The Design
What we can do with the nice setup where we can attempt to recover the Causal Effect from the Observed Association under a lot of stringent assumptions.
Treatment Assignment should not bear upon Potential Outcomes.
Treatment Assignment should not bear upon Potential Outcomes.
Treatment Assignment should not bear upon Potential Outcomes.
But the World is a messy place so you do the best you can to approximate this ideal.
Randomized Experiments
So far, a Big Consideration is to try and divorce potential outcomes from assignment. Randomization allows us to do this and gives us Exchangeability by design.
Observational Studies
This is where the pain begins. Sort of. People love Randomized Studies for a reason. The trick here is to make the Observational Study resemble a Randomized Study as much as possible during the design phase. More on this later.