"A democratic society needs Republican scientists."

So says this Slate article, which complains that only 6% of American scientists are Republicans. (55% are Democrats.)

The author, Daniel Sarewitz, argues:

For 20 years, evidence about global warming has been directly and explicitly linked to a set of policy responses demanding international governance regimes, large-scale social engineering, and the redistribution of wealth. These are the sort of things that most Democrats welcome, and most Republicans hate. No wonder the Republicans are suspicious of the science.

Think about it: The results of climate science, delivered by scientists who are overwhelmingly Democratic, are used over a period of decades to advance a political agenda that happens to align precisely with the ideological preferences of Democrats. Coincidence—or causation? . . .

How would a more politically diverse scientific community improve this situation? First, it could foster greater confidence among Republican politicians about the legitimacy of mainstream science. Second, it would cultivate more informed, creative, and challenging debates about the policy implications of scientific knowledge. This could help keep difficult problems like climate change from getting prematurely straitjacketed by ideology.

Sarewitz concludes that the United States "needs Republican scientists," but by his own account, we already have them: 6% of scientists. That's a lot of people. Why doesn't Sarewitz make any attempt to look at some of these scientists' work and explain how they've made a distinctive contribution that non-Republican (Democratic, independent, or apolitical) scientists wouldn't have made?

Scientists keep getting things wrong. Should we stop believing in science?

"Plenty of today’s scientific theories will one day be discredited. So should we be sceptical of science itself?" That's the teaser for a short, worth-reading article (via Arts & Letters Daily). Here's an excerpt:

Physicists, in particular, have long believed themselves to be on the verge of explaining almost everything. In 1894 Albert Michelson, the first American to get a Nobel prize in science, said that all the main laws and facts of physics had already been discovered. In 1928 Max Born, another Nobel prize-winner, said that physics would be completed in about six months’ time. In 1988, in his bestselling “A Brief History of Time”, the cosmologist Stephen Hawking wrote that “we may now be near the end of the search for the ultimate laws of nature.” Now, in the newly published “The Grand Design”, Hawking paints a picture of the universe that is “different…from the picture we might have painted just a decade or two ago”. In the long run, physicists are, no doubt, getting closer and closer to the truth. But you can never be sure when the long run has arrived. And in the short run—to adapt Keynes’s proverb—we are often all wrong.

Most laymen probably assume that the 350-year-old institution of “peer review”, which acts as a gatekeeper to publication in scientific journals, involves some attempt to check the articles that see the light of day. In fact they are rarely checked for accuracy, and, as a study for the Fraser Institute, a Canadian think-tank, reported last year, “the data and computational methods are so seldom disclosed that post-publication verification is equally rare.” Journals will usually consider only articles that present positive and striking results, and scientists need constantly to publish in order to keep their careers alive. . . . Historians of science call this bias the “file-drawer problem”: if a set of experiments produces a result contrary to what the team needs to find, it ends up filed away, and the world never finds out about it.

Despite all this and more, the author concludes the article by saying we should be generally credulous of science. Isn't this an outrageous paradox?

He thinks there's no other choice, since not to believe in science would be not to believe in anything, which would be paralyzing. "[S]cience is the only game in town."

Well, not really. If you're a professional scientist, it's certainly not that simple. You don't have a binary choice between "believing" or "not believing" in "science." You should be aware of enough of the complexities of your field that you can be more or less skeptical of different claims, using some kind of epistemic sliding scale.

And if you're a layperson, you usually don't have to believe in scientific theories at all in order to lead a productive life (as long as you're familiar with enough of the basics not to be embarrassed if they come up in conversation). But what if you're facing a specific problem and your best hope of a solution depends on science, such as taking medication? Well, you still don't need to be completely credulous about science in general or even the scientific claims behind that medication. You can take a gamble that the scientists are more likely than not to be right. This just means you think the odds that the medication is effective are greater than the odds of any other method you know of (including doing nothing); it doesn't mean you believe the odds that it's effective are 100%. You can use a working assumption that scientists are getting things right, based on your hunch that this usually turns out to be true -- but you can, at the same time, be skeptical of your own hunch. This needn't lead to paralysis. On the contrary, this semi-skeptical attitude can make it easier to move on once we discover, as we're sure to from time to time, that we were believing in bad science.

Can we close the gender gap in some sciences? Should we?

I've been critical of the New York Times in the past for its gender bias (I even have a tag for it), but I found this NYT article on the gender gap in the "math-oriented sciences" by John Tierney to be admirably balanced and well-reasoned (via).

This explanation for the gap strikes me as very plausible:

[M]en are more interested in working with things, while women are more interested in working with people. There's ample evidence — most recently in an analysis of surveys of more than 500,000 people — that boys and men, on average, are more interested in inanimate objects and "inorganic" subjects like math and physics and engineering, while girls and women are more drawn to life sciences, social sciences and other "organic" careers that involve people and seem to have direct social usefulness.

