Vinsamlegast notið þetta auðkenni þegar þið vitnið til verksins eða tengið í það: http://hdl.handle.net/1946/35707
The subject of this thesis is learning analytics in higher education, and different learning patterns and behaviours that can be seen as contributing factors when striving for good results in a course. Ultimately establishing an understanding regarding successful learning patterns and behaviour in higher education. The research questions are: 1. How do students in higher education use a learning management system as a part of their learning process, 2. Which patterns and behaviour contribute the most towards a high grade?
To answer the research questions this thesis is based on an analysis of a quantitative questionnaire (based on 273 responses), descriptive statistics as well as machine learning of click-logs generated from the learning management system Canvas, by analysing learning patterns, in connection to the final grade of students. The study included data from six undergraduate courses taught in spring and autumn of 2019 within the Department of Computer Science at Reykjavik University.
The main findings show the importance of defining the right variables when predicting final grades instead of redistributing the grade categories. The findings also indicate the importance of students remaining active throughout their courses and that through continuous engagement, high grades can be achieved.
|Learning Analytics - Discovering and understanding different learning patterns.pdf