What is it about?
Automatic grading systems have been widely deployed in university courses to enable course staff to spend more time interacting with learners and less time grading assignments. At the same time, automatic grading systems provide formative feedback to help students learn and improve. Unfortunately, some learners abuse these formative feedback systems by invoking them repeatedly to try to guess their way to the correct answer. In this work we explore applying regression penalties to encourage learners to more thoughtfully invoke the automatic grader.
Featured Image
Photo by JESHOOTS.COM on Unsplash
Why is it important?
By applying regression penalties we demonstrate that learners are more mindful of how they interact with the grading platform, reducing their use of the platform by over half. Learners spend more time thinking about and testing their own solutions, with minimal impact to their own performance in the course.
Read the Original
This page is a summary of: STOP THE (AUTOGRADER) INSANITY: Regression Penalties to Deter Autograder Overreliance, March 2021, ACM (Association for Computing Machinery),
DOI: 10.1145/3408877.3432430.
You can read the full text:
Contributors
The following have contributed to this page
