Substance vs. Cynicism

Focus on the facts when scientific integrity is at stake

A few days ago, I boiled down my recent criticism of Zandberg et al’s abortion restrictions-suicide article, which reports false findings, to a short comment. After the journal turbo-rejected it in less than an hour, I contacted the editor, Dost Öngür. He emailed me back an incredibly cynical response, now posted below my original correspondence. Here’s a quick summary, beyond the usual takeaway from the so-called science crisis: Science is done by human beings, and this is why we can’t have nice things. 

The substance: Zandberg et al report findings which are false according to their own published data and analyses. 

This is not up for debate. It’s not a matter of interpretation. The evidence is in the article and its supplement. You can read the relevant pages here. They show that the results don’t match the reported findings; the article instead misrepresents, omits, and inflates data, all in the same direction — of fitting the preferred narrative that abortion access restriction hurts women’s mental health. The authors meanwhile ignore a reliable ~2x suicide risk increase associated with abortion, and analyze data that do not match reproductive status with suicide.  

The reviewers who approved the article should have noticed the overt p-hacking, data omission, and misrepresentation of findings seen by comparing Table 1 in the article with eTable 5 in the supplement, and recommended revision or rejection. The editor should have noticed and not published the article for the same reason. Now that it’s been pointed out on the Internet, where in theory everyone can read it and see the merits of the facts, you might think the journal would retract.

The outcome: The outcome of interest to me originally was whether the first author would himself retract or engage in dialogue about retracting. I’m interested in intersections of science and activism, and science reform has loads of them. Why don’t scientists do the right thing in the interest of correcting the record? Here, retraction is also in the public interest, since dubious science may contribute to preventable harm to women. I have questions. But Jonathan stopped responding after I explained my concerns in the original post and asked him to please retract the article. So I figured I’d do my due diligence and be sure the journal knew what was up, by telling them. 

The ad hominem attack: In his response to my email, the editor, Dr. Öngür, didn’t engage with substance at all. Rhetorically speaking, that was a good move, since my argument is factually correct, based on evidence that’s now in the public domain, and makes it look like he didn’t do his job. Instead, he accused me of unfairly alleging research misconduct, and invited me to resubmit a comment for online publication on the journal’s website (not open access) that has original content and does not repeat this allegation. 

Look, I’ll be the first person to admit that I can be terrible with tone, need to dial it down, jump straight into the middle of methods and other conversations without a “how do you do,” overcompensate for insecurity and self-loathing with just the sort of overconfidence my methods rockstar idols criticize in statistics reform conversations (ironic) — while also channeling rage over injustice in the world like performers have always done — and slouch. In other words, I’m a human being with my own problems.

But no fault of mine (ad hominem) can change the substance at issue here, which is the evidence clearly showing research misconduct in this case. In fact, the euphemism “research misconduct” is a collegial umbrella that throws a shade of ambiguity this type of behavior doesn’t warrant. I’m reminded of Elisabeth Bik’s story of how angry it made her when she first discovered science’s photoshopping problem: “This was cheating, pure and simple... Cheating is a transgression against everything that science should be.”

Zandberg et al cheated, the published evidence shows. The journal published false findings as a result. The editor is responsible. And he’s more concerned about commanding deference to the tiny fiefdom of his paywalled platform, than the truth.

So What?

The interesting question, I think, is not what happens next. Authors who cheated have disincentives to admit it. Editors who published science fiction have disincentives to correct the record. Reputations are at stake. Retraction looks like an admission of guilt. 

So I had hoped Zandberg et al would accept the sort-of plea bargain of saying they made an honest mistake in p-hacking, even though the evidence is pretty clear that’s not what happened here. You don’t accidentally p-hack AND omit the rest of the relevant data AND misrepresent your findings AND inflate them AND ignore all the literature reporting the exact opposite finding from better data — with all these mistakes and more flowing in the lone direction of confirming your preconceptions that promote a preferred narrative and get you published (not to mention in the news). Oops!

About the news: Journalists who reported widely on these false findings did so because they fit the preferred narrative that abortion access restriction may harm women. They either didn’t understand what they were reporting on well enough to factually report it, or they didn’t bother reading the article and the supplement to see what the evidence really said. Apparently, nobody did.

Everyone looks bad if they say this.

So they don’t. The authors don’t do dialogue once you just say the facts. The journal and the media that published and widely disseminated the false findings shut out criticism. And the record stays wrong. 

This is really interesting because, ostensibly, we live in a time after gatekeeping. But it shows one way in which controls on information persist. Free online publishing doesn’t spell equal opportunity to disseminate information, even when the information you’re disseminating is more accurate than what’s been put out there before.

The truth has an SEO problem that affects both sides: People with institutional power would like to be able to better debunk misinformation. And people without it would, too.

People who want to do, read, benefit from, and benefit others with better science have an interest in thinking about this sort of problem more. What does it mean to promote scientific integrity in a system that disincentivizes and in some contexts even explicitly forbids talking about it? How do you advance as a conscientious scientist when the people who have power over your career are more concerned with heuristics than substance? 

Maybe you don’t advance within that system. Punk poetess-singer Patti Smith said “Freedom is… the right to write the wrong words.” Transgression doesn’t come with protection; then it wouldn’t be transgressive.

Speaking truth to power, breaking taboos that hurt the public interest, and promoting resistance to illegitimate authority will always be minority vocations. But you don’t have to play the game, tiptoe around the facts, or let your opponent shift the focus from the substance of your argument to the way you’re standing, to speak up. 

Say the wrong words when it’s right. Science doesn’t belong to institutions. It belongs to the human beings who create, recreate, and otherwise enjoy it. And there is life after academia. There is even science.