For a scientist testing a theory who wants to show causation

For a scientist testing a theory who wants to show causation, which of these is MOST important?

Question options:

Using a very large sample

Using random assignment

Using random selection

Solution

Using random assignment is most important to prove any casual link