Year: 2018
Author: Gonzalez, Carlos, Howard, Sarah K, Clasing, Paula, Montenegro, Maximiliano, Jack, Yang, Jun, Ma, Dany, Lopez
Type of paper: Abstract refereed
Abstract:
Learning analytics (LA) enables the use of data, generated from university students’ interactions with digital systems, to explore digital technology usage patterns. However, it has been criticised for a lack of connection with existing learning theories. In this study, we aim at exploring students’ digital technology usage patterns at one Chilean university. We combined data gathered using LA techniques and qualitative data. We interpreted them using the well-known Students’ Approaches to Learning framework. Thus, we intend to overcome criticisms and present a more holistic picture: providing insights both on students’ behaviour and on their underlying intentions and strategies.
We gathered one semester of data from undergraduates with complete records (18,042), from three sources: LMS, library and students’ records. We performed a k-means cluster analysis to identify groups with similar usage patterns. We then performed association rules mining for each of the clusters. Finally, students in each cluster were identified and invited to focus groups to explore usage patterns.
Three clusters were identified. Students in Cluster 1 borrowed more library print material and a higher than average amount of time accessing library digital resources, below average LMS use and above average grades. Students in Cluster 2 use the LMS more than average, borrowed more library print material than average, accessed library digital resources for a lower than average amount of time and had average grades. Finally, students in Cluster 3 showed lower than average LMS, library print material and library digital resources and had below average grades.
Association rules mining confirmed identified patterns and allowed to see that students in Cluster 1 were in their final years of study, while students in Clusters 2 and 3 were in their initial years. Interestingly, there is a group of first year students in Cluster 1.
Focus groups showed that students in Cluster 1 presented a deep-strategic approach to study: expanding and deepening class material with research papers or other relevant material and, at the same time, organizing time to achieve good outcomes. Students in Cluster 2 presented a strategic approach, maximizing time for doing good summaries or practicing math exercises. Finally, students in Cluster 3 tended to emphasize memorization and used others’ materials – summaries from other students or from other cohorts –. Regardless the cluster, all students described intensive use of digital platforms for learning others than the ones provided by the university (Google Drive, WhatsApp, Facebook, etc.). Implications will be discussed.
We gathered one semester of data from undergraduates with complete records (18,042), from three sources: LMS, library and students’ records. We performed a k-means cluster analysis to identify groups with similar usage patterns. We then performed association rules mining for each of the clusters. Finally, students in each cluster were identified and invited to focus groups to explore usage patterns.
Three clusters were identified. Students in Cluster 1 borrowed more library print material and a higher than average amount of time accessing library digital resources, below average LMS use and above average grades. Students in Cluster 2 use the LMS more than average, borrowed more library print material than average, accessed library digital resources for a lower than average amount of time and had average grades. Finally, students in Cluster 3 showed lower than average LMS, library print material and library digital resources and had below average grades.
Association rules mining confirmed identified patterns and allowed to see that students in Cluster 1 were in their final years of study, while students in Clusters 2 and 3 were in their initial years. Interestingly, there is a group of first year students in Cluster 1.
Focus groups showed that students in Cluster 1 presented a deep-strategic approach to study: expanding and deepening class material with research papers or other relevant material and, at the same time, organizing time to achieve good outcomes. Students in Cluster 2 presented a strategic approach, maximizing time for doing good summaries or practicing math exercises. Finally, students in Cluster 3 tended to emphasize memorization and used others’ materials – summaries from other students or from other cohorts –. Regardless the cluster, all students described intensive use of digital platforms for learning others than the ones provided by the university (Google Drive, WhatsApp, Facebook, etc.). Implications will be discussed.