Using Partial Identification Methods to Estimate the Effect of Violence Against Women on Their Children’s Health Outcomes

Non-experimental studies evaluate the robustness of their treatment effects by exploring the sensitivity of their estimates to the inclusion of additional variables. However, recent papers have shown that such heuristic approaches are insufficient. Instead, partial identification methods have been proposed to bound non-experimental estimates. I use proportional selection relationships to estimate the relative size of the unobservables needed to eliminate the estimated effects. I apply this method to test the effects that violence against women has on the health outcomes of their children, a research area that lacks credible identification strategies. Also, to expand the external validity of my analysis, I use data from five standardized nationally representative household surveys in Latin America. Consistent with previous studies, OLS estimates show large negative associations between violence against women and an array of child health outcomes. However, when accounting for omitted variable bias, at best, two-thirds of the estimates remain robust and they are concentrated on the outcomes with the largest cross-sectional estimated impacts.