One truth or many?


Let’s say researchers studied a question by conducting four studies in the same location, drawing from people with the same demographics, during the same period of time.

In that case, even if each study gave slightly different results, you could be pretty sure that the differences were only due to random chance and that each study was estimating a very similar, if not identical, effect.

For this kind of data, you’d use what’s called the “common effect” model.