Please take our brief survey

Blueprints Programs = Positive Youth Development

Return to Search Results

Promising Program Seal

Safe Dates

Blueprints Program Rating: Promising

A ten-session dating abuse prevention program to raise students' awareness of what constitutes healthy and abusive dating relationships, as well as the causes and consequences of dating abuse. It helps change adolescent norms about dating violence, equips students with skills and resources to develop healthy dating relationships, positive communication, anger management, and conflict resolution.

  • Vangie Foshee
  • University of North Carolina at Chapel Hill
  • Department of Health Behavior and Health Education
  • 317 Rosenau Hall, CB 7440
  • Chapel Hill, NC 27599-7440
  • (919) 966-6616
  • (919) 966-2921
  • foshee@email.unc.edu
  • www.sph.unc.edu
  • Sexual Violence
  • Violence
  • Violent Victimization

    Program Type

    • Community, Other Approaches
    • School - Individual Strategies

    Program Setting

    • Community (e.g., religious, recreation)
    • School

    Continuum of Intervention

    • Universal Prevention (Entire Population)
    • Selective Prevention (Elevated Risk)
    • Indicated Prevention (Early Symptoms of Problem)

    A ten-session dating abuse prevention program to raise students' awareness of what constitutes healthy and abusive dating relationships, as well as the causes and consequences of dating abuse. It helps change adolescent norms about dating violence, equips students with skills and resources to develop healthy dating relationships, positive communication, anger management, and conflict resolution.

      Population Demographics

      Safe Dates is implemented with middle school age youth (grades 8-9). It has been shown to be effective for both males and females, as well as with culturally and ethnically diverse audiences. It addresses both teens who abuse and teens who are abused.

      Age

      • Early Adolescence (12-14) - Middle School

      Gender

      • Male and Female

      Race/Ethnicity

      • All Race/Ethnicity

      Race/Ethnicity/Gender Details

      Safe dates had been proven equally effective for Caucasians and culturally diverse audiences.

      Risk: antisocial peer opportunities, antisocial peer involvement, negative gender stereotyping, abusive social norms.

      Protective: prosocial neighborhood opportunities, positive family involvement, good family management, positive school involvement, prosocial peer involvement, prosocial beliefs.

      • Individual
      • Peer
      • Family
      • Neighborhood/Community
      Risk Factors
      • Individual: Favorable attitudes towards antisocial behavior*
      • Peer: Interaction with antisocial peers
      Protective Factors
      • Individual: Prosocial involvement
      • Family: Attachment to parents
      • Neighborhood/Community: Opportunities for prosocial involvement

      *Risk/Protective Factor was significantly impacted by the program.

      See also: Safe Dates Logic Model (PDF)

      Safe Dates is a ten-session dating abuse prevention program for middle/high school students consisting of both school and community components. The school component has a curriculum that is implemented in schools by regular classroom teachers and targets primary prevention, while the community component targets secondary prevention by providing support groups and activities for youth as well as information for parents. The curriculum in the school component can also be presented by community resource people outside of the school setting. Each session is 45-50 minutes in length and includes the following topics: defining caring relationships, defining dating abuse, why people abuse, helping friends, overcoming gender stereotypes, equal power through communication, how we feel/how we deal, and preventing sexual assault. Booster sessions can also be offered after the initial administration of the curriculum.

      Safe Dates is a dating violence prevention program that provides male and female middle/high school students with the skills to prevent dating violence by changing dating violence norms and gender stereotyping, improving conflict-management skills, help-seeking behavior, and other cognitive factors associated with help-seeking.

