Jan 16, 2013


Seattle public elementary schools participated in an experimental design
prevention study on early childhood aggression. Parents and teachers of students
in the first grade were trained in family and classroom management skills,
respectively. The parents curriculum, “Catch ‘Em Being Good,” consisted of
training in monitoring children’s behavior, appropriate rewards and punishment,
effective communication, and family activities. Teachers were responsible for
tracking children’s progress throughout the study by assessing levels of
antisocial and problem behaviors. The study achieved mixed outcomes. No
significant impacts were reported for black students. White students achieved
significant impacts on only a few of the scales measured. White boys showed
lower rates of aggression and externalizing behavior, and white girls showed
lower rates of self-destruction. In general, the prevention appears to be most
effective for exceptionally antisocial behavior among white boys.


Target population: Teachers and parents of
students in the 1st/2nd grade of public elementary schools

Eight public elementary schools in Seattle were selected to
participate in a study on the prevention of early childhood aggression. In six
of the schools, 1st grade teachers and students were randomly
assigned to a control or experimental condition. The other two schools were
randomly assigned to be either fully control or fully experimental. The study
began in the fall of 1981 when students entered 1st grade and
continued through the spring of their 2nd grade year. Parents of
students participating in the experimental group were set to receive seven
consecutive, weekly sessions in family management training. The curriculum,
“Catch ‘Em Being Good” taught parents skills in monitoring and supervising a
child’s behavior, using appropriate rewards/punishments, and consistent
discipline practices, as well as effective communication skills and involving
children in family activities.

Teachers, on the other hand, were trained one year prior to
the study. The teacher training emphasized classroom management skills,
cognitive social skills training and interactive teaching methods. Through the
Interpersonal Cognitive Problem Solving curriculum, teachers learned about
proactive management skills to prevent behavior problems before they could
occur. Teachers were taught to promote appropriate classroom participation by
providing clear instructions and expectations about attendance, class activities
and rules. The training emphasized the importance of rewarding prosocial
behavior and employing the least disruptive disciplinary techniques. Teachers
also learned skills in communication, decision-making, conflict-resolution and
negotiation. Ultimately, teachers strived to increase students’ own
problem-solving abilities.


Hawkins, J.D., Von Cleve, E., & Catalano, R. (1991).
Reducing early childhood aggression: Results of a primary prevention program. Journal of the American Academy of Child and Adolescent Psychiatry, 30(2),

Evaluated population: Eleven 1st
grade teachers were randomly assigned to experimental classrooms, while ten 1st
grade teachers remained in control classrooms.

Approach: Studentswere then randomly assigned to either control or experimental classes upon entry
into the 1st
grade in the fall of 1981. 520 students were still in study schools in the
spring of 1983 when posttest data were collected. Posttest data were obtained
from 88.1% of those students. Selected demographic characteristics of both the
control and experimental groups are shown below:

Category Control Experimental
Male 50.3% 46.1%
Female 49.7% 53.9%
Free lunch eligibility 86.8% 84.6%
Lives with both parents 46.8% 50.5%
Nonwhite 52.9% 55.0%

Student self-report data collected at
baseline indicated that the control group was slightly higher on measures of
prosocial characteristics such as attachment to school and family, as well as
communication and family supervision.

Results:Parents of 122 experimental students (43 percent) who had
posttest data available participated in one or more sessions of parent training.
The average number of sessions attended was 5.4 per family. 48 of the families
attending sessions included fathers. Students’ posttest scores were not found to
be impacted by the number of classes parents attended except on a measure of
self-destructive behavior for females. In this case, however, it is not possible
to determine causality between the measure and parental class attendance.

Teachers’ implementation of the program was monitored in
two ways. They were required to complete a weekly Teacher Self-Report checklist
weekly. Teachers also participated in an Interactive Teaching Map (a structured
observation system). Teachers and students in both the control and experimental
groups were watched and recorded by trained observers. Each classroom was
observed for one 50-minute session for one day in the fall and one day in the
spring after students had entered the 2nd grade. No outcomes data
were collected during second grade.

As expected, experimental teachers implemented management
practices significantly more than control teachers. These teaching practices
were positively correlated with the time students spent actively engaged in
learning activities and negatively associated with students’ off-task behavior.

Student outcomes were evaluated at the end of the 2nd
grade year using Teacher Report Forms of the Child Behavior Checklist (CBCL).
The Checklist measured levels of children’s antisocial and problem behaviors.
CBCL was identical for boys and girls with the exception of two scales. Boys
were also evaluated for obsessive-compulsive behavior while girls were evaluated
for depressive behavior. In all cases, the reporting teacher must have known the
student for at least two months.

Although all boys in the experimental group reported better
posttest scores on the CBCL, significant outcomes were only found on the
Aggressive subscale ( [E] = 8.78, [C] = 21.51) and the Externalizing Antisocial
Behavior scale ( [E] =15.55, [C] = 21.51).

For females, significant outcomes were found on only one
measure: self-destructive behavior ([E] = 0.15, [C] = .65). In this case,
teachers rated experimental girls significantly less self-destructive than
control girls.

Because there were substantial subgroups of blacks and
whites within the sample, evaluators also analyzed outcomes according to these
racial characteristics. Of the total male sample, 33% were black and 46% were
white. No significant impacts were found among black males when comparing the
experimental and control groups. White males in the control group, however,
scored significantly higher on scales of aggressive behavior ([E] = 6.88, [C] =
13.4) and global externalized deviance than their white counterparts in the
experimental group ([E] = 13.18, [C] = 21.64).

Students scoring in the upper 11% on the CBCL are
considered in a clinical range and at highest risk. On externalizing antisocial
behavior, 6.8% of white males in the experimental group fell into this range,
whereas 20% of white males in the control group did. The intervention is
particularly effective, therefore, at the extreme end of the scale among
antisocial, white boys. Although no difference was found between black and white
males within the control group, the intervention was significantly less
effective on the aggressiveness and externalized deviance of black boys in the
experimental group then their white counterparts.

Of the total female population, 30% were black and 46% were
white. When comparing white girls in the experimental and control groups, white
girls in the experimental group scored significantly lower on the
self-destructive, depressive and nervous-overactive scales. Impacts were not
found for black girls.

Overall, it seems that the early childhood aggression
program benefited white students more than blacks. However, both male and female
black students were rated more negatively by their teachers than white students
on the whole. This may be the result of the differential perceptions of the
teachers. Of the 37 teachers participating, 31 were white. The difference in
scores between races could also mean that the intervention is better suited for
white students.



Hawkins, J.D., Von Cleve, E., & Catalano, R. (1991).
Reducing early childhood aggression: Results of a primary prevention program. Journal of the American Academy of Child and Adolescent Psychiatry, 30(2),

KEYWORDS: manual, co-ed, Aggression/Violence/Externalizing
Problems, White/Caucasian, Black/African American, Children, School-based,
parent or family component, parent training, Depression/Mood Disorders, Anxiety

Program information last updated 3/16/07


Subscribe to Child Trends

Short weekly updates of recent research on children and youth.