes the research model and hypotheses. Section 3 presents the research method used in this study. Section 4 analyzes the data and tests the model. Section 5 discusses the results. The last section summarizes and concludes this paper. 2. Theoretical background TAM has been one of the most well-known and influential models in IS acceptance studies (Chau, 1996; Liaw & Huang, 2003; Lin & Wu, 2004; Taylor & Todd, 1995; Venkatesh & Davis, 1996). TAM posits that user adoption of a new information technology is determined by the users intention to use the system, which in turn is determined by the users beliefs about the system. TAM further suggests two beliefs: perceived usefulness and perceived ease of use are instrumental in explaining the variance in users intention. Perceived usefulness is defined as the extent to which a person believes that using a particular system will enhance his or her job performance. Perceived ease of use is defined as the extent to which a person believes that using a particular system will be free of effort. Among these beliefs, perceived ease of use is hypothesized to be a predictor for perceived usefulness. Furthermore, both types of beliefs are influenced by external variables. The model is shown in Fig. 1. As Shirani, Aiken, and Reithel (1994) indicated, there are three important external variables: user characteristics, organizational characteristics, and information system characteristics. User characteristics include the user s expertise in computer-based technology and in the functional area for which he expects system support. Organizational characteristics include the structure, culture, and politics of the firm, and are essential to understanding satisfaction in context. If the characteristics of the system in development are not consistent with the user s expectations then disconfirmation is likely to occur. Negative disconfirmation could diminish the user s intention to use the IS. Goodhue and Thompson (1995) defined this construct as TTF. They proposed that information systems would have a positive impact on performance only when there was correspondence between their functionality and the user s task requirements. The TTF model is shown in Fig. 2. Igbaria et al. (1997) further divided the organizational characteristics into intra-organizational and extra-organizational factors. 3. Conceptual model and research hypotheses TAM offers a promising theoretical base for examining the factors contributing to EUC acceptance. This research adopted the TAM, from Igbaria et al. (1997), and integrated it Perceived Usefulness Attitude toward Using Intention to Use User Acceptance (Usage Behavior) Perceived Ease of Use Fig. 1. Technology acceptance model adapted from Davis et al. (1992).

4 J.-H. Wu et al. / Computers in Human Behavior 23 (2007) Task Characteristics Task- Technology Fit Technology Characteristics Precursors of Utilization: Beliefs, Affect, etc. Utilization Performance Impacts Fig. 2. Task-technology fit model. with the task-technology fit theory, network externality, subject norm, computer selfefficacy and computer enjoyment variables to investigate what determines EUC acceptance. The revised TAM is shown in Fig. 3. Dishaw and Strong (1999) indicated that TAM and TTF overlap in a significant way and, if integrated, could provide an even stronger model than either standing alone. For instance, TAM focuses on attitudes toward using a particular IT that users develop based on perceived usefulness and ease of use. TTF focuses on the ability of IT to support a task and match the user s tasks needs with the available IT functionality. This study links TTF with TAM to explain the EUC acceptance. Therefore, the TTF is integrated in the model. A product emerges as a standard not only because of its intrinsic features but also from the benefits deriving from a large user base. The nature of these benefits is termed network externalities. Shurmer (1993) pointed out that network externalities accruing to any given software package are derived from a number of different sources (add-ons, books, training courses and so forth). As more people use this computer software, users can easily obtain Individual factors Computer Self-efficacy Computer Enjoyment Internal factors Subjective Norm Management Support Internal Computing Support and Training Perceived Ease of Use Actual Use External factors External Computing Support and Training Network Externality Perceived Usefulness System factors Task-Technology Fit Fig. 3. Proposed research model for EUC.

