This study examined students' movement in and out of special education and predictors for later special education placement. The sample ( N = 556) came from a response to intervention (RTI) study, specifically, a cluster-randomized control field trial that undertook the development and study of a hybrid Tier 1 (classroom instruction) and Tier 2 (specific differentiated small group interventions) model in kindergarten during the 2007-2008 school year. On average, from kindergarten to third grade about 13% of the sample was classified for special education services. However, the students accounted for in that 13% varied greatly from year to year as many students exited special education while others entered it. The most movement out of special education occurred in the earlier grades. Specifically, 80% of the students in special education from kindergarten had been declassified from special education by the end of first grade, whereas only 0% and 2% exited special education from first to second grade and second to third grade respectively. Further, the students who exited special education were replaced by a significantly lower group of students from general education. Students with speech impairments, language impairments, specific learning disabilities, and developmental delays accounted for most of the declassification from special education. To determine students' responsiveness to instruction beyond placement in special education, a standard score of 90 on reading subtests of the Woodcock-Johnson Achievement Test-III was used as a cut point. In general, students who remained in or entered into special education had greater percentages of scores below 90, while students who remained in general education or exited special education had smaller percentages of scores below 90. Overall, the entire sample scored better on code-focused subtests (i.e., letter word identification and word attack) than on meaning-focused subtests (i.e., picture vocabulary and passage comprehension). Finally, logistic regressions run with Hierarchical Linear Modeling software revealed that no school- or classroom-level variables from the fall of kindergarten were significantly predictive of later placement in special education. Although treatment condition, which was assigned at the school level, was not predictive of later special education, children's performance during the fall of kindergarten on measures of letter naming fluency, picture vocabulary and cognitive processing speed were predictive of later special education placement. Also, being male significantly increased a student's odds of being in special education at the end of first grade, but not at any of the other grades. Further, although being African American never significantly predicted later special education placement, chi-square tests revealed that African Americans were significantly more likely to be placed in special education in first grade, but at no other grades. Recommendations for future research involving special education classification and declassification within RTI studies included using detailed child demographics, programmatic characteristics, well-described instructional practices, and longitudinal information. Implications for practice included: district reporting that demonstrates the movement both in and out of special education; policy requiring states to include separate special education classifications for students with speech impairments and language impairments; early literacy instruction that involves letter naming fluency; intensive vocabulary instruction during the reading acquisition years; and comprehension and meaning-focused interventions beginning by third grade. [The dissertation citations contained here are published with the permission of ProQuest LLC. Further reproduction is prohibited without permission. Copies of dissertations may be obtained by Telephone (800) 1-800-521-0600. Web page: http://www.proquest.com/en-US/products/dissertations/individuals.shtml.]