Learners’ Performance in a MOOC on Programming

Abstract

In every course, there are learners who successfully pass assessments and complete the course. However, there are also those who fail the course for various reasons. One of such reasons may be related to success in assessment. Although performance in assessments has been studied before, there is a lack of knowledge on the degree of variance between different types of learners in terms of scores and the number of resubmissions. In the paper, we analyse the performance in assessments demonstrated by non-completers and completers and by completers with different engagement levels and difficulty-resolving patterns. The data have been gathered from the Moodle statistics source based on the performance of 1065 participants, as regards their completion status, the number of attempts made per each programming task and quiz, and the score received per quiz. Quantitative analysis was performed with descriptive statistics and non-parametric tests. Non-completers and completers were similar in resubmissions per quiz, but the former, expectedly, made more resubmissions per programming task and received lower quiz scores. Completers made more attempts per task than per quiz. They could provide a correct solution with a few resubmissions and receive good scores already at a pragmatic engagement level. At the same time, the increased use of help sources in case of difficulties was also associated with a higher number of attempts and lower quiz scores received. The study may have implications in understanding the role of assessments in dropouts and how completers with different engagement and difficulty-resolving patterns cope with assessments.