Study
Catalano and colleagues (2003) used an experimental design to study 938 elementary students from first or second grade enrolled in 10 schools in the Pacific Northwest. These children were randomly divided into two groups: those receiving the Raising Healthy Children (RHC) prevention program and a control group.
The study sample was drawn from 10 suburban elementary schools that were paired on socioeconomic status and attendance patterns. One school from each pair was randomly assigned to the RHC experimental group (n = 5 schools) or to the control group (n = 5 schools). After asking for participation and obtaining consent, the final sample size was 938 first and second grade students—497 in the treatment schools and 441 in the control schools. The entire sample was 53 percent male, with an average age of 7.43 years. The vast majority of the sample was white (almost 82 percent), followed by Asian/Pacific Islander (7.4 percent), Latino (3.9 percent), African American (3 percent), Native American (3.3 percent), and other races (0.4 percent).
The RHC experimental schools received the complete intervention program. This included workshops for teachers that concentrated on classroom management, reducing academic risks, and enhancing protective factors. At the conclusion of each workshop, RHC staff offered classroom coaching for teachers. In addition, during the first year of the project, teachers participated in monthly booster sessions to refresh and instill the skills learned at workshops.
Parents received training from school-home coordinators (SHCs), who were classroom teachers or specialists in providing services to parents and families. Parents had multiple training options. There were five sessions of parenting group workshops, special topic workshops, and in-home problem-solving sessions. Monthly newsletters were also sent to participating parents to remind them of upcoming workshops and to reinforce learned content.
Students received the intervention through summer camps conducted by SHCs. Those attending summer camps were recommended by teachers or parents because of poor academic performance or behavioral problems. This was coupled with in-home services to reduce the student’s problematic behavior in class and so they could work on their academic skills.
Teachers, parents, and students in the control schools received treatment as usual. Because of the research design used, no programmatic elements of RHC were received by any of the control schools.
Data collection occurred four times during the 1½-year study period. Baseline measures were collected before the intervention, followed by another data collection in the spring of year 1 and two more collections in the spring and fall of year 2. Data was obtained from teachers, parents, and students. Teachers completed a student behavior checklist on each participant. Parents were phoned at home and completed a 45-minute interview in addition to the behavioral checklist. Lastly, students were administered a survey in the spring of years 1 and 2. Altogether this resulted in multiple waves of data from multiple sources over the course of the study.
Data from the teacher surveys was used to measure commitment to school, academic performance, social competency, and antisocial behavior. Items from the Teacher Observation of Classroom Adaptation—Revised and the Child Behavior Checklist—Teacher Report were used in conjunction with other study-developed questions to measure these concepts and behavior. Parents used a similar instrument to provide similar information on their child’s academic performance and behavior. Students completed a self-report survey that included a social competency scale and an antisocial scale.
Analyses were conducted 18 months after implementation and concentrated on academic and behavioral improvements within the school environment. Hierarchical linear models were used to determine the growth rate and level of each intervention across the multiple sources of data. This analysis incorporates two models of change. The first model determines whether the variance across individuals is due to sampling error. The second model examines whether the experimental or treatment condition predicts values for individuals, after controlling for baseline measures and descriptive information. The authors conducted subgroup analyses on gender and low-income status.
Study
Brown and colleagues (2005) used an experimental design to examine the long-term effects of RHC. This was a continuation of the study by Catalano and colleagues (2003) that included a slightly larger sample of students who had moved into the participating school districts after the study was initiated. Children were followed as they aged and entered middle school where they are exposed to different risk factors and encounter new life challenges. The main focus of this study was to examine the effects of RHC on adolescent substance use.
RHC uses a social developmental approach and the intervention adjusts as children age. The experimental schools received the complete intervention program. Workshops for teachers concentrated on classroom management, reducing academic risks, and enhancing protective factors. These workshops were given to teachers while students were in the first grade and again during the first year of middle school. The middle school workshops differ in content from the elementary school workshops in order to address the different classroom and behavioral challenges that teachers face. At the conclusion of each workshop, RHC staff offered classroom coaching for teachers. To monitor and enhance the fidelity of the project, teachers participated in monthly booster sessions to refresh and instill the skills learned at workshops.
Parents received training from SHCs and had multiple training options. There were parenting group workshops, special topic workshops, and in-home problem-solving sessions as well as monthly newsletters. These group and individual sessions occurred from grades 1 through 8. Like the teacher workshops, the parent workshops focused on different developmental issues as children aged and their lives changed. Middle school and high school workshops for parents spent more time discussing substance use and risky behavior than those given during elementary school. During high school, booster sessions were given at home, and students completed behavioral assessments.
The student intervention consisted of volunteer participation in afterschool tutoring sessions and study clubs from fourth through sixth grades. Similar to the rest of the intervention, elementary school sessions focused more on antisocial behavior and school commitment while sessions for middle school and high school students addressed alcohol and drug use. Students also received booster sessions and group workshops during the middle and high school years.
Teachers, parents, and students in the control schools received treatment as usual. Because of the research design used, no programmatic elements of RHC were received by any of the control schools.
Data collection occurred annually from sixth through tenth grades while students were in school. Those who missed school during a collection visit were contacted at home and completed their assessments by mail, phone, or in person with a member of the research team.
Measures of substance use were taken from self-reports of frequency of alcohol, marijuana, and cigarette use during the previous year and the previous month. Data analysis consisted of a two-part latent growth model. The first part of the model screened out nonusers from users. The second part of the model consisted of continuous indicator variables that represented the frequency of substance use amongst users. By using this method, students who never used any illicit substance did not contribute data to the growth rates and models of substance use. Only those students who had ever used illicit substances contributed information to the growth rate. The authors conducted subgroup analyses on gender, grade cohort, baseline antisocial behavior, and family income.
