Factors affecting intimate partner violence against women in Sri Lanka

Abstract

Intimate Partner Violence (IPV) against women has become a major health and social drawback in Sri Lanka that causes serious consequences for women’s mental and physical well-being as well as their reproductive health. The Sri Lanka Demographic and Health Survey (SLDHS) 2016 reported that a total of 17 percent of ever-married women age 15-49 in the country suffered domestic violence from their intim te partner. This study aimed to develop the Violence Index which measures the level of IPV experienced by a woman in Sri Lanka and to identify the factors that affect IPV against women in the country. A total of 2494 ever-married women age 15-49 who were found to be victims of IPV were selected from the database of SLDHS 2016 which was carried out by the Department of Census and Statistics in Sri Lanka. Characteristics of women, husbands, and their relationships that may influence IPV were considered as the explanatory variables while the estimated Violence Index was taken as the outcome variable. Multiple Correspondence Analysis and the theory of Composite Indicator were applied to develop the Violence Index. The association between the Violence Index and each explanatory variable was examined using the Kruskal-Wallis H test and the Mann-Whitney U test. Gamma regression analysis was carried out to determine the significant factors that affect IPV against women. Results show that the estimated Violence Index represented the lowest level of IPV experienced by a woman as 0.03 and the highest level of IPV experienced by a woman as 10.53. The values of the Violence Index showed a high right-skewed distribution. Religion, woman’s education level, woman’s married time, woman’s participation in making the household decisions, husband’s employment status, the age gap between husband and wife, enough money for daily expenses in the house, and household alcohol consumption were the most influential factors for IPV against women in Sri Lanka. The Gamma regression model provided a better prediction on the Violence Index with the mean squared error of 0.9447 and the mean absolute error of 0.5306. This study recommends raising public awareness on the risk and protective factors of IPV and the development of the current policies and new strategies to prevent IPV against women in Sri Lanka.