This study used Learning Analysis data to analyze the factors that influence cyber university students' dropout. This dropout prevention prediction model that incorporates Learning Analysis was presented for reducing dropout rates. For this purpose, H Cyber University students were divided into an experimental group and a control group, and a difference analysis (T-test) was performed. Logistic regression analysis was conducted on independent factors affecting dropout. The results of the study are as follows: First, for the experimental group, academic continuity (re-enrollment) was 93.3% (373 students/400 students), whereas for the control group, academic continuity (re-enrollment) was 79.3% (317 students/400 students). This indicates that the experimental group had a 14% higher academic continuity (re-enrollment) rate, with 56 more students re-enrolling. Second, six types of factors (immersion in the target institution, academic performance, faculty and staff interaction, peer group interaction, educational data mining, and learning analysis data) were identified as major factors in reducing dropout. Third, it was found that as the expected value of the major dropout prevention factors increases, academic continuity (dropout) decreases. It was determined that academic continuity (dropout) could be improved by applying a dropout prevention prediction model based on Learning Analysis. Lastly, the analysis results of this paper have limitations, as they are based on the results from a specific cyber university. The need for follow-up research on preventing dropout across all cyber universities was highlighted.
Read full abstract7-days of FREE Audio papers, translation & more with Prime
7-days of FREE Prime access