Evaluate the quality of the research against the standards for that type of research.
Select only studies with a sample size of at least 100 people.
Discard systematic reviews and meta-analyses since they are the weakest form of evidence.
Accept only studies that use a randomized controlled trial design.