Title: Using learning analytics to identify successful learners in a blended learning course
Authors: Sotiris Kotsiantis; Nikolaos Tselios; Andromahi Filippidi; Vassilis Komis
Addresses: Department of Mathematics, University of Patras, 26500 Rio, Patras, Greece ' Department of Educational Sciences and Early Childhood Education, ICT in Education Group, University of Patras, 26500 Rio, Patras, Greece ' Department of Educational Sciences and Early Childhood Education, ICT in Education Group, University of Patras, 26500 Rio, Patras, Greece ' Department of Educational Sciences and Early Childhood Education, ICT in Education Group, University of Patras, 26500 Rio, Patras, Greece
Abstract: In this paper, students' practices while using a Learning Content Management System in a blended learning environment were examined. This is a case study involving 337 students who attended an academic course based upon a blended learning approach over three years using Moodle. Eighteen variables depicting the students' perceptions of Moodle, as well as their interaction with it, were examined using four complementary data mining and statistical analysis approaches: visualisation, decision trees, class association rules and clustering. The analysis of the collected data shows that failure in the course was associated with negative attitudes and perceptions of the students towards Moodle. On the other hand excellent grades were associated with increased use of the LCMS. Requirements elicitation of a learning analytics dashboard, are also discussed.
Keywords: learning analytics; blended learning; learning content management systems; LCMS; case study; interaction data; higher education; Moodle; data mining; visualisation; decision trees; class association rules; clustering; student attitudes; student perceptions.
DOI: 10.1504/IJTEL.2013.059088
International Journal of Technology Enhanced Learning, 2013 Vol.5 No.2, pp.133 - 150
Received: 21 May 2013
Accepted: 24 Sep 2013
Published online: 23 Sep 2014 *