When do students with disabilities graduate high school?

With No Child Left Behind (NCLB), graduation rates have become an essential indicator of high-school success. High schools, districts, and states are now accountable for measuring, reporting, and improving their graduation rates for student subgroups, including students with disabilities. Policies expressed in the Individuals with Disabilities Education Act (IDEA), however, provide differing expectations for high-school graduation of students with disabilities. IDEA permits students with disabilities to receive services in high school until they are twenty-one years old. NCLB, on the other hand, emphasizes students with disabilities graduate within four years after high-school entry. To understand the appropriate expectation for high-school graduation, in this thesis, I examine when students with disabilities graduate and how graduation patterns differ for students based on selected student characteristics and educational factors. In Part 1 of the thesis, I utilized discrete-time survival analysis to estimate the conditional probability of high-school graduation in a given year after high-school entry, by disability category. I also examined how these graduation patterns differ for students fully-included in the general-education setting and students in substantially-separate educational settings. I utilized statewide data on students with disabilities from Massachusetts from 2005 through 2012. I found that students with high-incidence disabilities and sensory disabilities have elevated graduation-risk profiles and the majority of these students graduate within seven years, on average. Students with emotional disturbance, intellectual disabilities, and low-incidence disabilities, on the other hand, have lower graduation-risk profiles and about a third to a half of these students graduate. Additionally, I found that students with disabilities who are fully included have higher graduation-risk profiles than students with disabilities who are educated in substantially-separate settings. Next, in Part 2, I exploited a natural experiment that occurred in Massachusetts as a result of the exogenous cut-score on the MCAS, a high-stakes exit examination, to estimate the causal effect on the conditional probability for graduation, in a given year after high-school entry, of just failing the high-school exit examination. I found that just failing the MCAS for English or mathematics decreases the conditional probability of on-time graduation for students with disabilities by 6.2 and 9.4 percentage points respectively.