Abstract
OBJECTIVE: The application of structural equation modeling (SEM), a statistical modeling tool for scale construction and development, is becoming increasingly popular in the health sciences librarianship and information science research. This study explores the application of SEM to health science libraries by describing the development and validation of a modified LibQUAL scale within an Iranian health sciences library setting. METHODS: A literature search was performed across several information sources to identify candidate items to be included in the primary questionnaire. After translation, linguistic validation, and a pilot study, two cross-sectional studies were performed. SEM modeling framework was used for the assessment of the reliability and validity of the modified LibQUAL scale. The internal consistency of the scale was evaluated by measuring Cronbach's alpha coefficient and composite reliability. Exploratory and confirmatory factor analyses were used for the evaluation of the construct validity of the scale. Smart-PLS software was used for statistical modeling. RESULTS: Composite reliability and Cronbach's alpha coefficient for each scale ranged between 0.90 and 0.95, indicating adequate internal consistency with the LibQUAL scale. Confirmatory factor analysis confirmed the three-factor model of the LibQUAL scale. The convergent validity of the scale was supported, as the average variances extracted for all proposed factors were above 0.50. The discriminant validity was also confirmed using Fornel and Larcker and Heterotrait-Monotrait Ratio (HTMT) methods. CONCLUSION: Evaluation of psychometric properties of the translated and locally modified LibQUAL in the Persian language indicated adequate reliability, factorial validity, and stability of this instrument for Iranian health sciences libraries.