Constructing comparable intimate partner violence indicators across the DHS, MICS, and PMA health surveys.

Abstract

We construct comparable indicators that measure the prevalence of recent intimate partner violence (IPV) using publicly available, integrated microdata within the IPUMS data collections across many countries. The objective of this work is to increase opportunities for comparative research by leveraging vast quantities of harmonized data. We use consistent and comparable variables that measure domestic violence in international health surveys. The most consistent indicators of domestic violence measure physical, psychological, and sexual IPV in the last 12 months. We imposed a consistent reference period and restricted to a comparable subpopulation where these differed across surveys. Aggregating IPV variables across surveys, without careful attention to question wording and subpopulations, may produce inconsistent measures leading to bias, over- or under-estimation of IPV prevalence, or spurious trends and associations. Using comparable indicators in microdata and studying the level, distribution, and covariates of IPV in multiple settings over time, we can better understand these phenomena and identify effective policy interventions.