Introduction: Violence against women is more observed in families with some unique structures and conditions. This study aimed at exploring the impact of social capital and its components on violence against women with reference to the nature of family relationships.
 
Materials & Methods: This descriptive-correlative study was conducted on all married females with the age range of 15-50 years in Ilam. According to the information from the statistical center of Iran, there were 40846 females. Out of these women, 400 cases were selected based on Cochran's Formula, Strauss method, and multi-stage clustering technique. The data were collected using a questionnaire of violence against women developed by Strauss et al. (1996) and a researcher-made questionnaire of social capital and its components. The data were analyzed in SPSS software (Version 22) through the Pearson correlation coefficient and multivariable regression. Ethics code: IR.IAU.SRB.REC.1395.127
 
Findings: According to the results, women experienced mental, emotional, and economic violence more than other forms of violence against women. The findings also indicated that violence against women correlated significantly with women's age, the couple education, family income, the number of children, the spouses' socio-economic status, and the women's power in the family structure (P=0.001). The results of regression analysis revealed that 62% of the changes in violence against women can be explained by social capital and its components, including interpersonal trust, social support, social participation, family cooperation, social relations and interactions, social leniency, social consolidation, social solidarity ,life satisfaction, and legal awareness, P=0.001).
 
Discussion & Conclusions: Women enjoyment of social capital and its components, including social supports, social participation, social ties and relations, social trust, social awareness, and life satisfaction leads to the improvement of married life and decrease in the phenomenon of domestic violence.