Distributions of covariates within each group [95% confidence intervals].

Source: National Longitudinal Transition Study 2.Notes: Number of multiply imputed data sets = 50. Weighted to population levels. Variances adjusted for sampling method.aThe “A lot of trouble” and “No conversation” category estimates for ‘How well youth converses’ were combined for the Learning Disability group in compliance with U.S. Department of Education rules aimed at preventing data disclosure in instances where point estimates are based on very few underlying cases.bThe “Special” and “Other” category estimates for ‘Type of school’ were combined for the Speech-Language Impairment and Learning Disability groups in compliance with U.S. Department of Education rules aimed at preventing data disclosure in instances where point estimates are based on very few underlying cases.