When wars are unbalanced, so are classrooms: a global meta-analysis of conflict and schooling

Asymmetric conflicts have uneven impacts on educational outcomes. This study provides a systematic review and meta-analysis of existing quantitative research on the relationship between asymmetric conflict and education. Rather than generating new primary data or directly testing a causal relationship, it synthesizes effect sizes from previously published studies across diverse conflict-affected settings to assess the magnitude, consistency, and sources of variation in educational impacts. Using a multilevel random-effects meta-analytic framework, the analysis shows that exposure to asymmetric conflict significantly reduces years of schooling (−0.19 SD, p < .001) and, to a lesser extent, school enrolment (−0.06 SD, p = .024), with gender and migration status amplifying these effects. The findings reveal substantial heterogeneity, suggesting conflict typology, governance quality, and institutional resilience shape educational outcomes. Ethno-nationalist conflicts appear particularly damaging, while some protracted conflicts show evidence of adaptation and institutional coping. Supported by hierarchical modelling, pre-registration, and sensitivity analyses, this study demonstrates that asymmetric conflict produces patterned rather than uniform educational disruption. These findings highlight the need for conflict-sensitive and context-specific education policies, particularly for displaced and vulnerable populations, while underscoring the importance of adaptive, data-informed strategies for educational recovery and long-term resilience.