5 166 J.-H. Wu et al. / Computers in Human Behavior 23 (2007) information on the IT and are willing to use it. This factor can significantly influence users in choosing to use or not use computer software and thus, is added to the revised model. Computer self-efficacy represents an individual s perceptions of his or her ability to use computers in the accomplishment of a task (Chalmers, 2003; Compeau & Higgins, 1995). Computer enjoyment means that individuals that themselves experience immediate pleasure and joy from using the machine and perceive any activity involving using the computer to be enjoyable in its own right (Davis, Bagozzi, & Warshaw, 1992; Zhang & Li, 2004). Based on Fig. 3 and the foregoing discussion, the following hypotheses are proposed: H1. Perceived usefulness has a direct effect on actual use. H2. Perceived ease of use has a direct effect on perceived usefulness (H2a) and actual use (H2b). H3. Task-technology fit has a direct effect on perceived ease of use (H3a), perceived usefulness (H3b), and actual use (H3c). H4. Network externality has a direct effect on perceived ease of use (H4a), perceived usefulness (H4b), and actual use (H4c). H5. Computer self-efficacy has a direct effect on perceived ease of use (H5a), perceived usefulness (H5b), and actual use (H5c). H6. Computer enjoyment has a direct effect on perceived ease of use (H6a), perceived usefulness (H6b), and actual use (H6c). H7. Subjective norm has a direct effect on perceived ease of use (H7a), perceived usefulness (H7b), and actual use (H7c). H8. Management support has a direct effect on perceived ease of use (H8a) and perceived usefulness (H8b). H9. Internal computing support and training has a direct effect on perceived ease of use (H9b) and perceived usefulness (H9b). H10. External computing support and training has a direct effect on perceived ease of use (H10a) and perceived usefulness (H10b). 4. Methodology 4.1. Measurement development and pilot study To ensure that a comprehensive list of scales was included, works by previous researchers were reviewed. In the revised model, the construct for end user computing was based on the study by Brancheau and Brown (1993). Measures for perceived usefulness, perceived ease of use, and actual use were adapted from previous studies on TAM (Igbaria et al., 1997; Venkatesh & Davis, 2000). The measures for computer self-efficacy were based on Compeau and Higgins (1995) and computer enjoyment was adapted from the research by Davis et al. (1992). The construct for subjective norm was based on Venkatesh and Davis (2000). The measures for management support and internal/external computing support and training were refined based on the EUC context and prior research (Igbaria et al., 1997). The measures for network externality were adapted from the study by Katz and Shapiro (1985) and the task-technology fit was developed based on Goodhue (1998). Once the initial questionnaire was generated, an iterative personal interview process was conducted to refine the instrument. First, we interviewed faculty members, doctoral

6 students, and graduate students that had experiences in this area. Twenty-seven EUC users and 2 domain experts were then interviewed. These interviews enabled the researchers to gauge the clarity of the tasks, assess whether the instrument was capturing the desired phenomena, and verify that important aspects had not been omitted. The process was continued until no further modification to the questionnaire was found. Several iterations were conducted. Feedback served as a basis for correcting, refining, and enhancing the experimental scales, some of which were eliminated as they were found to represent essentially the same aspects with only slight wording differences and some measures were modified because the semantics appeared ambiguous or irrelevant to the characteristics of the EUC. The questionnaire consists of 38 items measuring the 12 latent variables. Table 2 summarizes the definition of each variable. The finalized questionnaire was then mailed to the subjects at the 100 companies, randomly selected from the Top 1000 firms in the manufacturing, sales, and marketing sectors (Common Wealth Magazine, 2000). The subjects were the EUC users and were not in the MIS department. 5. Data analysis and results 5.1. Descriptive statistics J.-H. Wu et al. / Computers in Human Behavior 23 (2007) We distribute eight hundred questionnaires and received 142 returned questionnaires. Twelve gave incomplete answers and were dropped. One hundred thirty were left for the statistical analysis, a 16% valid return rate. The data indicates that the majority of respondents had a college education. Nearly half of the respondents had experience using computers over nine years. The 130 respondents were equally distributed in every organization hierarchy. The demographic characteristics of the sample are summarized in Table 3. How respondents do their jobs with EUC and with what software are described in Table 4. The data indicates that the major jobs respondents do with EUC involve producing reports. The most frequently used software is Spreadsheets. Subjects have multiple choices over the way they perform their job and the software. The proposed research model was evaluated using structural equation modeling (SEM). The data obtained were tested for reliability and validity using confirmatory factor analysis (CFA). This step tested if the empirical data confirmed to the presumed model. The CFA was computed using the LISREL software. In CFA, factor loadings can be viewed as regression coefficients in the regression of observed variables on latent variables. The standard factor loadings for the observed variables (items) on latent variables (factors) were estimates of the validity of the observed variables. The internal consistency of a test was considered adequate when its reliability coefficients exceeded the 0.7 level (Fornell, 1982). Individual item loadings for the actual use, perceived usefulness, computer enjoyment, subjective norm, internal computing support and training, network externality and task-technology fit were all above 0.7. While each of the other constructs showed some weak loadings, the internal consistency reliabilities were all greater than 0.7 (see Table 5). Therefore, no items are dropped. The composite reliability was estimated to evaluate the internal consistency of the measurement model. The composite reliabilities of the measures included in the model ranged from 0.71 to 0.94 (see Table 5). All were greater than the benchmark of 0.60 recommended by Fornell (1982). This illustrates that all measures had strong and adequate reliability.

