Gender-based violence (GBV) against women remains a major public health concern; yet to date there has been little rigorous research evaluating the impact of interventions aimed to reduce and prevent GBV. This project will bring new evidence on violence-reduction policies by carefully evaluating the impact of an innovative and theory-driven policy intervention in Peru that aims to reduce GBV through a community-based approach that trains local leaders to become community health volunteers and to work within their communities on GBV prevention, monitoring, and reduction. In close cooperation with the Government of Peru, this project's research design will contribute to the empirical literature on the efficacy of GBV interventions by addressing pressing methodological concerns, provide feedback for programming of public institutions, and supply evidence on the theory underlying GBV.
External Link(s)
Citation
Field, Erica. 2018. "Training Local Leaders to Prevent and Reduce Domestic Violence in their Communities." AEA RCT Registry. March 29. https://doi.org/10.1257/rct.2629-1.0.
We will measure the casual impact of the intervention on IPV incidence, mental and physical health.
Primary Outcomes (explanation)
Secondary Outcomes (end points)
We will measure three main variables thought to operate as mechanisms: norms and attitudes, women's agency and empowerment, and take-up of GBV services.
Secondary Outcomes (explanation)
Experimental Design
We will conduct a cluster randomized controlled trial, with randomization at village level. Villages will be randomly assigned to either the treatment or control group.
Experimental Design Details
Randomization Method
Randomization done in office by a computer
Randomization Unit
Village
Was the treatment clustered?
Yes
Sample size: planned number of clusters
288 villages
Sample size: planned number of observations
1728 village members
Sample size (or number of clusters) by treatment arms
144 villages will receive treatment and 144 villages will be taken as control.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)