In our programme we work in a learning-outcome-driven way. We attach several rubrics to six assignments, and some learning outcomes (for example BC 01: collaboration) appear in multiple assignments. Each assignment links multiple learning outcomes to a single rubric.
We use the “weighted average” setting, with 50 % of the score coming from the most-recent result for each outcome. That works well—as long as you can grade the assignments in chronological order—until a student resubmits an earlier assignment.
If we then re-evaluate only one criterion (learning outcome) in that rubric, say just BC 01, Canvas treats every outcome in that rubric as “most recent”. This single action therefore overwrites all previous outcome scores in the result, even though they were not actually re-assessed. That makes the outcome tracking unreliable.
Request:
We propose that Canvas introduce timestamps at the learning-outcome-criterion level rather than at the rubric level. Then, when one (or several) outcomes are adjusted, only those outcomes receive a new date stamp. This would keep the “most recent score” calculation accurate and purposeful, even when assignments are resubmitted late.
Maybe as an option, so we can have a choice on how timestamps work?
