Poster #215 - The Contributions of High School Quality in Predicting Adolescent Self-Reported Socioemotional Risks and Strengths

• In Event: 2-055 - Poster Session 06
  In Poster Session: PS 06 Section - Social, Emotional, Personality

Integrative Statement

School climates have been shown to reduce student risk behaviors and promote positive youth development (Cohen et al., 2009). However, questions remain about the specific dimensions of school quality that best predict student social-emotional outcomes. In this project, we drew data from Chicago Public Schools (CPS) on school climate and linked these measures to student-level data that were collected as part of the Chicago School Readiness Project (CSRP; Raver et al., 2008). In each CPS school, students and teachers rated five distinct domains of school climate (“5Essentials”), allowing us to test if key school characteristics predicted student-reported socioemotional strengths and health-risk behaviors.

Student “risks” and “strengths” survey data was collected during the 2016-2017 school year, when the average participant was in 10th grade (M age = 15.61). Student socioemotional strengths (α = 0.63) represents an aggregated measure relating to participation in extracurricular activities and students’ civic engagement in school and community settings (Diemer et al., 2014). Health-risk behaviors measure (α = 0.83) constitutes items on unintentional injuries, violence, substance use, and sexual behavior (Riesch et al., 2006). The 5Essentials (UChicago UEI, 2008), included dimensions relating to Ambitious Instruction, Effective Leaders, Collaborative Teachers, Involved Families, and Supportive Environment. Each school’s 5Essentials score was merged with CSRP student-level data by school ID (267 students had valid 5Essentials scores).

Using a confirmatory factor analysis model, we tested the model fit of each factor of the 5Essentials (21 items), and found cohesive adequately fitting models of school characteristics based on the school quality data. We found a significant correlation (r = 0.91) between two of the 5Essentials aggregate factors, “Effective Leaders” and “Collaborative Teachers.” To avoid collinearity, the “Collaborative Teachers” factor was dropped from the analysis.

Table 1 shows the descriptive statistics of school characteristics factors (Panel A) including Ambitious Instruction (α = 0.83), Effective Leaders (α = 0.91), Involved Families (α = 0.93), and Supportive Environment (α = 0.88), and the correlations between them (Panel B).

Table 2 shows OLS estimates of the predictive roles of each school factor for tenth graders’ measures of “risk” and “strengths.” We included controls for demographic characteristics, and earlier “risks” and “strengths” measures. We found evidence that several school factors were statistically significant predictors of student-reported strengths (Table 2). Students enrolled in more academically demanding schools (i.e., schools with higher scores on the ambitious instruction, effective leaders, and involved families scales) reported significant gains in self-reported strengths since middle school (0.15, 0.19 and 0.20 SD’s, respectively). We also found that adolescents enrolled in schools with effective leaders reported significant gains in health risks (0.16 SD’s) over that same time period. This could be due to the role of selection where students in schools that are managed by more effective leaders experience more difficulties, and as such report more risk behaviors. Our analysis provides evidence that school quality is significantly related to student risks and strengths. Future work will attempt to further unpack this relationship by examining heterogeneity and specific dimensions of the “risks” and “strengths” scales.

Authors

• Chen Li, New York University
  Presenting Author

• Jill Gandhi, New York University
  Non-Presenting Author

• Alaa Khader, New York University
  Non-Presenting Author

• Tyler Watts, New York University
  Non-Presenting Author