Informal peer-assisted learning groups and dental exam performance: A network test
M. AbdelSalam, Maha El Tantawi, Asim Al-Ansari, Adel AlAgl, Ghada Al-Harbi
Medical Principles and Practice
Network design
Nodes
fourth-year dental students who responded to the peer-learning and friendship survey
Ties
undirected co-membership and reported friendship links within self-formed study groups of two to four students
Methods
The authors visualized the network in Gephi, calculated degree-based positions, described group composition, and used regression models to compare network membership and prior grades with examination performance.
Source: Medical Principles and Practice
Reviewed summary
Seventy-four of 92 invited fourth-year dental students reported self-formed peer-learning groups and friendship relations before a 22-item examination. Network and regression analyses found prior academic performance was the clearest predictor, while group structure and peer similarity did not show a dependable advantage. The study is a useful negative finding because it separates the existence of peer ties from evidence that those ties improve learning, discouraging educators from treating any visible study group as an effective intervention.
Prior grade remained the dominant predictor; belonging to a group with friends showed only weak evidence, and neither similarity nor the observed small-group structure consistently improved scores. This was one dental class with 74 respondents, self-reported relations, very small self-selected groups, a short observation period, gender segregation, and no randomized comparison. The analysis is correlational and cannot show that friendship or group membership caused examination outcomes.
SNA
How SNA was used
The study defined fourth-year dental students who responded to the peer-learning and friendship survey as nodes and undirected co-membership and reported friendship links within self-formed study groups of two to four students as ties. The authors visualized the network in Gephi, calculated degree-based positions, described group composition, and used regression models to compare network membership and prior grades with examination performance.
Source
Medical Principles and Practice, 26(4), 337-342
DOI: 10.1159/000477731
