Woke up to the good news that our paper (w/ @nreinholtz) on group sequential designs is accepted at @JCRNEWS !
Want to learn more about designing more efficient and more informative studies? This blog post summarizes the key insights from our paper:
https://t.co/yER4FOPPHE
Hilarious. This company is selling "transdermal nutrition support", better known as "placebo stickers" or "emotional support band-aids".
No, lactase enzyme can't be absorbed through the skin and make their way to the guts.
"AI Mode" in Google is constantly hallucinating plausible-sounding facts about scientific papers. I guess this is a consequence of the paper being paywalled, but this is bad for knowledge dissemination.
Google, please fix this. Don't offer AI Mode for paywalled papers.
Related working paper: When outcomes improve naturally over time, people become more confident that they understand how their actions affect outcomes. This is because they do things they believe will improve outcome, and the trend "proves them right". https://t.co/lVYhUcZJB2
Hilarious QJE paper: societies in places where it’s more likely to rain after a dry spell are 47% more likely pray for rain (bc it seems to work for them)
Excited to spend a week in NYC May 30th-June 7th: first for the MSI Young Scholars conference, then to explore this great city.
Recommendations welcome (preferably kid-friendly, my 3yo is tagging along)!
@erinhengel@squig@lessig@albertobisin She also won't to pay damages to all the people who tried to build on her findings. So yes, really, it's not a cruel or unusual punishment.
@erinhengel@squig@lessig@albertobisin She's also not being burnt at the stake. If she built her entire career on fraud (and evidence suggests she did), it's actually a pretty mild punishment to take this career away. She's not being asked to repay the money she was paid, or the grants she earned...
@erinhengel@squig@albertobisin Another thing to consider as you are weighing Gino's credibility. She made *very* specific claims that HBS used the wrong file in their analysis. These claims, if true, are ABSOLUTELY devastating to HBS's case.
https://t.co/Q8eRzwsubk.
1/2
@erinhengel@squig@albertobisin But she has now deleted these claims, and has not mentioned them at all on Lessig's podcast. Why is that...?
A hint: HBS is claiming to have forensics evidence that the file she claims exonerates her was fabricated and backdated.
@erinhengel@squig@albertobisin Another thing to consider as you are weighing Gino's credibility. She made *very* specific claims that HBS used the wrong file in their analysis. These claims, if true, are ABSOLUTELY devastating to HBS's case.
https://t.co/Q8eRzwsubk.
1/2
@erinhengel@lessig@albertobisin Again, a lot of terrible work was produced in psychology and behavioral sciences, some sloppy, some fraudulent. But there are honest people working hard to reform this field (Data Colada among them), and that's why it's important to clean house when the evidence is so clear.
@erinhengel@lessig@albertobisin A mix of sloppy, outdated training (at the time, no one knew that subtle manipulations don't produce gigantic effects like in Gino's paper), and probably some "let's not kill the golden goose" reasoning for some who stood to benefit from her productivity.
@erinhengel@lessig@albertobisin The other accused person at Harvard also faked her data. Again, clear evidence, but it's a single paper. There is also a lot of fraud! But that's why it's so important to pursue it harshly and not try to find excuses for fraudsters.
@erinhengel@lessig@albertobisin No, Ariely faked the data too. The evidence is incontrovertible. Duke just failed to do the right thing, as it often happens in cases of fraud. Gino's downfall was that there was repeated evidence in multiple papers across a decade. Much harder to play dumb like Ariely did.