A Latent Class Analysis of Perceived School Climate Among Middle and High School Students: Findings from the California Healthy Kids Survey

Purpose: A large body of research has found that school climate influences a wide array of youth outcomes (Thapa et al., 2013). These include academic outcomes (e.g. attendance, motivation, cooperative learning, test scores), social and emotional outcomes (e.g. bullying and victimization, low risky sexual, violence, alcohol, and drug use behaviors), and psychological outcomes (e.g. psychiatric problems, depression, anxiety). The current study utilizes LCA to identify and classify school climate perceptions among secondary students in California. This study also assesses how demographic factors such as race and gender are associated with school climate. To the author’s knowledge, no studies have utilized LCAs on statewide databases of perceived school climate. This study’s findings can potentially be utilized to understand school climate perceptions among different demographic groups and to inform school climate interventions in California.

Methods: Data from the 2011-2013 California Health Kids survey were used to generate classes of school climate. We constructed the following scales to represent school climate--sense of safety, caring relationships with school adults, school connectedness, and meaningful participation and included the following demographic variables—race/ethnicity, gender, and grade level among a statewide sample of 7^th, 9^th, and 11^th grade students during the 2009-2011 academic school years (n=577,026). LCA was utilized to generate school climate classes and multinomial regressions were conducted to assess relationships between school climate classes and demographic variables (race, gender, grade level).

 

Results: A four-class model provided the best overall fit to the data for school climate. The four classes were termed: Some caring/participation, low climate, high connectedness/participation, and high climate. Multinomial logistic regression analyses showed that race influenced school climate class membership. Compared to Whites, African Americans were almost three times as likely to be members of the low climate class (OR=2.90, 95% CI 2.58-3.25) and 1.65 times as likely to be in the some caring/participation class (OR=1.65, 95% CI 1.53-1.79). Compared to Whites, Latino students were twice as likely to be in the high caring/participation climate class (OR=2.42, 95% CI 2.25-2.61) and twice as likely to be in the some caring/participation class (OR=2.00, 95% CI 1.90-2.11). Gender did not significantly influence school climate membership.

 

Scholarly Significance: Findings from this study provide policy makers and leaders with a more complete understanding of school climate in middle and high schools throughout California. This study showed that there are racial differences in school climate perceptions (Thapa et al., 2013). One reason for racial differences in the latent classes of school climate may be traced to the context of school communities serving racial minority students. A large body of educational policy research has found that low SES Black and Latino students are in public schools and districts with limited financial resources and personnel to develop and implement school climate programs. In addition, communities surrounding schools with predominantly Black and Latino students, often have high levels of poverty and violence, which adversely affect a school’s social and emotional climate.