Silent bruises: Essays on determinants of intimate partner violence in India, and its consequences

Abstract

This dissertation investigates the determinants and consequences of intimate partner violence (IPV) in India. The research questions are: 1) what is the relationship between empowerment and IPV at the individual level, 2) what are the determinants of IPV at the community level, 3) does witnessing parental violence as a child determine the probability of the respondents’ victimisation as an adult, and 4) whether maternal exposure to IPV determines children’s anthropometric outcomes. The study uses data from the Demographic Health Survey (DHS). The first question is analysed using the third and fourth rounds of DHS data (2005-06 and 2015-16). The second question is analysed using spatial models at the district level; such data is available only for the fourth round of DHS. After completing the analysis for (1) and (2), the fifth round of DHS (2019-21) was publicly released and was used to examine the last two research questions.

The analysis of the relationship between IPV and empowerment, using a control function approach, found the relationship between empowerment and IPV to vary over the period of study; moreover, the relationship depended upon the form of the violence studied. Analysis using spatial regression models reveals a higher incidence of IPV in India’s eastern states than in their western counterparts. Districts characterized by low empowerment levels, high child-sex ratio, road connectivity, and a high proportion of socially marginalised groups are associated with higher IPV levels. Spatial spillovers are also observed in the incidence of IPV.

The study reveals that IPV has adverse consequences on the victim’s children. Non-parametric kernel-based estimates reveal that childhood exposure to spousal physical violence increases the respondent’s likelihood of experiencing all forms of violence as an adult; this effect varies with education, empowerment levels, and household asset holding. Further, a mother’s exposure to IPV can also lead to adverse nutritional outcomes (stunting, underweight, or a hyper-critical group comprising of children who are both stunted and underweight) in the victim’s children — as indicated by results of a Two Stage Least Square models and validated by a control function approach.

The study concludes by arguing in favour of empowering women, particularly at the community level, to reduce the incidence of IPV.