Peer Instruction in Hybrid Computing Courses: The Relationships Between Student Modality, Discussion, and Learning
Background: Peer instruction is an active learning pedagogy that emphasizes collaboration and learning from one’s peers. Prior research in computing education has shown that using peer instruction can improve students’ learning, reduce dropout and failure rates, and result in larger increases in students’ self-efficacy relative to standard lecture-based instruction. However, it has not investigated the use of peer instruction in hybrid courses, which have become increasingly popular in higher education due to their flexibility and ability to manage increases in enrollment. Objectives: We evaluated the effectiveness of peer instruction at facilitating students’ learning of computing concepts in in-person and online modalities of a hybrid course. Additionally, we examined how students’ discussions between rounds of peer instruction relate to their learning of computing concepts and whether these relationships vary when courses follow different attendance policies. Context and Methods: We analyzed students’ normalized learning gains from peer instruction in two offerings of a hybrid undergraduate data science course where students could choose to attend class meetings in person or synchronously online. In the first (open-policy) offering, students could self-select their modality for each class meeting and alternate modalities at will. In the second (closed-policy) offering, students needed to obtain instructor permission each time they wanted to attend class online, which drastically reduced the proportion of students who did so. Findings: Students who attended class online in the open-policy offering experienced smaller normalized learning gains than students who attended class in person in either offering. Additionally, in the open-policy offering, students who discussed the peer instruction questions in person experienced higher normalized learning gains than students who engaged in discussion online. Lastly, students who engaged in in-person discussion in the open-policy offering experienced higher normalized learning gains than students who did so in the closed-policy offering. Implications: Our results suggest that peer instruction may be more effective in person as opposed to online, particularly if there are differences in discussion environments across modalities. Additionally, they provide insight into how students’ discussion behaviors are related to their modalities. We encourage other instructors to take them into consideration when designing and implementing peer instruction in a hybrid computing course.
Authors
- Shao-Heng Ko (ORCID: https://orcid.org/0000-0003-0647-2837)
- Janet Jiang (ORCID: https://orcid.org/0009-0003-3237-0751)
- K Stephens (ORCID: https://orcid.org/0000-0002-3058-7418)
Institutions
- Duke University (US)
- Vanderbilt University (US)
- University of Washington (US)
Publication Details
- Journal
- ACM Transactions on Computing Education
- Published
- 2026-11-20
- DOI
- https://doi.org/10.1145/3864438
- Primary Topic
- Innovative Teaching Methods
- Type
- article
- Field-Weighted Citation Impact
- 0.00