Libraries of all types are using an increasing range of online applications for services to connect and communicate with their users. Their emerging use of social media for service delivery emphasizes the need for understanding user perceptions on service quality in order to meet user needs. However, studies about social media in libraries have found several limitations in identifying user perceptions, including outdated evaluation statements, fragmented quantitative information from applications, and a lack of measurement instruments. This study addresses some of these gaps by examining the applicability of the E-S-QUAL instrument, which was developed by Parasuraman et al. (2005), for measuring the service quality of library social media services. E-S-QUAL has never been applied in the library service field. Nine hypotheses address the applicability in terms of 1) the scale's reliability and validity, 2) the relationships between user-perceived service quality and three related variables to service quality (overall quality, customers' perceived value of information, and loyalty intentions), and 3) the scale's potential to identify differences between user-perceived importance for a library social media service and their-perceived performance of the service.;Online surveys were used to collect data in five academic university libraries in North America offering Twitter services. A questionnaire based on a modified version of E-S-QUAL was distributed through library social media accounts, and a total of 266 responses were analyzed. Multivariate statistical methods were used to analyze the data, including exploratory and confirmatory factor analysis, correlation, regression analysis, and t-test in order to identify reliability, dimensionality, and convergent, discriminant, predictive, nomological, and known-group validity using SPSS and Amos.;This study found that the modified E-S-QUAL has good reliability and relationships with criterion variables as an instrument to measure the service quality of library Twitter accounts in four dimensions. Although some psychometric properties of the scale were supported, dimensionality and validity tests diagnosed the possibility of within-measure and across-measure correlational systematic error. In order to reduce the possibility of such errors, rewording shared phrases in different constructs and different items' sequence on the questionnaire is suggested for application of the E-S-QUAL instrument for assessment of service quality of Twitter used by academic libraries. The results of this study help library managers apply the instrument to measure service quality, determine directions and strategies for social media services, and improve their performance to meet and exceed user needs. |