Path model of age 9 sleepiness as an intervening variable linking child SES to age 11 academic/cognitive performance. Child sex, zBMI, illness/asthma, ADHD status, sleep duration and quality, and age 9 academic/cognitive performance variables were covaried and allowed to predict the outcome variables. All academic/cognitive outcome variables were entered in the same model and were covaried. R2 values represent the variance accounted for by all predictors and covariates. Covariates and covariances between the outcome variables are not depicted for simplicity. Of note, although the association between sleepiness and ARMT scores was significant, the indirect effect linking SES to ARMT scores via sleepiness was not. Model fit: χ2/df = 3.00, CFI = .91, RMSEA = .08. *P ≤ .05, ***P < .001.