Principal Investigators: J. David Hawkins and Richard Catalano
Level of Intervention: Universal
Target Population: elementary aged children
References: Hawkins, Catalano, Morrison, ODonnell. Abbott, & Day (1992); ODonnell, Hawkins, Catalano, Abbott, & Day (1995); Hawkins, VonCleve, & Catalano (1991); Hawkins, Catalano, Kosterman, Abbott & Hill (in press).
Theory (Risk & Protective Factors Targeted):
The SSDP is based on the public health model of preventing adolescent health-risk behaviors by focusing on risk and protective factors associated with these behaviors. Research indicates that many of the same risk factors predict multiple poor outcomes including delinquency, substance abuse, teen pregnancy and school dropout. Bonding to school and family can serve to protect against this broad range of risk factors. SSDP was designed to increase bonding to school and academic success and to reduce early and persistent physical aggression, academic failure and poor family management practices including unclear rules, poor monitoring and harsh or inconsistent discipline. The intervention is designed to target multiple risk factors in multiple contexts (home & school) given research that risk for poor outcomes increases with multiple risk factors.
SSDP stems from the Social Development Model (Hawkins, Catalano, & Miller, 1992), and is designed to effect processes that result in three types of strong bonds (attachment, commitment, and belief) between families and schools. Strong bonds are hypothesized to serve a protective function by moderating the effect of social norms about maladaptive behavior (e.g. drug use, delinquent acts). Children bonded to family and school are more likely to subscribe to the prosocial norms of individuals in these contexts and to utilize refusal skills effectively.
According to the Social Development Model, childrens bonding is dictated by the degree of opportunity for active involvement in the family and classroom, the necessary skills to interact in these environments, and recognition or reinforcement for prosocial behavior in these groups. The Social Development Model also suggests that early and sustained intervention is necessary to change the developmental trajectory for these outcomes.
Description of Intervention:
SSDP is a combination of modified teaching practices in mainstream classrooms and parent training. The program is designed to be developmentally appropriate across the intervention period and to reduce academic failure, early conduct problems, and peer rejection, all thought to be antecedents of future maladaptive behavior or other poor outcomes.
In the classroom component, teachers are trained in proactive classroom management (Brophy, 1986), interactive teaching (Block, 1971, 1974; Brophy, 1987), and cooperative learning (Slavin, 1991). First-grade teachers were also trained to implement the Interpersonal Cognitive Problem Solving curriculum developed by Shure and Spivack (1988, see elsewhere in this report). This program focuses on developing childrens communication, conflict-resolution, and problem-solving skills. In the 6th grade, teachers also presented refusal skills training related to substance abuse and peer pressure.
The parent training component consists of parent classes offered as an optional program in 1st-3rd grade in collaboration with local school and parent councils. In first & second grade, the 7-session "Catch Em Being Good" program was offered. This program is designed to improve parental monitoring of child behavior, help parents provide expectations for child behavior, and support parents use of positive reinforcement and negative consequences for behavior. In 2nd and 3rd grade, the 4-session "How to Help Your Child Succeed in School" was offered in order to improve communication and involvement between parents and children, to help the parents support their childs learning, and to support collaborative relationships between parents and teachers. A 5-session program was also offered to help families develop family positions on drug use and to support childrens use of resistance skills.
The sample consisted of 643, 1st 5th graders (199 treatment, 709 control & unexposed). Ninety three percent of the sample was between 10 and 11 years old. The breakdown between gender was fairly even with 52% male and 48% female subjects; 44% of the students were Caucasian, 26% were African-American, and 22% were Asian-American. Approximately 56% of the students were low-income and qualified for the subsidized school lunch program. Also, because a policy of mandatory busing to achieve racial equality in the schools was in place during the study period, all schools in the study represented heterogeneous populations of students from at least two different neighborhoods each.
The most recent study examined the effects of the full intervention package provided to students throughout their elementary school years (grades 1-6), compared to a late intervention provided only in grades 5 and 6 and a no-treatment control group. Students in the sample were from eight schools in high crime areas within the Seattle Public School System. Two schools were administratively assigned to receive intervention, two schools were administratively assigned as control classrooms, and the remaining six were randomly assigned to intervention or control conditions. In the fifth grade all students in the 18 Seattle elementary schools were included in the study. When the participants were age 18 (in the spring of 1993) 598 of the original participants from all three conditions (93%) were successfully interviewed. Measures of internal validity showed the full intervention and late intervention groups were not significantly different on a number of socioeconomic and community contextual variables.
Teacher Instructional Practices (Kerr, Kent, & Lamb, 1985); Students perceptions of Proactive Family Management, Restrained Punishment, Family Communication, Family Involvement, and Attachment to Family; Students perceptions of School Rewards, Attachment to School, and Commitment to School; scores on standardized achievement tests; Students self-reported Beliefs and Norms, and self-reported Substance Use and Delinquency.
Results (Hawkins, VonCleve, & Catalano, 1991) found reduced rates of teacher-reported aggression (p<.01) and externalizing behavior (p<.05) only in Caucasian, male, treatment students compared to controls. Lower levels of self-destructive behaviors were also found only in Caucasian, female, treatment students over controls.
Post-Test results (beginning of 5th grade):
Intervention students reported more proactive family management by parents (p<.025) and greater family communication (p<.025) compared to controls. Intervention students report greater involvement by family (p<.05), bonding to family (p<.025), attachment to school (p<.025), commitment to school (p<.025), and found school more rewarding (p<.025). No significant differences were found on drug-related norms.
Follow-up results (end of 6th grade):
Teachers reported intervention students spent significantly less time with deviant peers compared to controls (p<.05).
Long-term followup at age 17&18:
Researchers examined the differential effects of the full and late intervention, as well as the potential mediating effects of gender and poverty. Full intervention students showed statistically significant positive outcomes related to commitment and attachment to school at age 18; self-reported achievement (though change in school-reported GPA did not reach significance and there was no effect on achievement test scores); self-reported involvement in school misbehavior; committing violent delinquent acts by age 18; heavy alcohol use in the past year at age 18 (though no significant effects were found for any of the lifetime measures of drug use); and engaging in sexual intercourse and having had multiple sex partners by age 18.
Though only the full intervention group showed significant effects, a dosage effect was observed on several of the variables, showing the full intervention group having the greatest positive change, followed by the late intervention group, and finally the control group. The only gender x intervention effects found were for repeating a grade and engaging in sexual activity, both favoring male students.
Strengths & Limitations:
The SSDP is a multi-component preventive intervention that targets multiple risk and protective factors across multiple domains. The study was not randomized, although extensive analysis of potential threats to internal validity and attrition effects provide no reason to expect bias. The significant outcome measures appear to be all self-report measures, although information on the actual constructs is not provided in the most recent publication. The distal measurement of effects and the heterogeneity of the sample lends to the generalizability of the outcomes. The long-term followup with little attrition makes this an exceptional study. Results indicated that only intervention that began in the early grades had long-term impact on post-graduation outcomes. No measures of implementation fidelity were discussed.