Several statistical techniques that can be used to ameliorate the difficulties inherent in the data analysis of longitudinal studies are presented. The first step in longitudinal data analysis is graphing. This permits visual inspection of the data, and with educated viewing can yield insights into the nature of the underlying mechanisms. The next level of sophistication is to apply regression analysis and change point analysis to the curves obtained from the graphical analysis. It is usually the case in longitudinal studies that the exact form of the curve is not known prior to the experimentation. The graphing of the data is useful in suggesting different mathematical models to apply to the curves. The results of the regression analysis will help determine the uniformity of the process across subjects. The next step is to use the form of the fitted equation to determine significant points on the curve. The shape of the curve will suggest change points in the subjects' behavior with respect to the dependent variable. In certain cases where problems arise, the use of principal components is called for. Practical advantages are that they explain the original curve best and will likely point to any existing major differences, and they occur mathematically and do not depend on the experimenter's ability to form a regression curve or pick important change points. When used in conjunction with each other, these techniques form a powerful package for analyzing longitudinal data. (RC)