Evaluating the Impact of a Kernel Designed to Build Executive Function Skills: A Randomized Trial of Brain Games

• In Event: 2-209 - Contextual Influences Promoting Resilience for Low-income Children: Evidence from Longitudinal and Experimental Data

Integrative Statement

Comprehensive social-emotional learning (SEL) programming is found to improve children’s social, emotional, behavioral, and academic outcomes as well as teacher practices and the culture and climate of schools (Durlak et al., 2011; Jones et al., 2017). Yet, SEL programs face a number of barriers that undermine efforts to effectively go to scale, including lack of resources and poor integration into standard practices and therefore low sustainability. There is a pressing need to develop and test less intensive SEL strategies that are easy to implement outside the context of a comprehensive program while still achieving meaningful outcomes for children – what we refer to as SEL kernels of practice (Jones & Bouffard, 2012; Jones, Bailey, Brush, & Kahn, 2017).

In this paper we describe findings from a randomized control trial (RCT) of one such kernel called Brain Games (BG) which was designed to build children’s executive function (EF) and self-regulation skills. This kernel was drawn from a comprehensive, school-based SEL intervention called SECURe. Studies of SECURe showed positive effects on classroom quality and indicated that SECURe classrooms, on average, had more children meeting benchmarks in cognitive, literacy, and social-emotional domains (Jones, Jacob, & Bailey, 2014). Importantly, teachers reported they rarely had time to complete SECURe lessons, but instead used a variety of SECURe routines, most often BGs. Over 75% of teachers played BGs at least once, and over 25% played four or more times, per week; and growth in EF skills partially mediated the impact of SECURe on children’s standardized test scores of math and literacy (Jones, Jacob, & Morrison, in preparation).

Building on these findings, in 2017-18 we conducted an RCT of Brain Games. The study was conducted in nine urban public and charter schools serving low-income Prek-4th grade students (n=1,650). Fifty-six classrooms/teachers were randomized to either the treatment or control group. All teachers assigned to the treatment group were trained in fall 2017 and implemented BGs for the full year. In the fall, winter, and spring, we collected the following data: (a) child direct assessments (n=330, a random subsample of children in classrooms) of EF skills, using a battery of measures (e.g., Dimensional Change Card Sort); (b) classroom observations conducted by trained blinded observers, focusing on classroom structure and management, discipline, teaching practices, and the percentage of children displaying well-regulated behavior; (c) teacher reports of EF and self-regulation skills for each child in their classroom; and (d) weekly implementation logs for BGs.

In this paper we address the following questions: (1) What is the impact of BGs on executive function skills and other key outcomes (e.g., discipline, teacher-student relationship, academic achievement) in spring of the intervention year controlling for fall scores? (2) Is BG implementation (e.g., the # and type of games) associated with variation in impact over time?

Implementation data reveal that teachers played, on average, three BGs per week, with the highest levels of implementation in the early grades. Preliminary results suggest positive effects of BGs on classroom practice, classroom regulation, and teacher-reported student EF skills.

Authors

• Sophie Barnes, Harvard Graduate School of Education
  Presenting Author

• Rebecca Bailey, Harvard Graduate School of Education
  Non-Presenting Author

• Stephanie Jones, Harvard Graduate School of Education
  Non-Presenting Author