E-Learning Information Quality and Student Satisfaction in Kenyan Public Universities: A Mixed-Methods Investigation

Public universities have increasingly embraced e-learning following the shift to remote learning during the COVID-19 pandemic. While digital platforms have expanded access to teaching and learning, empirical evidence on the relationship between e-learning quality and student satisfaction in universities remains limited. This paper’s objective was to establish the influence of e-learning information quality on student satisfaction in six public universities in Kenya. This paper sought to answer the following question: Does the quality of e-learning information influence student satisfaction? This study applied a mixed-methods research design – anchored in the Resource-Based View theory integrated with DeLone and McLean’s Information System Success Model – to assess how information quality influences student satisfaction. A total of 446 respondents, consisting of 400 fourth-year students and 46 key informants, were selected using stratified random and purposive sampling. Data were collected using questionnaires and interview guides. Descriptive and inferential statistics were used to analyse quantitative data, using SPSS v. 25. Thematic analysis was applied to analyse qualitative data. The results reveal a strong positive correlation between e-learning information quality and e-learning student satisfaction (r = 0.656, p < .05). The study concluded that there is a statistical relationship between e-learning information quality and students’ satisfaction. Moreover, multidimensional institutional strategies, including adequate learning resources, clear instructions, proper communication, and timely feedback, influence the effectiveness of the e-learning experience. The results suggest the importance of sound information quality systems and student-centred pedagogical approaches, such as self-paced and interactive learning, to enhance student satisfaction and facilitate scalable and high-impact e-learning models.