Beyond Access: A Cluster-Randomized Trial of a Participatory Inclusive Education Intervention on Children's Wellbeing and Learning In Rural Pakistan

Background and Purpose: Despite expanded school enrollment in Pakistan, gains in actual learning remain uneven, leaving the most marginalized children furthest behind. Children in rural areas navigate compounding disadvantages, including economic deprivation, entrenched gender norms, and disability-related exclusion, that together erode meaningful inclusion and participation in schooling. In these contexts, achieving inclusive, quality education requires interventions that simultaneously strengthen inclusion, community accountability, and pedagogical quality. This study evaluates the impact of a participatory intervention that integrated community-based system dynamics (CBSD) and child-centered pedagogical teacher training on children's psychosocial and academic outcomes in rural primary schools in Pakistan. The study is among the first to evaluate a combined participatory systems and pedagogical intervention in this context during a period marked by compounding disruptions, including COVID-19-related school closures, catastrophic flooding, and political instability.

Methods: A cluster-randomized controlled trial was conducted across 137 schools (63 intervention, 74 control) in four districts of Punjab and Sindh provinces, with 4,242 students (2,015 intervention, 2,227 control) assessed across three waves (baseline, follow-up 1, follow-up 2). Trained NGO facilitators led group model building workshops across four implementation rounds (2018–2024), engaging children, parents, teachers, and school committee members as co-analysts of their own educational systems. Each stakeholder group surfaced local barriers to inclusion and negotiated school-specific action plans. Alongside this community process, teachers participated in a week-long training in child-centered pedagogy, with emphasis on project-based learning. Intervention effects were estimated using linear mixed-effects models with random intercepts at the level of students nested within schools. Wave × intervention interactions were estimated. Adjusted models controlled for sex, grade, household assets, and livestock holdings. Subgroup models were stratified by sex and province.

Results: The intervention significantly improved psychosocial outcomes. Life skills were significantly higher in intervention schools at both follow-up 1 and 2, and resilience improved significantly by follow-up 2. Self-efficacy showed marginal positive effects at both waves. Academic outcomes were largely unaffected, with the exception of a small but significant negative effect on math at follow-up 2. Perceived stigma and school-based discrimination declined overall, but the intervention did not produce significantly greater reductions than the control condition. Subgroup analyses revealed important heterogeneity by sex and province. Girls in intervention schools showed significantly greater improvements in life skills and self-efficacy at both follow-ups, with effects strengthening by follow-up 2. Boys showed no significant psychosocial gains. Perceived stigma scores diverged by sex, with girls showing reductions and boys slight increases. Effects were concentrated in Punjab; Sindh showed more limited gains.

Conclusions and Implications: Participatory, community-driven school interventions can produce meaningful improvements in children's psychosocial outcomes in rural Pakistan, even amid severe contextual disruptions. The stronger, more sustained effects for girls suggest the intervention may be particularly effective in reducing gender-based barriers to inclusion. The absence of academic gains requires further exploration and may point to the need for longer implementation timelines or greater implementation intensity to translate psychosocial improvements into measurable cognitive learning gains. Findings underscore the importance of context-sensitive, community-accountable approaches to achieving education equity in fragile settings.