Long-Term Effects of Social-Emotional Learning on Academic Skills: Evidence from a Randomized Trial of INSIGHTS

Abstract

Social-Emotional Learning (SEL) programs are school-based preventive interventions that aim to improve children’s social-emotional skills and behaviors. Although meta-analytic research has shown that SEL programs implemented in early childhood can improve academic and behavioral outcomes in the short-term, there is limited work examining program effects on children’s math and language skills in the longer-term. Moreover, few studies have considered variation in impacts by children’s pre-intervention academic skills. Using an experimental design, the current study leveraged administrative data available through school records (N = 353) to examine the impacts of one SEL program—INSIGHTS into Children’s Temperament—implemented in early elementary school on math and language standardized test scores from third through sixth grade. Findings revealed positive average treatment effects on English/Language Arts (ELA) test scores in third and fourth grade, but not in fifth and sixth grade. Students who had higher academic skills at study enrollment showed lasting impacts on ELA scores in fourth, fifth, and sixth grade. There were no treatment impacts on math skills, and no variation in effects on math achievement by baseline skills. Implications are discussed.

Keywords:

• Social-emotional learning
• intervention
• language
• math
• baseline risk

Acknowledgments

The authors thank James Kemple and Patricia Chou from the Research Alliance for NYC Schools for all their work in helping the research team access the data used for this study. The authors also thank the NYC Department of Education for sponsoring and supporting this study for the past 13 years.

Notes

1 In their meta-analysis of SEL program impacts (N = 213), Durlak and colleagues (2011) found that 114 (53%) were primarily implemented by teachers, 44 (21%) were implemented by non-school personnel, and 55 (26%) were multi-component. Effects on academic performance were largest and statistically significant in the studies where teachers implemented the model directly (E.S. = .34) followed by multicomponent interventions (E.S. = .26). The effects of interventions on academic performance implemented by non-school personnel were not statistically significant (E.S. = .12). In the Taylor et al. (2017) meta-analysis examining long-term impacts of SEL programs, of the 82 studies reviewed, 32 (39%) examined programs implemented by teachers, 27 (33%) examined programs implemented by non-school personnel, and 23 (28%) examined multi-component interventions. The authors did not find any evidence that long-term program effects varied by intervention format.

2 Cohort 1 students enrolled in the study in kindergarten in the fall of 2008 and were enrolled in 3rd, 4th, 5th, and 6th grades between 2011–2012 (3rd grade) and 2014–2015 (6th grade). Cohort 2 students enrolled in the study in kindergarten in the fall of 2009 and were enrolled in 3rd, 4th, 5th, and 6th grades between 2012–2013 (3rd grade) and 2015–2016 (3rd grade). Cohort 3 students enrolled in the study in kindergarten in the fall of 2010 and were enrolled in 3rd, 4th, 5th, and 6th grades between 2013–2014 (3rd grade) and 2016–2017 (6th grade).

3 This distribution may be unique to NYC where students have middle school choice and there is a public transportation infrastructure to support students to attend middle schools across the full city.

Additional information

Funding

The research reported here was conducted as a part of a study funded by Grant R305A160177 from the Institute of Education Sciences to New York University with a subcontract to MDRC. The opinions expressed are those of the authors and do not represent views of the Institute or the U.S. Department of Education.