Program

Apr 19, 2012

OVERVIEW

The Good Behavior Game (GBG) intervention is intended to help reduce aggressive behavior in students in the early elementary grades. The program is one component of a two-part intervention administered in first and second grades.  The GBG uses behavior modification strategies to reduce levels of  aggression and poor conduct in the classroom.  In a random-assignment evaluation of the program, researchers found that that the GBG had impacts that were significant, and increased over time, but only for male students whose first-grade levels of aggression were high (above the median). However, another evaluation found a decrease in both aggressive and shy behavior, as rated by teachers, for both boys and girls. Yet another evaluation summarized here found a positive impact of GBG on ADHD, conduct disorder, and oppositional-defiant disorder symptoms at the end of treatment, for children with intermediate levels of symptoms.  Children in the intervention showed stable levels of symptoms over the course of two years, whereas control-group children showed an increase in symptoms over the course of the study.  Follow-up evaluations found that the intervention group showed a decrease in levels of aggression during transition times, through sixth grade, whereas in the control group aggression levels reached a plateau at third grade.  Follow-up data also indicate that boys who were in the GBG group at grades one and two were less likely to engage in smoking when they were early adolescents.  The game was also adapted for use in Dutch populations, and an evaluation of the Dutch version found it to have a significant impact on ADHD symptoms. Another evaluation also indicates that the GBG decreases suicide ideation and attempts through childhood, adolescence, and early adulthood.

DESCRIPTION OF PROGRAM

Target population: Elementary students in first and second grade

The Good Behavior Game is an intervention designed to reduce maladaptive behaviors in first and second grade students, based on the assumption that their behavior is malleable enough at this stage that the intervention will improve their behavior before they transition into middle school.

GBG is a classroom-based, teacher-led behavior management strategy, which rewards teams of children for good behavior. A team wins a game if at the end of the designated period its members have not exceeded a pre-established level of maladaptive behavior. In the early stages of the game, the designated “game time” is announced to students, and the length is fixed; rewards are given out immediately following the game. At later stages, the teacher does not announce the game time, and rewards are distributed at the end of the day.

In addition to reducing maladaptive and aggressive behavior among already aggressive first- and second-graders, the intervention is intended to prevent these behaviors from developing among children who do not have these early problems.

EVALUATION(S) OF PROGRAM

Van Lier, P.A., Muthen, Bengt, O., van der Sar, R.M., Crijnen A.A.M.  (2004)  Preventing disruptive behavior in elementary schoolchildren:  Impact of a universal classroom-based intervention.  Journal of Consulting and Clinical Psychology, 72, 467-478.

Evaluated population: 666 second-graders in 31 classrooms from 13 schools in the Netherlands.  The sample was 69 percent Caucasian, 10 percent Turkish, 9 percent Moroccan, 5 percent Surinam-Dutch Antilles, and 7 percent other ethnic groups.

Approach:  The GBG was adapted for Dutch populations. In the Dutch adaptation, the teams do not compete for weekly prizes, and teachers do not mention the names of students who violate the rules. Children are encouraged to actively support their teammates.

Thirteen schools in the metropolitan areas of Rotterdam and Amsterdam participated in the study. Within each school, classes were randomly assigned to intervention (GBG) or control groups. The study lasted for two years, so the intervention children received the GBG throughout first and second grades, and the control group received no intervention.

Teacher assessments of student behavior were collected twice in first grade (T1 and T2), at the end of the first year of intervention (T3), 18 months after the start of intervention (T4), and at the end of the second year (T5). At each assessment, teachers completed a written report on each child. At T3 and T5, teachers also participated in a structured interview about each child in their class.

Results: Children in the intervention condition showed a decrease in ADHD problems over time, while children in the control condition showed an increase in such problems over time. The difference between groups was statistically significant.

