Author ORCID Identifier

Date of Award


Degree Type


Degree Name

Doctor of Philosophy (PhD)



First Advisor

Gabriel Kuperminc

Second Advisor

Christopher Henrich

Third Advisor

Phyllis Holditch Niolon

Fourth Advisor

Cynthia Stappenbeck

Fifth Advisor

Daniel Whitaker


Although bullying and teen dating violence (TDV) have similar behavioral manifestations and are likely to affect the same individuals throughout development in similar ways, interventions typically address bullying and TDV separately. This study used longitudinal data from middle school participants (N = 1,504) to test aspects of the Dating Matters (DM) program’s theory of change, which aims to prevent bullying and TDV by teaching youth healthy relationship skills (HRS) and conflict resolution strategies. First, latent class analysis was used to identify classes of co-occurring bullying, dating, and TDV perpetration at baseline and assess the association of those classes with HRS and negative conflict resolution strategies (NCRS). A three-class model best fit the data: TDV & Bullying, TDV only, and Low Perpetration. No significant association was found between latent class of baseline perpetration and HRS; however, those in the TDV & Bullying class reported using NCRS significantly more than the other classes. Next, a latent growth curve model was used to assess HRS and NCRS development throughout the intervention, comparing the latent classes described above. Due to poor fit of the model for HRS, no conclusions can be drawn about the trajectory of HRS. The model for NCRS showed no significant linear or quadratic change in frequency of use throughout DM for the overall sample. However, there were significant differences in the NCRS trajectories among the latent classes of baseline perpetration, such that those in the TDV & Bullying class had consistently more frequent use of NCRS over time. Ultimately, two cross-lagged panel models found that NCRS and HRS were correlated with bullying and TDV cross-sectionally and longitudinally, suggesting that HRS can be a protective factor and NCRS a risk factor for future perpetration. Given the lack of significant effects on HRS, this study provided no evidence that HRS development drives DM’s treatment effects on bullying and TDV. This study’s findings suggest that NCRS is a determinant of later bullying and TDV. NCRS is an important target of youth violence prevention; however a higher intensity of intervention may be necessary for youth who perpetrate multiple forms of violence.


File Upload Confirmation


Available for download on Tuesday, August 01, 2023