Impact of egocentric networks on physical, emotional, and sexual violence among a sample of college students
Meg Patterson, Tyler Prochnow, Jordan Nelon, Carly McCord, Whitney GarneyIntroduction: Intimate partner violence (IPV) is an increasing concern on college campuses. IPV is defined as any behavior within an intimate relationship that results in physical, sexual, or psychological harm. While much research has identified individual-level risk factors related to IPV among college students (e.g., hooking up, alcohol use, gender), less has investigated interpersonal relationships and IPV. The purpose of this study is to determine the association between college students having a history of physical, emotional, and sexual abuse and egocentric network characteristics, including composition of networks that have also experienced IPV, while controlling for individual-level factors cited in the literature (e.g., risky drinking, grade classification, gender). In doing so, we will measure the unique variance in IPV explained by egocentric network variables. Methods: 697 students answered questions about demographics (age, gender, grade classification, race and/or ethnicity, and whether they were in a fraternity or sorority), history of IPV, alcohol use, hooking up, and egocentric networks via an online survey. Each respondent was asked to indicate the initials of up to “five people they felt closest to in their life,” and indicated their relationship to each alter, the gender of each alter, the race and/or ethnicity of each alter, and if to their knowledge alters had experienced physical, emotional/verbal, or sexual abuse. Finally, egos were asked to indicate if their alters knew one another. Hierarchical logistic regression analyses were conducted using individual-level variables and network variables to predict a history of physical, emotional, and sexual violence among this sample. Results: Egocentric network variables added 8.8% - 11.4% of explained variance in predicting the odds of sexual (Naglekerke R2= .303, p<.001), physical (Nagelkerke R2=.280 , p<.001), and emotional violence (Nagelkerke R2=.203, p<.001). Being connected to people who have a history of IPV significantly increased a student’s odds of indicating a history of physical (OR=1.034, p<.001), emotional (OR=1.023, p<.001) , and sexual violence victimization (OR=2.879, p<.010) themselves. Having more disconnected egonetworks was related to sexual violence victimization (OR=1.305, p=.017), while being connected to more people of the same gender was related to emotional violence victimization (OR=1.988, p=.043). Gender and hooking up were related to all types of IPV. Discussion: Egocentric networks were important in explaining history of IPV in this sample. Notably, people with a history of violence tend to populate one another’s personal networks. These findings add to the current literature that largely focuses on individual-level risk factors related to IPV. The way college students’ close networks are composed and structured help in understanding IPV in this population, and should be considered in prevention and reactionary efforts on college campuses.