Skip Navigation


International Journal for Quality in Health Care Advance Access originally published online on January 25, 2008
International Journal for Quality in Health Care 2008 20(2):123-129; doi:10.1093/intqhc/mzm074
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
20/2/123    most recent
mzm074v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Wu, J.-H.
Right arrow Articles by Bates, D. W.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Wu, J.-H.
Right arrow Articles by Bates, D. W.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author 2008. Published by Oxford University Press in association with the International Society for Quality in Health Care; all rights reserved

Testing the technology acceptance model for evaluating healthcare professionals' intention to use an adverse event reporting system

Jen-Her Wu1, Wen-Shen Shen2, Li-Min Lin3, Robert A. Greenes4 and David W. Bates5

1 Department of Information Management
2 Institute of Health Care Management, National Sun Yat-Sen University, 70 Lien-Hai Road, Kaohsiung 804, Taiwan, ROC
3 Department of Nursing, Mei-Ho Institute of Technology, Pingtung, Taiwan, ROC
4 Department of Biomedical Informatics, Arizona State University, Phoenix, Arizona, USA
5 Brigham and Women's Hospital, Harvard Medical School, Boston, MA, USA

Address reprint requests to: Robert A. Greenes, Department of Biomedical Informatics, Arizona State University, Phoenix, Arizona, USA. E-mail: greenes{at}asu.edu


   Abstract
Top
 Abstract
Methods
 Results
 Discussion
 Funding
 Appendix
 References
 
Background. Many healthcare organizations have implemented adverse event reporting systems in the hope of learning from experience to prevent adverse events and medical errors. However, a number of these applications have failed or not been implemented as predicted.

Objective. This study presents an extended technology acceptance model that integrates variables connoting trust and management support into the model to investigate what determines acceptance of adverse event reporting systems by healthcare professionals.

Method. The proposed model was empirically tested using data collected from a survey in the hospital environment. A confirmatory factor analysis was performed to examine the reliability and validity of the measurement model, and a structural equation modeling technique was used to evaluate the causal model.

Results. The results indicated that perceived usefulness, perceived ease of use, subjective norm, and trust had a significant effect on a professional's intention to use an adverse event reporting system. Among them, subjective norm had the most contribution (total effect). Perceived ease of use and subjective norm also had a direct effect on perceived usefulness and trust, respectively. Management support had a direct effect on perceived usefulness, perceived ease of use, and subjective norm.

Conclusion. The proposed model provides a means to understand what factors determine the behavioral intention of healthcare professionals to use an adverse event reporting system and how this may affect future use. In addition, understanding the factors contributing to behavioral intent may potentially be used in advance of system development to predict reporting systems acceptance.

Keywords: technology acceptance model, trust, patient safety, reporting systems

The Institute of Medicine has reported that more than 1 million preventable adverse events occur each year in the USA, of which 44 000–98 000 are fatal. More people die in a given year as a result of medical error than from motor vehicle accidents, breast cancer, or AIDS and adverse events are also expensive [1]. Thus, if medical errors can be prevented or reduced, enormous life and cost savings could be obtained. To enhance awareness and proactive participation in healthcare quality and safety initiatives, non-punitive adverse event reporting system implementation is one of the key tools [1, 2]. Although some adverse event reporting system approaches in healthcare rely on automatic monitoring of clinical actions in order to detect potential adverse events, self-reporting is useful as there are some events that can pretty much only be identified by self-reporting. However, many problems and barriers exist that make use of such reporting systems suboptimal [3–5]. Considering all fields of endeavor, only about one in 20 adverse events is reported [6], and this proportion is much lower for errors in the healthcare system [2]. Therefore, there is a need to understand the factors that might entice healthcare professionals to use an adverse event reporting system.

The theory of reasoned action is a well-established model that has been used broadly to predict and explain human behavior in various domains [7]. Davis proposed a technology acceptance model derived from the theory of reasoned action that has been tested and extended by numerous empirical research studies [7–12]. The original technology acceptance model was based on the factors relating to perceived ease of use of a system, perceived usefulness, attitude toward using, behavioral intention to use, and actual system use (Fig. 1) [9]. Behavioral intention to use is defined as the individual's interest in using the system for future work. Perceived usefulness is defined as the degree to which a person believes that using a particular system would enhance his or her job performance, while perceived ease of use is defined as the degree to which a person believes that using a particular system would be free of effort. Perceived usefulness has a direct effect on behavioral intention to use. Perceived ease of use has a direct effect on perceived usefulness and behavioral intention to use.


