|Cross-national and multilevel correlates of partner violence: an analysis of data from population-based surveys
|Lori L Heise, and Andreas Kotsadam
|Lancet, 3(6): e332–40; DOI: http://dx.doi.org/10.1016/S2214-109X(15)00013-3
More than one region
On average, intimate partner violence affects nearly one in three women worldwide within their lifetime. But the distribution of partner violence is highly uneven, with a prevalence of less than 4% in the past 12 months in many high-income countries compared with at least 40% in some low-income settings. Little is known about the factors that drive the geographical distribution of partner violence or how macro-level factors might combine with individual-level factors to affect individual women's risk of intimate partner violence. We aimed to assess the role that women's status and other gender-related factors might have in defining levels of partner violence among settings.
We compiled data for the 12 month prevalence of partner violence from 66 surveys (88 survey years) from 44 countries, representing 481?205 women between Jan 1, 2000, and Apr 17, 2013. Only surveys with comparable questions and state-of-the-art methods to ensure safety and encourage violence disclosure were used. With linear and quantile regression, we examined associations between macro-level measures of socioeconomic development, women's status, gender inequality, and gender-related norms and the prevalence of current partner violence at a population level. Multilevel modelling and tests for interaction were used to explore whether and how macro-level factors affect individual-level risk. The outcome for this analysis was the population prevalence of current partner violence, defined as the percentage of ever-partnered women (excluding widows without a current partner), aged from 15 years to 49 years who were victims of at least one act of physical or sexual violence within the past 12 months.
Gender-related factors at the national and subnational level help to predict the population prevalence of physical and sexual partner violence within the past 12 months. Especially predictive of the geographical distribution of partner violence are norms related to male authority over female behaviour (0·102, p<0·0001), norms justifying wife beating (0·263, p<0·0001), and the extent to which law and practice disadvantage women compared with men in access to land, property, and other productive resources (0·271, p<0·0001). The strong negative association between current partner violence and gross domestic product (GDP) per person (–0·055, p=0·0009) becomes non-significant in the presence of norm-related measures (–0·015, p=0·472), suggesting that GDP per person is a marker for social transformations that accompany economic growth and is unlikely to be causally related to levels of partner violence. We document several cross-level effects, including that a girl's education is more strongly associated with reduced risk of partner violence in countries where wife abuse is normative than where it is not. Likewise, partner violence is less prevalent in countries with a high proportion of women in the formal work force, but working for cash increases a woman's risk in countries where few women work.
Our findings suggest that policy makers could reduce violence by eliminating gender bias in ownership rights and addressing norms that justify wife beating and male control of female behaviour. Prevention planners should place greater emphasis on policy reforms at the macro-level and take cross-level effects into account when designing interventions.
What Works to Prevent Violence Against Women and Girls—a research and innovation project funded by UK Aid.