Exploring the role of automated feedback in programming education: a systematic literature review

Automated feedback systems have become increasingly central to programming education, where learners iteratively write and refine code. Despite rapid technical advances, research on automated feedback remains fragmented across technological, pedagogical, and interactional dimensions. This systematic literature review synthesizes 58 studies published between 2005 and 2024, following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) protocol. Studies were identified through searches of Scopus, ACM Digital Library, and IEEE Xplore, screened against predefined criteria, and analyzed using a five-category coding framework encompassing system architecture, feedback content, learner interaction, educational context, and evaluation. Across the reviewed studies, feedback systems were typically fully automated and delivered immediate feedback focused on task-level support for code correctness and error detection. Recent systems have begun to incorporate adaptive mechanisms and feedback generated by large language models; however, relatively few offer learners meaningful control over when or how feedback is received. Implementation was concentrated in higher education, often in blended course formats, and evaluation emphasized short-term gains, with limited attention to long-term learning, classroom integration, or effects on teaching. Together, these findings contribute a multidimensional understanding of automated feedback as a pedagogical and interactional process embedded in programming learning environments.