This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. For attribution, the original author(s), title, publication source (PeerJ PrePrints) and either DOI or URL of the article must be cited.
Cite this article
Yalcin MA, Gardner EE, Anderson LB, Kirby-Straker R, Wolvin AD, Bederson BB.2015. Analysis of consistency in large multi-section courses using exploration of linked visual data summaries. PeerJ PrePrints3:e964v1https://doi.org/10.7287/peerj.preprints.964v1
Higher education courses with large student enrollments are commonly offered in multiple sections by multiple instructors. Monitoring consistency of teaching activities across sections is crucial in achieving equity for all students, and in developing strategies in response to emerging patterns and outliers. To address this need, we present an approach to analyze the multivariate data of sections, assignments and student submissions collected by a learning management system (LMS) using a new data exploration framework that we call linked data summaries. Data summaries are a unit of exploration with uncluttered, analytical, comprehensible visualizations of aggregations of data records attributes. Data browsers link multiple summaries and record lists, and enable flexible and rapid data analysis through tightly coupled interaction. Our analysis approach, developed in collaboration between analytics researchers and university instructors, reveals patterns across many aspects, including assignment and section structures, submission grading and timeliness. We present findings from an analysis of three semesters of an introductory oral communication course with over 1,750 students and 90 sections per semester.
This paper has been submitted to PeerJ Computer Science for review.