Abstract
The Family Violence Option (FVO) was designed to help survivors of domestic violence (DV) more easily secure income support in the United States (U.S.), without placing them at risk of further abuse. The objective of this study is to determine whether the decision‐making of advocates responsible for determining waiver recommendations under the FVO is influenced by the relationship status of DV survivors. Recursive partitioning was used to analyse data from a sample of 237 survivor risk assessments from four New Jersey counties to determine which women receive waiver recommendations and which do not. Advocates completed risk assessments for the women and were instructed to make recommendations on waivers based on their assessment. Workers’ decision‐making was examined using classification and regression trees (CART) to determine what case factors made it more or less likely for survivors to be recommended waivers. The CART results were supplemented with logistic regression analyses to ensure validity. For two of three waivers, survivors who reported currently residing with their abuser or who had ended the relationship recently were less likely to receive waiver recommendations than those who had been out of the relationship for a longer period of time (OR = 0.09–0.21), even when accounting for the type and severity of DV and the impacts of the violence on survivors’ mental health. The results indicate that DV advocates’ decision‐making is complicated by factors independent of survivors’ case characteristics. This can affect the safety and well‐being of women attempting to leave violent relationships by affecting their access to resources.