Study of the impact of social learning and gamification methodologies on learning results in higher education

Abstract

In this work, as the last step of a longitudinal study of the impact of so- cial learning and gamification methodologies on learning results in higher education, we have recorded the activity in a software platform based on Moodle, especially built for encouraging online participation of the stu- dents to design, carry out and evaluate a set of learning tasks and games, during two consecutive editions of an undergraduate course. Our aim is to confirm the relationships of the patterns of accomplishment of the gam- ified activities and the network structure of the social graphs associated to the online forums with knowledge adquisition and final outcomes. For this purpose we have offered two learning paths, traditional and novel, to our students. We have identified course variables that quantitatively explain the improvements reported when using the innovative methodolo- gies integrated in the course design, and we have applied techniques from the social network analysis (SNA) and the machine learning/deep learn- ing (ML/DL) domains to conduct success/failure classification methods finding that, generally, very good results are obtained when an ensemble approach is used, that is, when we blend the predictions made by different classifiers. The proposed methodology can be used over reduced datasets and variable time windows for having early estimates that allow pedagog- ical interventions. Finally, we have applied other statistical tests to our datasets, that confirm the influence of learning path on learning results.