•  

Intervention Summary

PAX Good Behavior Game (PAX GBG)

The PAX Good Behavior Game (PAX GBG) is an environmental intervention used in the classroom with young children to create an environment that is conducive to learning. The intervention is designed to reduce off-task behavior; increase attentiveness; and decrease aggressive and disruptive behavior and shy and withdrawn behavior. The intervention also aims to improve academic success, as well as mental health and substance use outcomes later in life. PAX GBG evolved from the original Good Behavior Game developed and studied with fourth-graders.

The intervention includes a set of evidence-based strategies called "kernels" and a classroom game intended to increase self-regulation and cooperation and decrease unwanted behaviors called "spleems." The teacher first applies the kernels in the classroom. These kernels, some of which were developed for another NREPP-reviewed intervention, PeaceBuilders, include transition cues (PAX Quiet); written notes (Tootles) praising positive behavior; use of a timer to decrease the time needed for task completion (Beat the Timer); random calling of students during lessons (PAX Stix); and rewards in the form of brief and fun activities that are normally not allowed in the classroom, such as tapping a pencil on the desk or throwing paper balls (Granny's Wacky Prizes). The teacher also works with the students to establish a shared language and expectations about classroom behavior.

After these kernels are integrated into classroom activities, the game is played in two to five teacher-selected heterogeneous teams that are changed on a regular basis. Each day, the game is announced and played three times. Initially, the game is played for only a few minutes at a time when the children are engaged in simple tasks. As students improve at the game, the game is played for longer periods and during different activities and times of day. During the game, the teacher identifies and counts each unwanted behavior. At the end of the game, the teams with three or fewer infractions receive a reward, typically an activity selected from Granny's Wacky Prizes. In addition to the three announced games, one unannounced game is played each day. Roles (e.g., captains, coaches) can be assigned to children on each team. A booklet for parents and children explains the game and provides guidance on how parents can use elements of the game at home.

The study that was reviewed for this summary provided the foundation for the version of PAX GBG currently disseminated. Since the study was conducted, the game has been modified (e.g., it is played at different times, some games are unannounced, students can have roles) and elements have been added (e.g., parent booklet, kernels) to foster the generalization of self-regulation and peer cooperation across people, places, time, and activities. In addition, in this study, PAX GBG was used throughout first grade along with weekly classroom meetings to promote group problem solving, curriculum enhancements in language arts and mathematics, and additional support for children who did not respond adequately to the intervention. Although the study was conducted with first-graders, PAX GBG has been used with children of different ages.

Descriptive Information

Areas of Interest Mental health promotion
Outcomes
1: Conduct and problem behaviors
2: Academic success
3: Mental health service utilization
4: Initiation of substance use
Outcome Categories Alcohol
Drugs
Education
Mental health
Social functioning
Ages 6-12 (Childhood)
Genders Male
Female
Races/Ethnicities Black or African American
White
Settings School
Geographic Locations Urban
Suburban
Implementation History PAX GBG has been implemented in 32 States in the United States and 4 provinces in Canada, reaching more than 105,000 students in more than 650 schools. The intervention also has been used by several tribes in the United States and First Nations in Canada, as well as in Ireland. Six randomized trials have been or are currently being conducted within the United States and Canada.
NIH Funding/CER Studies Partially/fully funded by National Institutes of Health: Yes
Evaluated in comparative effectiveness research studies: No
Adaptations The training protocol has been adapted for use in afterschool settings, cafeterias, and playgrounds. It also has been adapted for use with tribal groups. All teacher and student materials have been translated into French, and many student materials have been translated into Spanish.
Adverse Effects No adverse effects, concerns, or unintended consequences were identified by the developer.
IOM Prevention Categories Universal

Quality of Research

Documents Reviewed

The documents below were reviewed for Quality of Research. The research point of contact can provide information regarding the studies reviewed and the availability of additional materials, including those from more recent studies that may have been conducted.

Study 1

Bradshaw, C. P., Zmuda, J. H., Kellam, S. G., & Ialongo, N. S. Longitudinal impact of two universal preventive interventions in first grade on educational outcomes in high school. Journal of Educational Psychology, 101(4), 926-937.  

