Migration of Clusters from Pre-session to Post-session: An Analysis of Elderly Students’ Perceived Digital Literacy

The Open University of Japan has offered a face-to-face digital literacy training session (DLT) every semester at all 50 study centers since 2010. Each student attending DLT completes a checklist at the beginning and end of their session. The objective of this study was to identify how students migrated from one cluster to another, representing their self-reported perceived digital literacy skills. We analyzed responses from 1,417 students who submitted checklists between Semester 2, 2014 and Semester 2, 2016 and completed all 20 items on both the pre- and post-session checklists. We developed clusters for pre- and post-session ability, based on each student’s reported ICT skill levels on the 20 checklist dimensions, using k-means with the Hartigan–Wong algorithm. We also analyzed how each of the pre-session clusters shifted to post-session clusters using a Sankey diagram. In addition, we analyzed students’ post-session checklist comments on DLT using a co-occurrence network method.