The Effect of Messaging on Project Completion Rates in an Introductory Computing Class utilizing Mastery Learning.

Inspired by Grading for Equity [1, 2], we embarked on an ambitious pilot for our introductory non-majors university course in the fall of 2022 (n=234) and spring of 2023 (n=107). We have six projects (five programming, one essay), and decided to remove all penalties for late submissions, in order to support students who learned the material at different rates and needed more time to complete them. This meant that students, in theory, could turn in all projects on the last day of class with no penalty. In the fall, we proudly announced ''No Late Penalties!!'' on the first day of class. In the spring, we continued to have the policy internally, but externally messaged that if they needed, they could either have a two week extension, or a one month ''emergency'' extension. If they needed more, they would just have to reach out and have a conversation with a staff member to check in -- this added friction was critical. After meeting with the student, we would always grant the extension request (but this ''always say yes'' wasn't made public). We compared the completion rates of projects across both semesters for the different messaging and found completion rates for the projects in the spring to be significantly higher than the fall. The lesson: messaging matters; it is better to hide a ''no late penalties'' policy under a layer of abstraction, and publicly message that there is an expected completion rate, with extra time available to anyone who needs it.