Analysis of Longitudinal Data II: Linear Mixed EffectsModels for Normal Response Variables]Analysis of Longitudinal Data II: Linear Mixed Effects Models for Normal Response Variables

This chapter focuses on a single class of methods, linear mixed effects models, suitable for responses that can be assumed to be approximately normally distributed after conditioning on the explanatory variables. The main objective in the analysis of data from a longitudinal study is to characterize change in the repeated values of the response variable and to determine the explanatory variables most associated with any change. Linear mixed effects models for repeated measures data formalize the idea that an individual's pattern of responses is likely to depend on many characteristics of that individual, including some that are unobserved. Correlation between observations from the same individual arises from unobserved or unmeasured characteristics of the individual that remain the same over time, for example, an increased propensity to the condition under investigation, or perhaps a predisposition to exaggerate symptoms. Repeated measurements of a response under different experimental conditions or over a period of time occur often in behavioral research.