Determinantes da dinâmica de risco de violência por parceiro íntimo utilizando análise categórica de sequências

Abstract

Intimate Partner Violence (IPV) is a serious social problem that has a complex temporal and mul- tifactorial dynamic. This study seeks to analyze the sequences of occurrence of IPV victimization

in the years 2016 and 2017 among women in Northeast Brazil, according to the categorization into four possible sequences: (0-0 for no violence, 0-1 for violence only in 2017, 1-0 for violence only in 2016, and 1-1 for violence in both years). The data used is the longitudinal panel from the Survey of Socioeconomic Status and Domestic and Family Violence against Women (PCSVDFWoman), with participants between 15 and 49 years of age. We sought to analyze the effect of the variables age, race, education, number of children, length of relationship and marital status on the chances of the woman belonging to one of the sequence categories through the multinomial logit model. The main results show that younger women and with a greater number of children have less chances of belonging to the 0-0 category and more chances of belonging to the 1-1 category.