In my experience, studies are typically not well constructed. See the numerous studies that cannot be replicated, the below-the-surface iceberg of unpublished

In my experience, studies are typically not well constructed. See the numerous studies that cannot be replicated, the below-the-surface iceberg of unpublished (unfavorable) studies paid for by for-profit institutions, or the shockingly small sample set or biased sample set in most studies. Most studies expose this information implicitly but it's up to the reader to determine the bias or replicate the study.