Figure 1
View larger version (5K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 1 The original technology acceptance model.

 
Implementing a self-reporting-based adverse event reporting system is not just a system issue. It requires making a cultural change in hospitals, to promote open discussion of errors and learning from failure. An environment of punishment and fear of legal prosecution will inhibit a healthcare professional from stepping forward after an unfortunate mistake is made [4, 13]. Leape [5] indicated that user confidentiality is an important component in building a successful reporting system. Lack of trust and bad publicity are critical barriers that can make reporters less likely to use a reporting system [3]. Trust is defined as the extent to which one is willing to ascribe good intentions to, and have confidence in, the words and actions of other people (or systems). Reporting will occur only if practitioners feel professionally comfortable and legally safe in reporting and view it as a culturally accepted activity within the healthcare community [13–16].

Work groups can provide social support that may protect individuals from the negative impact of environmental stressors, as well as provide a feeling of security, lowered anxiety and aggression, and more opportunity for relaxation. On the other hand, if individuals associate or identify with organization norms, goals and values, they are more likely to have a higher level of trust of the organization [17].

Management support is defined as the perceived level of general support offered by top management. Previous studies have proven that management support is one of the primary factors affecting system success [18]. Wu et al. [19] pointed out the significance of staff training and appropriated resources allocation in implementing and using an online system in clinical practice. Hence, we integrated into technology acceptance model two additional variables, trust and management support, to investigate which factors underlie adverse event reporting system acceptance by healthcare professionals.

In Taiwan, implementation of adverse event reporting systems has occurred in two stages. In the first stage, the Taiwan Joint Commission on Hospital Accreditation under the Department of Health launched a project to develop and promote a Web-based reporting system for hospitals to gather information about adverse events or errors occurring inside the hospital by self-reporting. In the second stage, another project was launched to establish a national information infrastructure to integrate individual reporting systems into a national adverse event reporting system in 2004 [20]. Shih et al. [20] reported that about 95% of hospitals (valid sample size 327) have implemented an adverse event reporting system in Taiwan. Among them, about 40% have had a reporting system for more than 5 years, 22% have had one for 2–5 years, and 38% have had a reporting system for less than 2 years. However, among these hospitals, only 6% have a computerized adverse event reporting system. Although very large hospitals have implemented a system for self-reporting by healthcare professionals, there still are large gaps between the volumes of self-reported events and the number actually occurring [20].


   Methods
Top
 Abstract
 Methods
Results
 Discussion
 Funding
 Appendix
 References
 
Conceptual model and research hypotheses
The technology acceptance model developed for a management information system may not necessarily apply to an adverse event reporting system context. In addition, the effects of perception of usefulness and ease-of-use on intention may change if other antecedents (e.g. trust and management support) are incorporated to the model [21]. Therefore, we test the following hypotheses:

H1: Perceived usefulness has a direct effect on behavioral intention to use the reporting system.
H2a: Perceived ease-of-use has a direct effect on behavioral intention to use the reporting system.
H2b: Perceived ease-of-use has a direct effect on Perceived usefulness
Many researchers have suggested that the technology acceptance model needs to be extended to incorporate additional constructs to enhance its explanation and prediction of acceptance behavior. Venkatesh et al. [22] formulated a unified model that integrated elements across the previous technology acceptance models and empirically validated the unified model. Their research indicated that subjective norm significantly influenced user acceptance. Subjective norm is defined as the degree to which an individual believes that people that are important to her/him think she/he should use the system.


Figure 2
View larger version (18K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 2 The model and empirical results with standardized path coefficients of this study.

