ABSTRACT
This chapter provides an overview of the data management tools necessary for understanding and working with the hierarchical data files used in multilevel modeling with IBM SPSS. It introduces several primary commands for manipulating data files to suit the needs of univariate and multivariate multilevel analyses. The chapter begins with a discussion of issues related to defining variables at each level of the analysis. The variables used in a typical data set may come from various sources, many of which are specific to a particular level of analysis and others which may be conceptualized as varying both within and between groups. The chapter covers the basics of how to prepare and build data files for multilevel analyses. For example, it illustrates restructuring the data from wide to long format, differentiating level-1 and level-2 files for developing descriptive statistics at each level of a data hierarchy, working with subject identifiers to merge data at different levels into one multilevel data set, aggregating individual cases, recoding, and computing new variables as well as centering level-1 predictors in facilitating the interpretation of the output, and checking the output for accuracy.
