Autism Spectrum and Developmental Disorders
Profiles of School Refusal Behaviors among Neurodivergent Youth
Jessica E. Granieri, M.A.
Graduate Student
Binghamton University
Kings Park, New York
Hannah E. Morton, Ph.D.
Postdoctoral Fellow
Oregon Health and Science University
Portland, Oregon
Derek Order, None
Undergraduate Research Assistant
Binghamton University
Port Jefferson Station, New York
Raymond G. Romanczyk, Ph.D.
Professor
Binghamton University
Binghamton, New York
Jennifer Gillis Mattson, Ph.D.
Professor
Binghamton University
Binghamton, New York
Neurodivergent (ND) youth (i.e., Autism Spectrum Disorder, Attention-Deficit/Hyperactivity Disorder) more frequently refuse school compared to their neurotypical (NT) peers (Kearney et al., 2014; McClemont et al., 2021). School refusal is associated with negative outcomes, including academic difficulties, school dropout, and psychopathology (Kearney, 2008; Munkhaugen et al., 2017). Factors associated with school refusal for ND youth are yet unknown.
Parent-report data was collected on 482 children (Mage = 11.51; SDage = 3.44). Autism traits were measured by the Social Responsiveness Scale, Second Edition (Constantino et al., 2012). The Vanderbilt ADHD Diagnostic Parent Rating Scale measured ADHD traits, internalizing and externalizing, and social and academic difficulties (Wolraich et al., 2003). Bullying was measured by the Assessment of Bullying Experiences questionnaire (Morton et al., 2021). A demographic questionnaire captured participant characteristics and school information. Latent Profile Analysis was conducted to identify school, parent, and child characteristics associated with subpopulations of school refusal, yielding 4 classes.
Class 1 (n = 122) and Class 2 (n = 144) were youth with ND traits above clinical cutoffs who refused school about 2-3x/month. Class 1 had more traits of ADHD (p = .002) and autism (p = .112) compared to Class 2; however, youth in Class 2 were 8.28x more likely to have a diagnosis of autism or ADHD (95% CI: 3.09, 22.22). Youth in Class 1 were less likely to have a 1:1 aide (p < .001) or behavior support plan (BSP; p < .001) and spent more time in general education ( >75% of the day) compared to Class 2 (50-75% of the day; p = .03), despite no difference in adaptive skills between classes (p = .55). Bullying frequency was similar between classes (p = .235); however, ND youth without supports were 2.17x more likely to refuse school due to bullying compared to ND youth with supports (95% CI: 1.26, 3.73).
Youth in Class 3 (n = 58) were also ND with autism traits above the clinical cutoff. These youth refused school every few months and spent 25-50% of the day in general education. Notably, Class 3 was less likely to refuse school compared to Class 1 (p = .006) or Class 2 (p = .006) despite no differences in autism traits, adaptive behavior, internalizing, externalizing, or bullying frequency (all p’s > .05).
Youth in Class 4 (n = 158) scored below cutoffs for ND traits. These youth had lower school refusal (p < .05) and externalizing symptoms (p < .05) compared to other classes, as well as the least social or academic difficulties (p < .05), and highest adaptive skills (p < .05). When they did refuse school, it was more likely due to mental health difficulties compared to other classes (ORs: 1.89-2.98, p’s < .05).
ND youth are more likely to refuse school compared to NT peers; however, a subset of ND youth do not frequently refuse school. School supports may partially buffer refusal (McClemont et al., 2021). Reasons for school refusal differ between ND and NT youth. Youth with ND traits above clinical cutoffs who are in general education but without a formal diagnosis or supports (e.g., 1:1 aide or BSP) are at high risk for school refusal due to bullying and may need additional support to maintain school engagement.