Developing A Data Driven Framework To Improve Feedback For Educators

This paper presents a framework for the use of weekly reflections to drive iterative course improvement throughout a term. This data includes weekly student reflections, where students anonymously self-assess both their behavior (through attendance and use of course resources) and their comprehension of course concepts. Results from a pilot of this framework for a large first-year introduction to computer programming course are used to illustrate the use case. Improvements to this framework from a previous term are made and discussed. Results demonstrate student reflections can offer actionable insights for targeted interventions in lectures and tutorials. Key observations include the impact of incoming experience, patterns of students falling behind, and differences across cohorts. By automating a portion of this data collection and analysis, student feedback can be integrated with minimal administrative burden. The authors plan to further automate this framework and develop predictive models based on specific course factors.