Improving Automated Group Assignments in an Academic Setting

A comparison of runtimes of our approach (GAT Random) compared to the prior state-of-the-art (Team-Maker).

Abstract

This research paper outlines and evaluates a method to efficiently assign students to groups in an academic setting. Group work is an important part of academic learning. Prior research shows that data-oriented group assignment methods outperform self selected and random methods. We introduce the Group Assignment Tool, an automated method to perform data-oriented group assignment in an academic setting. The Group Assignment Tool maximizes the benefits of instructor-assigned groups by performing probabilistic weighted optimization of groups based on student survey data. We show that the Group Assignment Tool produces comparable outcomes to prior hill climbing algorithms for group formation with significantly faster runtime, and produces groups which score within a small factor of optimal scores. We present the GAT via a traditional presentation as well as a demonstration.

Publication
2020 ASEE Virtual Online Conference

Click the Cite button above to demo the feature to enable visitors to import publication metadata into their reference management software.

Create your slides in Markdown - click the Slides button to check out the example.

Chris Miller
Chris Miller
Software Engineer & Incoming PhD Student

I’m a software developer at Microsoft working on Azure services, and an incoming PhD student at USC Viterbi.