August and colleagues (2002) examined the effects of the Early Risers ‘Skills for Success’ Program over a 3-year intervention trial in two semirural, Midwestern geographical sites consisting of white families of low to low–middle socioeconomic status. Within each site, 10 demographically comparable schools agreed to participate in the screening of kindergarten children for aggressive/disruptive behavior. After students were screened at each site, 5 schools were randomly assigned to the program group and 5 schools were randomly assigned to the control group, for a total of 10 program schools and 10 control schools. A total of 1,840 kindergarten children were screened for early-onset aggressive behavior, and 18.5 percent (341 students) met the high-risk criteria.
Parental consent was received for 124 children from program schools and 121 children from comparison schools, to form the original sample of 245 students. At the 3-year follow-up, there were 100 students in the intervention group and 99 students in the control group. The intervention group was mostly white (84.7 percent) and male (63.7 percent), with an average age of 6.6 years. The control group was also mostly white (93.4 percent) and male (73.6 percent), with an average age of 6.7 years. There were no significant differences between the students who were retained and those who dropped out of the study, except that retained students tended to be slightly older and were more likely to be white.
Outcome variables were four global competence domains: academic competence, behavior self-regulation problems, social competence, and parent investment. Data on academic competence, behavioral self-regulation, and social competence was measured through several scales. The Woodcock–Johnson Test of Achievement—Revised measured students’ basic reading and arithmetic skills. Teachers also completed the following:
- The Behavioral Assessment System for Children–Teacher Rating Scale, which rates dimensions of externalizing, internalizing, and adaptive behavior
- The Teacher’s Scale of Child’s Actual Competence and Social Acceptance, which rates cognitive competence
- The Teacher Observation of Classroom Adaptation—Revised, which provides additional information about classroom behavior
Parents completed the Parent Observation of Classroom Adaptation, which assessed child behavior in the home. Finally, parent investment in the child was measured through the Alabama Parenting Questionnaire, which assessed parenting practices. The data was collected from students, teachers, and parents during May and June in each of 3 subsequent years. This study examined the results at the end of the second year of the intervention.
Children in the intervention and control groups were categorized into three groups on the basis of their level of severity of aggression relative to the normative sample across all 3 time periods examined in the study. Children were categorized as either mildly, moderately, or severely aggressive based on their scales on the Aggression Scale of the Child Behavior Checklist–Teacher Reporting Forms. Approximately one third of the intervention and control groups fell into each of these categories.
In addition, a sample of kindergarten children was drawn from the same schools used to select program and control group participants, but the sample served as normative participants. The children were selected from the following year’s kindergarten class and did not receive intervention activities. The sample was drawn to be proportionally equivalent to the general population of kindergarten students in terms of gender and aggressive behavior. The resulting sample (n
=92) included 45 percent boys, and only 19 percent were in the at-risk range for aggressive behavior.
The outcomes were measured with composite variables computed as a mean of at least two scales. Assignment to group was randomized by school rather than by individual; thus, data was analyzed using a three-level mixed random regression model procedure, with time points nested within individual participants nested within schools. An intent-to-treat strategy was used in the analyses, meaning all available data was included, even data from missing participants. A subgroup analysis was conducted on dosage (which included both the level of intervention received and how many lessons parents attended), gender, and the child’s level of aggression.