How much do SEL programs actually reduce behavior problems?
The short answer is that SEL programs produce a small but meaningful reduction in problem behaviors on average, but the range is wide. A 2024 meta-analysis of 12 high-quality SEL programs involving over 83,000 students found an overall effect size of 0.15 for reducing antisocial behaviors [1]. To put that in plain terms: if a typical classroom had 10 students with noticeable behavior issues, an effective SEL program might bring that down to about 8 or 9 — a modest but real improvement across a whole school. The same meta-analysis also found that programs improved prosocial behaviors, academic performance, and social-emotional skills [1].
Some programs show much larger effects. The Social-Emotional Learning Foundations (SELF) curriculum for kindergarten and first graders at risk for emotional and behavioral disorders produced effect sizes ranging from 0.20 to 0.65 on measures of self-regulation and general behavioral functioning [2]. That means the most at-risk students in the SELF program improved by more than half a standard deviation compared to peers who received usual instruction — a substantial difference that could mean the difference between needing special education support and not.
However, not all findings are positive. A large randomized trial of a school-wide SEL program in Brazil found that after two years of implementation, students in the program actually showed slightly more behavior problems (effect size 0.09 to 0.12 in the wrong direction) and worse working memory compared to control schools [3]. The researchers noted that teachers who delivered more lessons saw worse outcomes, suggesting that poor implementation or cultural mismatch may backfire.
What separates effective SEL programs from ineffective ones?
The evidence points to three key factors: teacher training, program dosage, and targeting the right students. The 2024 meta-analysis found that training teachers' own social-emotional skills was a critical component for success, while simply adding more pedagogical activities or family engagement did not boost effectiveness [1]. Programs that delivered the recommended number of lessons (adequate dosage) were significantly more effective than those with low dosage [1]. This suggests that a well-trained teacher delivering a complete program matters more than piling on extra activities.
Targeting matters too. The SELF curriculum was specifically designed for kindergarten and first graders already identified as at risk for emotional and behavioral disorders, and it showed strong effects [2]. In contrast, universal programs that try to help everyone may show smaller or even null effects. For example, a Brazilian study of a universal SEL program in high-violence communities found no average benefit on behavior problems after one year, though there were small positive effects in low-violence areas [5]. This suggests that SEL programs may need to be adapted for high-stress environments.
A study of the CharacterStrong curriculum in middle schools found that the program reduced disruptive behavior and improved academic engagement, but the benefits were largest for students who had the most room for improvement [4]. This is an important nuance: universal programs may not help well-behaved students much, but they can make a real difference for those struggling the most.
Can SEL programs help with behavior problems after major disruptions like COVID-19?
Yes, and the evidence suggests SEL programs may be especially valuable after events that disrupt students' social and emotional development. A 2024 study of teachers returning to in-person schooling after COVID-19 closures found that teachers overwhelmingly reported worsened student behavior, but those from schools with SEL programs were significantly less likely to report academic motivation struggles [6]. This suggests that SEL programs may act as a buffer against the behavioral fallout from major disruptions.
The same study found that teachers reported academic motivation problems most often, followed by social and emotional struggles [6]. Schools with SEL programs in place before the pandemic seemed better equipped to help students re-engage with learning. This aligns with the broader finding that SEL builds foundational skills that help students cope with stress and change.
Sources used in this answer
Effective Components of Social Emotional Learning Programs: A Meta-analysis
A meta-analysis of 12 high-quality SEL programs (83,233 participants) found an overall effect size of 0.15 for reducing antisocial behaviors, with teacher training and adequate dosage as key success factors.
Efficacy of the social-emotional learning foundations curriculum for kindergarten and first grade students at risk for emotional and behavioral disorders
The SELF curriculum for at-risk K-1 students produced positive effects on self-regulation and behavior (effect sizes 0.20 to 0.65) compared to business-as-usual classrooms.
Exploring the Effects of a Social-Emotional Learning Intervention in Brazilian Primary Schools: Findings from Year Two of Implementation
A Brazilian SEL program showed small iatrogenic effects (d = 0.09-0.12) on behavior problems in year two, with more lessons linked to worse outcomes.
A multi-site cluster randomized trial of CharacterStrong’s Secondary Curriculum in middle schools
CharacterStrong curriculum in middle schools reduced disruptive behavior and improved academic engagement, especially for students with the most room for improvement.
Um Compasso Para Aprender: A Randomized Trial of a Social-Emotional Learning Program in Homicide-Affected Communities in Brazil
A Brazilian SEL program showed null average effects on behavior problems after one year, with small positive effects only in low-homicide communities.
The Return to In-Person School: Teacher Reports of Student Behavior and Social–Emotional Learning
Teachers from schools with SEL programs were significantly less likely to report academic motivation struggles after COVID-19 school closures.