ABSTRACT
This chapter provides a glimpse of the ways in which multilevel modeling can be used to investigate various policy issues. Multilevel modeling is fast becoming the standard analytic approach for policy research on schools due to its applicability to a broad range of research situations, designs, and data structures. Many examples of educational policy research involve the analysis of hierarchical, or nested, data structures. Recently, much policy research has focused on the analysis of existing, large-scale databases, because the data are more readily available. Linear models have had a long tradition in the social sciences for analyzing data from experimental, quasi-experimental, and nonexperimental research designs. Multiple regression is the most widely used statistical technique in social science research. The flexibility of the structural equation modeling (SEM) approach makes it an ideal modeling strategy for researchers to use in examining changes in school achievement over time after adjusting for measurement error.
