The federal No Child Left Behind Act (NCLB) holds schools, districts, and states accountable for improving the academic achievement of all students, including the nearly 14% of public school students who receive special education services because they have an identified disability. By 2014, 100% of students with disabilities are expected to perform at the "proficient" level on state tests--the same goal set for students overall and for other subgroups of students. NCLB also calls on schools and districts to close achievement gaps between students with disabilities and their non-disabled peers and to include these students in regular state testing programs to the maximum extent possible. This report by the Center on Education Policy (CEP) examines progress in raising achievement for students with disabilities. It also describes the factors that make it difficult to clearly discern achievement trends for this subgroup. The data for this analysis were collected by CEP with technical support from the Human Resources Research Organization, and come from the state reading and mathematics tests used for NCLB accountability in all 50 states. The trends discussed in this report cover fewer years--2006 through 2008--than the trends in the three earlier parts of this 2009 series of reports on student achievement. This is because federal policies for testing students with disabilities and determining their progress were revised in 2003 and 2005 in ways that likely affected the year-to-year comparability of test data. School year 2005-06 marked the first year that states administered tests under the revised rules, and so we now have enough data to establish a three-year trend line for students with disabilities in many states. Several findings emerged from this study: (1) Fuzzy data make it difficult to obtain a clear picture of achievement for students with disabilities; (2) Students with disabilities have made progress in grade 4 at all three achievement levels--basic-and-above, proficient-and-above, and advanced; and (3) Differences in test performance between students with disabilities and non-disabled students are very large. (Contains 5 tables and 7 footnotes.) [For the companion reports of this record, see the following: Part 1 (ED506121); Part 2 (ED506122); Part 3 (ED507909); and Part 5, (ED509023).]