Oftentimes, analysts are interested how a particular intervention differentially affects an observed population. How can we estimate the extent of causal effects which vary across a population?