Parental Training and CBC Implementation Quality: A Mediation Analysis of Engagement as Intermediary in Sotik Sub-County, Kenya

Kenya's Competency-Based Curriculum (CBC) designates parental training as the structural core of the Parental Empowerment and Engagement (PEE) framework. However, empirical evidence on the mechanism through which training influences CBC implementation quality remains limited. Existing studies report direct associations without testing whether parental engagement mediates that relationship. This study addressed the gap by testing the mediating role of composite parental engagement in the relationship between parental training and CBC implementation quality in public junior secondary schools in Sotik Sub-County, Bomet County, Kenya, and by proposing a model grounded in the mediation evidence. A convergent parallel mixed-methods design was employed. Stratified random sampling yielded 378 valid responses from 398 targeted respondents across five school zones (response rate = 94.97%), comprising headteachers, teachers, parents, and students. Quantitative data were analysed using descriptive statistics, Pearson product-moment correlation, and Baron and Kenny's (1986) causal steps mediation procedure with bootstrapped confidence intervals in IBM SPSS v.26. Qualitative data from semi-structured interviews and focus group discussions were analysed thematically following Braun and Clarke (2006). Parental training was positively correlated with CBC implementation quality, r(376) = .37, p < .001. Baron-Kenny mediation analysis confirmed that composite parental engagement — comprising guidance and counselling (r = .99), instructional materials provision (r = .93), and decision-making involvement (r = .87) — partially mediated this relationship. The total effect (β = .37, p < .001) was reduced to a residual direct effect of β = .19 (p = .041) when the mediator was controlled, and the indirect effect (β = .369, 95% CI [.28, .46]) was statistically significant and larger than the residual direct effect, indicating that a substantial portion of training’s influence on CBC quality operates through composite parental engagement. Five qualitative themes corroborated the statistical model. Variance inflation factor diagnostics confirmed the absence of problematic multicollinearity across all regression steps (maximum VIF = 2.14). Discriminant validity testing using the Fornell-Larcker criterion and heterotrait-monotrait ratio revealed that the G&C engagement and CBC quality constructs approach but do not fully meet discriminant validity thresholds, a limitation transparently reported and addressed through multi-informant design recommendations. The study proposes the Parental Training-Mediated Engagement Model for CBC Implementation (PTME-CBC) as a theoretically grounded and empirically derived framework for policy and practice