ABSTRACT
This chapter presents an overview of several data-related issues associated with modeling individual and group processes embedded in hierarchical data structures. It also provides some further data considerations in developing and testing multilevel models. First it provides an introductory discussion of the application of sample weights, focusing in particular on their potential impact at the group level of the study. Second, it turns the attention to issues related to sample size requirements in multilevel models and corresponding statistical power to detect hypothesized effects. Third, the chapter discusses some common issues related to missing data. Finally it provides some further data considerations in developing and testing multilevel models. This chapter also address import ancillary considerations that include sample weights, general sample size requirements, and corresponding statistical power to observe hypothesized effects, as well as working with missing data. Another issue to consider in thinking about multilevel analyses is the presence of missing data.
