(PS11-D82) Creating User-friendly Teacher Tools to Track Executive Function in the Classroom

Executive function (EF) challenges are well-documented in autistic youth, and increase in severity through adolescence. EF drives important outcomes such as academic success, employment, and mental and physical health, and is a critical component of transition planning. Schools are a primary source of intervention for autistic students, but the quality of transition planning in public schools is poor, and validated markers of EF skill progress are nonexistent for this population. The COVID-19 pandemic highlighted the need for a valid tool that captures EF problems in a high school classroom setting and does not require researcher presence. We adapted the Swanson, Kotkin, Agler, M-Flynn, and Pelham Scale (SKAMP), a teacher-report measure of EF-related classroom behaviors in ADHD youth, to expand on the range of school-related EF challenges, alongside a novel intervention-sensitive classroom observation tool to capture similar EF-related behaviors. The current study examined convergent/divergent validity in relation to the classroom observation tool, as well as other measures of EF.

Participants were 62 autistic teens (77% male; 68% white) aged 14-20 years (M=15.8, SD=1.3) with FSIQ ≥ 80 (M=102.2, SD=14.3) who met DSM-5 criteria and had an IEP classification of ASD supported by the Social Communication Questionnaire. Teachers rated the severity of EF-related classroom behavior problems using the adapted SKAMP. An observer rated the presence or absence of target EF-related behaviors on 11 items, which was summed to create a total score. Parents and teens rated severity of EF problems on the Behavior Rating Inventory of Executive Function-2 (BRIEF-2). Divergent validity was investigated using IQ and autism behaviors on the parent-reported Social Responsiveness Scale-2 (SRS-2). A Kuder-Richardson-20 reliability coefficient was calculated to determine the internal reliability of the SKAMP. Pearson correlations examined relations between total SKAMP scores, classroom observations, EF measures, and demographic variables. 

Internal reliability for the SKAMP was high (α = .92). The SKAMP was not significantly correlated with age, sex, FSIQ, or SRS-2 total T score. Classroom observations followed a similar pattern, except the total was correlated with IQ. SKAMP scores were significantly correlated with classroom observations (r=-.42, p< .001) but not with parent- or self-reported EF problems on the BRIEF-2.

The SKAMP demonstrated divergent validity from age, sex, FSIQ, and ASD behavior. The SKAMP also demonstrated a strong correlation with classroom observation. Non-significant relationship with parent- and self-reported EF problems may reflect differences in reporter-perspectives and settings where behaviors were observed. Parents may have reported on observed behaviors, while teens reported on their inner experiences across several settings (not restricted to school). Notably, the SKAMP demonstrated utility in the wake of COVID-19, indicating that it holds promise as a teacher-report tool for capturing EF-related classroom behavior problems across several classes that align with transition-planning metrics and could serve as an evaluative tool for progress on IEP goals. 

Item level data will be presented if accepted.