Assignment Performance Comparison

Overview:

Tool: ELMS-Canvas + Gemini*  

Time: 20 min  

What you need: ELMS-Canvas Course Analytics and your assignment descriptions

How to do it:

1. In your ELMS-Canvas course, click "Course Analytics" in the left-hand navigation (in blue). Select "Course Grade" from the top menu. You will see a graph of grade distribution across assignments over time.
2. Click individual assignments (blue circles) to see the spread of grades for each assignment - this will pop up in a sidebar on the right-hand side of the screen.
3. Identify two or three assignments that look unusual - either a very wide spread of grades, a cluster of low scores, or a bimodal distribution (lots of A's and lots of F's with few in between).
4. Open Gemini. Upload or paste the assignment prompt and rubric for an assignment you flagged.
5. Paste the prompt below along with a plain description of what the score distribution looked like:

"Here is the assignment description and rubric for [assignment name]. The score distribution showed [describe: e.g., "a cluster of low scores around 60%" or "a bimodal split"]. What does this distribution suggest about how clearly expectations were communicated, how the assignment was scaffolded, or whether the rubric was functioning as intended? Suggest two specific revisions."

Notes:  

This activity identifies patterns worth investigating, but it cannot explain causes without additional student context. A low-scoring assignment might mean the task was poorly designed, or that it followed insufficient preparation, or both.

Connect with the TLTC Academic Innovation Team to redesign specific assignments based on what you find.

*We encourage you to exclusively use UMD-approved GenAI tools, which are deployed in alignment with institutional security and compliance requirements.