Regression to the mean has always baffled me. Unless it works like this. People with the most extreme values at Time X will usually have less-extreme values at Time X+1, but others, initially less extreme, will drift out and replace them. So it's more like "shuffling at the extremes." Yes, no?
Bonus difficulty: it's a controlled substance. We can't just sip it ourselves.
Does anyone know of a blinded study in which a liquid medication (probably somewhat bitter) was matched to liquid placebo with any reported detail about how the respective flavors were checked? Every paper I've found has seemed to say "trust us, they matched."
"Did you come up with a snappy title for our paper?" "I sure did!" "Great! And you checked to make sure it hasn't been used before?" "Oh, this is guaranteed fresh."
"Convenience sample" is such a sneering term for something that's often excruciatingly effortful. I hear it in Dana Carvey's Church Lady voice.
This thing is making a reappearance on Twitter, and no one appears to know that the answer is zero. Good luck claiming your earned-income tax credit on those gains, cowboy.
You could say that scientists have a compulsion to fold things, but that would be unfair. Scientists have a compulsion to fold things a specified number of times. (Screenshot: quasirandom selection from papers I've recently read.)
We're neck and neck! We ran some long EMA studies in which we got a mean of 295 random-prompt entries per person (median 309, range 11-405) (doi 10.1111/j.1521-0391.2012.00230.x) plus EMA reports of drug use and craving (about 21 per person, doi 10.1001/archgenpsychiatry.2008.509).