Assessing student achievement over multiple years is complicated by students' annual matriculation through different classrooms. The process of matriculation, or annual classroom change, threatens the validity of statistical inferences because it violates the independence of observations necessary in a regression context. The current study investigates (1) current practices in educational research to address annual classroom change, (2) how different methods to address annual classroom change impact variance estimates of student achievement, and (3) what sample and cluster sizes are required to return estimates of student achievement with limited bias while accounting for annual classroom change. To address these questions, I conducted a systematic review of recent educational studies, a secondary data analysis of the Student Teacher Achievement Ratio study, and a simulation comparing growth modeling approaches. Findings suggest a majority of reviewed studies do not address annual classroom change and many do not acknowledge that annual classroom change presents a threat to statistical validity. Additionally, I find that a method introduced by Grady and Beretavs (2010) provides larger estimates of variance attributable to the classroom level over multiple years by including a coefficient representing the sum of students' classroom memberships over time into the calculation of the slope of student achievement. Finally, I present evidence that clusters of ten students or more may be sufficient for estimating student achievement while accounting for annual classroom change. Together this work helps improve the methodology needed to address shifting upper-level nesting structures and specifically annual classroom change. [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://bibliotheek.ehb.be:2222/en-US/products/dissertations/individuals.shtml.]