Furr-Holden, C. D. M., Ialongo, N. S., Anthony, J. C., Petras, H., & Kellam, S. G.. Developmentally inspired drug prevention: Middle school outcomes in a school-based randomized prevention trial. Drug and Alcohol Dependence, 73(2), 149-158.  

Ialongo, N., Poduska, J., Werthamer, L., & Kellam, S. The distal impact of two first-grade preventive interventions on conduct problems and disorder in early adolescence. Journal of Emotional and Behavioral Disorders, 9(3), 146-160.

Ialongo, N. S., Werthamer, L., Kellam, S. G., Brown, C. H., Wang, S., & Lin, Y. Proximal impact of two first-grade preventive interventions on the early risk behaviors for later substance abuse, depression, and antisocial behavior. American Journal of Community Psychology, 27(5), 599-641.  

Petras, H., Masyn, K., & Ialongo, N. The developmental impact of two first grade preventive interventions on aggressive/disruptive behavior in childhood and adolescence: An application of latent transition growth mixture modeling. Prevention Science, 12(3), 300-313.  

Supplementary Materials

Becker, K. D., Bradshaw, C. P., Domitrovich, C., & Ialongo, N. S. Coaching teachers to improve implementation of the Good Behavior Game. Administration and Policy in Mental Health, 40(6), 482-493.  

