Following Teachers to Test the Impact on Subsequent Student Cohorts: Effects of the GBG+MTP Program

Introduction: While many teacher training programs and classroom-based preventive efforts are explicitly or implicitly intended to have impacts that last to affect subsequent cohorts of students, there is a remarkable absence of studies that test this hypothesis. Almost all existing studies that are longitudinal follow only the students in the classroom at the time of the training. This inconsistency between theory of impact and modeling of effects could be an important limitation of school-based prevention. This presentation will focus on the importance of following teachers to track sustained effects of prevention/training efforts utilizing data from a randomized controlled trial testing the Good Behavior Game (GBG) when integrated with the MyTeachingPartner (MTP) model.

Methods:

Data from this study come from a group randomized trial of new teachers in three school systems, randomly assigned to serve as controls or receive training in an integrated program comprised of the two empirically tested components. Specifically, the trial provided training to new teachers (n =206) in grades K through 3 in GBG and related behavior management skills and leveraged coaching using the MTP model. Initial training consisted of a 3-day workshop plus ongoing coaching intended to support fidelity of implementation and use of sound teaching practices. We hypothesized that training in the GBG+MTP preventive intervention would not only benefit students in the classroom during the year of active training and coaching, but would result in sustained changes in the teachers’ practices beyond the intervention year; those impacts were intended to also improve outcomes for subsequent cohorts of students.

Results:

Analysis of data collected on students in the classroom the year post training will be utilized to test for differences in teacher motivation (distress related to teaching, motivation to teach), teacher classroom management practices (benchmarks drawn from program training foci), and related classroom student behavior (level of disruptive behavior, on-task behavior, and engagement in learning).

Conclusions: These findings build on our prior results regarding the immediate (i.e., pre-post) benefits of GBG+MTP on academic performance and student behavior in classrooms marked by initial levels of elevated student disruptive behavior. Key issues in statistical modeling of such longitudinal effects and practical considerations in design of school-based prevention trials will also be highlighted using data from this RCT.