Peer Scaffolds for Mathematical Confidence and Academic Persistence

Abstract

This study investigated how peer scaffolds contributed to mathematical confidence and academic persistence among Grade 6 learners in Santa Maria, Isabela. Using a predictive path-modeling cross-sectional design, data were obtained from 180 learners selected through proportionate stratified random sampling. A validated 36-item questionnaire measured peer scaffolds, mathematical confidence, and academic persistence, while partial least squares structural equation modeling was employed to test direct, indirect, and predictive relationships among the constructs. Results showed that peer scaffolds received the highest mean, while academic persistence emerged as the comparatively weaker learner outcome. The measurement model demonstrated satisfactory reliability, convergent validity, and discriminant validity. Structural analysis revealed that peer scaffolds strongly predicted mathematical confidence (β = .612, p < .001) and had a smaller but significant direct effect on academic persistence (β = .208, p = .003). Mathematical confidence significantly predicted academic persistence (β = .447, p < .001) and partially mediated the relationship between peer scaffolds and persistence, with an indirect effect of β = .274. The model explained 37.5% of the variance in mathematical confidence and 35.7% in academic persistence, with predictive assessment confirming useful out-of-sample performance. The findings indicate that peer support becomes more consequential when it strengthens learners’ belief in their own mathematical capability. Structured peer scaffolding that gradually transfers responsibility to learners is therefore recommended to promote confident and sustained mathematical engagement.