The failure mechanism of an item often can be linked directly to some sort of degradation process. This degradation process eventually weakens the item which then induces a failure. As system components have become highly reliable, traditional life tests, where the response is time to failure, provide few or no failures during the life of a study. For such situations, degradation data can sometimes provide more information for assessing the item's reliability. Repeated measures degradation is a form of degradation where the engineers are able to make multiple nondestructive measurements of the item's level of degradation. For some items, however, the degradation rates at nominal use conditions are so low that no meaningful information can be extracted. Thus the engineers will use accelerating methods to increase the degradation rate. Before a test can be performed, the engineers need to know the number of items to test, the points of time to make the measurements, and at what values of the accelerating variable should the units be exposed in order to achieve the best estimation precision possible. In this thesis we study the test planning methods for designing repeated measures degradation and accelerated degradation tests. First, Chapter 2 provides methods for selecting the number of units and the number of measurements per unit for repeated measures degradation tests without acceleration. Selection of these testing parameters is based on the asymptotic standard error of an estimator of a function of the model parameters. These methods can also be used to assess how the estimation precision changes as a function of the number of units and measurements per items. Chapter 3 describes methods for planning repeated measures accelerated degradation tests (RMADTs) where the engineers need to know the accelerated conditions at which the items should be tested. Chapter 4 is similar to Chapter 3, but uses a Bayesian approach for planning RMADTs.