      The Safe Dates program includes school (primary prevention) and community (secondary prevention) activities. School activities promote the primary prevention of dating violence perpetration by changing norms associated with partner violence, decreasing gender stereotyping, and improving conflict management skills. Community activities promote secondary prevention by changing those same variables and by also changing beliefs about the need for help, awareness of services for victims and perpetrators of partner violence, and help-seeking behavior. Community activities also enhance the availability of dating violence services from which adolescents can seek help. School activities include: a theater production performed by peers, a 10-session curriculum (45-50 minutes per session), and a poster contest. The 45-minute theater production, about how an adolescent victim of dating violence seeks help with her violent relationship, addresses many of the mediating variables related to help-seeking. The poster contest is described during day 10 of the curriculum, and interested students develop posters that address themes in the Safe Dates curriculum. The poster contest is designed to give adolescents another exposure to messages about dating violence. Posters are displayed in the classroom and judged by students to determine the top three posters in the school, each of which earns a cash prize.

      Booster sessions are administered three years post-intervention. The booster consists of an 11-page newsletter mailed to the adolescents' homes and a telephone call from a health educator approximately four weeks after the mailing. The newsletter contains worksheets based on the Safe Dates school curriculum for adolescents to complete. The health educators answer questions, provide additional information when needed, and assess completion of newsletter worksheets.

      Although the school component has primarily been implemented by regular classroom teachers as a part of required health education classes, the curriculum could also be delivered by community leaders or as a part of a youth-group activity, provided all of the sessions are completed and a high level of attendance is assured. Topics covered in the curriculum include: defining caring relationships, defining dating abuse, why people abuse, helping friends, overcoming gender stereotypes, equal power through communication, how we feel/how we deal, and preventing sexual assault. Community activities consist of special services for adolescents in abusive relationships (e.g., a crisis line, weekly support groups, materials for parents) and community service provider training.

      Primary prevention: Changes in norms coupled with improvements in prosocial skills serve as the theoretical base for primary prevention (school curriculum) activities. Norms are standards for acceptable behavior and thus have a significant effect on behavior and conformity. Weak conflict-management skills are associated with youth aggression in general and with partner violence specifically. Thus, school activities are expected to lead to the primary prevention of dating violence perpetration by (1) changing norms associated with partner violence, (2) decreasing gender stereotyping, and (3) improving conflict-management skills.

      Secondary prevention: Changes in norms, awareness of gender stereotyping, and conflict-management skills may also be important for adolescents in abusive relationships if they are to leave those relationships or stop being violent. In addition, secondary prevention activities encourage victims and perpetrators to seek help by addressing cognitive factors associated with help-seeking. Precaution adoption theory posits that decisions to seek help are influenced by (1) beliefs in the need for help and (2) beliefs in a given action to provide help.

      • Cognitive Behavioral
      • Skill Oriented
      • Social Learning

      Safe Dates was evaluated with 8th and 9th grade students in 14 public schools that were matched and each matched pair was then randomly assigned to treatment or control conditions. Evaluation has focused on the effects of the Safe Dates program on the primary and secondary prevention of dating violence. Primary prevention is achieved when the first perpetration of dating violence is precluded. Secondary prevention occurs when victims stop being victimized or perpetrators stop being violent. Data were collected from baseline (October 1994) through four years after program completion.

      At the one-month follow-up, Safe Dates prevented and reduced psychological abuse perpetration, reduced sexual dating violence perpetration, and positively changed cognitive mediating variables that were based on program content, such as dating violence norms, gender stereotyping, conflict management skills, and awareness of community services for dating violence.

      At the one-year follow-up, behavioral effects had faded but there were still effects on the cognitive mediating variables as well as on peer violence victimization and weapon-carrying. At the four-year follow-up, there were significant treatment and control group differences in the expected direction for physical, serious physical, and sexual dating violence perpetration and sexual victimization. Prior victimization moderated program effects on physical and serious physical victimization variables at almost all strata of prior victimization. The booster intervention did not improve the effectiveness of Safe Dates. When adolescents exposed to the booster were eliminated from the analysis, adolescents who only received Safe Dates reported perpetrating significantly less psychological, moderate physical, and sexual dating violence perpetration at all four follow-up Waves than those in the control group. In addition, those adolescents who reported no severe physical perpetration or average amounts of severe physical perpetration at baseline reported significantly less severe physical perpetration than control group adolescents at each of the four long-term follow-up waves.

