Design Issues in Multilevel Studies

Abstract

Discusses several design issues in multilevel studies. First, this chapter emphasizes sample design issues in multilevel studies. The authors describe the importance of representative sampling and how the choice of sampling plan depends on the study purpose. They also discuss optimal design, the choice of estimand (house vs senate model), and cohort vs repeated cross-sectional sampling in longitudinal studies. Next, this chapter examines experimental design issues. These topics include the level of randomization, split plot designs, and factorial designs. Finally, the authors discuss the estimation of power. (PsycINFO Database Record (c) 2008 APA, all rights reserved) (from the chapter)