 
There are few research studies that simultaneously incorporate the subjective norm and trust constructs into the technology acceptance model in the healthcare context. We believe it is desirable to ascertain the effect of these constructs on intention to use an adverse event reporting system. Therefore, we propose the following hypotheses (Fig. 2):
H3a: Subjective norm has a direct effect on the behavioral intention to use the reporting system.
H3b: Subjective norm has a direct effect on Trust.
H4: Trust has a direct effect on behavioral intention to use the reporting system.
Management support plays a vital role in promoting open discussion of errors in the hospital. After making a mistake, healthcare professionals may experience emotions such as shame, guilt, or depression. Litigation, complaints and actions by regulatory authorities can present additional burdens in a small subset of these situations. Particularly effective to address these issues can be actions by senior managers to talk openly about past mistakes and problems, to publicly endorse the concept of blame-free error reporting, and to offer encouragement and support to those directly involved [18, 23]. Changing this culture can be difficult and takes time. However, without management support, this type of cultural change in a hospital may not be possible. Hence, the following hypotheses are proposed (Fig. 2):
H5a: Management support has a direct effect on perceived usefulness of reporting systems.
H5b: Management support has a direct effect on perceived ease of use of reporting systems.
H5c: Management support has a direct effect on subjective norm of reporting systems.

Measures
A number of prior related studies were reviewed to ensure that a comprehensive list of measures was included. All measures for each construct were taken from the previously validated instruments. Those for perceived usefulness, perceived ease of use, and intention to use were adopted from Venkatesh and Davis [9]. The measures for subjective norm, trust and management support were based on the work of Igbaria et al. [7]. Once the list was generated, an iterative process involving personal interviews with domain experts (two of each of physicians, nurses, pharmacists, technology staffs and administration staff members) was carried out to gauge the clarity of the tasks presented, assess whether the instrument was capturing the phenomenon desired, and verify that important aspects of items were not omitted. This process was continued until no further modification to the questionnaire occurred.

A survey questionnaire was administered, consisting of two parts. The first part recorded the subject's demographic information. The second part recorded the subject's perception of each variable in the model. Data were collected using a five-point Likert-type scale, where –2 indicated strongly disagree; –1 showed disagreement to some extent; 0 stood for uncertain; 1 was for agree to some extent; and 2 indicated strong agreement.

Subjects and sampling
Subjects for this study were people that had the opportunity to use an adverse event reporting system, including physicians, nurses, medical technicians, pharmacists and administration staffs that worked for hospitals in Taiwan. Currently, the web-based adverse event reporting system implementation in Taiwan is still at an early stage. Only medical centers and a few regional hospitals have actually implemented or partially implemented the reporting system. We distributed 940 questionnaires to 144 hospitals that actually or partially implemented reporting systems. Data were collected via snowball and convenient sampling. Owing to the conventional expectation of low survey response rates in healthcare organizations, we endeavored to find a specific local contact person who would participate in the reporting system project for each target hospital. The contact person was asked to provide information on the reporting system project and number of end users, and to distribute self-administered questionnaires.

Three hundred and twenty-six questionnaires were received. Thirty-six participants gave incomplete answers and their results were dropped from the study. This left 290 sets of data for statistical analysis, a 31% valid return rate. Table 1 lists the respondents' profiles. The data indicated that the majority of the respondents were females (81%), had a college education (85%) and were nearly half nurses (49%). This likely is roughly representative of the underlying population of health professionals in Taiwan, given that a separate survey of all hospital employees indicated that about 50% of healthcare professionals are nurses in Taiwan.


View this table:
[in this window]
[in a new window]

  		Table 1 Respondents profile (N = 290)

 

   Results
Top
 Abstract
 Methods
 Results
Discussion
 Funding
 Appendix
 References
 
The proposed model was evaluated using structural equation modeling. The data obtained were tested for reliability and validity using confirmatory factor analysis. Both were analysed using LISREL software [24]. The model contained 17 items describing six latent constructs: perceived usefulness, perceived ease of use, management support, trust, subjective norm, and intention to use. The {chi}2/d.f. value was 1.82, less than two. The Goodness-of-Fit Index (GFI), Normed Fit Index (NFI), Non-Normed Fit Index (NNFI), Comparative Fit Index (CFI), Root Mean Square Residual (RMSR) and Root Mean Square Error of Approximation (RMSEA) values (see Appendix) were 0.93, 0.93, 0.96, 0.970, 0.033 and 0.051, respectively. Thus, the measurement model has a good fit with the data (as shown in Table 2), based on these assessment criteria [24]. The composite reliability was estimated to evaluate the internal consistency of the measurement model. As shown in Table 3, the composite reliability ranged from 0.72 to 0.94. All were greater than the benchmark of 0.60 recommended by Bagozzi and Yi [25]. In addition, average variance extracted provided evidence of discriminant validity among each of the latent constructs. The average variance extracted for all latent constructs also exceeded 0.5, which meant that more than one-half of the variances observed in the items were accounted for by their hypothesized constructs. Thus, all constructs had strong and adequate reliability and discriminant validity. The completely standardized factor loadings and individual item reliability for the observed variables are presented in Table 4. Fig. 2 shows the structural relationship among the research variables and the standardized path coefficients. All of the hypotheses were strongly supported (P < 0.05 or P < 0.1).


