Methods: This study analyzed data from a sample of 377 women who were residing in a domestic violence shelter. Data were collected from April 2016 to August 2018. To be eligible for the study, women must have experienced physical abuse (or threats), forced or coerced sexual abuse, or made to feel unsafe by an intimate or ex-intimate partner within the past 6 months. Survey questions that asked about actions women took drew from previous studies (Goodkind et al., 2004; Goodman, Dutton, Weinfurt, & Cook, 2003). Seven binary indicator variables captured women’s actions, including (1) Previously stayed in a shelter, (2) Sought housing assistance, (3) Visited healthcare provider, (4) Called the police, (5) Sought legal services, (6) Sought help for their abusive partner, and (7) Engaged informal networks for help. To investigate patterns in women’s actions, data were analyzed using latent class analysis (LCA) in STATA 15.
Results: Descriptive analyses revealed that this sample of women employed multiple strategies in response to IPV. Specifically, more than half of participants had engaged informal networks for help (85%, n=320), sought help for their abusive partners (63%, n=238), stayed in a shelter before (51%, n=194), or sought legal services (e.g., legal aid; 51%, n=191). Nearly half of participants had sought housing assistance to move to a safe location (48%, n=181). Fewer women, however, had called the police (42%, n=160) or visited a healthcare provider (40%, n=149). Results of LCA indicated three classes (AIC=3068, BIC=3458) for patterns in women’s actions. They could be labeled as (1) Broad engagement of formal and informal networks (51%, n=195), (2) Engagement of informal networks (12%, n=46), and (3) Avoiding medical and criminal justice systems (37%, n=139). The analysis also revealed that women’s race/ethnicity and number of children were associated with patterns in women’s actions. Specifically, African American women were more likely to engage informal networks (Class 2) than engage formal/informal networks more broadly (Class 1), as compared with White women (RRR=3.66, p<.05). An increase in the number of children increased the odds of avoiding medical and criminal justice systems (Class 3) in comparison to broad engagement of formal/informal networks (Class 1) (RRR=1.48, p<.05).
Conclusions: Study results suggested that there are patterns in women’s actions in response to IPV and that women tend to engage in multiple actions and strategies. Associated demographic factors point to persistent disparities in seeking formal services among women of color, as well as the salient role of children in determining strategies women employ. Implications for future research and practice will be discussed.