Abstract:

Nowadays, with the spread of the digital transformation process in education, digital tools have started to be used frequently in learning environments. Especially in the post-pandemic period, it is seen that online learning environments have undergone a rapid integration process into face-to-face learning environments. These environments, which are frequently used especially in higher education, are known as systems that support traditional learning environments. In this context, blended learning is known as hybrid systems in which face-to-face and online teaching content is presented to students through these tools. Among these tools, learning management systems, especially Moodle LMS, is one of the frequently used and preferred systems by institutions in higher education. On the other hand, with the use of learning management systems as online learning environments, students' interaction data on these systems have become one of the data sources that are taken into consideration because they provide an important (even if not complete) understanding of students’ actual learning behaviors. In this context, the field of learning analytics has been a field that has been employed in the process of collecting, analyzing and reporting online interactions data. From this point of view, in this study, the learning processes of first year computer engineering students who took courses with the same course design with blended teaching method in two different years (2021 and 2022) were compared. The learning designs of the two years were equivalent, but in the second iteration (2022) the course coordinator distributed to students weekly notifications with evolving feedback about the course learning design emphasizing next assignment deadlines. Thus, it was aimed to increase teacher-student interaction and to support students' self-regulation of their activities in alignment with the proposed learning design. The data of a total of 440 students who took the Introduction to ICT course given in Spanish and English languages in two different years log data analyzed using Python in order to reveal the weekly and component-based learning behaviors of the students. Box-plot, one of the data visualization techniques, was used. Data visualization techniques and quantitative analysis techniques were used and comparisons were made through the interaction events of students interactions of weekly learning activities in the Moodle LMSs. As a result of the data analysis, statistically significant differences were observed in the weeks of 2021 (4, 7, 10, and 8) and in the weeks of 2022 (1, 5, 6, and 11). As a result of the analysis of the data related to the learning design elements after 13-weeks of implementations, in the data belonging to 2021 years, forum, videos, quizzes, pages, and turnitin assignment elements were found to be significantly different in both courses. In the 2022 year assignment, files, file submissions, and URL are the most meaningful design elements in the English and Spanish courses. On the other hand, among the nine elements that received the most interaction, the elements were quizzes. Therefore, it can be said that messages sent to students on a weekly basis will increase their assignment submission behavior related to self-regulation skills by 2022. Overall, the results of this study may provide guidance to instructors, designers, and researchers who design LMSs in blended learning environments.

Keywords:

Blended learning, Learning analytics, Log data, Comparative study, Higher education.