Outcomes

Outcome 1: Conduct and problem behaviors
Description of Measures Conduct and problem behaviors were assessed using the following measures:
  • Teacher Observation of Classroom Adaptation--Revised (TOCA-R). A structured interview administered to teachers by trained assessment staff, the TOCA-R is designed to measure child behaviors that affect school and social adaptation. Using a scale from 1 (not at all) to 6 (always), teachers rate the frequency of child behaviors over the past 3 weeks on the following three subscales: authority acceptance/aggressive behavior (e.g., breaks rules, harms property, fights), social participation/shy behavior (e.g., plays with classmates, initiates interactions), and concentration (e.g., pays attention, stays on task, is easily distracted). Some analyses used a total TOCA-R score (a mean of all the subscale items), while others used only the score on the authority acceptance/aggressive behavior subscale. This instrument was administered in the fall of first grade and the spring of first, second, and third grades.
  • Peer Assessment Inventory (PAI). The PAI, designed to assess a child's adaptation to the demands of the classroom peer group, consists of 10 items taken from the Pupil Evaluation Inventory on the basis of their relevance to accepting authority/aggressive behavior, social participation/shy behavior, and likeability/rejection. For each item, a question is read aloud to the class (e.g., "Which children play alone a lot?"), and children circle the pictures of their classmates that they think are described in the question. Raw scores are converted to standard scores. Two composite scores from the PAI were used: bullying/victimization and social participation/shy behavior. This instrument was administered in the fall and spring of first grade.
  • Teacher Report of Classroom Behavior--Checklist Form (TRCB-CF). Based on the TOCA-R, the TRCB-CF is designed to obtain teacher reports of child conduct problems in the school setting. Each of 7 items presents a behavior consistent with DSM-IV criteria for conduct disorder that can be observed in school (i.e., starts physical fights with classmates, lies, hurts others physically, steals, damages other people's property on purpose, skips school, bullies classmates into getting his or her own way). The teacher rates the frequency of each behavior on a scale from 1 (not at all) to 6 (always). The teacher also reports whether the child has been suspended from school in the past year and whether the child is in need of mental health services. When more than one teacher assessed the child (e.g., English/language arts, reading, and math teachers in sixth grade), the teachers' ratings were averaged to create one TRCB-CF score, and responses to questions about suspension and mental health service needs were coded "yes" if any of the teachers responded "yes." This instrument was administered in the spring of 6th through 12th grades.
  • The Conduct Disorder Module of the Diagnostic Interview Schedule for Children (DISC-IV). The DISC-IV is a structured interview used to determine DSM-IV diagnoses. Child and parent reports on the Conduct Disorder Module were used to assess lifetime diagnosis of conduct disorder. A conduct disorder symptom was counted as present if either the parent or child reported it. This instrument was administered in the spring of sixth grade.
Key Findings Three first-grade classrooms in each of nine schools in Baltimore City were randomly assigned to one of three conditions:
  • PAX GBG. This condition included PAX GBG, provided over the duration of first grade, in combination with weekly classroom meetings to promote group problem solving, curriculum enhancements (new and supplementary curriculum materials in language arts and mathematics), and additional support for children who did not respond adequately to the intervention (e.g., tutoring, modifications in the curriculum to address individual learning styles).
  • Family-School Partnership (FSP). FSP, provided over the duration of first grade, included training for teachers, workshops for parents, and weekly home-school activities aimed to enhance parent-teacher communication and provide parents with effective teaching and child behavior management strategies. As FSP was not compared with PAX GBG in the analyses, no findings on FSP are presented.
  • Control, in which children received instruction as usual.
Findings from this study included the following:
  • In intent-to-treat analyses that controlled for pretest scores on the TOCA-R, PAX GBG students had fewer teacher-reported problem behaviors on the TOCA-R than control group students in the spring of first grade (p = .03 for boys; p = .01 for girls) and spring of second grade (p = .001 for boys; p = .0001 for girls).
  • In intent-to-treat analyses that controlled for pretest scores on the PAI, boys in the PAX GBG group had fewer peer nominations for aggression on the PAI bullying/victimization composite than boys in the control group (p = .02) in the spring of first grade. No significant effects were found for boys on the social participation/shy behavior composite, and no effects were found for girls on the bullying/victimization or social participation/shy behavior composite.
  • In the spring of sixth grade, relative to students in the control group, PAX GBG students had lower levels of teacher-reported conduct problems on the TRCB-CF (p < .001). According to teacher reports, PAX GBG students were also less likely to be in need of mental health services (odds ratio = 0.37, p < .05) and to have been suspended from school (odds ratio = 0.73, p < .05). PAX GBG students were less likely than control group students to have a lifetime conduct disorder diagnosis based on parent or child report on the DISC-IV Conduct Disorder Module (odds ratio = 0.42, p < .05).
  • In first through third grades, PAX GBG students were three times more likely than control group students to be in the low-aggressive/disruptive behavior trajectory (i.e., displaying low, stable levels of aggression) based on teacher reports on the TOCA-R authority acceptance/aggressive behavior subscale (odds ratio = 3.117, p < .01 for boys; odds ratio = 3.059, p < .05 for girls). A significant difference between groups was sustained into middle and high school for boys (odds ratio = 0.497, p < .01) but not for girls.
Studies Measuring Outcome Study 1
Study Designs Experimental
Quality of Research Rating 3.5 (0.0-4.0 scale)
Outcome 2: Academic success
Description of Measures Academic success was assessed using the following measures:
  • Comprehensive Test of Basic Skills (CTBS). This standardized achievement battery assesses verbal skills (word analysis, visual recognition, vocabulary, comprehension, spelling, and language mechanics and expression) and quantitative skills (computation, concepts, and applications). This instrument was administered in the fall of first grade and the spring of first and second grades.
  • Kaufman Test of Educational Achievement (KTEA). The KTEA is a measure of educational achievement assessing skills in reading, math, and spelling. Only the reading and math domains were used in the study. This instrument was administered in 12th grade.
  • Special education service use. Official records were provided by the school district for students who had an Individualized Education Program (IEP) at any time during 1st through 12th grades.
  • High school graduation. Data were obtained from the school district to determine whether the student had graduated from high school or had passed the General Educational Development (GED) test.
  • College attendance. When they were 19, participants were asked during a phone interview whether they had attended college (e.g., 4-year college, junior college).
Key Findings Three first-grade classrooms in each of nine schools in Baltimore City were randomly assigned to one of three conditions:
