This is often what happens with meta-analyses: many people have convinced themselves there is a pattern based on limited-but-good data, have asserted so in the

This is often what happens with meta-analyses: many people have convinced themselves there is a pattern based on limited-but-good data, have asserted so in the literature such that it's almost accepted as fact, and then a meta-analysis actually crunches the numbers and finds little to no statistical support for the claim. A recent well-documented case is that of jellyfish blooms in the oceans: http://theconversation.com/how-time-poor-scientists-inadvertently-made-it-seem-like-the-world-was-overrun-with-jellyfish-61564 http://theconversation.com/how-time-poor-scientists-inadvert... Just to clarify, I'm not saying these people are stupid. But rather that Feynman was very right when he said that the easiest person to fool is yourself.