View this table:
[in this window]
[in a new window]

  		Table 2 Model evaluation overall fit measurement

 


View this table:
[in this window]
[in a new window]

  		Table 3 Assessment of the construct reliability

 


View this table:
[in this window]
[in a new window]

  		Table 4 Standardized factor loading and individual item reliability

 

   Discussion
Top
 Abstract
 Methods
 Results
 Discussion
Funding
 Appendix
 References
 
The results indicated that perceived usefulness had a significant effect on the behavioral intention to use the reporting system; perceived ease of use had a direct effect on both behavioral intention and perceived usefulness. These findings are consistent with most prior research on this topic [9]. These mean that a user's perceptions of usefulness and ease of use are important determinants of a healthcare professional's willingness to use an adverse event reporting system, and a professional's perception of usefulness is also influenced by the perception of ease of use. When use of the reporting system is supported by healthcare professionals' existing values, prior experiences and practice needs, they will not only feel more confident in using the system, but also exhibit a higher degree of perception of the advantages of the system. Therefore, they will be more likely to use it.

With respect to the subjective norm, the data showed that it not only had a direct effect both on behavioral intention and trust but also had the most contribution (total effect) on a professional's intention to use an adverse event reporting system. These factors imply that in practice subjective norm is the most significant antecedent for system success and must be taken into account while promoting and implementing an adverse event reporting system.

Regarding trust, the data indicated that it had a significant effect on behavioral intention. This result supports previous research [5] that shows that fear of lack of security is one of the key factors affecting the use of the reporting system. Users are concerned about the level of security present when providing sensitive information online and are only willing to participate in the interactions if a certain level of trust is present.

The data showed that the management support had a direct effect on perceived usefulness, perceived ease of use and subjective norm, respectively. The results also indicate that management support is an important factor directly affecting the subjective norm. This suggests that strong management support represents a key to building a more conducive environment for system success, including diminishing users' unenthusiastic attitudes toward the adverse event reporting system, and fear of retaliation or punishment from others.

The external environment can also play an important role. In Taiwan, healthcare professionals can be legally prosecuted if they are found to be responsible for an adverse event caused by an error, even if it was unintentional, and a number of nurses and physicians have been punished. This obviously will have an impact on use of an adverse event reporting system. In addition, the positions of state boards for professionals such as nursing, physicians, and pharmacists are important, since, if they are unduly punitive, this may have a chilling effect.

The contributions and implications of this study include that:

Adverse event reporting should become a culturally accepted activity within the healthcare community. Our findings indicated that subjective norm not only directly affects behavioral intent, but also has the most contribution to a professional's intention to use an adverse event reporting system. These findings imply that practice subjective norm is the most significant antecedent of system success and must be taken into account while promoting and implementing an adverse event reporting system. These findings also suggest that the manager should promote the system and let everyone feel that he/she should use the adverse event reporting system when adverse events, medical errors or accidents occur.

An adverse event reporting system should be easy to use and provide useful and timely reports. An adverse event reporting system's user interface should be user-friendly and include key functions to minimize the user's efforts in reporting. One way recommended by others to make reports useful is to have experts that understand the clinical circumstances evaluate reports and analyse them promptly, so that useful recommendations are rapidly disseminated or shared with those that need to know.

Reporting should feel comfortable and assured to be free of negative consequences. When implementing an adverse event reporting system, the manager should promote the system with emphasis on the fact that users will be free of retaliation or punishment from others as a result of reporting. One way recommended is that the identities of the patient, reporter, and institution (or division) should not be revealed to a third party. When participants have higher confidence about using an adverse event reporting system, there can be expected to be a higher degree of system acceptance.

