An Ecological View of Predictors of Children's Perceptions of School Interpersonal Climate

Introduction: A positive school climate is gaining attention as an essential component of school improvement policies and practices (Piscatelli & Lee, 2011; Education Week, 2013). Student ratings are commonly used to assess school climate, but variation in these ratings are usually characterized as rater bias. In line with ecological systems theory (Bronfenbrenner & Morris, 1998), this variation may be related to qualities inherent to the child, the classroom, and the school, or to interactions between these. Identifying these characteristics could contribute to a better understanding of how and under what conditions whole-school interventions effectively improve children’s school experiences. Empirically, this knowledge helps more precisely control for or model factors that contribute to individual variation in perceptions. This study examines the direct and interactive effects of multiple child characteristics, classroom- and school-level social climate on children’s perceptions of school interpersonal climate.

Method: Data come from year 2 of a three-year school-randomized evaluation of Social and Character Development (SACD) Programs in 83 elementary schools in 7 sites and 6 states. Participants included 4,016 4^th grade children in 311 classrooms and their 4^th grade teachers. A series of 3-level hierarchical linear models were estimated in HLM (Raudenbush & Bryk, 2002). The two outcome measures, assessed in spring, were negative social climate and feeling afraid. Student predictors included measures of social and academic competence, behavior, and socio-demographic risk. The interpersonal climate predictors were, at the classroom level, teachers’ individual perceptions of student respect and safety, and at the school-level, aggregated child perceptions of negative climate, teacher perceptions of student respect and safety, and teacher affiliation. Predictors were assessed in fall and fall levels of the outcomes were included as controls. Cross-level interactions between student characteristics and group-level climate variables were examined.

Results: Almost all child characteristics explained variation in one or both of the outcomes. Further, children in schools with less respect among students, less safety, and less teacher affiliation perceived the school climate more negatively. Interactions suggested that a misfit between child and environment was associated with worse school experiences.

Conclusion: Results suggest that improving children’s social and behavioral skills might improve their experiences of the social climate. Further, a more positive social climate may be more harmful for some low-functioning students. Whole-school efforts to improve school social climate should perhaps be tailored to meet the needs of these struggling students.