      • Safe Dates is effective in preventing and reducing violence perpetration among teens already perpetrating dating violence.
      • Safe Dates resulted in less acceptance of dating violence, stronger communication/anger management skills, less gender stereotyping, and greater awareness of community services.

      Compared to the control group, participants in the treatment group schools showed the following improvements at one-month follow-up:

      • 25% less psychological perpetration;
      • 60% less sexual violence perpetration;
      • 60% less violence perpetrated against a current dating partner.

      At one-year follow-up (Foshee et al., 2014), participants in the treatment group showed the following improvements compared to participants in the control group:

      • 12% lower rates of reported peer violence victimization
      • 23% lower rates of reported peer violence perpetration among a subsample of minority students
      • 31% lower odds of carrying a weapon to school

      Follow-up results at one year showed a relapse in behavioral outcomes, but at four year follow-up, the following results were found among Safe Dates participants compared to the control group participants:

      • Between 56 - 92% less reported physical, serious physical, and sexual dating violence perpetration and victimization;
      • Adolescents who received Safe Dates reported perpetrating significantly less psychological, moderate physical, and sexual dating violence perpetration at all four follow-up waves;
      • Safe Dates-only adolescents who reported no severe physical perpetration or average amounts of severe physical perpetration at baseline reported significantly less severe physical perpetration than control group adolescents at each of the four follow-up waves.

      Assumptions cannot be made regarding the generalizability of this study outside of a rural setting. The generalizability of the findings to groups other than those represented in the sample is unknown. Due to the high rate of attrition over the course of the study (Baseline to Wave 6), resulting in very small subsample sizes by the four-year follow-up, caution should be used when attempting to generalize the findings to similar groups, particularly for the sustainability of effects.

      Behavioral effects of the Safe Dates program had diminished by the one-year follow-up. Significant behavioral effects were noted in the Safe Dates -only group (without booster) at the four-year follow-up, although the treatment group was very small (n = 124), and attrition was quite high (52%) among the four-year follow-up sample from baseline to the follow-up (Foshee, Bauman et al., 2004). In addition, the four-year follow-up data relied solely upon adolescents from the initial eighth grade cohort, further limiting the interpretation of the findings. The ninth grade cohort was not included in the four-year follow-up data collection due to graduation from high school. However, a more comprehensive analysis on long-term results was subsequently conducted (Foshee et al., 2005) that included the 9th grade cohort, limiting the analysis to three years to capture the students before they graduated, and used multiple imputation procedures to account for the missing data due to attrition.

      • Blueprints: Promising
      • Crime Solutions: Effective
      • OJJDP Model Programs: Effective
      • SAMHSA: 2.9-3.3

      Foshee, V. A., Bauman, K. E., Arriaga, X. B., Helms, R. W., Koch, G. G., & Linder, G. F. (1998). An evaluation of Safe Dates, an adolescent dating violence prevention program. American Journal of Public Health, 88, 45-50.

      Foshee, V. A., Bauman, K. E., Ennett, S., Linder, G. F., Benefield, T., & Suchindran, C. (2004). Assessing the long-term effects of Safe Dates and a booster in preventing and reducing dating violence victimization and perpetration. American Journal of Public Health, 94, 619-624.

      Foshee, V. A., Bauman, K. E., Ennett, S. T., Suchindran, C., Benefield, T., & Linder, G. F. (2005). Assessing the effects of dating violence prevention program "Safe Dates" using random coefficient regression modeling. Prevention Science, 6(3), 245-257.

      Foshee, V. A., Bauman, K. E., Greene, W. F., Koch, G. G., Linder, G. F., & MacDougall, J. E. (2000). The Safe Dates program: One-year follow-up results. American Journal of Public Health, 90, 1619-1622

      Foshee, V. A., Benefield, T. S., Ennett, S. T., Bauman, K. E ., & Suchindran, C. (2004). Longitudinal predictors of serious physical and sexual dating violence victimization during adolescence. Preventive Medicine, 39, 1007-1016.

