Bending Science in Service of Book Promotion

If you don’t have a Ph.D. (or its educational equivalent), you shouldn’t consider yourself a serious researcher. The Ph.D. degree prepares (and qualifies) you to do solid empirical research that can stand up to peer-review.

You can certainly do science with any degree (heck, you don’t even need a degree to do science!), so naturally medical doctors (M.D.s) can do science. But it’s also why you see the really serious medical researchers go on and get a Ph.D. as well. The Ph.D. provides the deep didactic training in research methods and statistics you need to actually do serious research, not just interpret it.

That’s why I’m not a serious researcher — I know my limitations. I can interpret research until the cows come home, but I do very little of it myself.

That’s also why it was enjoyable reading Vaughan Bell’s post over at Mind Hacks about neuropsychiatrist’s Louann Brizendine new article over at CNN (which apparently doesn’t bother do any editorial vetting of the content it publishes). In the blog entry, Bell points out at least one ridiculous contention Brizendine makes in the CNN piece. Brizendine says:

Our brains are mostly alike. We are the same species, after all. But the differences can sometimes make it seem like we are worlds apart.

The “defend your turf” area — dorsal premammillary nucleus — is larger in the male brain and contains special circuits to detect territorial challenges by other males. And his amygdala, the alarm system for threats, fear and danger is also larger in men. These brain differences make men more alert than women to potential turf threats.

To which Bell replies:

Male and female humans are indeed the same species, but we are not a species which has a dorsal premammillary nucleus because it’s only been identified in the rat.

Furthermore, there is no reliable evidence that amygdala size differs between the sexes in humans and a recent study that looked specifically at this issue found no difference.

It makes you wonder — how much do authors “bend” scientific facts in order to fit their theories?

Not to worry, Brizendine is not alone in this practice. In fact, it appears that virtually any professional who’s on the hook for a followup book learns that they may need to bend the science in order to make a second book viable and interesting.

Christopher Shea, writing for the Boston Globe earlier this month, noted how Kaiser Fung, a professional statistician, was working his way through the book SuperFreakonomics, and found some questionable passages (SuperFreakonomics is the follow-up book to Freakonomics). He’s not the first to call out the book’s authors, Steven D. Levitt and Stephen J. Dubner, about some of their hypotheses that they forward in the book, without mentioning the contrary or contradictory evidence. Indeed, the books’ authors have had their findings debunked since their first book, demonstrating their seemingly clean, simple conclusions really were too good to be true. Because, in fact, they were.

Clean, simple conclusions sell books. Even if the conclusions are based upon shoddy analyses that breeze over contradictory evidence.

This explains why scientists primarily publish in peer-reviewed journals. When they forgo the peer-review process and instead just publish a popular non-fiction book, you can expect many corners to be cut in the name of what I call the Three S’s of Non-Fiction Success: Sales, Sexy conclusions, and Simplicity:

• Sales – First and foremost, the publisher wants a successful successor to the first blockbuster. This means authors have to make even more dramatic conclusions than in their first book to maintain Sales. Because if the second book doesn’t do better Sales than the first, a third book may be a difficult sell.
• Sexy Conclusions – People love sex. People love reading about sex. And people really love reading about how everything can be explained by sex or some gender differences. If it’s something about rats, that’s fine… You can gloss over the fact that it’s only been proven in rat studies by simply leaving that to the footnotes or references.
• Simplicity – You need to get to the point. Nobody’s buying the book to read dry academic studies. So authors make sure they season their book with lots of little anecdotes about small, unpublished classroom studies, or something overheard at a dinner party. People like stories because they’re simple and engaging. Stories have little scientific value, but they will “prove” the points nonetheless, at least in most reader’s eyes.

These are fun, interesting reads. But they often cut factual corners in order to make their point, and leave out all those messy studies that contradict the authors’ own hypotheses. That’s why researchers generally ignore books. They can act as a nice synthesis of our knowledge, but they often do so at the price of accuracy, forwarding a very specific author bias or point of view.

So next time you pick up a copy of Super Duper Freakonomics, or the Male Brain That’s Trapped Inside a Female Body, enjoy the read! But take it with a grain of salt, because everything you read in the book may not actually be true, despite the myriad of references to studies.

Read the original article over at Mind Hacks: Brizendine, true to stereotype