7 168 J.-H. Wu et al. / Computers in Human Behavior 23 (2007) Table 2 Definition of the variables Construct Definition Reference End-user computing The adoption and use of information Brancheau and Brown (1993) technology by personnel outside the information systems department to develop software applications in support of organizational tasks. Perceived usefulness The degree to which a person believes that Davis (1989) using a specific application system will increase his or her job performance within an organization context Perceived ease of use The degree to which a person believes that Davis (1989) using computer technology would be free of effort. Actual use The frequency and the actual amount of time Igbaria et al. (1997) spent on the computer systems usage. Computer self-efficacy Computer self-efficacy represents an Compeau and Higgins (1995) individual s perceptions of his or her ability to use computers in the accomplishment of a task. Computer enjoyment Individuals that themselves experience Davis et al. (1992) immediate pleasure and joy from using the machine and perceive any activity using the computer to be enjoyable in its own right. Subjective norm The degree to which an individual believes Fishbein and Ajzen (1975) that people that are important her/him think she/he should perform the behavior in question. Management support The perceived level of general support offered by top management. Venkatesh and Davis (2000) Internal computing support and training External computing support and training Network externality Task-technology fit The technical support and the amount of training provided by individuals or groups with computer knowledge internal to the company. The technical support and the amount of training provided by friends, vendors, consultants, or educational institutions external to the company. The utility that a user derives from usage the products or services increase if the number of other users that use them increase is defined as network externalities. The degree to which an organization s information systems functionality and services meet the information needs of the task. Venkatesh and Davis (2000) Venkatesh and Davis (2000) Katz and Shapiro (1985) Goodhue (1998) The average variance extracted measures the amount of variance for the specified indicators accounted by the latent construct. Higher variance extracted values occur when the indicators are truly representative of the latent construct. As shown in the Table 5, the average variance extracted for all measures exceeded 0.5. The measurement model test presented a good fit between the data and the proposed measurement model. For instance, the comparative fit index (CFI) value was 0.9, close

8 J.-H. Wu et al. / Computers in Human Behavior 23 (2007) Table 3 Demographic characteristics of the sample Percent (%) Gender Male 66.9 Female 33.1 Age Education level Senior high school 10 Vocational school 26.9 College degree 53.1 Master degree 10 Organization hierarchy Support staff Professional staff Management Computer experience <1 year <2 years <3 years <5 years <9 years 22 >9 years 46.3 Industry type Manufacturing 60 Sales/marketing 40 No. of employee P to the standard for model fit. According to Browne and Cudeck (1993), an RMSEA value of 0.05 indicates a close fit, white a value of up to 0.08 represents a reasonable fit. The various goodness-of-fit statistics are indicated in Table 6. Overall, the results showed that the measurement model exhibits a good level of fit based on the assessment criteria such as RMR, GFI, v 2 /df, NNFI, CFI and RMSEA. In addition, the explained variance in perceived usefulness, perceived ease of use, and actual use were 41%, 67% and 56%, respectively. This means that the model explains 56% of the actual use variance, which is higher than the measured value in prior researches (e.g., Igbaria et al., 1997). Fig. 4 displays the results from the final structural model, including the estimated path coefficients. The results are as follows. Consistent with Hypotheses 1 and 2b (H2b), perceived usefulness and perceived ease of use are both positively related to actual use. The