      Foshee, V. A., Linder, G. F., Bauman, K. E., Langwick, S. A., Arriaga, X. B., Heath, J. L., . Bangdiwala, S. (1996). The Safe Dates project: Theoretical basis, evaluation design, and selected baseline findings. American Journal of Preventive Medicine, 12, 39-47.

      Foshee, V. A., Reyes, L. M., Agnew-brune, C., Simon, T. R., Vagi, K. J., Lee, R. D., & Suchindran, C. (2014). The effects of the evidence-based safe dates dating abuse prevention program on other youth violence outcomes. Prevention Science, 15(6), 907-916.

      Safe Dates
      15251 Pleasant Valley Road, P.O. Box 176
      Center City, MN 55012-0176
      Phone: (800) 328-9000 ext. 4324
      Email: kmcelfresh@hazelden.org
      Website: http://www.hazelden.org/web/public/safedatesproduct.page
      www.violencepreventionworks.org

      Study 1

      Foshee, V. A., Bauman, K. E., Arriaga, X. B., Helms, R. W., Koch, G. G., & Linder, G. F. (1998). An evaluation of Safe Dates, an adolescent dating violence prevention program. American Journal of Public Health, 88, 45-50.

      Foshee, V. A., Bauman, K. E., Ennett, S., Linder, G. F., Benefield, T., & Suchindran, C. (2004). Assessing the long-term effects of Safe Dates and a booster in preventing and reducing dating violence victimization and perpetration. American Journal of Public Health, 94, 619-624.

      Foshee, V. A., Bauman, K. E., Ennett, S. T., Suchindran, C., Benefield, T., & Linder, G. F. (2005). Assessing the effects of dating violence prevention program "Safe Dates" using random coefficient regression modeling. Prevention Science, 6(3), 245-257.

      Foshee, V. A., Bauman, K. E., Greene, W. F., Koch, G. G., Linder, G. F., & MacDougall, J. E. (2000). The Safe Dates program: One-year follow-up results. American Journal of Public Health, 90, 1619-1622

      Foshee, V. A., Benefield, T. S., Ennett, S. T., Bauman, K. E ., & Suchindran, C. (2004). Longitudinal predictors of serious physical and sexual dating violence victimization during adolescence. Preventive Medicine, 39, 1007-1016.

      Foshee, V. A., Linder, G. F., Bauman, K. E., Langwick, S. A., Arriaga, X. B., Heath, J. L., . Bangdiwala, S. (1996). The Safe Dates project: Theoretical basis, evaluation design, and selected baseline findings. American Journal of Preventive Medicine, 12, 39-47.

      Foshee, V. A., Reyes, L. M., Agnew-brune, C., Simon, T. R., Vagi, K. J., Lee, R. D., & Suchindran, C. (2014). The effects of the evidence-based safe dates dating abuse prevention program on other youth violence outcomes. Prevention Science, 15(6), 907-916.

      Evaluation Methodology

      Design: Program activities took place from November 1994 through March 1995. The booster took place between the two- and three-year follow ups. Fourteen public schools in Johnston County, North Carolina with students in the eighth or ninth grade were stratified by grade and matched by school size. One member of each matched pair was then randomly assigned to a treatment or control condition. Treatment adolescents were exposed to school and community activities (the Safe Dates program), whereas control adolescents were exposed to community activities only. Adolescents who had been in the treatment schools and who had parental permission completed the two-year follow-up data collection and then were randomly allocated to booster and non-booster conditions. As a part of the community activities, 20 workshops were offered to community service providers, including social service, mental health, crisis line, and health department staff, school counselors, sheriff's deputies, and officers from the nine police departments in the county. Students in the 7 treatment schools first saw the theater production, then received the 10-session curriculum two months later, and finally participated in the poster contest one month after the curriculum was implemented. Baseline data were collected in October 1994 from 1,886 (81%) of the 2,344 eligible students. Follow-up data were collected from the treatment and control adolescents at 1 month (Wave 2), 1 year (Wave 3), 2 years (Wave 4), 3 years (Wave 5), and 4 years (Wave 6) after completion of the Safe Dates program using the same procedures as in the baseline data collection. Students who were absent for school data collection, including those who had dropped out of school, were mailed a questionnaire and 50% of those questionnaires were completed and returned.

