User satisfactions on digital library: a correlational study/ Syasya Syazreen Suharto and Mohd Razilan Abdul Kadir

• Main Authors: Suharto, Syasya Syazreen, Abdul Kadir, Mohd Razilan
• Format: Article
• Language: English
• Published: Faculty of Information Management 2021
• Subjects: Information services. Information centers; Digital libraries
• Online Access: https://ir.uitm.edu.my/id/eprint/65521/1/65521.pdf; https://ir.uitm.edu.my/id/eprint/65521/; https://ijikm.uitm.edu.my/
• Institution: Universiti Teknologi Mara

Academic Digital Library (DL) is an online system providing access to a wide variety of academic content and services. It is designated to help students to access digital resources to fulfill their academic needs as an alternative to printed library materials through network environment. Covid19 pandemic has in large affecting and restricting teaching and learning activities on premises thus making it to be done through online. DL as an enablement system, is capable of channelling its benefits to students to get access and retrieve relevant academic resources digitally. The aim of this paper is to investigate the effectiveness of the academic DL in fulfilling students’ needs by correlating their satisfaction through usage experience with the following predictors: perceived of use, perceived usefulness, system quality, service quality and digital library collection. A quantitative study is opted by surveying 200 students of public university in Selangor. The study findings indicate that all of the posited determinants possess highly significant positive relationships with students’ satisfaction, ranging from r = 0.45 (the lowest) to r = 0.70 (the highest). Nevertheless, a serious multicollinearity issue is expected to arise in modelling part as most predictors are correlated to each other. It could lead to unreliable and unstable estimates of linear generalized regression coefficients. Thus, these bivariate relationships results serve as basis towards a larger scope of research to embark on structural path model rather than using generalized model to avoid the variance of model’s coefficient inflated due to linear dependence with other predictors when modelling this relationship.