9 170 J.-H. Wu et al. / Computers in Human Behavior 23 (2007) Table 4 Application for EUC Percent (%) Job task Producing report 83 Note 40 Data retrieval 69 Decision making 15 Data analysis 28 Planning 30 Problem resolving 31 Budget 35 Work control Others 3 Computer software Spreadsheets (e.g., MS Excel) 100 Database (e.g., MS Access) 42.6 Statistics (e.g., SAS, SPSS) 15.7 Programming (e.g., VB) 37 Graphics (e.g., AutoCAD) 46.3 Application packages Video conference software (e.g., Net meeting) 24.1 Table 5 Assessment of the construct reliability and discriminate validity Variables The composite reliability (>0.6) Average variance extracted (>0.5) Actual use Perceived usefulness Perceived ease of use Computer self-efficacy Computer enjoyment Subjective norm Management Internal computing support and training External computing support and training Network externality Task-technology fit Table 6 Model evaluation measures of overall fit Root mean square residual (RMR) (<0.05) Goodness of fit index (GFI) (>0.9) v 2 /df (<5) Non-normed fit index (NNFI) (>0.9) Comparative fit index (CFI) (>0.9) Root mean square error of approximation (RMSEA) (< ) data shows that task-technology fit has a strong direct effect on perceived ease of use (H3a: c = 0.38, p < 0.01). The data also confirms that network externality positively and directly influences perceived ease of use (H4a: c = 0.34, p < 0.05). Consistent with Hypotheses 5a,

10 J.-H. Wu et al. / Computers in Human Behavior 23 (2007) Individual factors Computer Self-efficacy Computer Enjoyment 0.77** Internal factors 0.57** Subjective Norm 0.22* Management Support Internal Computing Support and Training 0.35* Perceived Ease of Use 0.32* Actual Use External factors External Computing Support and Training 0.34* 0.48** Perceived Usefulness 0.64** Network Externality 0.38** System factors Task-Technology Fit Fig. 4. The empirical results of this study. Computer self-efficacy has direct effects on perceived ease of use (c = 0.57, p < 0.01). Computer enjoyment positively influences actual use (H6c: c = 0.77, p < 0.01). Consistent with H7a and H7b, the subjective norm has positive direct effects on perceived ease of use (c = 0.22, p < 0.05) and perceived usefulness (c = 0.48, p < 0.01). Fig. 4 also shows that internal computing support and training has a significant effect on perceived ease of use (H9a: c = 0.35, p < 0.05). In summary, the structural model tests showed that computer enjoyment, perceived ease of use and usefulness are the dominant factors affecting system usage. Subjective norm had the strongest effect on perceived usefulness. The data also showed that computer self-efficacy, subjective norm, internal computing support and training, network externality, and task-technology have direct effects on perceived ease of use. The effect of every factor on actual use is summarized in Table Discussions Both perceived ease of use and perceived usefulness are important factors that encourage actual EUC use. The perceived usefulness effect is higher than that for perceived ease of use. This may suggest that users are driven to accept EUC primarily based on usefulness because of the function EUC performs for them. This result is consistent with the finding from prior research (e.g., Venkatesh, 2000). The results show that perceived usefulness, perceived ease of use and computer enjoyment all indirectly influences actual EUC usage. The most important determinant for usage is computer enjoyment. This suggests that organizations should enhance the computer environment for the end user to promote EUC use. If users enjoy using EUC, they may tend to underestimate the means or process involved in using a new system because they simply enjoy the process and do not perceive it as being effortful (Venkatesh, 2000).

