School climate and student-based contextual learning factors as predictors of school absenteeism severity at multiple levels via CHAID analysis

Abstract

Abstract School absenteeism is an important target of prevention science frameworks within the context of contributing/risk factors and tier-based intervention strategies. Little research has been done with respect to how specific aspects of school climate, academic mindset, and social emotional learning relate to different levels of absenteeism severity. Ensemble analysis, and specifically chi-square adjusted interaction detection analysis, was conducted on a measure of these constructs across multiple levels of absenteeism severity (3+%, 5+%, 10+%, 15+%, 20+%) for 128,381 students (Mage = 13.98; SD = 2.48). Pathways revealed some school climate and academic mindset items to be unique at higher levels of absenteeism severity, though item homogeneity was noted regarding key split points. The latter included items related to turning in assignments on time, liking school, and safety concerns. The findings reveal the need to examine school climate in an integrated fashion with student-based contextual learning factors, may support a dimensional approach to conceptualizing school absenteeism, and may suggest demarcations for tier-based intervention strategies. The findings may also have implications for cohesive school-based initiatives for academics and behavior. Items generated from the present study could serve as targets for school climate intervention components to enhance curriculum-based skill development, teacher care and classroom structure for students, student decision-making, personalized sessions for certain students, and acceptable school grounds. Item-level analysis of school climate may also be preferable in some cases to school-average reports given absenteeism disparities among marginalized students.