Beyond Academic Meritocracy: Mentees’ Multi-Disciplinary Evaluation of aPeer Mentoring Program and the Relational-Scaffolding Mentorship Model (RSMM)

Background Peer mentoring serves as a pivotal educational scaffolding architecture across undergraduate health and basic science education. However, the precise empirical mechanisms through which mentor attributes and relational dynamics shape mentee appraisal across heterogeneous disciplines remain under-theorized. This study evaluated first-year undergraduate mentees’ multi-dimensional appraisal of a formal peer mentoring program and examined relational predictors of mentoring efficacy. Methods A descriptive-analytical cross-sectional study was conducted during the 2025–2026 academic year at Islamic Azad University, Yazd, Iran (N = 90). Mentees across medicine (n = 27), nursing (n = 26), psychology (n = 18), anesthesiology (n = 13), and biology (n = 6) completed a validated multi-domain instrument evaluating three functional dimensions: communicative/relational, academic support, and psychosocial support, alongside global satisfaction and open-ended feedback. Data were analyzed using independent-samples t-tests, one-way ANOVA with post-hoc LSD tests, and Pearson correlation coefficients in SPSS v27.0. Results The overall mentor evaluation score was 52.96 ± 12.40 (out of 65). The communicative domain received the highest proportional appraisal (8.91 ± 1.77), followed by academic (16.62 ± 3.74) and psychosocial support (27.42 ± 7.65). Global mentee satisfaction was favorable (4.44 ± 0.90 out of 5). Encounter duration significantly and positively correlated with all mentoring dimensions (r = 0.337 to 0.397 0, p ≤ 0.001), whereas meeting frequency showed no significant correlation. Mentor Grade Point Average (GPA) was not associated with overall evaluation (r = 0.082, p = 0.443), but exhibited a counterintuitive inverse correlation with communicative competence in the medical cohort (r = − 0.551, p = 0.003). Disciplinary disparities were prominent, with basic biology students reporting significantly lower scores across domains compared to clinical peers (p < 0.05). Qualitative thematic synthesis identified mentor communication inertia (17.8%) as the primary operational vulnerability. Conclusion Mentoring efficacy is governed by the substantive depth of relational dialogue rather than superficial meeting frequency or mentor academic standing. Synthesizing these empirical findings with Vygotskian scaffolding and safe-failure paradigms, we propose the Relational-Scaffolding Mentorship Model (RSMM) as an actionable framework advocating for non-cognitive mentor selection, temporal engagement, and relational safety containers in higher education.