You missed the point. For all intents and purposes this is (data) cherry picking. Slice and dice enough and you can find whatever it is you’re trying to prove.

You missed the point. For all intents and purposes this is (data) cherry picking. Slice and dice enough and you can find whatever it is you’re trying to prove. Done properly, that’s not how it works. An obvious and classic case of lies, damn lies and statistics. Also file under: there’s a sucker born everyday. Again. Interesting and almost thought provoking but ultimately inconclusive cherry picking. Re: “ They're trying to exclude participants who can introduce possible errors.” Are they? How can we be sure? Or are they crafting the subset to get the conclusion they’re looking to “prove”? I’m not saying they’ve been nefarious. I’m saying it’s painfully obvious we can’t be sure (due to all the heavy handed hand crafting of the subset).