Association for Public Policy Analysis & Management

The educational landscape for students with disabilities has undergone vast changes with the advent of the Individuals with Disabilities Education Act (IDEA) in 1975 and its subsequent reauthorizations. By 2012-13, the proportion of students with disabilities receiving special education had risen to nearly 6.4 million K-12 schoolchildren per year or13 percent of total public school enrollment in the US. In many cities the proportion is even higher, sometimes over 20% (NCES, 2015). While research shows that the receipt of special education services is not proportional across race or gender (Morgan, et al., 2015; Hibel, et al., 2010), there is less evidence about the rate at which students exit special education. Research highlights educational (academic performance) and financial (costs associated with financing special education services) factors as important considerations when students with disabilities are declassified from special education services. Academic performance is dependent on students’ timing and duration in special education and is linked to two proposed theories in literature. On the one hand, students who remain in special education year after year and continue to fall further behind would seem to exemplify the Matthew Effect (or cumulative growth model). On the other hand, students in special education who “catch up” to their peers in general education and are therefore declassified from special education reflect the compensatory growth model. (Morgan, et al., 2011). Yet, surprisingly little is known about the differential exit rates among students by race, gender, or income.

This paper aims to fill this gap in literature by examining when students with disabilities transition out from the special education services to general education and how patterns differ for students based on selected student characteristics and educational factors. We control for disability classification since differential declassification rates could be related to the incidence of disability classification of the student (low versus high incidence disability).

Using a nationally representative data from the Early Childhood Longitudinal Study, Kindergarten Class of 1998-99 and 2010-2011 (ECLS-K), and rich administrative student data from New York City public schools from 2006 to 2012, we conduct discrete time survival analyses to estimate the conditional probability of transitioning to general education services (from special education services) in a given year after entry into elementary school.

Preliminary results for NYC suggest that, across all disability categories, the peak conditional probability for transition occurs at three years after entry into elementary school. Moreover, minorities’ students with disabilities (Hispanic, Black, Asian, etc.) are less likely to exit special education services in each period compared to their White peers. Similarly, males and those from lower socio-economic status are less likely to transition to general education services in each period.

These findings yield crucial insights for state policymakers, school districts and local education agencies related to their special education populations as well as instructional practices to improve special education services. It would be beneficial for educators and policymakers to better understand why students have different transition probabilities and whether they get sufficient support to transition to general education and achieve their academic targets.