To assess the impact of the GBG on children with different levels of problems, the researchers used mixture modeling to identify three trajectories of children with ADHD symptoms: high (Class 1), intermediate (Class 2), and low (Class 3). Children were classified into one of the three trajectories, and intervention effects were assessed for each trajectory. Class 1 children had high levels of ADHD problems at baseline and their problems declined significantly throughout the intervention period. The GBG did not have any significant impacts for children in this group. Class 2 children had fewer initial symptoms than Class 1 children, but still exhibited intermediate levels of problems. Control-group children in Class 2 had increasing problems throughout the intervention period, but intervention children did not. This difference was statistically significant. Class 3 children exhibited low levels of ADHD symptoms at baseline and throughout the intervention. There were no significant differences between intervention and control group children in Class 3.

Lastly, the researchers assessed the effect of the GBG on conduct problems and oppositional defiant disorder (ODD). For children in Class 1, there was no significant difference between control group and intervention children for ODD. Class 1 intervention children did exhibit lower conduct problems than control-group Class 1 children, and this finding approached significance (p = .06). For Class 2, children in the control group showed increases in conduct problems and ODD symptoms, whereas children in the intervention group did not. This effect was significant. Class 3 children showed low levels of ODD and conduct disorder problems, and there were no differences between intervention and control groups.

Kellam, S.G. Rebok, G.W., Ialongo, N. & Mayer, L. S. (1994). The course and malleability of aggressive behavior from early first grade into middle school: Results of a development epidemiologically-based preventive trial. Journal of Child Psychology and Psychiatry. 35(2), 259-281.

Evaluated population: 693 first-graders in 41 classrooms from 19 schools in five diverse urban Baltimore public school districts. The sample was 49 percent male and 65 percent African American. The authors noted that the population in this study was considered representative of elementary public school children in the eastern part of Baltimore who remained in the same school for two years, and remained in a Baltimore public school for six years.

Approach: Within each of the five urban areas, one school was assigned to receive the GBG intervention and another to receive an achievement-focused intervention called Mastery Learning (ML), which was used in this study as an active control for the GBG classrooms. Schools were matched on characteristics such as the school’s average achievement level, socio-economic status of families, and ethnicity of students. The rest of the schools within each urban area were assigned to be external controls with no experimental intervention. Within each school assigned to either experimental intervention, half the classrooms were randomly assigned to receive the intervention and half were used as internal, school-specific, controls.

Students’ level of aggression was assessed through a Social Adaptation Status (SAS) measure, which consisted of a semi-structured interview with the child’s teacher. The teacher was asked to rate students on social tasks and aggressive behavior. All students were assessed at baseline with this measure. Further assessments were conducted on representative samples of the experimental group twice a year, through sixth grade. A second conduct-disorder measure, the Diagnostic Interview Schedule for Children (DISC), was used in the sixth-grade follow-up with students who reported three or more conduct-disorder behaviors.

Researchers hypothesized that students who exhibited aggressive behavior and conduct problems in first grade and participated in the Good Behavior Game would show fewer aggressive behaviors in subsequent years (between first and sixth grades). Further, it was hypothesized that students who exhibited fewer aggressive and problem-conduct behaviors in first grade would be less likely to develop them in subsequent grades.

Results: Researchers found no overall impacts on sixth-grade aggressive behavior for students in the GBG condition compared with the internal, external, and active Mastery Learning controls. They did, however, find significant GBG impacts, which increased over time, for male students with higher levels of aggressive behavior on the SAS measure as reported in first grade. Among these aggressive males, trajectories of aggressive behavior among experimental and control groups remained the same until third grade. After third grade, the level of aggressive behavior among control group males reached a plateau while males in the GBG group showed a continued decrease in aggressive behavior up to the transition into middle school. The decline in aggression was not seen in females or in GBG males whose baseline levels of aggressive behavior at the first-grade assessment were below the median.

Dolan, L. J., Kellam, S. G., Brown, C. H., Werthamer-Larsson, L., Rebok, G. W., Mayer, L. S., Laudolff, J., & Turkkan, J. S. (1993). The short-term impact of two classroom-based preventive interventions on aggressive and shy behaviors and poor achievement. Journal of Applied Developmental Psychology, 14, 317-345.