  • PAX GBG. This condition included PAX GBG, provided over the duration of first grade, in combination with weekly classroom meetings to promote group problem solving, curriculum enhancements (new and supplementary curriculum materials in language arts and mathematics), and additional support for children who did not respond adequately to the intervention (e.g., tutoring, modifications in the curriculum to address individual learning styles).
  • Family-School Partnership (FSP). FSP, provided over the duration of first grade, included training for teachers, workshops for parents, and weekly home-school activities aimed to enhance parent-teacher communication and provide parents with effective teaching and child behavior management strategies. As FSP was not compared with PAX GBG in the analyses, no findings on FSP are presented.
  • Control, in which children received instruction as usual.
Findings in early reading achievement, as measured by the CTBS, included the following:
  • In intent-to-treat analyses that controlled for pretest levels of reading achievement, boys in the PAX GBG group with pretest levels of reading achievement at or above the 40th normal curve equivalent (NCE) had higher levels of reading achievement than their counterparts in the control group in the spring of first grade (p = .01).
  • In the spring of second grade, boys in high-implementation PAX GBG classrooms (those with more than 50% of the intervention implemented with high fidelity) had higher levels of reading achievement than boys in the control group (p = .001).
  • No significant effects in reading achievement were found for girls.
Findings in early math achievement, as measured by the CTBS, included the following:
  • In the spring of first grade, boys in high-implementation PAX GBG classrooms with pretest levels of math achievement at or above the 40th NCE had higher levels of math achievement than their counterparts in the control group (p < .001).
  • In intent-to-treat analyses that controlled for pretest levels of math achievement, boys in the PAX GBG group with pretest levels of math achievement at or above the 60th NCE had higher levels of math achievement than their counterparts in the control group in the spring of second grade (p = .02).
  • In the spring of first grade, girls in high-implementation PAX GBG classrooms with pretest levels of math achievement at or above the 40th NCE had higher levels of math achievement than their counterparts in the control group (p = .01).
  • In the spring of second grade, girls in high-implementation PAX GBG classrooms with pretest levels of math achievement at or above the 60th NCE had higher levels of math achievement than their counterparts in the control group (p = .04).
Subsequent analyses to assess the longitudinal impact of PAX GBG controlled for academic readiness as measured by teachers on the TOCA-R in the fall of first grade. Findings included the following:
  • In 12th grade, PAX GBG students had higher reading and math performance than control group students as measured by the KTEA (p < .01 for reading; p < .01 for math).
  • PAX GBG students were about half as likely as control group students to have received any special education during 1st through 12th grades (odds ratio = 0.494, p < .05).
  • PAX GBG students were more likely than control group students to graduate from high school (odds ratio = 1.702, p < .05) and attend college (odds ratio = 2.222, p < .05).
Studies Measuring Outcome Study 1
Study Designs Experimental
Quality of Research Rating 3.3 (0.0-4.0 scale)
Outcome 3: Mental health service utilization
Description of Measures Mental health service utilization was assessed using the following measures:
  • Service Assessment for Children and Adolescents Parent Report (SACA-P). This structured interview, designed to accompany the DISC-IV, is used to obtain information from parents on lifetime child and adolescent mental health service utilization, including use of inpatient and outpatient services provided by mental health professionals. This instrument was administered in the spring of sixth grade.
  • School Mental Health Professional Report (SMHPR). This checklist, developed by researchers for use in the study, was completed by school mental health professionals who indicated whether they provided mental health services to a student in the past year and what type of services were provided (i.e., individual or group counseling, psychological assessment, consultation). This instrument was administered in the spring of sixth grade.
Key Findings Three first-grade classrooms in each of nine schools in Baltimore City were randomly assigned to one of three conditions:
  • PAX GBG. This condition included PAX GBG, provided over the duration of first grade, in combination with weekly classroom meetings to promote group problem solving, curriculum enhancements (new and supplementary curriculum materials in language arts and mathematics), and additional support for children who did not respond adequately to the intervention (e.g., tutoring, modifications in the curriculum to address individual learning styles).
  • Family-School Partnership (FSP). FSP, provided over the duration of first grade, included training for teachers, workshops for parents, and weekly home-school activities aimed to enhance parent-teacher communication and provide parents with effective teaching and child behavior management strategies. As FSP was not compared with PAX GBG in the analyses, no findings on FSP are presented.
  • Control, in which children received instruction as usual.
By the sixth grade, PAX GBG students were less likely to have received mental health services than control group students according to parent reports on the SACA-P (odds ratio = 0.53, p < .05) and school reports on the SMHPR (odds ratio = 0.56, p < .05).
Studies Measuring Outcome Study 1
Study Designs Experimental
Quality of Research Rating 3.4 (0.0-4.0 scale)
Outcome 4: Initiation of substance use
Description of Measures Initiation of substance use was assessed using standardized questions regarding first use of tobacco, alcoholic beverages, inhalants, marijuana, and other illegal drugs (i.e., powder cocaine, crack, heroin). Students self-reported their initiation of substance use through audio computer-assisted self-interview (ACASI) methods during the spring of sixth through eighth grades.
Key Findings Three first-grade classrooms in each of nine schools in Baltimore City were randomly assigned to one of three conditions:
  • PAX GBG. This condition included PAX GBG, provided over the duration of first grade, in combination with weekly classroom meetings to promote group problem solving, curriculum enhancements (new and supplementary curriculum materials in language arts and mathematics), and additional support for children who did not respond adequately to the intervention (e.g., tutoring, modifications in the curriculum to address individual learning styles).
  • Family-School Partnership (FSP). FSP, provided over the duration of first grade, included training for teachers, workshops for parents, and weekly home-school activities aimed to enhance parent-teacher communication and provide parents with effective teaching and child behavior management strategies. As FSP was not compared with PAX GBG in the analyses, no findings on FSP are presented.
  • Control, in which children received instruction as usual.
In an analysis that controlled for teacher-rated problem behaviors in the fall of first grade, PAX GBG students were less likely than control group students to have initiated use of tobacco (relative risk = 0.55, p = .013) and illegal drugs other than marijuana (relative risk = 0.29, p = .028) by eighth grade.
Studies Measuring Outcome Study 1
Study Designs Experimental
Quality of Research Rating 3.4 (0.0-4.0 scale)

