Co-Principal Investigator: Erik Girvan

Purpose: The purpose of this project is to develop a professional development intervention (ReACT) to reduce racial/ethnic disproportionality in school discipline and special education referrals. Disproportionality in school discipline remains a long-term and pressing concern in education, and exclusionary discipline (e.g., suspensions, expulsions) exposes students to increased risk of academic failure, referrals for special education, and school dropout. Professional development is needed to help schools address these issues. Project ReACT will address this gap by developing a comprehensive program that is based on the promising theory of implicit bias in education (i.e., the unconscious attitudes or stereotypes that may affect teachers' understanding, action, and decision-making in schools) and provides educators with feasible strategies to increase equity in school discipline and special education referral decisions.

Project Activities: The ReACT intervention will be iteratively developed and pilot tested across three phases. In Phase 1, a range of national experts and school personnel will help to develop and refine ReACT and its training materials. In Phase 2, the research team will conduct a feasibility study of ReACT with all staff across two schools. In Phase 3, the researchers will conduct a randomized controlled trial to document the feasibility and potential efficacy of ReACT for reducing racial/ethnic disproportionality in school discipline and special education referrals.

Products: The products of this project will include a fully developed professional development program that reduces disproportionality in school discipline and referrals for special education, as well as peer-reviewed publications and presentations.

Structured Abstract

Setting: The project will take place primarily in racially and ethnically diverse urban elementary schools in Oregon. Educators from school districts in Nebraska, North Carolina, and Texas will also participate in usability testing.

Sample: Five national experts and approximately 200 educators in Nebraska, North Carolina, and Texas will participate in Phase I, and approximately 300 educators and their students in 10 public elementary schools in Oregon will participate in all phases of the project.

Intervention: ReACT includes a series of schoolwide in-person professional development trainings and ongoing coaching for teachers. The components of ReACT include (a) working with school leadership to examine their available school data to identify specific situations that increase the likelihood of implicit bias affecting discipline decision making, (b) adapting existing behavior systems to better meet the needs of students from various cultural backgrounds, (c) assisting all teachers in developing personal strategies to neutralize the effects of implicit bias on school discipline decisions, and (d) supporting teachers' use of these strategies through ongoing coaching and feedback.

Research Design and Methods: In Phases 1 and 2, the research team will use an iterative process to develop ReACT. To start, the researchers will form a Design Team with administrators and teachers from two elementary schools (Design Schools), who will participate in all phases of the research. Phase 1 will focus on developing ReACT, its training materials and fidelity measures, with input from national experts, personnel from a wide range of schools, and the Design Team. At the start of Phase 1, the Design team will examine ReACT and its materials (e.g., training content, fidelity of implementation measures) and provide feedback regarding ReACT component selection, usability, and refinement. The researchers will then conduct focus groups with two groups—Design School staff and national experts—to assess usability across a range of users. Phase 2 includes full implementation of the refined intervention in the Design Schools to assess usability, feasibility, fidelity of implementation; preliminary associations between ReACT and decreased disproportionality in office disciplinary referrals and suspensions; and teacher-reported motivation to avoid exhibiting bias. In Phase 3, the research team will assess the promise of ReACT through a small cluster randomized controlled trial, comparing outcomes related to disproportionality in schools receiving ReACT to those in the control schools.

Control Condition: In the randomized controlled trial, educators in control schools will engage in business-as-usual instruction, support, and professional development.

Key Measures: Measures will include semi-structured focus group interview protocols and surveys to assess the usability and feasibility of the intervention's content and delivery. Treatment acceptability will be assessed using the Primary Intervention Rating Scale. The research team will use researcher-developed tools to assess fidelity of training, coaching, and implementation of the ReACT professional development activities, as well as teachers' use of self-selected behavioral strategies to neutralize the effects of their implicit bias. Researchers will assess the primary outcomes using recommended metrics for calculating school- and teacher-level racial disproportionality in (a) office discipline referrals, (b) suspensions, (c) referrals for special education evaluation, and (d) special education placement (e.g., percent of Black students with emotional and behavioral disorders [EBD] divided by percent of White students with EBD). Additional teacher outcome measures include the Motivation to Respond without Prejudice scale to assess teacher motivation and the Affect Misattribution Procedure to assess implicit bias.

Data Analytic Strategy: Focus group data from Phase 1 will be analyzed using qualitative data analytic techniques (i.e., the critical incident technique). Survey and observational data from Phases 1 and 2 will be analyzed using descriptive and inferential statistical techniques, including means, standard deviations, and paired t-tests. Data from the Phase 3 randomized controlled trial will be analyzed through mixed-model analysis of covariance to assess effects of the intervention on disproportionality outcomes. These analyses will account for nesting of teachers within schools.