Adolescent Dating Violence: An Exploratory Factor Analysis of Perpetration and Victimization Scales Developed for Youth Empowerment Solutions for Healthy Relationships (YesHR)

Background and Purpose: The prevalence of teen dating violence (TDV) is a significant concern due to its profound impact on individuals' behavior, well-being, and future trajectories. Assessing both perpetration and victimization is crucial for understanding TDV dynamics. Few validated scales are utilized to measure these aspects among adolescents. This study draws upon items from four validated scales to develop an assessment tool to capture TDV. The main aim of the current study is to assess the factor structure of the items that were administered in the survey instrument to assess if the compiled scale provides a comprehensive and cohesive measure of sexual violence experiences among youth.

Methods: Data for the study was drawn from a longitudinal study of a universal, primary prevention project to reduce sexual violence and harassment among youth, the Youth Empowerment Solutions for Healthy Relationships (YES-HR). The present study included data from the first cohort of participants collected at the follow-up assessment. The research team developed the scales by compiling items from four different validated scales to measure both perpetration and victimization scales. Items were included based on the literature and the team’s collective experience in this field to comprehensively measure experiences with violence. Both the perpetration and victimization scales had 25 items each. A total of N=1381 tenth-grade youth were recruited to participate in this study. An Exploratory Factor Analysis (EFA) was conducted utilizing Varimax rotation to identify factors and structure of these 25 items of the scale, followed by calculating Cronbach’s Alpha to identify the internal reliability of the factors.

Results: The results showed that the assumptions of sufficient sample size, normality, and linearity for the EFA were met, and Bartlett’s test of Sphericity was also calculated, which was satisfied and was significant for both Perpetration (KMO = .84, p<.001) and Victimization ( KMO = .84, p<0.001) which indicated good sampling adequacy for both scales. Five main factors were identified for the Perpetration scale (1^st subscale α= .93, 2^nd subscale α= .84, 3^rd subscale α= .84, 4^th subscale α= .87, 5^th subscale α= .54), and five factors were identified for the Victimization scale (1^st subscale α= .88, 2^nd subscale α= .86, 3^rd subscale α= .68, 4^th subscale α= .83 5^th subscale α= .77)., which explained 30.4% and 30.5% of the variance, respectively. Cronbach’s Alpha was calculated for both scales and identified good reliability for Perpetration (α =.86) and Victimization (α=.87).

Conclusions and Implications: The present study developed and validated scales for assessing the perpetration and victimization of TDV. Findings suggest that both scales have good internal reliability, highlighting their potential use in future projects to assess TDV. The factors loading on the scales were similar for perpetration and victimization, with factors forming sub-scales that were conceptually coherent. In the next part of the study, a confirmatory factor analysis will be conducted to strengthen the scales' validation in assessing victimization and perpetration among adolescents.