Co-Principal Investigator: Woodbridge, Michelle

Purpose: This replication study will test the effectiveness of BEST in CLASS, a manualized tier 2 intervention designed to prevent and reduce challenging behaviors and promote appropriate behaviors in children ages 3–5. BEST in CLASS was developed and initially tested for efficacy by the developers with IES funding, and demonstrated that BEST in CLASS increases teachers' use of effective behavioral and instructional practices, management skills, and sense of self-efficacy to enhance the classroom quality of early childhood classrooms, as well as reductions in children's problem behavior and increases in positive teacher-child interactions and relationships. The purpose of this project is to test BEST in CLASS under routine conditions to determine how, where, at what cost, and with whom the intervention is likely to produce positive social, emotional, and behavioral outcomes.

Project Activities: The project will use a randomized controlled trial (RCT) with 360 children identified as at risk for social or emotional developmental delays who attend Head Start programs in California, Florida, and Virginia.

Products: This project will result in evidence of the efficacy of BEST in CLASS when implemented under routine conditions for young children in Head Start programs, as well as information about the cost and cost-effectiveness of the program. The project will also result in a final dataset to be shared, peer-reviewed publications and presentations, and additional dissemination products that reach education stakeholders, such as practitioners and policymakers.

Structured Abstract

Setting: The research will take place in diverse Head Start service areas in suburban and urban communities in California, Florida, and Virginia.

Population/Sample: The sample comprises 140 Head Start centers, 180 teachers, and 360 children (ages 3–5) identified as at risk for delays in social or emotional development.

Intervention: BEST in CLASS is a strengths-based, manualized intervention model that targets the prevention and amelioration of problem behaviors demonstrated by young children at risk for or with emotional/behavioral disorders in early childhood classrooms. This intervention combines evidence-based behavioral strategies with a coaching model designed to optimize implementation by teachers of preschool-age children. Intervention components include seven teacher learning modules and a coaching model. The modules include: (1) Basics of Behavior and Development; (2) Home-School Communication; (3) Rules, Expectations, and Routines; (4) Behavior Specific Praise; (5) Precorrection and Active Supervision; (6) Opportunities to Respond and Instructional Pacing; and (7) Teacher Feedback. The coaching model provides comprehensive instructions for coaches to support and enhance teachers' acquisition, implementation, and maintenance of the intervention strategies. Participating teachers receive a 1-day overview training followed by 14 weeks of classroom-based coaching

Research Design and Methods: Researchers will use a multisite clustered RCT, blocking by sites across three cohorts. Teachers will be the unit of random assignment. Within sites (Head Start centers), teachers will be randomized to BEST in CLASS or comparison condition. Cohorts (consisting of 30 intervention and 30 BAU teachers each) will participate in the study for one year, resulting in a total of 90 intervention and 90 BAU teacher participants. The immediate and short-term (i.e., one month following participation) effects of BEST in CLASS will be evaluated.

Control Condition: The comparison condition consists of business-as-usual Head Start classrooms with teachers who will not receive BEST in CLASS training and practice-based coaching.

Key Measures: Children's social, emotional, and behavioral problems and competencies The Social Skills Improvement System and the Caregiver-Teacher Report Form. Children's preacademic skills will be measured with the Bracken Basic Concept Scale: Receptive. Researchers will use the Classroom Assessment Scoring System, Pre-K to observe teachers' instruction and classroom quality. Teacher-child interactions and relationships will be observed with the Teacher-Child Interactions Direct Observation System, the Individualized Classroom Assessment Scoring System, and the Student Teacher Relationship Scale. Teacher self-efficacy will be assessed with the Teachers' Sense of Self-Efficacy Scale and Teachers' Efficacy Beliefs System–Self. Classroom characteristics as well as teacher and child demographics will also be collected. The team will also collect measures of cost, cost-effectiveness, and implementation fidelity.

Data Analytic Strategy: Researchers will conduct two-level hierarchical linear modeling to account for the multilevel, nested data. These analyses will include child and teacher covariates to reduce residual error and increase power. Analysis of covariance will be used to estimate program impact on teacher outcomes, controlling for baseline scores and other demographic characteristics. To test for mediators, structural equation modeling, test complete models, and partial mediation models will be used. to test for moderators, appropriate within- and cross-level intervention-by-covariate interaction terms will be added to the multilevel models.

Cost Analysis: Researchers will use the ingredients method to account for the total and net costs incurred during implementation, gather opportunity costs in both study groups, and derive a cost-effectiveness ratio to clarify the resources needed to produce the intervention's intended effects.

Related IES Projects: Promoting Social, Emotional, and Behavioral Competence in Young High-Risk Children: A Preventative Classroom-Based Early Intervention Model (R324A080074); Efficacy of the BEST in CLASS Intervention for Young Children at High Risk for Emotional and Behavioral Disorders (R324A110173)