Another shoe just dropped in the ongoing Dan Ariely scandal: JMR, a top 4 marketing journal, has issued a formal “Expression of Concern” for Mazar, Amir, & Ariely (2008), the infamous Ten Commandments study

🧵 This is Ariely’s most cited paper (4000 cites!), and I’ve long suspected that it would be the next domino to tumble.

The EoC highlights several major issues with the paper: first, a massive replication study of the MAA experiment run in 2018 failed spectacularly.

This is a cool application of Miller and Sanjurjo’s 2018 ECMA paper overturning the Hot Hand Bias. The basic intuition is that HH subsequences can overlap with themselves, while HT cannot. So in every set of finite coin toss sequences, there are an equal amount of HH and HT… 1/

https://twitter.com/littmath/status/1769408478139785497

…subsequences, but HT subsequences appear across more of the sequences because they cannot be “packed in” as efficiently. Thus, in a *finite* set of coin tosses, sequences w/HT are more likely to appear, even though the *expected* number of HH and HT subsequences is the same. 2/

You can't make this shit up. Lessons from the Sages of Dishonesty

The real issue is not about the original data being missing, though that’s also bad, especially for this prominent of a study. As @AaronCharlton has rightly emphasized, it’s about the fact that none of the original authors “remember” key details of how the data was gathered… 1/

https://twitter.com/cjwich/status/1678897459349553152

…and that their ostensible “attempts to recall” just happen to involve trying to establish that another academic, who is *not even a coauthor (!!)* supposedly collected the data for them. 2/

No English subtitles unfortunately, but some *very* interesting info in this Israeli Channel 13 reporting from last year about the ongoing Dan Ariely scandal.

A few things I didn’t know: 1/

13tv.co.il/item/news/hama… Ariely apparently left MIT in ‘08 after conducting an experiment that administered electric shocks to undergrads, *w/o IRB approval (!!)*. When confronted, he tried to throw his RAs under the bus. He received a 1 year suspension from running experiments, then moved to Duke. 2/

I've been using Mendeley Desktop daily for years, & have an annotated library of thousands of pdfs. Today, I switched to Zotero. Here's a 🧵of tips on how to do this, if you're also tired of Elsevier's BS and want to make the switch w/minimal effort while maintaining workflow. First off, why switch? In short, Mendeley has been in a severe decline for years. Elsevier already discarded the mobile app, & there hasn't been a feature update in living memory. The straw that broke the camel's back for me is that they're discontinuing the desktop app on 9/1.

I guess I have to do a response to the response now, because that's how arguing on the internet works. I'll keep it brief, given that I made most of my points in the original thread.

https://twitter.com/lymanstoneky/status/1253219512591106051

Let me be blunt in response: this argument is just plain dumb. Just because the infections variable is measured with noise, doesn't mean an analysis employing it is useless. Furthermore, as I discussed before, *deaths are also measured with noise*.

https://twitter.com/lymanstoneky/status/1253219743042920449?s=20

This hot take has been getting some attention today, so I think it's worth talking about why I think it's specious, and why the headline it comes with ("Lockdowns don't work") is probably wrong (and irresponsible) given the current empirical evidence.🧵

https://twitter.com/lymanstoneky/status/1252795142026608640

First off, @lymanstoneky asserts that no nationwide evidence from the US currently exists that lockdowns are effective. This is materially false. Here's our nationwide analysis showing lockdowns are associated w/dramatic declines in county infection growth rates in the raw data.