Applications of longitudinal data analysis techniques: Metabolic syndrome trial

Longitudinal data arise when a response is measured at several measurement occasions on the same experimental unit. Special methods of statistical analyses are required to analyze longitudinal data due to the unique properties these data exhibit. In particular, the responses collected for each individual tend to be correlated. Longitudinal data analysis techniques are examined and applied to an experimental study of humans with metabolic syndrome. Initially, exploratory analyses are conducted by creating several graphs to visualize the data and make preliminary inferences. Next, simple methods of analysis for longitudinal data are applied to the data. Finally, more advanced techniques including mixed model methodology are examined.