In higher education, academic libraries are increasingly recognised as key sites for data-informed student support. Despite the growing interest in learning analytics, research in this area remains fragmented, with inconsistent methodologies and limited synthesis, particularly regarding the personalisation of library services. Following the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) guidelines, this systematic review examines how learning analytics are operationalised in academic libraries to personalise services and enhance student learning outcomes. Guided by the Learning Analytics Cycle (LAC), the study synthesises 32 peer-reviewed empirical studies published between 2016 and 2025. It categorises data sources, including library usage logs, learning management system analytics, and artificial intelligence (AI)-enhanced interaction data, and examines corresponding analytical methods such as descriptive statistics, predictive modelling, and natural language processing. Findings reveal a strategic shift towards proactive, personalised interventions, including targeted information literacy instructions and early risk detection mechanisms. However, implementation is often constrained by ethical concerns, governance gaps, technical infrastructure limitations, and staff analytics literacy. The review concludes that while learning analytics hold transformative potential for academic libraries, full integration requires investment in ethical frameworks, system interoperability, and inclusive policy design. Recommendations are provided to support the development of scalable, student-centred analytics ecosystems within library services.
扫码关注我们
求助内容:
应助结果提醒方式:
