THE PREDICTORS OF LEARNERS’ TEST SCORES IN AN ONLINE EXAM PREPARATION SYSTEM: AN EDUCATIONAL DATA MINING APPROACH

Date

2024-1-24

Author

Günay Gökben, Ayşe

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

2
views

0
downloads

Cite This

The study was carried out with a three-pronged purpose. The primary objective of this study was to elucidate the relationship between students' interaction within the online learning environment and their exam performance. The secondary aim was to analyze how student participation in learning activities at different times of the day correlates with their exam performance. The third aim was to examine the patterns of students’ learning behavior throughout the online learning process in relation to their levels of exam performance. The study analyzed the activities of 667 students enrolled in online learning courses over a 10-month period. The dataset comprised 51 independent variables reflecting the diverse aspects of students' interaction behaviors. A nationwide standardized examination defined the dependent variable. According to results, taking practice tests, live class attendance, studying with video recordings, and mobile app usage were significantly associated with improved exam performance, while progress inventory usage showed a negative correlation. Time-of-day study behaviors, particularly video viewing in the morning and solving practice tests towards evening, were positively associated with exam performance. Process mining results indicated that high achievers predominantly initiated their learning with lecture notes, proceeded to practice tests and video recordings, and then attended live classes in a prepared manner, with notable bidirectional activity. On the other hand, medium and low achievers showed varied patterns of interaction. The findings led to the development of recommendations aimed at enhancing the institution's online learning environment, thereby contributing to the advancement of pedagogical practices in this domain.

Subject Keywords

Educational Data Mining, Online Learning Environment, Interaction Types, Exam Performance, Process Mining

URI

https://hdl.handle.net/11511/110076

Collections

Graduate School of Natural and Applied Sciences, Thesis