False Positives, FWER, and FDR Explained

False positives are something we must think about when making inferences, but not everyone is aware that the probability of getting a false positive skyrockets as more comparisons are made. In this blog post, I discuss the problem of false positives and controlling error rates.

Meta-Analysis: Choose Your Model Wisely

Meta-analyses are incredibly common in the literature and they can be hard to understand. In this post, I explain why we conduct meta-analyses in the first place, the effects of different assumptions on the calculations, and traps to avoid when choosing a model.

Myth: Covariates Need to Be Balanced in RCTs

Many people believe that the purpose of randomization is to perfectly balance out both known and unknown covariates in order to reach causal inferences, however, this is simply not true and is a misunderstanding of the process of randomization.