Utilizing Learning Analytics to Support Students' Academic Self-efficacy and Problem-Solving Skills

Abstract

The use of the flipped classroom (FC) model in higher education is becoming increasingly common. Although the FC model has many benefits, there are some limitations using this model for learners who do not have self-directed learning skills and do not have a developed learner autonomy. One of these limitations is that students with low academic self-efficacy (ASE) cannot improve their problem-solving skills (PSS). In the FC model, it is thought that these problems can be solved and supported by providing learning analytics to the learners and making recommendations and guidance based on these results. The aim of this study is to investigate the effect of recommendations and guidance based on learning analytics on ASE and PSS. The research was conducted according to experimental design with pretest-posttest control group and mixed methods approach (QUAN + qual) was used. This study was carried out on 44 university students randomly assigned to experimental group (LA group) and control group (non-LA groups). While feedback messages showing learning analytics to LA groups students were sent via learning management system (LMS) on a weekly basis, non-LA groups students did not receive any feedback messages. The experimental process of the study continued for 7 weeks within the FC model. Research data were obtained through ASE scale, problem-solving inventory and student opinion form. The results of the research show that sending feedback messages showing learning analytics to students has a statistically significant effect on their ASE and PSS. The qualitative findings of the study confirm the results obtained from the quantitative findings. Based on the current implementation, various suggestions were summarized for instructors, instructional designers, and researchers.