Patriarchal Ideology and Violence Against Women: A Theoretical Contribution Using Longitudinal, Individual-Level Analyses

Feminist researchers have recently highlighted the need to revive patriarchy as a theoretical tool in regards to violence against women. Patriarchy is typically considered to be a structural concept, but a theory of patriarchy for violence against women must also include an individual-level component of patriarchal ideology. Patriarchal ideology has not been clearly conceptualized and is rarely operationalized. Very little research has assessed patriarchal ideology as a dependent variable and almost none has done this longitudinally. This research aims to fills these gaps. The current study also seeks to identify significant predictors of change in patriarchal ideology, an issue of tremendous importance for a theory of violence against women. Stronger theories that can appropriately incorporate patriarchy may lead to more effective proactive policies rather than the existing reactive policies based on poor theoretical understanding. The data used for this dissertation comes from the Longitudinal Study of Violence against Women: Victimization and Perpetration among College Students in a State-Supported University in the United States, 1990-1995 (White, Smith, and Humphrey 2001). Since the data was from a sample of college men, the first wave of data is just before individuals transitioned into college. This was followed by three subsequent waves of data assessing patriarchal ideology after the first, second, and third years of college. Because of the various challenges posed by longitudinal data and specific challenges posed by this data, two major analyses were performed. The main goals of these analyses were to: come up with reliable operational measures of patriarchal ideology, determine their measurement invariance over time, assess predictors of patriarchal ideology, evaluate the change/stability in patriarchal ideology, and account for the predictors of change/stability. The first major analysis operationalized individual patriarchal ideology using attitudinal measures over two waves of data in a traditional test/re-test panel design. The second major analysis operationalized patriarchal ideology using vignettes that were proxy measures of patriarchal ideology over three waves of data. Latent growth-curve modeling was used in order to assess the intra-individual and inter-individual changes in patriarchal ideology over time. The analyses were the first of their kind to assess patriarchal ideology as an outcome variable overtime. Overall, findings suggest patriarchal ideology is a multidimensional concept that can be measured using attitudinal measures and vignettes, with some support suggesting vignettes may be preferred. During the transition from precollege to the end of one's freshman year, patriarchal ideology was stable. After this time period, over the next three years of college patriarchal ideology declined for the sample as a whole. Interestingly, there was inter-individual change but no infra-individual change in patriarchal ideology. The means that regardless of one's initial levels of patriarchal ideology, everyone's levels declined over these three years. These results are discussed further regarding their theoretical, methodological, and policy implications. Suggestions for future research assessing the role patriarchal ideology plays in theories of violence against women are discussed.