This chapter highlights methods for estimating causality in the health and behavioral sciences, with an emphasis on methods that have been utilized in the study of recovery from alcohol use disorder. Emphasis is placed on the role of design as a necessary component in teasing out causal relationships, with the ideal approach being an experimental approach with a randomization component. In the absence of experimental design, researchers often turn to observational studies. In such cases, it is necessary to turn to quasi-experimental designs, two of which are highlighted herein: regression discontinuity and interrupted time series designs. Additionally, disadvantages of propensity scores are discussed, psychometric network modeling is described, and software packages for implementing these methods are highlighted.