This paper explores social-emotional learning (SEL) practices in ten middle schools and their impact on students' outcomes. Despite the growing recognition of the importance of SEL, many schools struggle to implement effective strategies. The paper focuses on schools with strong student-reported data on SEL outcomes, particularly for African American and Latinx students.

This paper uses mixture IRT models to evaluate whether students experience confusion due to negative wording of items on a social-emotional learning assessment. It examines the consequences of confusion on student- and school-level scores and correlations with student-level variables, finding that confusion is present and most prevalent in lower grades, and is related to reading proficiency and ELL status. The study recommends the use of positively oriented items in future SEL measures, and suggests bias corrections based on the studied mixture model to maintain measurement continuity.

This report discusses the validity and reliability of CORE Districts’ social-emotional learning (SEL) student-report surveys. Through a pragmatic approach, the report answers four guiding questions that explain different facets of validity for school leaders. The aim is to provide guidance around the use of SEL surveys within and outside of the CORE districts to facilitate decision-making for educational leaders.

The use of social-emotional learning (SEL) and school culture/climate (CC) measures is a promising way to understand school performance. SEL and CC measures are reliable, distinguish between schools, and relate to academic and non-academic measures. They can identify areas for improvement within schools, such as subgroup gaps. Incorporating these measures into higher stakes accountability systems requires further research.

This report presents benchmarking data on self-report student surveys measuring social-emotional learning (SEL) from nearly half a million students in grades 4 through 12 across 8 CORE districts in California. The data provide means and standard deviations by construct, grade level, and subgroup, and can serve as a proxy for a nationally-normed sample for other schools across the country looking to administer the CORE survey.

This study explores ten "outlier schools" in California's CORE districts that have strong social-emotional learning outcomes. The brief and infographic summarize the various practices found in these schools and the common implementation challenges faced. The findings offer lessons that can help other schools and districts implement social-emotional learning at scale.

This study examines how social-emotional skills develop from Grade 4 to Grade 12 and vary by gender, socioeconomic status, and race/ethnicity. Based on self-report student surveys administered to around 400,000 students in California, the study finds that social-emotional skills do not increase uniformly and vary across subgroups. Females have higher social awareness but lower self-efficacy than males. Economically disadvantaged students show improvement in high school. White students report higher social-emotional skills than African American and Latinx students.

This study examines the impact of target setting for non-academic indicators on school quality ratings in CORE districts' elementary schools. The authors found that non-academic measures of school quality vary over time, are not consistently associated with demographics, and are sensitive to small changes in rating thresholds. The results suggest a need for more research on target setting for non-academic indicators in school quality systems.

This report examines the stability of school effects on social-emotional learning (SEL) over two years in California's CORE districts. The correlations among school effects in the same grades across different years are positive but lower than those for math and ELA. While these effects measure real contributions to SEL, their low stability draws into question whether including them in school performance frameworks and systems would be beneficial.

This paper investigates the effect of student satisficing, the act of providing suboptimal survey responses, on data quality in a large-scale social-emotional learning survey. The study examines the prevalence and impact of satisficing among 409,721 students and finds that its effect on data quality is less significant than expected. The paper offers a practical approach to defining and calculating satisficing, useful for researchers, policymakers, and practitioners working with large-scale datasets.

This study examines the properties of California's CORE Districts' SEL survey, which measures social-emotional learning in students. The survey was given to over 400,000 students in grades 3-12. The study uses both classical test theory and item response theory frameworks to analyze the data and make recommendations for modeling and scaling SEL survey data. Policy implications are discussed for educators, administrators, policy makers, and other stakeholders.

This paper examines which social-emotional skills are most important for student success and how they vary among student subgroups. It uses longitudinal data from over 200,000 fourth through seventh grade students in California to explore growth mindset gaps across student groups and shows the significant predictive power of growth mindset for academic achievement gains, even when controlling for other social-emotional skills and background characteristics.

This paper examines the use of social-emotional learning (SEL) measures to evaluate school-level growth in student outcomes. The study finds substantial differences across schools in SEL growth, suggesting that schools may contribute to students' SEL. However, caution is recommended in interpreting measures as the causal impacts of schools on SEL due to potential measurement error and omitted variables bias.

This paper uses a large dataset to confirm that self-management skills predict student success and are a better predictor of student learning than other socio-emotional skills. Students with higher levels of self-management experience almost 80 days of additional learning. The paper also describes self-management gaps across student groups.

This brief examines the consistency of the drop in academic self-efficacy during middle school across various student groups, such as demographics, achievement level, and school of attendance. The study finds that the drop is essentially universal across all groups, despite some showing lower self-efficacy than others. Additionally, schools vary in their students' level of self-efficacy, but do not differ much in this trend. The drop in academic self-efficacy is steeper for female students, resulting in lower self-efficacy for girls than boys throughout middle and high school.

This study uses value-added models to explore whether social-emotional learning (SEL) surveys can measure effective classroom-level supports for SEL. Results show that classrooms differ in their effect on students' growth in self-reported SEL, suggesting that classroom-level effects may be larger than school-level effects. However, the low explanatory power of the SEL models means it's unclear if these are causal effects. The SEL measures also capture growth not measured by academic test scores, indicating that classrooms within schools may impact measurable student growth in SEL.