Corrected: An earlier version of this story misidentified Joseph Durlak’s university. He is a professor emeritus of psychology at Loyola University Chicago.
Includes updates and/or revisions.
From role-playing games for students to parent seminars, teaching social and emotional learning requires a lot of moving parts, but when all the pieces come together such instruction can rival the effectiveness of purely academic interventions to boost student achievement, according to the largest analysis of such programs to date.
In the report, published Feb. 4 in the peer-reviewed journal Child Development, researchers led by Joseph A. Durlak, a professor emeritus of psychology at the University of Chicago, found that students who took part in social and emotional learning, or SEL, programs improved in grades and standardized-test scores by 11 percentile points compared with nonparticipating students. That difference, the authors say, was significant—equivalent to moving a student in the middle of the class academically to the top 40 percent of students during the course of the intervention. Such improvement fell within the range of effectiveness for recent analyses of interventions focused on academics.
Compared with their peers, participating students also significantly improved on five key nonacademic measures: They demonstrated greater social skills, less emotional distress and better attitudes, fewer conduct problems such as bullying and suspensions, and more-frequent positive behaviors, such as cooperation and help for other students. Also, the effects continued at least six months after the programs ended.
“We learned this is very practical for schools and doable in schools,” Mr. Durlak said. “There can be a payoff academically for these kids that compares to a lot of straightforward academic interventions, which is really sort of amazing.”
Social and emotional education seeks to provide a foundation for academic instruction by teaching students skills in self-awareness and self-management, getting along with others, and decision-making. Programs vary from afterschool courses taught by outside providers to schoolwide efforts incorporating curriculum, teacher professional development, school activities, and parent training.
For this study, researchers distinguished SEL programs intended to teach social skills broadly from programs focused on fixing specific behaviors, such as bullying.
The research team identified individual practices in 213 school-based studies, covering 270,034 students in kindergarten through 12th grade. About half the studies used randomized controlled trials. The team informally announced preliminary findings in 2007 that suggested SEL programs increased academic achievement. (“Social-Skills Programs Found to Yield Gains in Academic Subjects,” December 19, 2007.)
For the final report, the researchers narrowed the focus to include only programs that had no academic component, and were provided during the school day as universal programs, rather than those targeted to specific students with behavior problems.
The researchers found the SEL programs most likely to be effective followed what Mr. Durlak called the SAFE model: “sequenced,” step-by-step instruction; “active” learning, such as role-playing; with sufficient time “focused” on each lesson; and “explicit” learning goals. The most effective programs used all four practices together.
Lost Teaching Time
Corinne Gregory, the president and founder of the Seattle-based schoolwide SEL program SocialSmarts, suggested the improvement in those soft skills likely caused the rise in academic achievement, in part because educators could teach more efficiently with calmer, more cooperative students. A 2003 brief by the New York City-based policy-research group Public Agenda found teachers reported losing as much as 30 percent of instructional time to deal with behavior problems in class, and Ms. Gregory said some of the schools participating in the SocialSmarts program reported increasing students’ time on task by more than 40 percent.
She agreed with the researchers’ decision to test programs that focused on social-skills instruction, rather than preventing specific behaviors like bullying.
“We focus all of our efforts on that nasty endpoint of the social-emotional continuum, bullying, rather than preventing all the other problems that lead up to that,” such as disrespect in the classroom or cheating, Ms. Gregory said. “By then, it’s almost too late in the game.”
Yet one finding ran counter to both the researchers’ expectations and prior research: Simple teacher-led programs vastly outperformed multifaceted programs involving schoolwide activities and parent involvement. While classroom-based programs showed significant improvements across all five social measures and academics, comprehensive programs showed no significant effect on students’ social-emotional skills or positive social behavior, and were less effective at improving academic performance.
The finding bears echoes of an Institute of Education Sciences study released last fall, which found that seven of the nation’s most common character education programs failed to yield significant social or academic improvements.
Other previous studies have suggested students respond better to programs that reinforce social learning throughout the school and at home, but this study found that programs with more components were less likely to follow any—not to mention all—of the best practices of sequenced, active lessons based on explicit goals and sufficient focus on each goal.
“The more-comprehensive and broader programs tended to have more implementation problems,” Mr. Durlak said. “Trying to do more in the schools tends to be harder, takes more coordination, involves more people—they’re a lot harder to pull off.
SocialSmarts is such a comprehensive program, and while Ms. Gregory said it has shown positive academic and social effects, she acknowledged that coordinating implementation “has been a challenge.”
“Everyone from the janitor to the principal has to be responsible for this,” Ms. Gregory said. “We’ve had schools that said we can’t send this [parent activity] home; the parents won’t read it. I say well, you can be sure they’ll never read it if you don’t send it home.”
To offset the implementation issues, SocialSmarts requires schools to purchase a license for the program, rather than buying a curriculum. It also provides a newsletter, training for teachers and parents, and online platforms where teachers share practices.
Mr. Durlak said the researchers would continue to add to the research database, which now includes studies only as recent as 2007, and hopes to dig into the best practices of effective programs in more depth.
