ABSTRACT
This chapter provides an introduction to addressing the non-independence among a set of observations that are nested within a hierarchical structure. Nested data structures are a fundamental part of many areas of psychology and related disciplines. In mixed-effects models for nested data structures, in addition to regression coefficients quantifying the overall effect of certain variables, there is generally an interest in partitioning the model into within and between components. The major advantage of the mixed-effects modeling approach is that it can easily accommodate complex data structures such as those that include additional predictor variables of interest. One other way in which the mixed-effects modeling approach differs from the analysis of variance (ANOVA) approach is the way in which it implements the normality assumption. An important virtue of the mixed-effects approach to analyzing data from nested designs is that the same models can be used whether sample sizes are equal or unequal.
