Chapter 8 focuses on statistical approaches to evaluate and quantify the stability and sensitivity of causal model selection. Stability analysis addresses questions concerning the robustness of DDA results against sample composition. DDA model selection can be considered robust when causal decisions are stable across replicates of the initial data. In contrast, DDA statistics can be expected to show considerable variability, when outliers and overly influential observations contaminate the data. Sensitivity analysis focuses on the robustness of DDA against hidden external influences. A Monte-Carlo based sensitivity algorithm is introduced that can be used to test the sensitivity of DDA against artificially induced hidden confounding. While robust causal models can be expected to be fairly immune against additional hidden confounding, competing causal models that are already affected by latent external factors tend to become indistinguishable even when a small amount of external confounding is added to the data. Simulated and real-world data examples are presented to illustrate stability and sensitivity approaches in the context of probing the causal direction of effects.