Program Profile: Ending Violence Curriculum

Study Title: Impact of a School-Based Dating Violence Prevention Program Among Latino Teens: Randomized Controlled Effectiveness Trial (which is associated with Outcome 1, Outcome 2, Outcome 3, and Outcome 4)

Research Design

Jaycox and colleagues (2006) conducted a randomized controlled trial of the impact of the Ending Violence curriculum on dating violence outcomes with 9th-grade students from the Los Angeles Unified School District. The evaluation took place over a 3-year period (2001–2004). This study recruited from tracks within large urban high schools with at least 80 percent Latino students.With one exception, all schools operated on a year-round calendar in which students were assigned to one of three schedules or “tracks.” Tracks were randomized within schools to intervention conditions to increase the power to detect changes (relative to randomization of schools), while controlling for potential contamination (relative to randomization of classrooms). Six schools participated in the first cohort, five in the second cohort, and six in the third cohort (including five schools that had participated in the first cohort, and one that had participated in the second). The schools that participated in more than 1 year of the intervention were only randomized once. 

Sample

A total of 2,540 9th-grade students from 110 classes in 10 high schools in the Los Angeles Unified School District were included in the sample. Health classes were randomly selected within tracks to participate in either the intervention group (1,384 students), who received the Ending Violence curriculum, or the control group (1,156 students), who received the standard health curriculum. 

The majority of the total study sample was of Latino descent (92 percent), 75 percent were U.S. born, roughly half were female (52 percent), and the average age was 14.4 years. At baseline, there were no statistically significant differences between the treatment and control groups, except that the control group reported less acceptance of female-on-male aggression and a higher likelihood of seeking help related to violence.

Data Collection

Both the intervention group and the control group were followed for six months. For intervention classes, pretest surveys were given on Monday, the Ending Violence intervention was administered Tuesday through Thursday, and the posttest surveys were administered on Friday. Control groups followed the same schedule, with the traditional health class curriculum being taught in place of the Ending Violence intervention. Follow-up surveys were administered in class six months after the curriculum ended. Although parents were notified of the intervention, consent was not required for students to participate in the health class because violence prevention is a part of the school district’s curriculum. However, parental consent and student assent were required for participation in the survey. 

Outcome Measures

The outcomes examined were knowledge of dating violence and the laws, the approval of unprovoked male-on-female aggression, the approval of unprovoked female-on-male aggression, dating violence perpetration, and dating violence victimization. All measures were self-reported and were standardized as z-scores, having a mean of zero and a standard deviation of one at baseline. The authors developed separate scales to assess knowledge of dating violence and the law, and the approval of unprovoked male-on-female aggression and unprovoked female-on-male aggression. The Revised Conflict Tactics Scale assessed victimization and perpetration in relationships among those students who dated. Because dating violence is reportable as child abuse in California, the authors gathered these data anonymously. The outcome of acceptance of person-on-person aggression was a combined measure of approval of unprovoked male-on-female aggression and approval of unprovoked female-on-male aggression.

Statistical Analysis

The effects of the program on outcomes were tested by examining the differences between the adjusted means for students in the treatment and control groups. Because the anonymous dating violence data lacked identifiers for linking students’ outcomes across waves or to classrooms, the authors tested these data for an intervention effect at the follow-up

period using analysis of covariance, controlling for the preintervention-track mean and adjusting for clustering of students within track-years. No subgroup analysis was conducted.

Citation:

Jaycox, Lisa H., Daniel McCaffrey, Beth Eiseman, Jessica Aronoff, Gene Shelley, Rebecca Collins, and Grant Marshall. 2006. “Impact of a School-Based Dating Violence Prevention Program among Latino Teens: Randomized Controlled Effectiveness Trial.” Journal of Adolescent Health 36:694–704.

This study is a high-quality randomized controlled trial.