I think what is missing in this interesting episode is explaining the role of p-value - false positive value. As a standard it is <5 % for saying results are statistically significant, whatever the effect size is. So if you analyse or divide data on average in 20 different ways, you should find something "significant". https://xkcd.com/882/

I've browsed the paper "False-Positive Psychology: Undisclosed Flexibility in Data Collection and Analysis Allows Presenting Anything as Significant".
I'm surprised that among recommendations there is no duty to publish the p-value, sometimes papers write only whether p-value is below a threshold.

The other side of p-value is when after combining all research results with thousands of participants, we get something like p=0.1, which is oversimplified in that there is no evidence. I think it is better to act considering all the information gathered, also those that have higher chances of being wrong, like 10 %, especially when being wrong is low cost compared to benefits. Like whether the public should wear masks during the pandemic or whether supplementation with minerals and vitamins has any benefits.

I thought it was amazing to do piece on a subject like this and not mention the academic fraud papers by Pluckrose, Lindsay and Boghossian. They were an amazing look into the suspect nature of the academic journals. It blew up that side of academic review. Makes me think that the political implications of their work played a hand in this omission, which makes their work even more prescient.

I'd love it if the feds paid for a vastly increased peer-review testing of academic papers. Make that a basic requirement for grad school students to participate in.

This is one of the most fascinating episodes (about academic fraud) in Freakonomics. The funniest thing that Mr. Dubner said was that he has been naive, he always thought that academia was held to higher standards for research and repeatability, etc. He seemingly comes to the shocking (sarcasm included) conclusion that there is fraud in the research, etc. And that this is done for other reasons beyond academic integrity.

Come on man - the entire Mann "hockey stick" lie about man-made global warming has cost this world millions and BILLIONS of dollars. Mr. Dubner, do you not see that?

Excellent episode, as always. It makes me think of this Wittgenstein's letter to Norman Malcolm in 1940:
“Congratulations to your Ph.D.! And now: may you make good use of it! By that I mean: may you not cheat either yourself or your students. Because, unless I'm very much mistaken, that's what will be expected from you. And it will be very difficult not to do it, & perhaps impossible; & in this case: may you have the strength to quit.”

What was the nature of the fraud? The Freakonomics audio statement was the 31:20:00 – 33:15:00; the indication was the 4 Tawain's professionals journal was retracted from the journal article entry describe as the Judging and Decision-Making journal article. The authors used correlated data from another citation which did not make sense.

Who committed the fraud? The details were that three private data detectives public finding fraud of 10 years ago, Yeary announce that the data set 9.1 9.2,9.3, to 9.9 but however there were –1.7 so the data set was wrong or broken.

How was the fraud discovered? It was discovered by a chart comparison from one article to another.

What was the punishment? No punishment was rendered, or sanctions adjoined.

Please help create more dialogue, what important details i'm missing?

I need to existing while this system is experienced and relevant and truthful. Fraud is real.

I am a community college English professor, but my husband is an economics professional. Since day one of our 23 year relationship, I have called BS on the entire economics sphere. This episode has me in absolute tears from laughing so hard. I’ve always found the invisible hand to be ridiculous and any system designed to objectify people into datasets is also ridiculous. There is value in quantifying things, but as the episode explains using math to explain things does not equate with objectivity. Thank you for a truly enjoyable discussion of our human flaws.

This is priceless. Let's investigate the research behind all the Covid-19 mRNA jabs..