Automated Feedback Generation in Programming Environments: A Systematic Review

• In Event: AERA Virtual Poster Session 3
  In Poster Session: Systematic Review and Meta-Analysis SIG Poster Session

Abstract

Programming is a skill that students find challenging to learn, and teaching novices how to program can be a difficult task, especially in large courses. Feedback can speed up the learning process and encourage students to keep learning, but high-quality feedback is difficult to generate at scale. This systematic literature review explores the recent literature on automated feedback generation. It focuses on generating different types of feedback, and it includes 20 peer-reviewed records published in the past five years. Findings revealed that almost half of the studies: (1) used rule-based artificial intelligence approaches; (2) did not evaluate the impact of delivering automated feedback to students; (3) required manual effort or the collection of students’ historical data to enable feedback generation.

Authors

• Mohammad Karimiabdolmaleki, University of Alberta

• Carrie Demmans Epp, University of Alberta

• Maria Cutumisu, University of Alberta