Date of Award

Spring 2020

Document Type


Degree Name

Doctor of Philosophy (PhD)




Health Psychology

Committee Director

Michelle L. Kelley

Committee Member

James F. Paulson

Committee Member

Randy Gainey


Intimate partner violence (IPV) is a serious, pervasive problem affecting over 30% of young adults. Although early research focused on men as perpetrators, it is commonly found that women are just as likely, if not more likely, to perpetrate violence in relationships. Some studies have categorized violent couple dyads into unidirectional (perpetration only) or bidirectional violence (reporting both perpetration and victimization). The current study identified predictors of the two types of violent profiles, specifically among female perpetrators. Included in the regression model were early family variables as dictated by the literature (e.g., child abuse, parental violence, early aggressive behavior, and early dating violence) as well as individual-level proximal variables (e.g., alcohol problems, anger, hostility, emotion dysregulation, and psychological aggression) thought to differentiate unidirectional and bidirectional violence.

Participants (M age = 20.49 years; SD = 2.54) were recruited from a large, southeastern university. Study criteria were: must be female, between the ages of 18 and 29, and report being in a current or recent (within the previous 12 months), heterosexual, and in a dating relationship lasting at least 3 months. After reading a notification statement and agreeing to participate, participants completed an anonymous online survey.

Data were analyzed using multiple logistic regression models. Logistic regression models examined which variables predicted the likelihood of dating violence perpetration compared to no perpetration. Results indicated that witnessing mother-to-father interparental violence, experiencing dating violence in adolescence, emotion dysregulation, and psychological abuse each significantly and uniquely predicted the likelihood of dating violence perpetration (compared to no violence). A second logistic regression analysis distinguished the two dating violence perpetrators (unidirectional and bidirectional) to determine if predictors of interest differentially increased the likelihood of unidirectional violence compared to bidirectional violence. Results of the analysis indicated that the variables hostility and alcohol problems each significantly and uniquely predicted the likelihood of bidirectional violence, such that an increase in each of these variables increased the odds of bidirectional violence compared to unidirectional violence. The aim of this research was to better understand the variables that predict perpetration in a sample of college women and furthermore, to identify variables that differentiate women who report only perpetration in their relationship compared with women who report both perpetration and victimization.


In Copyright. URI: This Item is protected by copyright and/or related rights. You are free to use this Item in any way that is permitted by the copyright and related rights legislation that applies to your use. For other uses you need to obtain permission from the rights-holder(s).