Abstract:

The concept of learning analytics emerges as a reflection of big data technology in the field of education to improve the quality of learning. Students leave large amounts of data from digital traces, especially during e-learning activities. These data can be processed to facilitate e-learning processes or to use for other educational and administrative purposes. Techniques such as predictive modeling, social network analysis (SNA), user, usage tracking, content analysis, semantic analysis, suggestion systems are among learning analytics techniques and learning analytics are used to identify and improve current learning process performance. Due to today's learning environments are digitalized, students are more lonely, their motivation is lower during e-learning activities, so managing the learning process is more difficult. Therefore, learning analytics applications come to the fore in today's education systems, just like virtual mentors. It will be easier to improve and manage the e-learning process, which has become more individualized through virtual mentors, compared to traditional education. In this study, an application study done before will be discussed in terms of how learning analytics can be adapted to educational institutions, and the adaptation of learning analytics application design (Learning Analytics Application Design, LAID) principles will be discussed with research questions. LAID consists of conceptual and logistical coordination, as well as the principle of coordination, comparison, and customization. The application data consists of the answers given to the learning tests applied in the Mathematics course by Sakarya University Electronic Technology distance education students at the associate degree level in the fall semester of the 2010-2011 academic year. The developed learning tests were first applied to Computer Programming students with similar characteristics, and after the item analysis was completed, updated tests consisting of the best questions were applied to the Electronic Technology students. Since participation in the tests is not compulsory, students who want to learn about their own learning performance participated in the published tests, and the number of participants varies between 88 and 107. Respectively, 107 students participated in sets tests, 93 students participated in numbers test, 101 students in algebra tests, 89 students in inequalities and equations tests, 105 students in functions tests, and 88 students participated in logarithm and trigonometry tests. Findings obtained, evaluation of the application in terms of learning analytics perspective, suggestions for improvement are given in the results section.

Keywords: