Program Profile: Social Aggression Prevention Program (SAPP)

Study Title: The Prevention of Social Aggression Among Girls. (which is associated with Outcome 1, Outcome 2, Outcome 3, Outcome 4, and Outcome 5)

Research Design/Sample

Cappella and Weinstein (2006) conducted a longitudinal field experiment with a mixed methods design to examine the effect of the Social Aggression Prevention Program on female students’ empathetic behaviors, social problem-solving, and both aggressive/hostile and prosocial social behaviors. Participants came from six schools in one urban school district in northern California. 

Fifth-grade teachers and teachers of combined fourth/fifth-grade and fifth/sixth-grade classes were asked to participate. Thirteen teachers volunteered for the study, approximately one to three teachers per school. One-hundred and thirty-four parents provided consent for their children to participate. The sample was all female and predominately made up of fifth-grade students (89 percent) while 11 percent of the students were in fourth- or sixth-grade in combined-grade classrooms. Participants ranged in age from 9 to 12 years, with a mean age of 10.5 years. The student sample was 26 percent white, 25 percent Black, 24 percent Latino, 19 percent Asian, and 6 percent another race/ethnicity. Students were randomly assigned within classrooms to participate in the social aggression prevention program treatment group or in a small-group reading club control group that matched the intervention in structure and time, but did not include content related to relationships or conflicts. There were no statistically significant differences in demographic characteristics between the groups at baseline; however, students in the treatment group had statistically significantly higher baseline reading achievement compared with students assigned to the control group.

Data Collection

Data collection occurred 2–4 weeks prior to exposure to the intervention, and 2–4 weeks post-intervention. At both times, self- and peer-report instruments developed by the researchers based on existing instruments were used to obtain data on participants’ empathetic behavior, social problem-solving, and social behavior.

Peer assessments were conducted for each child in a private setting via a one-to-one interview with a researcher. Teachers were also asked to complete behavior rating scales for each student in the classroom during the same week in which researchers collected data from the students.

Outcome Measures 

The empathic behavior scale—short form assessed the teacher’s perspective of participants’ empathic behavior (e.g., “goes to the help of someone who has been hurt”). To measure social problem-solving, scenarios were designed to assess students’ knowledge of strategies to resolve socially aggressive conflicts. Responses on the victim perspective scenarios (imagining oneself as the target of social exclusion and nasty gossip) were used to assess prosocial problem-solving. Responses on the perpetrator perspective scenarios (designed to elicit social aggression in response to an uncomfortable situation) and the follower perspective scenarios (where the respondent is placed in the middle of social exclusion and nasty gossiping) were used to assess antisocial problem-solving. Responses were coded as prosocial/assertive, neutral/unsophisticated, and antisocial/aggressive. 

Two revised measures were used to assess students’ use of social behavior with peers. A revised version of the children’s social behavior scale–teacher form asked teachers to rate students’ use of behaviors in the past two weeks on subscales for social aggression (e.g., “This child spreads rumors or gossips about some peers”); physical/verbal aggression (e.g., “This child threatens to hit or to beat up other children”); and prosocial behavior (e.g., “This child is helpful to peers”). Additional questions were added to these scales to capture nonverbal, socially aggressive expressions (e.g., “This child rolls her eyes or snubs her nose at other children”); and verbal, directly aggressive expressions (e.g., “This child calls peers mean names”). 

The second measure of social behavior with peers was a revised version of the children’s social behavior scale–peer report. On a five-point scale (1 = never to 5 = almost always), students were asked to think about their female classmates’ behavior over the past two weeks and rate the frequency with which they used a behavior. As with the teacher report, this measure included subscales for social aggression, physical/verbal aggression, and prosocial behavior and receipt. Both teacher- and peer-reported social and physical/verbal aggression make up the aggressive/hostile social behaviors outcome.

Statistical Analysis

Chi-square analyses and analysis of variance procedures were used to establish baseline equivalence between groups in regard to ethnic measures and baseline dependent variables: empathy, social behavior (aggressive/hostile behaviors and prosocial behaviors), and social problem-solving strategies (both prosocial and antisocial). Main effects of the intervention were assessed using analysis of variance and analysis of covariance techniques. Subgroup analysis was conducted by baseline levels of social problems.

Citation:

Cappella, Elise, and Rhona Weinstein. 2006. “The Prevention of Social Aggression Among Girls.” Social Development 15(3): 434–62.

This study is a high-quality randomized controlled trial.