A big-push community intervention reduced rates of child marriage by 80%
Summary
This study evaluates Pathways to Choice, a locally tailored “big-push” community programme tested in northern Nigeria using a paired cluster-randomised trial across 18 communities. Two years after the programme started, rates of marriage among adolescent girls fell from 86% in control communities to 21% in treatment communities — an ~80% reduction. The intervention boosted girls’ re-enrolment in school, but the authors find education alone does not fully account for the decline in child marriage. The paper argues the whole-community design reduced social backlash and that bundling multiple interventions produced effects greater than single interventions alone. Data and replication files are available via the Harvard Dataverse.
Key Points
- Pathways to Choice was evaluated with a rigorous paired cluster-randomised trial in 18 communities in northern Nigeria.
- Remaining unmarried two years after programme start: 86% (control) vs 21% (treatment) — roughly an 80% decrease in child marriage rates.
- The programme substantially increased girls’ re-enrolment in school, but schooling does not fully explain the marriage reduction.
- A whole-community, bundled approach appears to lower the risk of social backlash and shift entrenched social norms.
- Sample and analysis details (including Stata replication files) are publicly available on the Harvard Dataverse.
Content summary
The paper documents a multifaceted intervention — Pathways to Choice — combining safe-space activities, community engagement and supports aimed at keeping girls in school and altering norms around early marriage. Implemented by the Centre for Girls Education with support from partners and funders, it was tested using a paired cluster-randomised design. Two-year follow-up data show a very large decline in child marriage in treated communities. The authors conduct heterogeneity and mediation analyses: increased schooling explains some but not all effects, and community-level changes in attitudes and social dynamics appear important. The authors situate their results against the broader literature on education, cash transfers and norm-change, and provide extensive supplementary and extended-data material.
Context and relevance
Child marriage remains widespread worldwide and especially common in parts of northern Nigeria; delaying marriage is linked to better education, health and economic outcomes. Prior interventions typically produce small percentage-point changes; this trial’s large effect suggests that bundled, community-focused programmes can produce much bigger shifts in strongly normative behaviours. Results matter to policymakers, NGOs and funders designing programmes to reduce child marriage and improve girls’ life chances.
Why should I read this?
Short version: this paper shows a properly designed, community-wide package can actually move the needle — a lot. If you work on girls’ education, child protection or community development, it’s worth skimming the methods and the theory-of-change to see what was bundled and how they avoided backlash. Saves you time: someone else ran the hard trial and the outcome is strikingly positive.
Source
Article date: 11 March 2026
Authors: Isabelle Cohen, Maryam Abubakar & Daniel Perlman