Coaching Teachers to Improve the Implementation Quality of Evidence-Based Programs: Linking Coaching with Fidelity in the Paths to Pax Project

Introduction: Although randomized trials suggest that coaching is more effective than the absence of this form of support, well-articulated guides of the coaching practices that support teacher implementation of prevention programs are lacking in the literature (Stormont et al., 2012). It is also unclear whether all teachers benefit equally from coaching or if coaching needs to be tailored in terms of types of supports, amount of time, or number of contacts. The primary goal of the current paper is to describe the practices employed by a team of coaches aiming at optimizing implementation of two widely-used classroom-based prevention programs: the PAX Good Behavior Game (PAX GBG; Embry et al., 2003) and the Promoting Alternative Thinking Strategies curriculum (PATHS; Kusche & Greenberg, 1995). A second goal of this paper was to examine the association between specific coaching activities and improvement in implementation quality over the course of the school year.

Method: Data came from a randomized trial contrasting use of the PAX GBG alone and in combination with the PATHS curriculum (referred to as PATHS to PAX), relative to a control condition. Teachers received training and on-site support for the interventions from coaches who worked for the project and were familiar with the interventions. This study explored the association between coaching and the implementation of the PAX GBG by 129 urban elementary school teachers. Data on coaching contacts with teachers included the duration of the contact and the practices employed by coaches (e.g., modeling) over the course of a school year. Teacher implementation quality was assessed through observations of teachers delivering their assigned interventions through 7 months post-training.

Results: Analyses indicated that coaches strategically varied their use of coaching strategies (e.g., modeling, technical assistance/feedback, needs assessments) based on teacher implementation quality and intervention condition. In general, teachers delivering the PATHS to PAX intervention with low quality implementation received the most coaching. Additionally, particular coaching practices were associated with improved implementation quality of the PAX GBG. Specifically, less time spent modeling of general teaching or behavior management practices and more time spent using “other” coaching activities were associated with improved implementation quality of the PAX GBG.

Discussion: Together these findings are intended to inform the implementation of coaching models, and the potential for tailoring coaching efforts to optimize impact. This study lays the foundation for future research examining ways to enhance coach decision-making about teacher implementation.