      Across all follow-up waves, between 4 and 9% of the students completed questionnaires by mail. A total of 48.2% (n = 460) of the adolescents from the eighth grade cohort who provided baseline data also provided data at the two-year and four-year follow-up, which included measures of booster effects (control = 201; Safe Dates = 124; Safe Dates plus the booster = 135). Additional analyses were limited to adolescents who completed a baseline questionnaire and who were in either the original control or the treatment group that received Safe Dates but not the booster. The booster-group adolescents were eliminated from subsequent analyses because the focus was on evaluating the effects of Safe Dates, rather than the booster, over time.

      There were no significant differences on the baseline measures of dating abuse victimization and perpetration between adolescents who received Safe Dates only and those who received Safe Dates and the booster. Attrition from baseline to Wave 3 was only 12%, but by Wave 5 was approximately 50%. There were no statistically significant treatment group differences in the amount of attrition at each follow-up wave.

      Sample: Adolescents were eligible for the study if they were enrolled in the eighth or ninth grade in Johnston County, North Carolina on September 10, 1994 (n = 2,344 eligible students). Of these eligible adolescents, 1,700 completed the one-month post-test questionnaire. Fifty-one percent of the initial post-test sample were female, 19.1% were African American, and 77.1% were White. Ages ranged from 11 to 17 years, with a mean of 13.8. At both baseline and the first follow-up, about 70% of the adolescents reported dating. Of the adolescents who were dating at baseline, 34.9% of the girls and 38% of the boys reported being a victim of dating violence at least once. At baseline, dating girls were significantly more likely than boys to report perpetration of nonsexual dating violence, but dating boys were more likely than dating girls to report sexual dating violence perpetration. Caucasian adolescents reported less victimization of nonsexual dating violence than either African-American adolescents or adolescents in the "other" racial group category, and reported significantly less nonsexual violence perpetration than African-American adolescents. Adolescents in the "other" racial group category reported more sexual violence perpetration than Caucasian adolescents.

      The 1,700 students were divided into three subsamples based on dating violence experience. The primary prevention subsample included dating adolescents who reported at baseline that they had never been a victim or a perpetrator of dating violence (n = 862), the victim secondary prevention subsample included dating adolescents who reported being a victim of dating violence at baseline (n = 438), and the perpetrator secondary prevention subsample included dating adolescents who reported they had been a perpetrator of dating violence at baseline (n = 247). Most of the adolescents in this study reporting experience with dating violence reported being both a victim and a perpetrator, thus many of the same adolescents were in both the victim and perpetrator subsamples. There were no significant differences between treatment and control schools at the individual or school level. All of the study samples and subsamples were equivalent at baseline.

      Measures: Data were collected from students during 50-minute in-school sessions through self-administered questionnaires or via the mailed questionnaires used in subsequent Waves of data collection. Four victimization and four perpetration variables were measured using a Likert scale: psychological abuse victimization, nonsexual violence victimization, sexual violence victimization, and violence in the current relationship. Moderate physical, severe physical, and sexual dating violence perpetration were measured by responses (0 = never; 3 = 10 or more times) to 18 listed acts of direct perpetration. From these responses, composite measures for moderate physical, severe physical, and sexual dating violence perpetration were created. The five mediating variables were also assessed at each Wave. In addition, four variables measuring dating violence norms were created: acceptance of prescribed norms, acceptance of proscribed norms, perceived positive consequences of dating violence, and perceived negative consequences of dating violence. The same Likert scale format was used to measure all four constructs. Gender stereotyping and beliefs regarding need for help were also measured using the above format. Students were also asked if they were aware of any services for victims and perpetrators of dating violence, and whether they had asked for help from others as either a victim or perpetrator. Finally, four conflict management variables were measured with Likert scale responses: constructive and destructive communication skills, and constructive and destructive responses to anger.