Study Populations

The following populations were identified in the studies reviewed for Quality of Research.

Study Age Gender Race/Ethnicity
Study 1 6-12 (Childhood) 53.2% Male
46.8% Female 		86.8% Black or African American
13.2% White

Quality of Research Ratings by Criteria (0.0-4.0 scale)

External reviewers independently evaluate the Quality of Research for an intervention's reported results using six criteria:

  1. Reliability of measures
  2. Validity of measures
  3. Intervention fidelity
  4. Missing data and attrition
  5. Potential confounding variables
  6. Appropriateness of analysis

For more information about these criteria and the meaning of the ratings, see Quality of Research.

Outcome Reliability
of Measures 		Validity
of Measures 		Fidelity Missing
Data/Attrition 		Confounding
Variables 		Data
Analysis 		Overall
Rating
1: Conduct and problem behaviors 3.3 3.4 3.8 3.3 3.4 3.8 3.5
2: Academic success 3.2 2.9 3.8 3.1 3.3 3.6 3.3
3: Mental health service utilization 3.3 2.9 3.8 3.3 3.3 3.8 3.4
4: Initiation of substance use 3.3 3.3 3.8 3.0 3.3 3.8 3.4

Study Strengths

An impressive array of instruments was used to provide a comprehensive picture of the variables examined. Most of the measures used were established and tested by independent researchers and shown to have good reliability and validity or were adaptations of these tested measures. Implementation fidelity was maximized through training and certification for teachers, ongoing expert support, monthly meetings, standardized manuals, and support materials (e.g., outlines, checklists). Fidelity was assessed through the systematic collection of data using psychometrically tested measures of classroom setup, classroom observations, and reviews of classroom records. Five of the nine intervention classrooms were identified as high-implementation classrooms, and all but two of the nine intervention teachers implemented more than 50 percent of the protocol. There was minimal attrition, which is particularly noteworthy considering the period of follow-up and breadth of data collected. There were no significant differences between the intervention and control groups in rates of attrition or sociodemographic characteristics of children with missing data. Sophisticated methods were used to account for missing data. Classrooms were randomly assigned to condition, and there were no significant differences between intervention and control groups in age, gender, ethnicity, or free lunch status. Teacher differences at pretest were controlled for in subsequent analyses. Data analyses were sophisticated, and sample size and power were adequate.

Study Weaknesses

Several intervention classrooms did not implement the intervention with fidelity.
Readiness for Dissemination

Materials Reviewed

The materials below were reviewed for Readiness for Dissemination. The implementation point of contact can provide information regarding implementation of the intervention and the availability of additional, updated, or new materials.

