Program

Second Step: A Violence Prevention Curriculum

May 13, 2015

OVERVIEW

Second Step is a set of violence prevention curricula for elementary and middle school-aged children, consisting of 28 to 30 specific lessons that teach social skills related to anger management, impulse control, and empathy. In an RCT evaluating the elementary school Second Step program, reductions in levels of observed aggressive behavior during social problem-solving tasks, and increases in neutral and prosocial behavior were found two weeks after the program’s end. However, no impacts were found for teacher or parent reports. Some of these impacts remained six months after the completion of the program. In an RCT evaluation of the middle school version of the program, a positive impact on physical aggression was found at the end of sixth grade. At the end of seventh grade, no full-sample program impacts were found, although state-specific analyses revealed evidence of effectiveness in Illinois middle schools. The program was also found to reduce bullying perpetration among students with disabilities.

DESCRIPTION OF PROGRAM

Target population: Elementary school-aged children in grades 1-3, and middle school-aged children in grades 6-7

Second Step is used to prevent aggressive behavior by increasing prosocial behavior. The elementary school version of the program consists of 30 lessons that last for about 35 minutes each.  During each lesson, students are shown a photograph accompanied by a scenario that forms the basis for the activities of the lesson.  Lessons are arranged in 3 units that include empathy training, impulse control, and anger management.  Role-playing is a large component of the program.  During role-play, the teacher models the skill, students practice it, feedback is given, and reinforcement is used when a skill is demonstrated appropriately.  The middle school version consists of 28 lessons covering similar topics and applying the same activities, along with information on substance abuse prevention. Lessons can be delivered in a 50-minute session once weekly or in two 25-minute sessions each week. The pre-K and middle school program kits include training materials, while the elementary school programs include individual, computer-based staff training.

Costs for the materials are:

  • $379 for pre-K kit
  • $379 each for the kindergarten and first grade kits
  • $329 each for  the second and third grade kits
  • $359 each for the fourth and fifth grade kits
  • $369 each for the sixth through eighth grade kits

EVALUATION OF PROGRAM

Grossman, D.C., Neckerman, H.J., & Koepsell, T.D, et al. (1997).  The effectiveness of a violence prevention curriculum among children in elementary school: A randomized controlled trial.  JAMA, 277(20).  1605-1611.

Evaluated population: 790 2nd and 3rd grade students were evaluated.  Of the 790 students, 418 were in intervention schools and 372 were in control schools.  Of the total sample, 54 percent were male. In addition, 588 of the 790 subjects (12 subjects from each participating classroom) were randomly selected to be observed in the classroom, on the playground, and in the cafeteria.  The average age was 8 years old and the sample included 79 percent white, 7 percent African American, 8 percent Asian, 4 percent Hispanic, and 1 percent Native American children.

Approach: Data were collected before the start of the program, two weeks after the program, and six months after the program.  The impact of the curriculum was measured by teacher ratings, parent ratings, and direct observations by trained observers (The Achenbach Child Behavior Checklist and Teacher Report Form, the School Social Behavior Scale, and the Parent-Child Rating Scale).

Results: The results of the parent and teacher surveys found little difference between the intervention and control schools at both evaluations, but the observations did show curriculum impacts.  Several impacts were observed at the 2-week follow-up.  In the classroom, differences between the intervention and control groups for physical negative and overall negative behaviors were borderline significant.  Rates of negative behavior, measured as episodes witnessed per hour of observation, decreased in the intervention group (0.51 to 0.48 episodes per observation-hour) but increased in the control group (0.36 to 0.71 episodes per observation-hour).  In the playground and cafeteria settings, there were significant differences between the two groups.  As in the classroom, rates of physical negative behavior decreased in the intervention group (2.20 to 1.56 episodes per observation-hour) but increased in the control group (1.82 to 2.56 episodes per observation-hour).  The trends for verbal negative behavior were similar but not significant.  Another significant finding was that the number of observed neutral/prosocial behaviors in the playground and cafeteria increased by 17.1 more observations per hour in the intervention group than in the control group.

At the 6-month follow-up, there were no significant differences in negative behavior between the control group and intervention group.  Both groups experienced a decline in negative behavior.  However, with regard to physical aggression in the classroom setting, the rate was significantly lower in the intervention schools than in the control schools (intervention=0.33 episodes per observation-hour and control=0.57 episodes per observation-hour).   There were no differences for neutral/prosocial behavior rates.

Evaluators note that the impact of this intervention on behavior outside of the school environment is not known.  Additionally, it is noted that program impacts may be greater if the curriculum is implemented with a whole-school approach over several years.

Frey, K.S., Nolen, S.B., Edstrom, L.V.S. & Hirschstein, M.K. (2005).  Effects of a school-based socio-emotional competence program: Linking children’s goals, attributions, and behavior.  Journal of Applied Developmental Psychology, 26(2).  171-200.

Evaluated population:  15 western Washington elementary schools, with 1,253 children between the ages of seven and eleven were evaluated.  Located in both urban and suburban districts, school populations ranged from 52% to 89% European-American. Asian-Americans and African-American comprised 18% and 12% of the sample, respectively.  The sample was approximately even by sex (48.2% female).  620 students were in the intervention group, and 615 students were in the control group.

Approach: Eleven schools were recruited and randomly placed in either an intervention or a control group in the pre-study year.  Four more schools were recruited and randomly assigned during the first year of data collection.

Second- and fourth-grade teachers sent letters to students’ parents obtaining permission for study participation.  Sixty-three percent of the students (1,253) received permission to participate in the study; however, the program was given to all the students in treatment classrooms, whether or not they were participating in the study.  Teachers participated in a two-day training session prior to implementation in order to familiarize them with program content.  During implementation, program consultants met with teachers twice monthly to discuss issues.

Teachers taught one or two 25-40 minute lessons per week from October to March over two years.  The three units in the program include empathy training, impulse control and problem solving, and anger management.  Empathy training focuses on teaching children to recognize contextual cues through emotional understanding, prediction, and communication.  Impulse control and problem solving emphasizes safety, fairness, efficacy, and mutually rewarding interaction.  Anger management uses techniques such as self-talk and attention control.  All behavior skills practiced are meant to increase children’s social problem-solving abilities.

Teacher-rated student behavior was assessed each October, and aggression and attribution surveys were administered in October of the first year.  Teacher ratings and student surveys were administered each April.  In May and June of year two, pairs of children participated in tasks requiring negotiation of prizes and a prisoner’s dilemma game, which were intended to create conflicts between self-interest and pro-social goals.  Data in this analysis come from the second cohort and the control group, at the same time.

The estimated program cost, including class materials, staff training, and training materials, is less than $9 per student.

Results: Second Step participation was associated with positive student behavior, goals (effect size = 0.17), and social reasoning.  The intervention group had less aggression and need for adult intervention during prize division when compared with the control group.  Differences between the treatment and control groups in teacher-reported behavior were strong during the first year of intervention (effect size = 0.2), but not during the second (effect size = 0.1).  Higher-level negotiation strategies increased for girls in the treatment group (effect size = 0.17).  There were no differences between the groups in regards to the survey measures, attributions, and behavioral intentions. The intervention group was significantly less likely to behave aggressively during negotiations than the control group (effect size = 0.14); however, the groups showed no difference in demanding behavior.  Compared with those in the control group, students in the intervention group were significantly more satisfied with outcomes of the prisoner’s dilemma game and prize division (effect size = 0.14).

Students rated as highly antisocial at baseline showed the largest decrease in antisocial behavior (effect size = 0.25), but decreases in antisocial behavior were also significant for students with low ratings of antisocial behavior at baseline (effect size = 0.17).

Note: Randomization was done at the school level, but analysis was done at the student level.  Analyses were not designed to adjust for the impact of clustering within schools.

Espelage, D.L., Low, S., Polanin, J.R., & Brown, E.C. (2013). The impact of a middle school program to reduce aggression, victimization, and sexual violence. Journal of Adolescent Health, 53. 180-186.

Espelage, D.L., Low, S., Polanin, J.R., Brown, E.C. (2014). Clinical trial of Second Step© middle-school program: Impact on aggression & victimization. Journal of Applied Developmental Psychology, 37. 52-63.

Espelage, D.L., Rose, C.A., & Polanin, J.R. (2015). Social-emotional learning program to reduce bullying, fighting, and victimization among middle school students with disabilities. Remedial and special education, 0741932514564564.

Evaluated population: Thirty-nine middle schools were approached to participate in this study. Ultimately, 36 middle schools in Illinois (n=24) and Kansas (n=12) were randomly assigned to intervention or control groups, providing a study sample of 3,658 sixth-grade students. On average, students were 11.24 years old at baseline. The overall study sample was 34 percent Hispanic, 26 percent African-American, 25 percent white, and 15 biracial or another race. The students in control group schools were significantly more likely to be Hispanic than those in intervention group schools (36.6 percent versus 32.0 percent). The majority of students (74.1 percent) were eligible for free or reduced-price lunch. Twelve schools provided data on students with disabilities (n=123). In this subgroup, 53 percent were African-American, 31 percent were white, six percent were Hispanic, and 10 percent were biracial.