You can argue how much of this difference is due to biology and how much to society, but could you really affect it by sending scientists and engineers off to the workshops mandated by the bill now in Congress?

(The article has more info on the pending bill, if you're interested.)

In addition to those studies on male vs. female preferences, there's also evidence in the form of gender percentages in other scientific fields:

Aided by . . . continuing federal grants, researchers and advocates have developed theories that women are being held back from pursuing careers in engineering and physics by “stereotype threat,” by “implicit bias” and by a shortage of female role models and mentors. Yet none of these theorized barriers prevented girls and women from dominating the fields that most interested them.

The life sciences and social sciences were once male bastions, yet today women make up a majority of working biological scientists, and they earn nearly three-quarters of the doctorates in psychology. Now that women are earning a majority of all undergraduate and graduate degrees, it’s odd to assume they’re the gender that needs special help on campus. If more women prefer to study psychology and medicine than physics and engineering, why is that a problem for Washington to fix?

As I've said, quoting William Hazlitt, "It is essential for the triumph of reform that it shall never succeed."

I’d love to see more girls pursuing careers in science (and more women reading science columns), but I wish we’d encourage their individual aspirations instead of obsessing about group disparities. I can’t see how we’re helping them with scare stories about the awful discrimination they’ll face.

That last point is a very sensible one but also one that's sure to be ignored. There's a certain mindset that thrives on exposing discrimination, and this naturally leads to plenty of false positives. There are innumerable disparities in the world; some of them result from discrimination, but the mere existence of a disparity doesn't prove there's discrimination. But leaping to the discrimination conclusion can feel so energizingly righteous that you can be blinded to the downsides of false positives. Pointing out real discrimination is a noble thing, of course, but I think it's wrong (in the moral and factual sense of the word) to tell a group of people that they're being discriminated against to a greater extent than they are. I don't see how this is any less wrong than telling an individual that he or she is despised or being conspired against by people around him or her, unless this is actually true.

Getting back to Tierney's point about the "things"/"people" gender difference, does anyone really believe it should be off limits to posit that men and women are simply, ineradicably different in some ways? I hope not. We have to be open to this possibility in order to have an intellectually honest discussion of any kind of gender gap. A gender gap isn't necessarily a problem that needs to be solved; it could be just life.

(I've focused on only some of the article's data about, and possible explanations for, the gap, so I recommend reading the whole thing.)

Should children be medicated to deal with their psychological issues?

In this book review that's probably a more worthwhile read than the book, Alison Gopnik writes:

Within the past few years more and more children have been given powerful brain-altering drugs to deal with a wide range of problems. . . .

You can sympathize with the impulse of parents to do something, anything at all, to help their children. But that doesn't alter the fact that the scientific evidence just isn't clear about what to do. On balance, though, the evidence suggests that we should be conservative about prescribing drugs to children, and much more conservative than we actually are. Even the scientists who advocate some use of drugs acknowledge that they are overprescribed and badly managed. Brains are complex enough, children's developing brains are even more complex, and determining the long-term effects of drugs that alter those brains is especially difficult. Children are different from adults, often in radical ways, and many childhood problems resolve just as part of development.

On top of that, each generation of doctors discovers that the last generation was disastrously misguided in its medical interventions, from lobotomies to estrogen replacement, at the same time that they assure the patients that this time is different.

I'm also glad to see that the review highlights the importance of compatible "levels of description":

[Judith] Warner's book [We've Got Issues] also reflects a common confusion in popular writing about psychology. She writes as if there are just two kinds of explanations for human behavior. Either the everyday narratives are right—so that children are unhappy because their parents don't care about them, or they fail at school because they are lazy. Or else the right answer is that the children's problems are the result of "something in their brains." Warner's logic seems to be that since the parents do care about their kids, the problem must be in the children's brains and therefore drugs will fix it.

But everything about human beings, cultural or individual, innate or learned, is in our brains. Loss and humiliation change our serotonin levels, education transforms our brain connections, social support affects our cortisol. Neurological and psychological and social processes are inextricable. The work of psychological science is to identify causes at many levels of description—social, cultural, individual, and neurological.

There was a very insightful blog post (on Psychology Today's website) that makes a similar point about evolutionary psychology: "Is it evolutionary, or is it . . . ?" (That post speaks of "levels of causation," which is the same thing as the "levels of description" in the above block quote.) I've also blogged this concept before: "Can you give a neurological or evolutionary explanation of love without debunking the whole idea of love?"

By the way, that book review has several points that could be added to my list of ways blogs are better than books. The bias in favor of conclusions that come from riveting stories is huge.