Becker, K. D., Bradshaw, C. P., Domitrovich, C., & Ialongo, N. S. Coaching teachers to improve implementation of the Good Behavior Game. Administration and Policy in Mental Health, 40(6), 482-493.  

Becker, K. D., Darney, D., Domitrovich, C., Keperling, J. P., & Ialongo, N. S. Supporting universal prevention programs: A two-phased coaching model. Clinical Child and Family Psychology Review, 16(2), 213-228.  

Embry, D. PAX to the MAX manual for behavioral supports for users of the PAX Good Behavior Game. Baltimore, MD: Johns Hopkins Center for Prevention and Early Intervention.

PAXIS Institute. PAX Good Behavior Game reproducibles. Tucson, AZ: Author.

PAXIS Institute. Handouts for PAX Good Behavior Game training. Tucson, AZ: Author.

PAXIS Institute. PAX Plus Good Behavior Game kit. Tucson, AZ: Author.

PAXIS Institute. Purrfect PAX rubric. Tucson, AZ: Author.

PAXIS Institute. Logic model for PAX Good Behavior Game implementation and dissemination. Tucson, AZ: Author.

PAXIS Institute. Teacher training materials [USB drive]. Tucson, AZ: Author.

Set of five PAX posters

Willmann, M. Green County implementation survey.

Willmann, M. Green County spleems observation.

Willmann, M. Collecting and using PAX data for coaching PAX [PowerPoint slides].

Readiness for Dissemination Ratings by Criteria (0.0-4.0 scale)

External reviewers independently evaluate the intervention's Readiness for Dissemination using three criteria:

  1. Availability of implementation materials
  2. Availability of training and support resources
  3. Availability of quality assurance procedures

For more information about these criteria and the meaning of the ratings, see Readiness for Dissemination.

Implementation
Materials 		Training and Support
Resources 		Quality Assurance
Procedures 		Overall
Rating
4.0 4.0 4.0 4.0

Dissemination Strengths

Implementation materials are creative and engaging and contain a variety of tools to aid in the delivery of the intervention. Instructions for teachers are clear and materials are intuitive. The required training is supported by high-quality training materials and assorted multimedia trainer tools. An online learning community offers easily accessible opportunities for implementers to maintain skills and obtain support. Training and implementation materials emphasize quality assurance, particularly through the collection of outcome data that offer immediate feedback to teachers and partners. A comprehensive quality assurance protocol is provided that includes clear instructions for using all tools and guidance for interpreting results. Standard summary reports with graphics can help sites in reporting to financial stakeholders.

Dissemination Weaknesses

No weaknesses were identified by reviewers.
Costs

The cost information below was provided by the developer. Although this cost information may have been updated by the developer since the time of review, it may not reflect the current costs or availability of items (including newly developed or discontinued items). The implementation point of contact can provide current information and discuss implementation requirements.

Item Description Cost Required by Developer
PAX Good Behavior Game kit for school support staff and administration (includes the manual, Tootles note pads, portable scoreboard, OK/Not OK desk cards, Granny's Wacky Prizes booklet and box, My Wonderful PAX School reproducible masters, digital timer, PAX Quiet harmonica, PAX Stix, lanyard, pen, bracelets, and subscription to the online learning community and supports) $235 each Yes
PAX Plus Good Behavior Game kit for classroom teachers (includes materials in the PAX Good Behavior Game kit, scoreboards, pinch slates, PAX to the MAX music CD, PAX-It note pad, and large posters) $299 each Yes
Pocket apron (to carry PAX materials) $10 each No
T-shirt $11 each No
Small PAX Quiet harmonica $5 each No
1-day, on-site training $2,900 for up to 40 people, plus travel expenses Yes
1-day, on-site booster training $2,900 for up to 40 people, plus travel expenses No
Custom training and support packages Varies depending on site needs No
3-day, off-site training of trainers for PAX partners (held in various locations) $2,500 per person, plus travel expenses No
On-site custom training for PAX partners Varies depending on site needs No
Phone support Varies depending on site needs No
Annual Data Monitoring System Robust Results License $250 per school, plus $50 if off-line data entry is needed Yes