The Growth of Social-Emotional Competence Among Diverse Subgroups of K-2 Students: Quasi-Experimental Evidence of Intervention Effect on Developmental Trajectories

Background/Purpose. The implementation of universal social and emotional learning (SEL) programs in schools creates new opportunities for school social workers to become key contributors to achieving the Grand Challenge to Ensure Healthy Development for All Youth. This Grand Challenge aims to reduce not only the overall incidence and prevalence of youth behavioral health problems, but also racial and socioeconomic disparities in behavioral health problems through effective preventive interventions. Although research evidence supports the benefits of universal SEL among the overall student body, the question of whether it works equally across diverse student subgroups remains unanswered. Some claim that universal SEL may reduce disparities, believing that it provides more benefits to marginalized students experiencing less favorable conditions for optimal social-emotional development. Others argue that it might maintain or exacerbate existing disparities, criticizing SEL for reflecting white, middle-class values that reinforce conformity, assimilation, and structures of dominance. In the context of this debate, the current study (1) explores the extent to which disparities in social-emotional competence (SEC) exist by student gender, age, race/ethnicity, English language learner (ELL) status, special education (SPED) status, and free/reduced-price lunch (FRL) status, and (2) examines the extent to which an SEL intervention disrupts or maintains any observed disparities.

Methods. Data come from the 2015-16 TOOLBOX Implementation Research Project, a quasi-experimental study of the school-based universal SEL program, TOOLBOX™. The sample consists of 1,766 K-2 students within the same school district (66.3% under TOOLBOX condition; 48.6% female; Mean baseline age = 6.05 years; 55.3% Hispanic, 13.6% Asian, 10.9% Black, 7.5% White; 48.2% ELL; 8.0% SPED; 67.8% FRL). Student SEC was measured three times over the academic year using the DESSA-Mini, a teacher-completed behavior rating scale. Multigroup latent growth modeling was conducted to compare student SEC growth trajectories across sociodemographic subgroups between TOOLBOX and non-TOOLBOX conditions. The Wald test of parameter equality was conducted to examine whether the sociodemographic differences in SEC growth trajectories differed by intervention condition.

Results. Under both conditions, girls (vs. boys) and older (vs. younger) students had higher baseline SEC; Black (vs. full-sample average) and SPED (vs. non-SPED) students had lower baseline; and Asian students (vs. full-sample average) showed more growth. Only Black students showed a statistically different growth rate between two conditions (χ²(1)=3.73, p=.05). In the intervention condition, Black students' growth rate did not differ from the average, whereas Black students in the comparison condition showed less-than-average growth.

Conclusions/Implications. This study suggests that student SEC growth trajectories can differ by gender, age, race/ethnicity, and SPED status. It also suggests that a universal SEL intervention may prevent disparities from widening for Black youth throughout the school year. This could imply that universal SEL might enable some marginalized racial groups, who otherwise might have shown a less favorable social-emotional developmental trajectory, to grow at the same rate as their classmates. We call for more rigorous replication studies and the general disaggregation of intervention effects by student sociodemographic subgroups. Just examining “average” outcomes is likely insufficient to the strategic pursuit of greater equity in youth behavioral health.