Dallan F Flake and Dr. Renata Forste, Sociology

Violence against women is one of Latin America’s most pressing social problems. Over the past twenty years, it has come to be recognized not only as a significant public health concern, but as a human rights issue as well. Latin America is notorious for its high rates of political and social violence. Much less understood is the violence that occurs behind closed doors—between husbands and wives, parents and children. With so much attention centered on Latin America’s corruption, crime, and political instability, it is easily overlooked that the family is perhaps this region’s most violent social institution.

Despite impressive advances in domestic violence research, our understanding is hampered by two methodological constraints. First, the lack of cross-cultural research impedes our knowledge of domestic violence outside the traditional regions of study in the United States and Europe. Second, much of the research has been dominated by single-factor theories that fail to reflect the full complexity of real life.

My research seeks to fill the gaps in domestic violence literature by examining multiple factors associated with domestic violence in a non-Western setting. The analysis is based on data from four countries: Bolivia, Colombia, Nicaragua, and Peru. Two types of datasets, the Family Interaction and Children’s Well-being Survey (FICW) and the Demographic and Health Survey (DHS), are employed in the analysis. The FICW was conducted by BYU in Bolivia and Colombia, in 2000 and 1998, respectively. Mothers in rural and poor urban communities were surveyed on key family issues. DHS datasets from Bolivia, Colombia, Nicaragua, and Peru are included in the research because of they are larger, nationally representative samples. A total of 29,951 women are included in the statistical analyses.

Six independent variables are included in the model, each of which has been empirically linked to wife abuse in past research. The variables are categorized as either personal or relationship factors, in order to understand how different ecological levels influence the likelihood of abuse individually, simultaneously, and interactively. The personal variables included in the model are partner alcohol consumption, woman’s education level, and woman’s age at marriage. The relationship variables are marital status, decision-making authority, and household size.

The dependent variable in the model is a dichotomous measure (yes/no) of whether the woman has ever been physically hit by her partner. Logistic regression techniques are employed in determining the likelihood that a certain variable increases the risk of abuse. Three different models were run in order to understand the differences in how personal and relationship factors influence domestic violence. The first model measures the effect of personal factors; the second measures relationship factors. The third model combines personal and relationship variables in order to understand how the additive effect influences domestic violence.

Basic descriptive statistics reveal that domestic violence affects a significant percentage of Latina women. In Bolivia, 26.3 percent of women reported physical abuse; in Colombia, between 17.8 percent (FICW) and 19.4 percent (DHS) have been hit. More than 25 percent of Nicaraguan women reported abuse, as did 40.5 percent of Peruvian women.

Of the personal variables included in the model, partner alcohol consumption is most strongly correlated with abuse. Women whose partners drink are between 1.7 and 2.4 times more likely to experience abuse and women whose partners do not drink. Age at marriage and woman’s education level are also negatively correlated with abuse. One of the most intriguing findings from this research is that education level most strongly reduces the likelihood of abuse when the woman has attended college. For example, the risk of abuse for Nicaraguan women decreases 16 percent for women with primary education, 20 percent for women with secondary education, and 40 percent for women with post-secondary education.

Relationship factors are more strongly correlated with domestic violence than personal variables. Marital status, decision-making authority, and household size all play significant roles in the likelihood of abuse. Women who marry are between 27 and 48 percent less likely to be abused than women who cohabit. As cohabitation is an extremely common type of union in Latin America, this finding is particularly important. Decision-making also influences the risk of violence. The likelihood of abuse is much lower in relationships where the couple makes decisions together than in those in which either the man or woman makes all the decisions. Finally, for each child a woman has, the risk of abuse increases between 15 and 30 percent. This might best be explained by the added economic stress that comes with children.

These findings provide empirical support for the hypothesized models of male violence towards women. This project helps to fill the gaps in domestic violence literature by applying a multidimensional model to a non-Western sample. From the analysis emerges a clearer profile of the abused Latina. She enters into the relationship young—most likely before her 18th birthday. She may have had some schooling but didn’t go to college. She cohabits rather than marries, and her partner most likely drinks alcohol. The decisions made by the couple tend to be non-egalitarian. In addition, she is likely to have several children.

This study offers four important contributions to domestic violence research. First it begins to bridge the aforementioned gap in abuse literature by examining the causes of abuse in a crosscultural context. Next, by conceptualizing domestic violence as an interplay among personal, situational, and sociocultural factors, multiple causation is recognized. Thirdly, this research brings us one step closer to unraveling the complex maze of domestic violence by going beyond the simplicity of single-factor theories. Finally, and most importantly, it provides information that women in Latin America can use to help them recognize, prevent, and ultimately protect themselves from abusive relationships.