Evaluation Population: 864 students from 19 schools in Baltimore.

Approach: This study examined the impact of preventive interventions on children’s aggressive and shy behaviors.  Classrooms in the intervention schools were randomly assigned to GBG or Mastery Learning (ML) interventions.  In addition, entering first-graders were randomly assigned to classrooms.  The Mastery Learning intervention is an enrichment of the reading curriculum that includes goal-setting, communication of high expectations, small instructional units, formative and summative testing, and immediate feedback.  Students in classrooms randomly assigned to receive no intervention were in both intervention and non-intervention schools and acted as a control group.  Students’ behavior was assessed through teacher reports, peer assessments, and standardized tests.

Results: Compared to students in the external school controls, but not the internal school controls, GBG produced a significant decrease in boys’ aggressive behavior, as reported by teachers.  Compared with students in the internal control condition, but not the external controls, GBG produced a significant decrease in girls’ teacher-reported aggressive behavior.  Over time, levels of aggressive behavior among boys declined  as a result of GBG, but there was no impact for girls.  For all students, GBG was associated with a larger declines in teacher ratings of aggression among students who began the year with higher aggression levels.

Both boys and girls in GBG showed a decrease in teacher-rated shy behaviors .  The ML intervention was associated with increased reading achievement scores only for low-achieving male students and high-achieving female students.  Overall, GBG seemed to decrease aggressive behavior, while ML was associated with increased reading achievement.  An internal validity issue for this study was that the teacher provided both the intervention and the behavior ratings and was aware of specific outcome targets.  Peer ratings may also have been skewed by teacher’s expectations of improved behavior.

Kellam, S. G. & Anthony, J. C. (1998).  Targeting early antecedents to prevent tobacco smoking: Findings from an epidemiologically based randomized field trial.  American Journal of Public Health, 88, 1490-1495.

Evaluated Population: 1,604 first-grade children from five urban areas of Baltimore.  Students were recruited in two consecutive cohorts in 1985 and 1986 (Wave 1 N = 818, Wave 2 N = 786).  808 students were male, and 796 were female.

Approach: Three to four elementary schools were selected in each of 5 urban areas of Baltimore.  The schools were randomly assigned to the Good Behavior Game treatment, the Mastery Learning comparison treatment, or the control group which received the school’s normal curriculum.  In each intervention school, one classroom was randomly assigned to the treatment condition, and one served as a within-school control classroom.  Children were interviewed for 40-70 minutes each spring, for 6 years following the implementation of treatment, to assess their tobacco and drug use.

Results: Boys in the GBG program were less likely to smoke tobacco after the age of ten than were boys in the control group.  Girls in the GBG program, however, were equally likely to smoke as those in the control group.  There were no overall differences between the GBG and the Mastery Learning program in terms of students’ tobacco use.

Kellam, S.G., & Rebok, G.W. (1992). Building developmental and etiological theory through epidemiologically based preventive intervention trials. In J. McCord& R.E. Tremblay (Eds.), Preventing antisocial behavior: Interventions from birth through adolescence (pp.162-195). New York: The Guilford Press.

Evaluated Population: 501-first grade children from five urban areas of Baltimore.

Approach: Three to four elementary schools were selected in each of 5 urban areas of Baltimore.  The schools were randomly assigned to the Good Behavior Game treatment, the Mastery Learning comparison treatment, or the control group which received the school’s normal curriculum.  In each intervention school, one classroom was randomly assigned to the treatment condition, and one served as a within-school control classroom. Children’s concentration problems, aggressive and shy behavior, depressive symptoms, and academic achievement were assessed in the fall and spring of the first-grade year.

Results: Positive impacts for GBG were found for teacher ratings of aggressive and shy behavior, and for peer nominations of aggressive behavior for both boys and girls. Girls in the Good Behavior classrooms were less likely to be rated by their peers as having few friends after the intervention. There was also an increase in on-task behavior in the Good Behavior Game classrooms compared with control classrooms.

