When more than one independent variable is analysed in a regression analysis, a multiple regression analysis is performed. Regarding multiple regression analysis a distinction has to be made between association models and prediction models. The aim of an association model is to estimate the relationship between a main independent variable and a particular outcome variable as good as possible. The latter means that confounding has to be taken into account and that effect modification has to be investigated. The aim of a prediction model on the other hand is to predict a particular outcome variable as good as possible by a set of independent variables. For both the building of an association model and a prediction model different procedures are available and different procedures can lead to different results and conclusions. The final part of this chapter deals with a discussion of quality indicators for a prediction model. Regarding the quality of a prediction model a distinction is made between internal quality, i.e. how good the model fits the data and external validity, i.e. how good the model behaves in an external dataset.