Canvas & Ratio
Choose your destination platform format
Layout Template
Choose a content structure for your slides
Preset Themes
Typography & Sizing
Brand Kit Customization
AGENCYConfigure brand assets for headers & footers
Outro Slide CTA
Customize your closing call-to-action slide
Background Pattern
Build Your Carousel
Drag and drop any post card below onto a slide, or use the quick buttons to insert content/images instantly!
I just updated one of my articles a second time. It has to do with Justice Jackson's comments that when Black newborns are delivered by Black doctors, they're much more likely to survive, justifying racially discriminatory admissions. We now know the study contained fraud🧵
The original article claimed that, when Black babies are attended to by Black physicians, their infant mortality rates decline substantially relative to when they have a White physician. Justice Jackson cited this in the Supreme Court, even though it was implausible.
A few months back, we learned that the original finding was driven by the authors failing to include a required control variable. Not only that, but they seemingly knew they this variable was important. <a target="_blank" href="https://x.com/cremieuxrecueil/status/1835802211395268913" color="blue">x.com/cremieuxrecuei…</a> <a target="_blank" href="https://x.com/cremieuxrecueil/status/1835802211395268913" color="blue">x.com/cremieuxrecuei…</a>
Now, the organization @donoharm has done a Freedom of Information Act request that turned up pre-publication emails between the authors and an early draft of the paper. In that early draft, we see evidence that they actively lied to the public and to the editors of @PNASNews.
My updated article explains how this happened. I'll start with their least bad offenses, before moving to the worst of it. Firstly, the authors were warned not to make causal claims because they lacked the fixed effect I said they needed. They made those claims anyway.
Secondly, we had hints that these authors knew about the critical role of birthweight and neglected that control anyway, but now we have confirmation that they neglected it. We now know for certain they were asked to look into it.
And finally, fraud. The authors published that racial concordance did not help White newborns, only Black ones. But in the now available earlier draft of their paper, we see that they did not find that: They found that White doctors with White newborns had lower mortality too.
Why, then, would they have put their quote about there being "little benefit for White newborns" into their paper? Because they're liars. One of the authors deleted the line because they'd "rather not focus on this" as it "undermines the narrative".
As a reminder, the narrative they aimed to promote was a racial blood libel. They wanted to promote the view that America needs more racial discrimination in college admissions, harming Whites and Asians, in favor of lower-performing Black students.
In my opinion, undermining blood libel is good. If someone lies, that should be undermined, not supported, as these authors clearly wanted. The result of their research was that tons of research money was wasted in following up a dead-end based on scientific dereliction.
The next steps are clear: Debar these people. No more grants, no more academic positions, and retract the paper now. With any luck, the damage can be mitigated from here onward, but given how little retractions tend to matter, I doubt it.
Unfortunately, what I think will happen is not very good. I think these authors will get off with a slap on the wrist. That's what tends to happen with academic frauds, anyway.
Citation for my point on retractions: <a target="_blank" href="https://www.cremieux.xyz/p/sometimes-papers-contain-obvious" color="blue">cremieux.xyz/p/sometimes-pa…</a> Article explaining this saga and linking the FOIA'd documents: <a target="_blank" href="https://www.cremieux.xyz/p/missing-fixed-effects-dont-justify" color="blue">cremieux.xyz/p/missing-fixe…</a>