      Foshee et al. (2014) examined three additional measures of violence: peer violence perpetration, peer violence victimization and weapon carrying to school. Although data were gathered and baseline and then at 1-month, and 1, 2, 3, and 4 years post-intervention, only 1-year follow-up results were presented. Peer violence perpetration (baseline alpha = .80; follow-up alpha=0.86) was measured using questions such as “How many times have you ever done the following things to someone of the same sex and the same age as you: ‘beat them up,’ ‘hit them with my fist,’ ‘threatened them with a weapon,’ and ‘used a weapon on them.’” Response options ranged from never (0) to 10 or more times (3) and were summed to find a total score. Similar questions were used to measure peer violence victimization (baseline alpha = .75; follow-up alpha=0.81). Weapon-carrying was measured by asking adolescents if they had ever brought a gun or any other kind of weapon to school (0=no; 1=yes).

      Analysis: Logistic regression was used to evaluate the multivariate relationship of study variables to dropout status through odds ratios. Three variables, White racial identity, gender stereotyping, and nonsexual violence victimization were significantly associated with dropout status. White students were slightly more likely to drop out of the study than other students. The odds of dropping out increased with gender stereotyping and nonsexual violence. Accordingly, all three variables were controlled in multivariate analyses.

      In the full sample and each subsample, the treatment and control groups were compared at baseline and follow-up on demographic, mediating, and outcome variables using the school as the unit of analysis while taking the matching design into consideration. Matching allowed consideration of each matched pair as a primary sampling unit. School wide means for each outcome of interest were compared using the nonparametric Wilcoxon signed rank test for differences from matched pairs. When identifying the variables that mediated program effects, analyses were performed at the individual level.

      Sample sizes were too small to conduct analyses on subsamples of "pure perpetrators" and "pure victims."

      Linear mixed regression models were used to assess program and booster effects (where included) and moderators of effects.

      For the analysis of the Safe Dates non-booster and control groups, data were used from Waves 1 through 5. Wave 6 data were not used because the ninth grade cohort had graduated from high school and therefore no data were collected from them. Within-subjects multiple imputation procedures were used to decrease the likelihood that the findings would be biased by missing data.

      Long-term analysis by Foshee et al. (2005) to determine primary (prevention) and secondary (intervention) program effects over time up to 3 years post-intervention used random coefficient regression modeling. Multiple imputation procedures were used to deal with missing data values.

      To examine the three additional measures of violence, Foshee et al. (2014) used multivariate logistic regression models (for weapon-carrying) and negative binomial regression models (for peer violence victimization and perpetration). Models adjusted standard errors for nesting within schools due to the clustered design. Only adolescents without missing data were used in the analysis, resulting in an analyzed sample size of 1620 for peer victimization and perpetration. Further, adolescents who had reported carrying a weapon to school at baseline were also dropped from the analysis for the weapon-carrying outcome, resulting in a sample size of 1397.

      Outcomes

      Post test: At follow-up, there were no significant differences in the victimization of psychological abuse, nonsexual violence, sexual violence, or violence in the current relationship between the treatment and control groups in any of the samples.

      Full sample: At the one-month follow-up, adolescents in the treatment condition reported significantly less psychological abuse perpetration and significantly less perpetration of violence against a current dating partner than those in the control condition. Adolescents in the treatment group, as compared with those in the control group, were less supportive of prescribed dating violence norms, were more supportive of proscribed dating violence norms, perceived fewer positive consequences from using dating violence, used more constructive communication skills and responses to anger, were less likely to engage in gender stereotyping, and were more aware of victim and perpetrator services.

