p-value
The probability of finding the coefficient in the sample if the null-hypothesis were true
Counterfactual outcome
Unobserved outcome
The treatment one didn’t receive
Identifiability conditions
For causal effects
Positivity
Exchangeability
Consistency
Positivity
Have to have people in both treatmentgroup as controlgroup
–> having a control group
Consistency
Well-defined intervention
One’s outcome is not affected by whether someone else was treated
Exchangeability
Treatment groups are exchangeable
–> randomize
Disadvantages of RCT
Limited generelizability
Practicaland ethical considerations