Interaction Map: A Visualization Tool for Personalized Learning Based on Assessment Data

Abstract

Personalized learning is the shaping of instruction to meet students’ needs to support student learning and improve learning outcomes. While it has received increasing attention in education, limited resources are available to help educators properly leverage assessment data to foster personalized learning. Motivated by this need, we introduce a new visualization tool, the interaction map, to foster personalized learning based on assessment data. The interaction map approach is engineered by the latent space item response model, a recent development in assessment data-leveraging social network analysis methodologies. In the interaction map, students and test items are mapped into a two-dimensional geometric space, in which their distances tell us about the student’s strengths and weaknesses with individual or groups of test items given their overall ability levels. Student profiles can be generated based on these distances to display individual student strengths and weaknesses. Finally, we introduce a user-friendly, free web-based software IntMap in which users can upload their own assessment data and view the customizable interaction map and student profiles based on settings that users can adjust. We illustrate the use of the software with an educational assessment example.