Latent Transitions in Sexual Violence Perpetration in a Nationally Representative Sample of Individuals Age 16-22

Introduction: Compelling research suggests that sexual violence (SV) emerges in adolescence yet little is known about how this behavior persists and desists over time. To begin examining this critical question, we use national, longitudinal data from the Growing up with Media study.

 

Methods: Data are from Waves 4 – 6, collected between 2010 and 2012, when extensive SV questions were added to the study. A total of 887 youth between the ages of 16-22 were included. Respondents were asked about their past year engagement in six types of SV perpetration: sexual harassment, sexual harassment expressed exclusively online, attempted rape, rape, coercive sex, and sexual assault. Latent Class Analysis was used to empirically determine the number of latent profiles. Transitions were then examined across two age groups: 16-17 year olds and 18-22 year olds. Given the non-nested nature of models with different numbers of classes, recommended information criteria (e.g., the Bayesian Information Criterion) and likelihood-based ratio tests (e.g., the Vuong-Lo-Mendel-Rubin likelihood ratio test), in addition to considerations of interpretability and parsimony, were used to identify the optimal number of classes.

 

Results: Most of the 887 participants (85.2% for 16-17 year olds and 79.6% for 18-22 year olds) did not engage in SV perpetration across the three waves assessed. Two risk profiles were identified, however: 13.1% of the 16-17 year olds and 16.8% of the 18-22 year olds engaged in acts characterized mostly by sexual harassment. 1.7% of the 16-17 year olds and 3.7% of the 18-22 year olds were classified in a third profile, characterized by high probabilities of committing all types of SV perpetrations.

Interestingly, the probability for online-exclusive sexual harassment was higher for 18-22 year olds compared to 16-17 year olds assigned to the sexual harassment class. Among those in the third profile, 18-22 years compared to 16-17 year olds tended to have a higher probability for coercive sex and a lower probability of other types of SV perpetration. Over time, most youth tended to deescalate or remain in the no-perpetration profile (84.1%). Furthermore, among those in the third profile however, 51% transitioned to the second profile and the remaining 49% transitioned into the first, low-risk profile.

 

Conclusions: The vast majority of youth – more than four in five – never perpetrate SV during adolescence. Among those who do, most engage in sexual harassment. The presentation will expand on these results by talking more about those who transition across profiles and their associated characteristics.