11 172 J.-H. Wu et al. / Computers in Human Behavior 23 (2007) Table 7 Prediction of EUC acceptance: actual use Variables Network externality has a direct effect in perceived ease of use. The network externalities accrued to any given software package are derived from a number of different sources (add-ons, books, training courses and so forth) (Shurmer, 1993). When EUC become more conventional, users can easily obtain the needed information from them or perform their work efficiently, making EUC easier to use. The results also showed that task-technology fit directly influenced perceived ease of use. This is consistent with prior research (e.g., Dishaw & Strong, 1999). When the degree of fitness between the task and the system becomes higher, users will perceive the system to be easier to use for that task. The most striking finding was that perceived ease of use does not significantly influence perceived usefulness. Prior research indicated that perceived ease of use has a direct and significant effect on perceived usefulness. This effect subsides over time. In past years, the user interface for the information system mainly was text mode. Presently, most computer software uses a graphic user interface with an intuitive interface design. These features have significantly improved the functionality of systems and thus enhanced the ease of use. Because of the pervasion of PCs, users have more software knowledge and are familiar with software operation. This provides a good reason that supports this striking finding. Scarborough and Zimmer (2000) addressed that the introduction of IT has three stages: substitution, adaptation and revolution. When the substitution stage approaches its end, users are familiar with using the information technology. In this study, about 70% of the subjects had five years or more computer experience. Thus, ease of use for them is not an important issue and perceived usefulness outweighs ease of use. 7. Conclusion Actual use Direct effect Total effect Indirect effect Individual factors Computer self-efficacy Computer enjoyment 0.78 * * Intra-organizational factors Subjective norm Management support Internal computing support and training Extra-organizational factors External computing support and training Network externality System factors Task-technology fit Perceived ease of use 0.75 * * Perceived usefulness 1.12 * * R * * p < This study proposed a revised TAM that adopted the TAM, from Igbaria et al. (1997), and integrated it with the task-technology fit theory, network externality, subject norm,

12 computer self-efficacy and computer enjoyment variables to investigate what determines EUC acceptance. The results showed that perceived usefulness, perceived ease of use, and computer enjoyment all directly influence actual usage. The essential determinant for actual use is computer enjoyment. If users enjoy using the EUC, they may tend to underestimate the difficulty or process involved in using a new system because they simply enjoy the process and do not perceive it as being effortful (Venkatesh, 2000). Network externality has a direct effect on perceived ease of use. When EUC become more conventional, users can easily obtain information and they become easy to use EUC. The results also indicated that task-technology fit has a direct influence on the perceived ease of use. This is consistent with prior research (Dishaw & Strong, 1999). When the degree of fitness between the task and the tool becomes higher, users perceive the tool to be easier to use for that task. The above findings provide an expanded understanding of the factors that promote EUC acceptance. There were some limitations in our study. First, the number of the returned questionnaires was not broad enough. Because the subjects were limited to employee outside the MIS department, many respondents did not actually know what EUC is. This might have decreased their willing to complete the questionnaires. This study proposed an integrated model and increased the explained variance for EUC acceptance. As information technology diffuses throughout companies, EUC develops a more important role helping workers to complete their work efficiently. In a further study, we could select one company and use our model to help them introduce EUC. The effects from EUC use could be examined. We will leave this issue for further research. The combination of information obtained in this study and from future studies would be valuable to educators, researchers and managers. Acknowledgement This research was supported by the National Science Council of Taiwan under the Grant NSC H References J.