You could design an experiment that looks at people who were accepted into a program correlated to higher wages but went to a different program instead. (E.g.,

You could design an experiment that looks at people who were accepted into a program correlated to higher wages but went to a different program instead. (E.g., accepted to an engineering school but went to a psychology program).