On the Evaluation of Structural Equation Models

Bagozzi, Richard P.; Yi, Youjae

doi:10.1177/009207038801600107

Cited by 6,300 publications

(5,359 citation statements)

References 39 publications

Supporting

195

Mentioning

4,948

Contrasting

Unclassified

211

Order By: Relevance

“…As shown in Table 4, each model-fit index is above the recommended values [47,49], exhibiting an adequate fit to the collected data. …”

Section: Resultsmentioning

confidence: 92%

“…Convergent validity of the scales is examined using three standards suggested by Bagozzi and Yi [47]: (1) Loadings of each indicator should be higher than 0.70 [48]; (2) CR should be above 0.70; and (3) Average variance extracted (AVE) of each construct should be surpassed the variance because of the measurement error of that construct (i.e. AVE should be exceeded 0.50).…”

Section: Resultsmentioning

confidence: 99%

See 1 more Smart Citation

Modeling the adoption of personal health record (PHR) among individual: the effect of health-care technology self-efficacy and gender concern

Dutta

Peng

Sun

2018

Libyan Journal of Medicine

View full text Add to dashboard Cite

Background: With the development of information technology (IT) and medical technology, medical information has been developed from traditional paper-based records into up-to-date medical information exchange system called personal health record (PHR). Empowering PHR provides health awareness and intention for health promotion.Objective: The purpose of this study was to present a research framework to examine individuals’ intention to PHR use.Methods: This cross-sectional study used the questionnaire to collect data from the individual in Taiwan. Individual’s intention to use PHR has been examined by a framework based on extended technology acceptance model (TAM), with gender and health-care technology self-efficacy (HTSE) as external variables. Additionally, gender differences were explored in perceptions and relationships among factors influencing an individual’s intention to PHR use. The research framework was evaluated by structural equation modeling (SEM) and represented by Analysis of a Moment Structures (AMOS).Results: A total of 234 valid responses were used for analysis. The results suggest that the extended TAM model explains 40.6% of the variance of intention to PHR use (R2 = 0.406). The findings also supported that perceived usefulness, perceived ease of use, and attitude toward using PHR significantly influenced individual’s intention to PHR use. Additionally, results also indicated that women were more strongly influenced by perceptions of HTSE.Conclusions: The extended TAM model contributes reasonable explanation for interprets and anticipates of individuals’ intention to use and adopt PHR. Moreover, the results have provided support for HTSE and gender as significant variables in TAM. However, the study identified three relevant factors directly and one factor indirectly influencing on individuals’ intention to PHR use. Thus, health care providers and hospital authorities must take these factors and gender difference into consideration in the development and validation of the theories regarding the acceptance of PHR. Based on the findings, the theoretical and practical implications are discussed.

show abstract

“…As shown in Table 4, each model-fit index is above the recommended values [47,49], exhibiting an adequate fit to the collected data. …”

Section: Resultsmentioning

confidence: 92%

Section: Resultsmentioning

confidence: 99%

Modeling the adoption of personal health record (PHR) among individual: the effect of health-care technology self-efficacy and gender concern

Dutta

Peng

Sun

2018

Libyan Journal of Medicine

View full text Add to dashboard Cite

show abstract

“…Each of the 33 observable indicators loaded significantly (alpha=.001) on their intended factors, indicating convergent validity among the items of each scale. The item loadings and measurement errors are in acceptable ranges (Bagozzi and Yi 1988); these values are listed in the Appendix. Discriminant validity between the scale measures is assessed using the stringent test of Fornell and Larcker (1981).…”

Section: Measurement Estimationmentioning

confidence: 99%

“Pie Sharing” in Complex Collaboration Contexts

Jap

2001

Journal of Marketing Research

300

292

View full text Add to dashboard Cite

Recently, there has been a growing interest in the development of collaborative relationships between organizations. Much attention has been given to how organizations 'expand the pie' of benefits between them; however, there is little that addresses the ensuing issue -how organizations divide the expanded pie. This research examines the relational impact of pie sharing in complex collaboration contexts marked by uncertainty in resources and outputs, information asymmetries, intangible aspects, and noncomparable factors and processes. A conceptual framework is developed that examines how the use of equity and equality sharing principles in conjunction with various resource and organizational conditions can be used to systematically impact relational outcomes. Survey results of 300 R&D managers, scientists, and engineers indicate that sharing principles can have a positive or negative effect on the relationship depending on the type of sharing principle used, and the characteristics of the resources and organizations. In particular, sharing processes should be responsive to the goals of the collaboration. The results underscore the strategic nature of the sharing phenomenon as well as the importance of relational concerns in complex and uncertain interorganizational settings.

show abstract

“…CFA assumes each manifest variable to be a distinct indicator of an underlying latent construct, whereby different constructs are permitted to be inter-correlated. The appropriateness of a specific Measures of local fit evaluate whether each construct can be reliably estimated from its indicators [32,33] and whether the constructs within the model are sufficiently distinguishable [34]. Because coefficient alpha as a measure of construct reliability wrongly assumes that all indicators contribute to reliability equally [35] we chose composite reliability, which draws on the unstandardized regression weights and measurement error components for each indicator [34,35].…”

mentioning

confidence: 99%