Women's economic independence and physical intimate partner violence (IPV) during separation

Abstract

A growing number of studies have addressed the issue of intimate partner violence (IPV). While some studies have shown that women's economic independence reduces the risk of physical IPV, the empirical evidence is still inconclusive. In particular, little is known about the factors that are associated with the likelihood of experiencing physical IPV during the immediate separation process. We use data from the German Family Panel (pairfam) for the years 2009-2022. The analytical sample consists of women who went through a separation since the last interview (n = 786 person-years). We use logistic regression models to examine the determinants of physical IPV during separation. We find that mothers with minor children have significantly higher odds of experiencing physical IPV during separation than women who do not have minor children (OR=1.96, p=.006), controlling for key sociodemographic characteristics. Low education as well as non-employment are associated with higher risks of physical IPV during separation for both women with and without minor children. However, non-employment is a particular risk factor for mothers with lower levels of education. Divorce and separation, especially when minor children are involved, can be a highly stressful time when partner conflict escalates and, in some cases, culminates in violence against the partner. This paper highlights the importance of maternal labor market integration and economic independence as a lever to potentially reduce physical IPV and conflict during separation.

Fig 1. Results from the logistic regression model with physical IPV during separation as the dependent variable.

Odds ratios and 95%-CI. Note. pairfam, waves 2-14, unweighted own estimates. Clustered standard errors. Low education includes “less than post-secondary education” and high education includes “post-secondary education or higher.” Employment is measured in the year before separation.

Fig 2. Average predicted probabilities and 95%-CI from the logistic regression model with physical IPV during separation as the dependent variable.

a. Women without minor children (Model 3a). b. Women with minor children (Model 3b). Note. pairfam, waves 2-14, unweighted own estimates. Clustered standard errors. Low education includes “less than post-secondary education” and high education includes “post-secondary education or higher.” Employment is measured in the year before separation.