Repeated Measures Data

Summary

Specializes mixed models to repeated-measures data where the cluster is a person observed across time, emphasizing autocorrelation as the temporal structure that distinguishes longitudinal data from generic clustering. Develops growth-curve models with random intercepts and slopes for time, the intercept-slope correlation as a measure of trajectory heterogeneity, and extensions through quadratics, splines, and piecewise linear models for known change points. Covers five within-person covariance structures from compound symmetry through autoregressive AR-1 to unstructured, then time-varying covariates with within-versus-between-person decomposition, missing data and dropout, and a brief preview of cross-lagged panel and latent change score models.