The association of personal, interactional, and environmental factors with leaving or staying in a violent relationship among women who stayed at a shelter

Violence against women is an entrenched global problem, with between 13% and 61% of the female populations of all nation states reporting that they have experienced some form of violence.

Violence towards one’s spouse, as a manifestation of relationship domination, often takes the form of mental, emotional, economic and sexual abuse. The profound consequences of these forms of violence, on the woman victim and her children, prompts the question of why some women return to a cycle of domestic violence, or remain in an abusive relationship; researchers have tried to conceptualize the factors that may explain this phenomenon. Other than this, relatively few research studies have been conducted in this field.

Informed by the relative paucity of the literature, the current research sought to test the role played by personal factors (i.e., emotional distress, post-traumatic distress, and distress), interactional factors (severity of violence in the relationship, sense of danger, previous breakups from spouse, previous stay at a shelter or refuge), and environmental factors (social and family support) in the decision of women staying in shelters for victims of domestic violence, whether to leave or remain in the violent relationship.

This research is quantitative and correlative in nature, based on an existing database. The data were collected as part of a comprehensive research project conducted by three researchers from Bar- Ilan University: Dr. Anat Ben Porat, Prof. Rachel Dekel and Prof. Haya Itzhaki. The research was conducted in 12 out of 14 shelters for female victims of domestic violence across Israel between 2009 and 2014, and with a total of 221 reserch participants. The research tool deployed for the research study was a self-reported questionnaire, with data collected over three time periods. The research used Pearson tests (for sequential variables) and Spirman tests (for dichotomic variables) to test the correlation between the independent research variables and the decision to either leave or stay with the spouse in the wake of domestic violence. In addition, in connection with the dependent dichotomic variable “logistics regression” analysis was employed to analyze the overall contribution of all of the research variables to the decision to leave or stay in a relationship.

The research findings indicated that the study hypothesis was partially confirmed. With regard to the environmental factors, a positive association was found between family support and ending the violent relationship; however, no connection was found between social support and leaving. A correlation was found between the background, education and employment variables and leaving the spouse. Moreover, the employment factor was found to be the most influential in predicting the likelihood of the research participant leaving the spouse.

With regards to the study questions, no correlation was found between the “emotional distress” variable (post-traumatic distress and depression) and ending the relationship. In addition, the correlation between the interactional variables and the dependent variable was determined to be only partially significant. With regard to the severity of the violent experiences, and the sense of danger, no correlation was found with regard to predicting the dependent variable; additionally, the previous breakup factor, identified as significant during a preliminary study, did not present as a contributory factor to the dependent factor in the regression model. Nevertheless, the previous “staying at a shelter” variable did contribute to the likelihood of “not leaving” the spouse. 

In conclusion, the current research findings identify employment, education and family support as variables that contribute to predicting the outcome of the dependent variable—the female victim of domestic violence leaving her spouse.

This study makes a major contribution to the process of mapping the factors related to the decision of female domestic violence victims to either leave or stay with a violent spouse. In addition, the findings of this study will assist in developing therapeutic programs for women who take refuge in domestic violence shelters.