Challenges and Future Directions in Measuring Social and Emotional Learning Practices in Schools

• In Event: What We Learned Measuring Educator Well-Being During the 2021–2022 School Year: Social and Emotional Learning Assessment Science and Practice

Abstract

Although the development of SEL programming and the measurement of SEL competencies has been growing at a fast pace, the development of practical measures to assess SEL implementation has significantly lagged behind. This is concerning given the essential role of implementation in promoting effective SEL. Without accurate measures of SEL implementation, we cannot assess which the components of SEL practice are most important. In this paper, we focus on the challenges and suggest future directions regarding creating practical measures of social and emotional learning (SEL) practices in schools. We examine this issue based on our experience with a multi-year process of developing a self-report measure of emotion-focused SEL practices that has gone through multiple iterations. Initially, our item development and statistical modeling was guided by a theoretical approach undergirding an established emotion-focused SEL program (RULER). This theoretical approach focused on capturing the core elements of practices aimed at producing positive SEL and climate outcomes (positive classroom climate, emotional literacy, emotion self-regulation, and social problem-solving; see Table 1). Importantly, the measure is designed to be program-neutral in that survey items use no program-specific language. Method. We used multilevel confirmatory factor analysis to assess the fit of the data with the aforementioned theoretical model. Results. Contrary to fitting a multidimensional model, we found that the measure had inadequate differentiation between constructs with substantial cross-loadings. We revised items to increase differentiation and re-tested the outcomes-focused model. Again, this approach again resulted in weak differentiation and with substantial cross-loadings of items on constructs. Finally, using the same dataset and data analytic approach, we tested an alternative theoretical model focused on pedagogical modality. These domains consisted of: a) describing personal and real-life experiences; b) discussing class expectations and rules; c) identifying and strategizing SEL strategies; d) teaching SEL strategies; d) using a display tool; e) demonstrating/modeling social emotional skills; f) sharing social-emotional experiences; and g) using strategies in the moment. This resulted in even less differentiation between constructs than the first approach, again suggesting that the measure may best be represented by a single dimension. Significance. Our findings challenge some existing theoretical models of SEL implementation and suggest a need to consider other conceptualizations, methods, and means of analysis. We discuss the implications of these findings for the future of measuring SEL practices, including: (1) exploring other theoretical models that may better describe SEL instruction; (2) other data analytic approaches, such as person-centered analyses, that may shed new light on SEL practices; and (3) the pros and cons of a unidimensional measure.

Authors

• Michael Strambler, Yale University
  Presenting Author

• Almut K. Zieher, Yale University
  Presenting Author

• Joanna L. Meyer, Yale University
  Non-Presenting Author

• Maegan Genovese, Yale University
  Non-Presenting Author