Social science research presents an opportunity to study phenomena that are multilevel or hierarchical. Multilevel modeling is an attractive approach to exploring relationships between individuals and their social groupings. Multilevel modeling facilitates the incorporation of substantive theory about individual and group processes into the clustered sampling schemes (e.g., multistage stratified samples) or hierarchical data structures (e.g., individuals clustered in organizations, repeated measures nested within individuals) found in many existing data sets. This chapter develops a context and rationale for using multilevel models in the social and behavioral sciences. It highlights relevant conceptual and technical issues in defining and investigating multilevel and longitudinal models using the linear mixed-effects modeling procedures for continuous and categorical outcomes in IBM SPSS. The chapter then presents a general modeling strategy for investigating a series of multilevel models (e.g., fixed and random effects, cross-level interactions).