      Mediation: After gender, variables associated with attrition, and baseline values of the dependent variables had been controlled, placement in the treatment condition was significantly associated with changes in psychological abuse perpetration, sexual violence perpetration, and violence perpetrated in the current relationship. The association between treatment condition and psychological abuse perpetration was mediated by changes in prescribed norms, gender stereotyping, and awareness of victim services. The association between treatment condition and sexual violence perpetration was mediated by changes in prescribed norms, gender stereotyping, awareness of victim services and awareness of perpetrator services. The relationship between treatment condition and violence perpetrated in the current relationship was mediated only by changes in prescribed norms.

      Primary prevention subsample: At the one-month follow-up, adolescents in the treatment condition, as compared to those in the control condition, reported initiating significantly less psychological abuse perpetration. Treatment adolescents were more supportive than controls of proscribed dating violence norms, perceived more negative consequences from using dating violence, and engaged in less gender stereotyping.

      Mediation: Controlled analyses indicated that placement in the treatment condition was significantly associated with initiation of psychological abuse. The relationship remained statistically significant when proposed mediators were controlled, indicating that this program effect occurred through mechanisms other than those proposed.

      Victim subsample: Adolescents in the treatment group, relative to control group adolescents, were less accepting of prescribed dating violence norms, less accepting of traditional gender stereotypes, and more aware of victim services.

      Perpetrator subsample: At the one-month follow-up, there were trends toward significance in the predicted direction in differences in psychological abuse perpetration and sexual violence perpetration. Adolescents in the treatment group, as compared to those in the control group, perceived more negative consequences for using dating violence and were more aware of services for perpetrators.

      Mediation: Placement in the treatment condition was associated with psychological abuse perpetration and with sexual violence perpetration. The association between treatment condition and psychological abuse perpetration was mediated by changes in awareness of perpetrator services, while the association between treatment condition and sexual violence perpetration was mediated by changes in perceived negative consequences for using dating violence and awareness of perpetrator services.

      Long term: At the one-year follow-up (8th/9th grades), no significant differences were found between the treatment and control groups in any of the behavioral outcomes in any of the samples. Some of the mediating variables were maintained and are described below:

      Full sample: Adolescents in the treatment group were less accepting of dating violence, perceived more negative consequences from engaging in dating violence, reported using less destructive responses to anger, and were more aware of victim and perpetrator services than were adolescents in the control group.

      Primary prevention subsample: No significant differences were found in any of the mediating variables between treatment and control groups at follow-up.

      Victim subsample: Adolescents in the treatment group were less accepting of dating violence, perceived more negative consequences from engaging in dating violence, reported using less destructive responses to anger, reported less gender stereotyping, and were more aware of victim services than were adolescents in the control group.

      Perpetrator subsample: Adolescents in the treatment group reported using less destructive responses to anger, and were more aware of perpetrator services than were adolescents in the control group.

      Additional measures of violence (Foshee et al., 2014): Adolescents in the treatment group were less likely to report peer violence victimization (p=.04) and less likely to carry a weapon to school (p=.005) at follow-up compared to adolescents in the control condition. While there was not a significant effect of the program on peer violence perpetration at follow-up for the full sample, participation in the program was significantly negatively associated with peer violence perpetration for minority adolescents (p=.001) but not for white adolescents (p=.48).

      Four-year follow-up at 12th grade (Foshee et al., 2004): These analyses were restricted to the eighth grade cohort at four years post-intervention. There were significantly more females in the four-year follow-up sample than in the sampling frame. Analysis of differential attrition revealed no significant differences between treatment condition and dropout status. Two baseline variables were associated with dropout status in both treatment and control groups -- gender and serious physical violence victimization -- although these interactions were only marginally significant. Males were more likely than females to drop out of the study, and the odds of dropping out decreased with increased serious physical violence victimization.

