The topics of this study were (1) identification of children for early intervention and special education, (2) declassification of children who were no longer eligible for early intervention and special education services, and (3) outcomes for children identified for services under IDEA. Under an earlier contract, a design team had reviewed sources of national data already collected on special education and early intervention through other studies or as part of ongoing data collection systems, identified the sources that could be useful in addressing these three topics, and determined which topics were likely to be addressed with the available data and which would require collecting new data (Abt Associates and Westat 2007). Building on the work of the design team, this study addressed the following research questions using the extant data—that is, data already collected through other studies or as part of ongoing data collection systems—as a cost-effective means of addressing the following descriptive research questions for the National Assessment:
Key questions related to identification:
Key questions related to declassification:
Key questions related to outcomes:
The research questions were examined for the three age groups covered under IDEA: infants and toddlers (birth through age 2), preschool-age (ages 3 through 5), and school-age children and youth with disabilities (ages 6 through 21).
In addition, the research questions addressed both one point in time and trends over time. The time frame most relevant for addressing research questions at one point in time was the most recent year of data available. For analyses of changes over time, 1997 was chosen as the starting point because it was the year of the last IDEA reauthorization before the 2004 reauthorization.
Two sets of analytic activities were conducted to address the research questions. A review of relevant literature was conducted to identify published sources of data and analyses of pertinent data sources from which findings were drawn. On the basis of this initial review, extant databases were selected to conduct new analyses addressing the research questions for this study. As a result, the following 14 datasets were selected that targeted the age ranges of interest and the time frame most relevant for this study:
Population data on children identified for services under IDEA:
Population data used for identification and graduation ratios:
Sample data from four longitudinal studies that followed nationally representative samples of children of different age groups identified for services under IDEA:4
Sample data on the outcomes of the general population for comparison with the outcomes of children identified for services under IDEA:
Analyses using population data were considered to be descriptive, and no statistical testing was conducted. When analyses included sample data, statistical testing was conducted. When sample data were used for comparison of outcomes, t tests for differences in mean values were applied to calculate the statistical significance of the comparison, and when appropriate the Benjamini-Hochberg (1995) procedure was used to control the false discovery rate. The alpha level was set to .05 for each family of comparisons.
For the research questions concerning identification, data include the number of children identified for services under IDEA; the percentage of children from the total population who were identified for services under IDEA are presented by age, race/ethnicity, and state; and the gender composition of children identified for services under IDEA Part B. Declassification data are presented on the percentage of children no longer eligible for early intervention or special education services and their outcomes as compared with children who continue to receive services. Data on outcomes for children identified for services under IDEA include academic and developmental outcomes and trends over time. Analysis results include comparisons over time between children identified for and not identified for services under IDEA, between state results and national averages, with the general population means, across IDEA eligibility categories, and across ages. A description of school completion by disability category and cluster is also presented.
All comparisons of outcomes between children identified for services under IDEA and other children and between children identified for services under IDEA across states and over time are presented for descriptive purposes only. These comparisons were not designed and are not suitable, to measure the impacts of IDEA on child outcomes.
The following sections highlight results for each of the three age groups—infants and toddlers, ages birth through 2, who were served under Part C of the law; preschool-age children, ages 3 through 5, served in Part B preschool programs; and school-age children and youth ages 6 through 21 served in Part B programs. For each age group, the patterns of identification and the academic and developmental outcomes for children with disabilities are addressed. Declassification information (the loss of eligibility for services) is presented for infants and toddlers and school-age children and youth.
Exhibit ES.1. National number of children identified for services under IDEA, by age (2005)