Professional Development to Support and Sustain a Classroom Behavior Management Strategy
Co-Principal Investigators: Hendricks Brown, Anja Kurki, Carla Ford, and Wei Wang
Purpose: The Good Behavior Game (GBG), a fully developed classroom behavior management strategy, is one of the few preventive interventions targeting aggressive/disruptive behavior that has shown both short-term and long-term improvements for classroom behavior in randomized field trials. In this study, the research team addresses a fundamental challenge in moving research to practice, the need to understand the level of professional development required for teachers to implement and sustain new program practices so that benefits are seen both for their current students and for students they teach in subsequent years.
Project Activities: Using a randomized control trial, the research team will test the relative efficacy and sustainability of the GBG under two models of professional development compared to one another and a control condition. In the first professional development (PD) model, GBG Basic provides group-based pre-implementation training to teachers supplemented by a group-based booster session. In the second more intensive PD model, GBG w Coach, the group-based activities are enhanced by a coach who works directly in the classroom with the teacher. Within each of 20 schools, two consecutive cohorts of first-grade children are randomly assigned to all first-grade classrooms in the fall of the school year. Teachers are then randomly assigned to GBG Basic, GBG w Coach or a control group and remain in their assigned group for 2 years to determine which PD model is most conducive to sustaining high fidelity GBG practices over time. Effects of GBG under these two PD conditions on student behavior and teacher practices will be measured and described.
Products: The expected products of this study include published reports on the relative benefits of providing coaching to teachers in addition to the standard group-based training for learning and implementing the GBG with high levels of fidelity. Such information has the potential to provide key information about the necessary support and costs involved to both introduce and maintain high-quality implementation of effective interventions. Such information is critical for bringing effective interventions to scale.
Setting: Twenty elementary schools in a large public school district in Texas are participating in the study.
Population: Participants are 60 first-grade teachers and approximately 3,000 children.
Intervention: The GBG is a behavior management strategy used with all children in a classroom to reduce aggressive, disruptive behavior and promote on-task behavior. The "game" has four core components. (1) Teachers begin by explicitly identifying acceptable and unacceptable classroom behavior. Classroom rules are posted on a wall chart and teachers explain how certain behaviors represent violation of these rules. (2) Once teachers have had a few weeks to observe their students and establish classroom rules, teachers divide students into several teams five to eight children that are heterogeneous with respect to observed behaviors and engagement with learning, and with equal representation of boys and girls. (3) Over the course of the school year, students and their teacher play the game during periods of independent work. Teams do not compete against one another; rather, children must work cooperatively to maintain good behavior within their teams to win the game. When a student breaks a classroom rule, the teacher identifies the behavior and the rule that was violated and gives a point to that student's team. When the game is over, each team that has four or fewer points is rewarded. (4) Positive reinforcement is the final key element of this intervention. At the start of the year, teachers use tangible rewards such as school supplies or small toys to reward teams. As the school year progresses, rewards become more intrinsic and natural to the school setting, such as extra time to read during the school day.
Teachers will learn how to implement these GBG practices under one of two models of professional development. GBG Basic involves a two-day group-based training session for teachers that occurs at the start of the first year of implementation. During this session, teachers learn the four core elements through demonstrations of strategies and procedures and guided practice of GBG activities. Teachers also learn techniques for generalizing GBG practices to settings outside the classroom, such as the hallway or lunch room. This initial pre-intervention workshop is supplemented with 6 hours of booster training mid-year. In GBG w Coach, the group-based activities are enhanced by a coach who works directly in the classroom with the teacher throughout the school year. This coach acts as a mentor to the teacher and provides technical support on good implementation practices. In the first half of the school year, the coach works with the teacher in the classroom for 90 minutes every other week. In the second half of the year, coaching is individualized based on the needs of the teacher. GBG w Coach is most like the professional development used in previous prevention trials of the GBG in which positive benefits were seen relative to a control group. GBG Basic is more like the professional development that schools and districts request when implementing a new program, and has not yet been tested in a randomized control trial.
Research Design and Methods: Two levels of random assignment are used to evaluate the relative efficacy of the two PD models; within each school children are first randomly assigned to classrooms, then teachers/classrooms are randomly assigned to the three study conditions: GBG w Coach, GBG Basic, or a control condition. Teachers will maintain their assigned condition for 2 years to evaluate sustainability of practices over time. The research team will select schools within the district that have the highest rates of discipline problems and the lowest academic achievement to provide the most stringent test of GBG Basic against GBG w Coach. Classroom observations of student behavior and teacher classroom management practices will be conducted three times per year (baseline, midpoint, and end of year), teachers will be interviewed twice each year (fall and spring), and school records will be collected at the end of each school year.
Control Condition: Teachers in the control group receive whatever the district usually provides in the way of professional development with regard to classroom behavior management.
Key Measures: Key outcomes are student behavior and teacher practices related to classroom behavior management. Data will be collected through classroom observations, teacher interviews, and school records. Student behavior will be assessed using the Teacher Observation of Classroom Adaptation (TOCA-R), a modified version of the classroom observation schema developed by the University of Texas' Center for Academic and Reading Skills (CARS), and school records of referrals for discipline and special education and mental health services. Teacher practices will be measured using the modified version of the CARS classroom observation schema and the Good Behavior Game Fidelity Checklist observation form. Implementation fidelity will be assessed using The Good Behavior Game Fidelity Checklist, teacher logs on game play, coaches' training logs to document participation in training and on-site coaching, and The Coaching Practices Checklist. Moderators of intervention impact will be assessed using The Teacher Sense of Self-Efficacy Scale, The Behavior Management Self-Efficacy Scale, The Classroom Atmosphere Rating Scale, The Organizational Climate Description Questionnaire for Elementary, and school district archival data on student mobility.
Data Analytic Strategy: The central questions will be addressed within the framework of generalized additive mixed models which allow for intervention gains to vary smoothly as a function of baseline risk, and to accommodate potential non-linearity, non-normality and clustering effects at the classroom and school levels. Growth models and structural equation models will be used to assess change in children's behavior over the school year, and to determine how teachers' practices change over time. Variation in the impact and sustainability of the GBG will be examined mainly by mediation analysis. A sensitivity analysis will be used to understand variation in program costs such as trainer/consultant fees and operating/administrative costs between the study conditions.
Publications from this project:
Poduska, J., Gomez, M., Capo, Z., Holmes, V. (2011). Developing A Collaboration With The Houston Independent School District: Testing The Generalizability Of A Partnership Model. Administration and Policy in Mental Health and Mental Health Services Research, 39 (4): 258–267.