ABSTRACT
To this point, we have focused on multilevel models in which a single measurement is made on each individual in a sample and the individuals are in turn clustered. However, as explained in Chapter 2, multilevel modeling can utilize varying data structures in a number of contexts. This chapter will focus on using multilevel modeling to analyze longitudinal data generated when a series of measurements are made on each individual in a sample, usually over a set period of time. While longitudinal data can be measured on bases other than temporal (e.g., measurements at multiple locations on a plot of land), we will focus on the most common-time-based-type of longitudinal data. In this chapter, we will first demonstrate the application to the special case of tools we have already discovered and then briefly describe the correlation structures that are unique to longitudinal data. We will conclude the chapter by describing advantages of using multilevel models with longitudinal data.