Managers should support and motivate reporting publicly. Hospitals may provide relatively efficient technical support, training, and an awareness program for all staff involved and sufficient resources to implement the adverse event reporting system. To help reporting to become a culturally accepted activity within the healthcare community, management needs to show publicly that they are keen to see that people are happy to use the system.

In this study, we follow a rigorous procedure of research model design, measure development, sampling, and statistical analysis. Therefore, good internal validity can be assured. However, the current study has several limitations that also represent opportunities for future research. First, while the findings of this study only apply to the adverse event reporting system setting and the model was validated using the sample data gathered in Taiwan, the interpretation of the findings should be made with caution when generalizing to other systems or countries.

Second, another factor that was not evaluated is that an adverse event reporting system depends on intrinsic motivation of a participant to use it. An adverse event reporting system is not integral to direct patient care, it requires extra effort, and there are no positive rewards for using it, only possibly negative consequences, which one attempts to mitigate with management support and reinforcement. As a consequence, even the most highly supportive, positive environment will not be able to overcome limits caused by other heavy demands on user time or energy. This will no doubt result in some irreducible minimum level of non-use that needs to be assessed in future studies.


   Funding
Top
 Abstract
 Methods
 Results
 Discussion
 Funding
Appendix
 References
 
This research was supported by the National Science Council of Taiwan under the grant NSC 96-2416-H-110-009-MY3 and was partially supported by Aim for the Top University Plan of the National Sun Yat-Sen University and Ministry of Education, Taiwan.


   Appendix
Top
 Abstract
 Methods
 Results
 Discussion
 Funding
 Appendix
References
 

Table A1 Overall Goodness-of-Fit measures for structural equation modeling in this study

Goodness-of-Fit Description
GFI GFI is a measure of the relative amount of variances and covariances jointly accounted for by the model. GFI is independent of the sample size and relatively robust against departures from normality GFI = 1 – tr[(Formula–1 S – I)2]/tr[(Formula–1 S)2] for maximum likelihood
NFI NFI is a measure ranging from 0 (no fit at all) to 1.0 (perfect fit). It is a ratio of the difference in the {chi}2 value for the proposed model and a null model divided by the {chi}2 value for the null model NFI = {chi}null2 – {chi}proposed2/{chi}null2
NNFI NNFI uses a similar logic but adjusts the NFI for the number of degrees of freedom in the model NNFI = ({chi}null2/dfnull) – ({chi}proposed2/dfproposed)/({chi}null2/dfnull) – 1
CFI CFI is based on the non-central parameter, which can be estimated as {chi}2 – df. It also ranges between 0 and 1, with values exceeding 0.90 indicating a good fit to the data CFI = 1 – {chi}proposed2 – dfproposed/{chi}null2 – dfnull
RMSR RMSR is the square root of the mean of the squared residuals—an average of the residuals between individual observed and estimated variance and covariance terms RMSR = {surd}2{Sigma}i=1k{Sigma}j=1i (sij – Formulaij)2/k(k + 1)
Root Mean Square Error of Approximation (RMSEA) Similar to RMSR, RMSEA is based on the analysis of residuals, with smaller values indicating a better fit to the data RMSEA = {surd}({chi}proposed2/dfproposed) – 1/n – 1

Where Formula, the estimate of a structured covariance matrix; S, an unbiased sample covariance matrix; I, an identity matrix; tr[ ], the trace of the matrix, i.e. the sum of the diagonal elements; {chi}proposed2, the non-centrality parameter for the model tested; dfproposed, the degrees of freedom for the model tested; {chi}null2 and dfnull, the non-centrality parameter for the null model; sij, an element in the observed covariance matrix; Formulaij, an element in the fitted covariance matrix (estimated); k, the total number of observed variables in the model; n, sample size.


   References
Top
 Abstract
 Methods
 Results
 Discussion
 Funding
 Appendix
 References
 

  1. Kohn LT, Corrigan JM, Donaldson MS. To Err is Human: Building a Safer Health System (1999) Washington, DC: Institute of Medicine.

  2. Mariner WK, Miller FH. Medical error reporting: Professional tensions between confidentiality and liability. Massachusetts Health Policy Forum (2001) 13:1–35.

