Poster #89 - A Person-Centered Approach to Measuring School Climate: Capturing Between- and Within- School Variability in Student Experiences

• In Event: 1-005 - Poster Session 01
  In Poster Session: PS 01 Section - Education, Schooling

Integrative Statement

School climate is recognized as a valuable indicator of school health - and is generally comprised of the primary school-level characteristics responsible for determining adult and student outcomes (National School Climate Council, 2007). Although the benefits of evaluating school climate are well-studied, there is considerable debate as to how best to measure this construct with existing evaluation tools. A recent meta-analysis of school climate challenged the assumption of reporter homogeneity while evaluating school climate (Wang & Degol, 2016). Most studies of school climate to date use variable-centered approaches, which assume consistent responses across participants (Bergman, 2001; Bergman & Anderson, 2010). Person-centered approaches would account for variability across subpopulations – it is likely that different students experience their school climate in different ways (see Wang & Degol, 2016 for review). Person-centered approaches support the heterogeneity of views to be captured from students in a school community. Our study takes a person-centered approach to explore individual student perceptions and their relationship to specific dimensions of school climate. To capture this varia¬¬¬bility in student experience, we developed the School Climate Walkthrough Tool, an 85-item student-report measure of observable domains of their school climate. The tool was iteratively developed through a combination of literature reviews and focus groups with students and teachers. School climate domains were based on the dimensions outlined by the National School Climate Center (2007); Safety, Teaching and Learning, Interpersonal Relationships, and Institutional Environment. These four dimensions are further organized into eleven domains. High school students (n = 507) from nine diverse schools, varying in size and location, completed the measure. Participants rated items on a two-point scale ranging from “not observed” (0) to “observed” (1). Students were also given the choice of “not applicable”. We hypothesized that there would be differing levels of agreement between students on the school climate domains. To explore levels of agreement, variance scores were calculated by taking the absolute value of the difference between the percent of students that reported observing that item and the percent of students that did not observe that item. Scores could range from 0 (suggesting high variance in agreement) to 100 (suggesting low variance in agreement). Variance scores from the total sample were then ranked from high variance to low variance as presented in Table 1. Variance scores ranged from least amount of agreement among reports of Social and Emotional Safety (30.02) to greatest amount of agreement in reports of Support from Adults (69.01). In addition, we explored if the level of agreement associated with each school climate domain varied by school (see Table 2). As expected, ranking of domains by school based on variance of agreement differed from the total sample. Implications for person-centered approaches that account for heterogeneity of both experiences between- and within- schools when measuring school climate, including how subgroup analyses determined by agreement can be used to inform school climate interventions will be discussed.

Authors

• Christina Bradley, Yale University
  Presenting Author

• Julie McGarry, Yale University
  Non-Presenting Author

• Jessica Hoffmann, Yale University
  Non-Presenting Author