Wilcox, H.C., Kellam, S.G., Hendicks Brown, C., Poduska, J.M., Ialongo, N.S., Wang, W., & Anthony, J.C. (2008). The impact of universal randomized first- and second-grade classroom interventions on young adult suicide ideation and attempts. Drug and Alcohol Dependence, 95, S60-S73.

Evaluated Population: Participants were 1,196 first-grade students in 19 schools and 41 classrooms in the Baltimore City Public School System.  The schools were located within five urban areas with socioeconomic status ranging from very poor to moderate.  Sixty-six percent of students were African-American, and a large majority of the remainder were non-Hispanic white.

Approach: Within each of the five urban areas, schools were sorted into homogenous sets of three or four ,and randomly selected for the trial.  All entering first- graders were eligible participants.  Within each intervention school, all first-grade students were assigned so as to balance classrooms with respect to gender, kindergarten experience, and academic and behavioral performance.  Classrooms and teachers were randomly assigned to theMastery Learning (ML) intervention or ML-control groups, or The Good Behavior Game (GBG) or GBG-control groups.  External control groups were also created within control schools by assigning youth in a balanced fashion.  The interventions lasted for two years, which required students to move up to the second grade in grouped units.  At baseline, and twice in young-adulthood follow-up face-to-face interviews measured Suicide Ideation and Suicide Attempts.

Results: Students in the GBG intervention group were significantly less likely to experience suicide ideation, compared with youth in control classrooms.  Suicide ideation was not significantly less for students in the ML intervention classrooms than for controls.  There was not a significant difference between suicide attempts between intervention groups and controls, once baseline characteristics were controlled.  There were significantly fewer occurrences of suicide ideation in the GBG classrooms compared with controls, after age 16.  This difference was not significant between ML and control groups. In a replication trial with the second cohort, the impact of GBG was substantially smaller.

SOURCES FOR MORE INFORMATION

Link to program curriculum: www.channing-bete.com/GGC

References:

Dolan, L. J., Kellam, S. G., Brown, C. H., Werthamer-Larsson, L., Rebok, G. W., Mayer, L. S., Laudolff, J., & Turkkan, J. S. (1993). The short-term impact of two classroom-based preventive interventions on aggressive and shy behaviors and poor achievement. Journal of Applied Developmental Psychology, 14, 317-345.

Kellam, S. G. & Anthony, J. C. (1998).  Targeting early antecedents to prevent tobacco smoking: Findings from an epidemiologically based randomized field trial.  American Journal of Public Health, 88, 1490-1495.

Kellam, S.G., & Rebok, G.W. (1992). Building developmental and etiological theory through epidemiologically based preventive intervention trials. In McCord, J., & Tremblay, R.E. (Eds.), Preventing antisocial behavior: Interventions from birth through adolescence (pp.162-195). New York: The Guilford Press.

Kellam, S. G., Rebok, G. W., Ialongo, N. & Mayer, L. S. (1994). The course and malleability of aggressive behavior from early first grade into middle school: Results of a development epidemiologically-based preventive trial. Journal of Child Psychology and Psychiatry. 35(2), 259-281.

Van Lier, P.A., Muthen, Bengt, O., van der Sar, R.M., Crijnen A.A.M.  (2004)  Preventing disruptive behavior in elementary schoolchildren:  Impact of a universal classroom-based intervention.  Journal of Consulting and Clinical Psychology, 72, 467-478.

Wilcox, H.C., Kellam, S.G., Hendicks Brown, C., Poduska, J.M., Ialongo, N.S., Wang, W., & Anthony, J.C. (2008). The impact of universal randomized first- and second-grade classroom interventions on young adult suicide ideation and attempts. Drug and Alcohol Dependence, 95, S60-S73.

KEYWORDS: Children (3-11), Elementary, Males and Females, Urban, School-based, Manual is available, Other Behavioral Problems, Tobacco Use, Aggression, Conduct/Disruptive Disorders, Black/African American, Depression

Program information last updated 4/19/12.