Comments on: That Chocolate Study http://slatestarcodex.com/2015/05/30/that-chocolate-study/ In a mad world, all blogging is psychiatry blogging Fri, 24 Jul 2015 13:00:33 +0000 hourly 1 http://wordpress.org/?v=4.2.3 By: That Chocolate Study | Beyond Labels http://slatestarcodex.com/2015/05/30/that-chocolate-study/#comment-211790 Mon, 15 Jun 2015 15:41:17 +0000 http://slatestarcodex.com/?p=3658#comment-211790 […] http://slatestarcodex.com/2015/05/30/that-chocolate-study/ […]

Report comment

]]> By: Chocolate and Disillusionment | closetpuritan http://slatestarcodex.com/2015/05/30/that-chocolate-study/#comment-211030 Fri, 12 Jun 2015 20:46:57 +0000 http://slatestarcodex.com/?p=3658#comment-211030 […] Slate Star Codex also discussed the chocolate study, and how some people are taking unwarranted conclusions from it. I think some of his commentary about Conclusion 1 is kind of questionable–he’s combining findings of blood pressure reduction, flow-mediated dilation, insulin sensitivity, absolute BMI, and weight gain. Only the non-meta-analyses had findings related to BMI or weight gain, and even with meta-analyses we should be cautious, since the bias against publishing negative findings can affect them too, and from what I can tell from a quick pass, the effects were statistically significant but small in magnitude. (And similarly, in Conclusion 2, I think he overstates how much certainty we can have about chocolate’s health benefits.) […]

Report comment

]]> By: Alexander http://slatestarcodex.com/2015/05/30/that-chocolate-study/#comment-210115 Tue, 09 Jun 2015 22:29:05 +0000 http://slatestarcodex.com/?p=3658#comment-210115 One must note that being able to ignore the news is a privilege enjoyed by those living in the developed world. If I ignored the news where I’m living in the past 3 years, I would be definitely much, much poorer, and also conceivably dead.

Report comment

]]> By: A week of links | EVOLVING ECONOMICS http://slatestarcodex.com/2015/05/30/that-chocolate-study/#comment-208738 Fri, 05 Jun 2015 09:02:04 +0000 http://slatestarcodex.com/?p=3658#comment-208738 […] perspectives on the chocolate study – Scott Alexander and Andrew Gelman. I would say that the chocolate study didn’t tell us anything that we […]

Report comment

]]> By: Adam http://slatestarcodex.com/2015/05/30/that-chocolate-study/#comment-208643 Thu, 04 Jun 2015 19:13:20 +0000 http://slatestarcodex.com/?p=3658#comment-208643 The basic idea is that a study purposefully looking for one thing that attains a p < 0.05 has only a 5% chance of incorrectly rejecting the null. A study looking at 18 hypotheses, on the other hand, has a 60% chance of incorrectly rejecting at least one null (assuming iid binomial). Hence the point of using FWER to find a p-value that brings that probability back down to 5%.

You're right, however, that it's also true that out of any 18 studies, there is a 60% chance that at least one of them incorrectly rejected a null, which is basically what this author is taking issue with in the first place.

Report comment

]]> By: Doug S. http://slatestarcodex.com/2015/05/30/that-chocolate-study/#comment-208532 Thu, 04 Jun 2015 07:26:07 +0000 http://slatestarcodex.com/?p=3658#comment-208532 Wine is spoiled grape juice.

Report comment

]]> By: RCF http://slatestarcodex.com/2015/05/30/that-chocolate-study/#comment-208518 Thu, 04 Jun 2015 05:34:28 +0000 http://slatestarcodex.com/?p=3658#comment-208518 “You measure what percentage of trouble tickets are close in two hours, you’ll get more closures at 2 hours.”

Well, that’s rather incomprehensible.

Report comment

]]> By: RCF http://slatestarcodex.com/2015/05/30/that-chocolate-study/#comment-208517 Thu, 04 Jun 2015 05:32:46 +0000 http://slatestarcodex.com/?p=3658#comment-208517 It can be viewed as 18 studies performed concurrently. And you didn’t answer me question. How does having several variables make the weight one invalid?

Report comment

]]> By: Steve http://slatestarcodex.com/2015/05/30/that-chocolate-study/#comment-207928 Tue, 02 Jun 2015 21:00:05 +0000 http://slatestarcodex.com/?p=3658#comment-207928 Granted, there is also the matter of Niel’s plagiarism of quotes (don’t have the link off hand, but he made up some ridiculous things “said by religious nuts” and then road that skepticism/slander train to celebrity (among the athiest/skeptic/’scientist’ community and in the hearts of really stupid “nerds” on tumblr everythwere), and general willingness to suck every available media opportunity out of any cock which is presented to him.

Report comment

]]> By: NZ http://slatestarcodex.com/2015/05/30/that-chocolate-study/#comment-207835 Tue, 02 Jun 2015 17:02:55 +0000 http://slatestarcodex.com/?p=3658#comment-207835 I dunno, the general “we”? I obviously don’t ascribe sacredness to journalism. Should I have said “y’all”?

Anyway, you “need” them to be honest. Okay. Does that mean you expect them to be honest? That you hope they’re honest? That you think honesty is somehow ontologically related to what journalists do for a living?

A diversity of dishonesty is exactly what I was calling for when I said that journalists should turn over their output not to the public but to writers and bloggers who are expert in just a few areas, who both will be able to scrutinize what the journalists give them and will not try to feign disinterested objectivity.

Report comment

]]>