SIGCSE TS 2025 (series) / Posters /
Using LLMs to Detect the Presence of Learning Outcomes in Submitted Work Within Computing Ethics Courses
This program is tentative and subject to change.
Fri 28 Feb 2025 15:00 - 17:00 at Exhibit Hall C - Posters - Posters #3
This study investigates the role of large language models (LLMs) in identifying learning outcomes from student submitted work in a computing ethics course. We use grounded theory to craft a codebook to spot key learning outcomes, such as the usage of critical reasoning and awareness of various social issues. Using Cohen’s kappa to assess interrater reliability, we compare human reviewers’ coding to outputs from models like GPT-4o and GPT-3.5- turbo, finding that GPT-4o performed just as well as the agreement between human reviewers. We use the model outputs to identify specific course readings that students engaged particularly deeply with to better inform our computing ethics instruction.
This program is tentative and subject to change.
Fri 28 FebDisplayed time zone: Eastern Time (US & Canada) change
Fri 28 Feb
Displayed time zone: Eastern Time (US & Canada) change