  3. Yourstone SA, Smith HL. Managing system errors and failures in health care organizations: suggestions for practice and research. Health Care Manag R (2002) 27:50–61.

  4. Chiang M. Promoting patient safety: creating a workable reporting system. Yale J Regul (2001) 18:383–408.

  5. Leape LL. Reporting of adverse events. N Engl J Med (2002) 347:1633–8.[Free Full Text]

  6. Cullen DJ, Bates DW, Small SD, et al. The incident reporting system does not detect adverse drug events: a problem for quality improvement. Jt Comm J Qual Improv (1995) 21:541–8.[Medline]

  7. Igbaria M, Zinatelli N, Cragg P, et al. Personal computing acceptance factors in small firms: a structural equation model. MIS Q (1997) 21:279–305.[CrossRef]

  8. Hu PJ, Chau PYK, Sheng ORL, et al. Examining the technology acceptance model using physician acceptance of telemedicine technology. J Manag Inf Sys (1999) 16:91–112.

  9. Venkatesh V, Davis FD. A theoretical extension of the technology acceptance model: four longitudinal field studies. Manag Sci (2000) 46:186–204.[CrossRef]

  10. Chau PYK, Hu PJH. Information technology acceptance by individual professionals: a model comparison approach. Dec Sci (2001) 32:699–719.[CrossRef]

  11. Chau PYK, Hu PJH. Investigating healthcare professionals' decisions to accept telemedicine technology: an empirical test of competing theories. Inf Manag (2002) 39:297–311.[CrossRef]

  12. Karsh BT. Beyond usability: designing effective technology implementation systems to promote patient safety. Qual Saf Health Care (2004) 13:388–94.[Abstract/Free Full Text]

  13. Firth-Cozens J. Organizational trust: the keystone to patient safety. Qual Saf Health Care (2004) 13:56–61.[Abstract/Free Full Text]

  14. Cohen MR. Why error-reporting system should be voluntary. Brit Med J (2000) 320:728–9.[Free Full Text]

  15. Gefen D, Karahanna E, Straub D. Trust and TAM in online shopping: an integrated model. MIS Q (2003) 27:51–90.

  16. Lee JD, See KA. Trust in automation: designing for appropriate reliance. Hum Factors (2004) 46:50–80.[CrossRef][Web of Science][Medline]

  17. Williams LL. Impact of nurses' job satisfaction on organizational trust. Health Care Manag R (2005) 30:203–11.

  18. McElhinney J, Heffernan O. Using clinical risk management as a means of enhancing patient safety: the Irish experience. Int J Health Care Qual Assur (2003) 16:90–8.[CrossRef]

  19. Wu JH, Wang SC, Lin LM. Mobile computing acceptance factors in the healthcare industry: a structural equation model. Int J Med Inf (2007) 76:66–77.[CrossRef][Web of Science][Medline]

  20. Shih CL, Hou SM, Hsueh YS, et al. Incident reporting system and reporting barriers. Formosa J Med (2005) 9:63–70.

  21. Plouffe CR, Hilland JS, Vandenbosch M. Inf Sys Research (2001) 12:208–22. Research report: richness versus parsimony in modeling technology adoption decisions—understanding merchant adoption of a small card-based payment system.[CrossRef]

  22. Venkatesh V, Morris MG, Davis GB, et al. User acceptance of information technology: toward a unified view. MIS Q (2003) 27:425–78.

  23. Vincent C. Understanding and responding to adverse events. N Engl J Med (2003) 348:1051–6.[Free Full Text]

  24. Hair JF, Black WC, Babin BJ, et al. Multivariate Data Analysis (2005) NJ: Prentice Hall.

  25. Bagozzi RP, Yi Y. On the evaluation of structural equation models. J Acad Marketing Sci (1988) 16:74–94.

Accepted for publication December 13, 2007.


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 Qual Saf Health CareHome page
K. G Shojania
The frustrating case of incident-reporting systems
Qual. Saf. Health Care, December 1, 2008; 17(6): 400 - 402.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
20/2/123    most recent
mzm074v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Wu, J.-H.
Right arrow Articles by Bates, D. W.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Wu, J.-H.
Right arrow Articles by Bates, D. W.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?