-H. Wu et al. / Computers in Human Behavior 23 (2007) Adams, D. A., Nelson, R. R., & Todd, P. A. (1992). Perceived Usefulness, case of use, and usage of information technology: a replication. MIS Quarterly, 16(2), Brancheau, C., & Brown, V. (1993). The management of end-user computing: Status and directions. ACM Computing Surveys, 25(4), Browne, M. W., & Cudeck, R. (1993). Alternative ways of assessing model fit. In K. A. Bollen & J. S. Long (Eds.), Testing structural equation models. Newbury Park, CA: Sage Publications. Chalmers, P. A. (2003). The role of cognitive theory in human computer interface. Computers in Human Behavior, 19, Chau, P. Y. K. (1996). An empirical assessment of a modified technology acceptance model. Journal of Management Information Systems, 13(2), Common Wealth Magazine (2000). Top 2000 companies in Taiwan, Common Wealth Magazine, 28. Compeau, D. R., & Higgins, C. A. (1995). Computer self-efficacy: Development of a measure and initial test. MIS Quarterly, 19(2), Davis, D. (1989). Perceived usefulness, perceived ease of use, and user acceptance of information technology. MIS Quarterly, 13(3), Davis, F. D., Bagozzi, R. P., & Warshaw, P. R. (1992). Extrinsic and intrinsic motivation to use computers in the workplace. Journal of Applied Social Psychology, 22(14),

13 174 J.-H. Wu et al. / Computers in Human Behavior 23 (2007) Dishaw, M. T., & Strong, D. M. (1999). Extending the technology acceptance model with task-technology fit constructs. Information and Management, 36(1), Fishbein, M., & Ajzen, I. (1975). Belief, attitude, intentions and behavior: An Introduction to theory and research. Reading, MA: Addison-Wesley. Fornell, C. (1982). A second generation of multivariate analysis: An overview. In C. Fornell (Ed.), A second generation of multivariate analysis (pp. 1 21). New York: Praeger. Goodhue, D. L., & Thompson, R. L. (1995). Task-technology fit and individual performance. MIS Quarterly, 19(2), Goodhue, D. L. (1998). Development and measurement validity of a task-technology fit instrument for user evaluations of information systems. Decision Sciences, 29(1), Henderson, R., & Divett, M. J. (2003). Perceived usefulness, ease of use and electronic supermarket use. International Journal of Human Computer Studies, 59(3), Igbaria, M., Parasuraman, S., & Baroudi, J. (1996). A motivational model of microcomputer usage. Journal of Management Information Systems, 13(1), Igbaria, M., Zinatelli, N., Cragg, P., & Cavaye, L. M. (1997). Personal computing acceptance factors in small firms: A structural equation model. MIS Quarterly, 21(3), Katz, M. L., & Shapiro, C. (1985). Network externalities, competition, and compatibility. American Economic Review, 75(3), Legris, P., Ingham, J., & Collerette, P. (2003). Why do People use information technology? A critical review of the technology acceptance model. Information and Management, 40(3), Liaw, S.-S., & Huang, H.-M. (2003). An investigation of user attitudes toward search engines as an information retrieval tool. Computers in Human Behavior, 19(6), Lin, F.-H., & Wu, J.-H. (2004). An empirical study of end-user computing acceptance factors in small and medium enterprises in Taiwan: Analyzed by structural equation modeling. Journal of Computer Information Systems, 44(3), Scarborough, N. M., & Zimmer, T. W. (2000). Effective small business management (6th ed.). NJ: Prentice-Hall. Shirani, A., Aiken, M., & Reithel, B. (1994). A model of user information satisfaction. Data Base, 25(4), Shurmer, M. (1993). An investigation into sources of network externalities in the packaged PC software market. Information Economics and Policy, 5(3), Taylor, S., & Todd, P. A. (1995). Understanding information technology usage: A test of competing models. Information Systems Research, 6(2), Venkatesh, V. (2000). Determinants of perceived ease of use: Integrating control, intrinsic motivation, and emotion into the technology acceptance model. Information Systems Research, 11(4), Venkatesh, V., & Davis, D. (2000). A theoretical extension of the technology acceptance model: Four longitudinal field studies. Information Systems Research, 46(2), Venkatesh, V., & Davis, F. D. (1996). A model of the antecedents of perceived ease of use: Development and test. Decision Sciences, 27(3), Zhang, P., & Li, N. (2004). An assessment of human computer interaction research in management information systems: topics and methods. Computers in Human Behavior, 20, Jen-Her Wu is Professor of Information Management and Director of Institute of Health Care Management at National Sun Yat-Sen University. He has published a book (Systems Analysis and Design) and more than 30 papers in professional journals such as Information & Management, Decision Support Systems, International Journal of Technology Management, Expert Systems, Knowledge Acquisition, International Journal of Expert Systems: Research & Applications, International Journal of Intelligent Systems in Accounting, Finance and Management, and Journal of Computer Information Systems, and others. His current research interests include various aspects of information systems development and management, human computer interaction, and knowledge management. Yung-Cheng Chen is an engineer at the Inotera Memories Inc. He earned a MBA degree in Information Management. His current research interests include end user computing, human computer interaction, and information systems development and management. Li-min Lin is Instructor of Mei-Ho Institute of Technology. She earned a MBA degree in Human Resource Management. Her current research interests include human resource management, end user computing, and human computer interaction.