50,51Generalizability might also be limited becausethe participants who remained in the cohort(e.g., English-speaking only, lived with bothparents, reported making grades of A and B,less likely to have had a boyfriend or girlfriend)might be at lower risk for engaging in datingviolence than students who did not remainin the cohort. Fourth, although we used multi-level modeling to adjust the regression esti-mates and standard errors for ICC amongstudents within the same school, a largernumber of schools would be preferable andwould produce more accurate estimates ofthe variance components in the model. How-ever, even small higher level units (schools),such as those in this study, were shown toproduce unbiased estimates of thefixed ef-fects.52Fourth, only 1 item was used to assessphysical dating violence; however, this itemwas similar to the 1 item used to assess thisconstruct in the Centers for Disease Controland Prevention Youth Risk Behavior Survey.1Lastly, self-reported data were used; how-ever, using the audio---computer-assisted self-interviews approach helped to increase studentaccuracy in responses and their perceptionsof confidentiality.53ConclusionsThis study’sfindings indicated thatIYGwasan effective program for preventing datingviolence among ethnic-minority middle schoolyouths. Its current use among many students asan effective adolescent pregnancy preventionprogram could increase the likelihood thateven more positive effects could be seen fordating violence prevention. Additional study,however, is needed to determine ifIYGshouldbe widely disseminated in dating violenceprevention efforts.jAbout the AuthorsAll authors are with the Center for Health Promotion andPrevention Research, The University of Texas HealthScience Center at Houston (UTHealth) School of PublicHealth.Correspondence should be sent to Melissa F. Peskin, PhD,Center for Health Promotion and Prevention Research,UTHealth School of Public Health, 7000 Fannin St., Suite2658, Houston, TX 77030 (e-mail: [email protected]uth.tmc.edu). Reprints can be ordered at by clicking the“Reprints”link.This article was accepted January 18, 2014.ContributorsM. F. Peskin directed the study and conceptualized andwrote the article. C. M. Markham conceptualized anddesigned the study. R. Shegog was involved in theconceptualization of the intervention. E. R. Baumlerconducted all data analyses. R. C. Addy assisted withdata analyses. S. R. Tortolero obtained the funding andconceptualized and designed the study. All co-authorsoffered critical revisions to the article and approved thefinal version.AcknowledgmentsThis study was funded by the National Institutes ofMental Health (NIMH) (R01 MH66640-01).We thank Lionel Santibáñez for his editorial assistance.Human Participant ProtectionThis study was approved by the UTHealth institutionalreview board and the school district’s Office of Researchand Accountability.