School climate is a multidimensional construct that reflects the quality of school life for key audiences and includes academics, school safety, interpersonal relationships, and the quality of the physical building and materials. Individual perceptions (level-1) are often aggregated to represent school perceptions (level-2) for comparison. This is problematic without adequate evidence that perceptions are experienced similarly at both levels. Cross-level invariance is one such form of evidence that the same measurement structure exists at the individual and school levels. This aggregation without supporting evidence creates additional statistical and theoretical complications that must be considered and addressed when choosing an appropriate measure of school climate.

Chapter One of this dissertation is a systematic literature review of school climate measures with evidence of invariance reported in peer-reviewed literature over the last ten years. In addition, information regarding study characteristics, measurement characteristics, and invariance methodologies were extracted and coded for review. Practitioners can use this information to reference when determining if a measure is appropriate for a given need. Chapter Two of this dissertation utilized multilevel factor mixture models to assess the within and between-level measurement invariance of the Georgia School Personnel Survey (GSPS) of school climate with regards to gender and ethnicity. Dichotomous gender (i.e., male, female) and race/ethnicity (i.e., White, People of Color) were both found to have adequate within-level measurement invariance. Both were found to adequately support partial scalar measurement invariance at the between-level. This evidence extends the research literature supporting the use of the GSPS to compare individuals at the person and school-levels.