The Challenges of Conducting School-Based SEL Intervention Research: Strategies for Conducting Small Scale but High Quality Evaluations

Introduction

The purpose of this paper is to present a conceptual model of school-based SEL that can be used to guide the design of high-quality evaluations that will add to the intervention evidence-base. The focus is on small scale evaluations that are feasible to conduct with modest funding since this is what is more commonly available to non-profit organizations, individual program developers, or research teams conducting expensive assessment protocols or exploring innovative methods.  Such efforts, however, present new challenges to conducting rigorous intervention evaluations given smaller sample sizes, lack of true random assignment, and potential confounding factors at the individual, classroom, and school level that have been shown to influence students’ social and emotional development.

 

Methods

One strategy for managing baseline group differences among classrooms and schools is to create propensity scores that include multiple, relevant covariates. This score can either be included in models as a covariate to maximize the power of analyses or it can be used to create intervention and control groups that are matched on the propensity score to conduct group comparisons. This paper will present and discuss the propensity matching strategy with data from the current study.  This method produced results that both control for these contextual differences and also highlight the role they play in directly influencing outcomes of interest.

Results

Social and emotional skills are influenced most immediately by the social and learning characteristics of the classroom and school environment. As students learn social and emotional skills, it is important that they have opportunities to practice and apply the skills in actual situations and be recognized for using these skills across a variety of settings. Given the importance of practice for skill mastery and the influence of adults and peers outside of the school, most SEL interventions suggest ways to coordinate classroom instruction in social and emotional development with school, family, and community activities.

 

Conclusion

It is likely that a combination of improvements in student SEL skills, the school environment, teacher practices and expectations, and student-teacher relationships contribute to students’ behavior change and academic performance. The goal of presenting this model is to provide evaluation teams with a framework for measuring the most common confounds that undermine the validity of impact analyses that are conducted without the benefit of randomization.