      Safe Dates effects on perpetration: Adolescents who received only Safe Dates (and not the booster) reported perpetrating significantly less physical, serious physical, and sexual dating violence perpetration at the four-year follow-up than did those in the control group. Safe Dates effects on psychological abuse perpetration were moderated by prior involvement in dating violence.

      Booster effects on perpetration: There were no significant effects of the booster on physical, serious physical, or sexual dating violence when comparing the Booster to the Safe Dates only group. Prior involvement in psychological abuse perpetration, however, moderated the effect of the booster on psychological abuse perpetration. There were no significant differences between the booster and control group in follow-up physical, serious physical, or sexual dating violence perpetration.

      Safe Dates victim subsample: Adolescents who received only Safe Dates reported significantly less sexual victimization from dates at follow-up than those in the control group. There were no effects on psychological victimization. Safe Dates had effects on physical and serious physical victimization, but the effects were moderated by prior involvement with the behavior. The Safe Dates group reported less physical abuse victimization in all three strata at follow up than the control group. The pattern was similar for serious victimization: in all three strata of prior serious physical victimization, adolescents exposed to Safe Dates reported significantly less victimization from serious dating violence than did adolescents in the control group. Safe Dates had both primary and secondary prevention effects on all six of the outcomes and the program was equally successful by gender and race.

      Booster victim subsample: The effects of the booster on physical, serious physical, and sexual victimization were all moderated by prior victimization. Adolescents exposed to the booster reported more serious physical and sexual victimization in all three strata at follow-up than those exposed only to Safe Dates. These differences were statistically significant only when prior involvement in dating violence was high. There were no effects of the booster on psychological abuse victimization. There were no significant differences between the booster and control group at follow-up for psychological abuse victimization. Adolescents exposed to the booster reported significantly less victimization than controls in serious victimization and serious sexual victimization when there was no prior serious victimization.

      Non-booster subsample: When the booster-group was eliminated from all Waves of analysis, the findings indicated that adolescents who only received Safe Dates reported perpetrating significantly less psychological, moderate physical, and sexual dating violence perpetration at all four follow-up Waves than those in the control group. The lack of any significant treatment condition by baseline outcome interactions indicate that treatment effects were the same for those who did and did not report using those forms of violence against a dating partner prior to intervention exposure. Those adolescents who reported no severe physical perpetration or average amounts of severe physical perpetration at baseline reported significantly less severe physical perpetration than control group adolescents at each of the four follow-up Waves.

      Significant treatment outcomes for mediating variables were also found for dating violence norms, gender-role norms, and beliefs in need for help. The Safe Dates Program was successful in decreasing acceptance among adolescents in prescribed dating violence norms and traditional gender-role norms, and increasing the belief in the need for help at all four follow-up periods, compared to control group youth. The Safe Dates Program was also successful at increasing awareness of community services, with gains most significant at Wave 2. Dating violence norms was found to have mediating effects on all program measures.

      These outcomes were mostly confirmed in an additional analysis that included both 8th and 9th grade students at three years post-intervention (Foshee et al., 2005), using random coefficients models with multiple imputation of missing data. In summary, Safe Dates students, relative to controls, reported less psychological, moderate physical, and sexual dating violence perpetration and less moderate physical dating violence victimization at all four follow-up periods. There was a marginal program effect on sexual victimization and no effects on psychological or severe physical victimization at any of the four follow-ups.

      Further research (Foshee, Benefield et al., 2004) points to several variables that were found to predict future victimization. Male victims of serious physical dating violence were more likely to have been hit by an adult with the intention of harm, have low self-esteem, and have been in a physical fight with a peer. Along with these variables, having a friend who has been a victim of dating violence, alcohol use, and being white, predicted chronic victimization for males. For females, the onset of serious physical dating violence victimization was predicted by having been hit by an adult. That variable plus coming from a single-parent household predicted chronic victimization from serious physical violence. Onset of sexual victimization for females was predicted by having a friend who has been the victim of dating violence and being depressed. Those variables plus gender stereotyping predicted chronic victimization from sexual dating violence.