Approach: Schools were placed into matched pairs, with one member from each pair randomly assigned to an intervention group that would receive the middle school Second Step curriculum, and the other assigned to a wait-list control group that was provided a copy of the P3: Stories of Us bullying program (although only one control school adopted the P3 curriculum, with the rest choosing no substitute). There were seven primary outcomes examined in the study: perpetration of verbal or relational bullying, peer victimization, physical aggression, homophobic name-calling (both perpetration and victimization) and sexual harassment and sexual violence (both perpetration and victimization). For the full-sample analysis, outcomes were dichotomous; physical aggression, verbal and relational bullying, victimization by peers, and homophobic perpetration and victimization were reported as occurring if participants reported experiencing or engaging in two or more items on their scales, while sexual violence perpetration and victimization were reported as occurring if participants reported experiencing or engaging in one or more items on the scale. The subgroup analysis for students with disabilities evaluated the outcomes of bullying perpetration, victimization by peers, and physical aggression, and used scale scores rather than a dichotomous indicator.

Outcomes were measured through student surveys. The baseline survey was conducted in the fall of sixth grade, and changes were measured with surveys in the springs of sixth, seventh, and eighth grades. The multi-level analysis used adjusted for clustering, and groups did not differ significantly on any outcome measures at baseline.

Results: In the spring of sixth grade, intervention group students had a 42 percent lower probability of reporting physical aggression than students in control group schools, a significant difference. However, none of the other outcomes of interest were significantly different between students in intervention group or control group schools at this time. At the spring of seventh grade, no overall program impacts were found. However, analysis by state found that the program was more effective in Illinois schools than in schools in Kansas. When restricting the analysis to just schools in Illinois, students in intervention group schools reported significantly lower levels of homophobic name-calling victimization and sexual violence perpetration. Students in Illinois intervention group schools were 56 percent less likely to report being the victim of homophobic name calling than those in Illinois control group schools, and 39 percent less likely to report perpetrating sexual violence.

Analyzing program impacts for students with disabilities, a positive impact on bullying perpetration was found. The bullying perpetration scale scores for intervention group students with disabilities significantly decreased across the four time periods compared to scores for control group students with disabilities. However, no positive impacts were found on victimization by peers or physical aggression.

SOURCES FOR MORE INFORMATION

Link to program curriculum: http://www.cfchildren.org/orderf/index_html

References

Grossman, D.C., Neckerman, H.J., & Koepsell, T.D, et al. (1997).  The effectiveness of a violence prevention curriculum among children in elementary school: A randomized controlled trial.  JAMA, 277(20).  1605-1611.

Frey, K.S., Nolen, S.B., Edstrom, L.V.S. & Hirschstein, M.K. (2005).  Effects of a school-based socio-emotional competence program: Linking children’s goals, attributions, and behavior.  Journal of Applied Developmental Psychology, 26(2).  171-200.

Espelage, D.L., Low, S., Polanin, J.R., & Brown, E.C. (2013). The impact of a middle school program to reduce aggression, victimization, and sexual violence. Journal of Adolescent Health, 53. 180-186.

Espelage, D.L., Rose, C.A., & Polanin, J.R. (2015). Social-emotional learning program to reduce bullying, fighting, and victimization among middle school students with disabilities. Remedial and special education, 0741932514564564.

Espelage, D.L., Low, S., Polanin, J.R., Brown, E.C. (2014). Clinical trial of Second Step© middle-school program: Impact on aggression & victimization. Journal of Applied Developmental Psychology, 37. 52-63.

KEYWORDS: Elementary School, Middle School,  Males and Females (Co-ed), Skills Training, Social and Emotional Health and Development, Aggression, Bullying, Behavioral Problems, Education, Academic Achievement, Disruptive Behavior Disorder, Middle Childhood (6-11), Children, School-Based, Urban, Suburban, White or Caucasian, Black or African American, Hispanic or Latino, American Indian or Alaskan Native, Asian, Native Hawaiian or Pacific Islander.

Program information last updated 5/13/15


Subscribe to Child Trends

Short weekly updates of recent research on children and youth.