Longitudinal data analysis growth modeling

Analyze change over time using growth curve models. Data structure: repeated measures nested within individuals (Level 1: time, Level 2: person). Models in R lme4 or HLM software: 1. Unconditional growth model: test for linear change over time. 2. Conditional growth models: add predictors of intercept and slope. 3. Piecewise growth: different slopes for different time periods. 4. Nonlinear growth: quadratic, exponential, or logistic growth patterns. Model building: 1. Plot individual trajectories to visualize patterns. 2. Test unconditional means model for ICC. 3. Add time variable, test linear growth. 4. Add predictors systematically. 5. Test model assumptions (linearity, normality, homoscedasticity). Missing data: use maximum likelihood estimation (handles MAR data). Report: fixed effects (average growth), random effects (individual variation), model fit indices (AIC, BIC).