Office of Epidemiology and Research, Division of Research
Advancing Applied MCH Research
Division of Research
Strategic Research Issues
Applying for a Grant
Podcasts & Webinars
News & Events
MCH Research Listserve
NOFO FAQsFrequently Asked Questions
(SDAS) The Effectiveness of Special Education Services for Children with Autism: A National Longitudinal Study
Project Number: R40MC15595
Grantee: University of North Carolina at Chapel Hill
Department/Center: Maternal and Child Health / Gillings School of Global PH
Project Date: 09/1/2009
Principal Investigator: Edward Michael Foster
- Early Childhood (3-5 years)
- Middle Childhood (6-11 years)
- Adolescence (12-18 years)
No nationally representative, longitudinal data are currently available on children and youth with Autism. However, recent data from the U.S. Department of Education's Office of Special Education Programs include adequate numbers of autistic youth and represent an important new opportunity to learn about these children and how they fare in the educational system. These data include the Pre-Elementary Education Longitudinal Study (PEELS), the Special Education Elementary Longitudinal Study (SEELS), and the National Longitudinal Transition Study-2 (NLTS2). These studies represent the experiences, special services, and outcomes of children throughout their school years and beyond. While each study is representative of all children in special education in the relevant age range (PEELS: age 3-5; SEELS: 6-12; NLTS2: 13-16), each includes adequate numbers of children with autism (N= roughly 200, 1,100, 1,100, respectively). The proposed study will address three questions about the education of these children: 1) Do children receive high-quality treatment from the schools, as indicated by whether children are taught by appropriately certified teachers; educated in inclusive settings; identified early; and having appropriate plans to make transition to adulthood? 2) If so, which children receive such an education? For example, how does service quality vary by race/ethnicity and gender? Along the autism spectrum? By district? These analyses will rely on multilevel modeling to reflect the fact that children and youth are nested within school districts. Informative in and of themselves, these analyses also will inform aim 3 and the modeling used to separate the causes of quality from its consequences. 3) Do these indicators of service quality improve outcomes? The key methodological issue is that the effects of services in observational data are confounded with a range of individual and district characteristics. We will employ the latest tools of causal inference from biostatistics and econometrics to assess services impact. Those analyses will provide both cross-sectional estimates as well as models that capture the dynamic interplay between service use, a child's functioning and key outcomes. These dynamic models (structural nested and marginal structural models) have seldom been applied in education research of any type. (See1 for an exception.) Our analyses will allow us to describe children's receipt of high-quality school services nationwide and to determine patterns (e.g., by age) in who receives those services. Those service quality indicators include: teacher certification, inclusive settings; early identification; and the transition planning. Personnel preparation remains a weak element of effective intervention for children with ASD; inclusive setting has been a controversial issue people debate about whether certain skills serve as prerequisite for beneficial effects; early identification represents a critical development window for the interventions; and transition planning is related to the ultimate goal of special education for children with ASD-personal independence and social responsibility. These analyses will provide insights into key policy questions, such as the benefits of teacher certification and of early identification. Outcomes considered include key measures of academic, social, behavioral, functional, and motor skills. These data do have limitations. First, the data include only those in special education who have been identified by school systems as needing services. Second, the data provide little information on the content of the services (e.g., the specific behavior approach that teachers use). What the study will assess is the effect of broad services policy makers have used and will use to affect the instruction available to these children.
Listed is descending order by year published.
Foster EM, Pearson E. Is inclusivity an indicator of quality of care for children with autism in special education? Pediatrics 2012; 130:Supplement 2 S179-S185.
Autism, School Outcomes & Services, Capacity & Personnel, YSHCN Transition to Adulthood, Developmental Disabilities, Special Health Care Needs