Often, when confronting new research, I have to remind myself of one of my favorite aphorisms ever, from the much-lamented former blogger (and current scientist) Charles Murtaugh:
The universe is not here to please you.
It would be nice if the empirical evidence offered definitive proof that each and every one of my policy ideas was Pareto Efficient, but there you are: the universe is not here to please you.
Which is funny, because he said this in response to being called a racist for suggesting that there might be persistent heritable inter-group differences in IQ. Charles Murtaugh is pretty darn liberal, and as you can imagine, his fellow liberals did not take kindly to his statement that there might be something there; hence his issuance of the dictum that has wormed its way into my heart. And this new study on the neurological differences between liberals and conservatives made me think of nothing so much as the debate over race and IQ.
It certainly could be true; perhaps conservatives and liberals were programmed from birth. But I have two problems with it. First, I've spent too much time watching groups agree with each other not to think that one's peer group has very large effects on one's beliefs. Would a doctrinaire liberal from the Upper West Side who gets angry at the very notion of questioning the wisdom of affirmative action really look very much different from a doctrinaire conservative in Kansas who gets angry at the very notion of questioning the sanctity of marriage?
Almost any study of this type is going to be done in a place which skews one way or another. In that place, you will have two types of people:
1) People who agree with the dominant orientation of the locality
2) People who disagree with the dominant orientation of the locality
The people who disagree will probably look very different from the people who agree, because the mental qualities needed to maintain strong disagreement with one's neighbors are probably somewhat singular. But that doesn't mean their mental model caused the beliefs, and if you switched to a different location with a different orientation, you might get results showing the opposite.
My other problem is that the west has a long history of research on race and IQ, and gender and IQ, and so forth, that generally finds--quel surprise!--that the dominant group is genetically superior. This could be true; to discuss a subject I feel less queasy bringing up, I find it possible that male IQ's are distributed with fatter tails, so that there are more male geniuses (and cognitively disabled people) than female. But it's also true that scientists, like everyone else, have a tendency to find what they are expecting. Often, it's a case of not asking yourself the right question that might disconfirm your findings: like "Is the West Village a good sample of American political philosophy?" Evolutionary biology stories about gender are often plausible, but then, I can generally tell an equally plausible story that cuts the other way.
So when I read about research that has confirmed that a group the researchers themselves belong to is smarter, or more flexible, or just generally groovier, than some other group of folks, I can't help but reach for the salt shaker.
The people who disagree will probably look very different from the people who disagree
It's not a typo, it's a McArdle.
I am lost. You know the political persuasion of the researchers? How, exactly?
Ms McArdle links to a mushy Yahoo story. A much better account, including some of the methodology is here (LA Times):
http://www.latimes.com/news/science/la-sci-politics10sep10,0,5982337.story?coll=la-home-center
That's true, the NYU political science department certainly could be right-leaning...relative to the NYU theater department or something
So Jack weighs in with "all academics are leftists." Boring, trite, silly.
Besides, Jack, it's from the psychology department, not the political science department. But, hey, don't let a little fact stand in the way of a knee-jerk wingnut reaction!
No, I weighed in with "political science professors at NYU are left-leaning". Now, maybe I'm getting a little over my head because the only information I'm going on is having 6 highschool buddies who recently graduated from NYU, i.e. anecdotes, but would you really lay money on that being wrong?
"New York University political scientist David Amodio"
Whatever. Chuck, you realized in a scant few lines of posting you've used the words "boring," "trite," "silly," "knee-jerk" and "wingnut"? It's like you're trying to win a rousing game of ad-hominem bingo
Yeesh! Only on the internet would this conversation immediately devolve into name calling.
Cleverly anticipating leftist denial of bias, the lead author says:
Remember Satan's greatest trick: to convince the World he did not exist. I smell a liberal.
"New York University political scientist David Amodio"
Wrong. See above. Poor sourcing by McArdle. Caveat emptor.
Jack, your sampling of six is meaningless. What real support leads you and Megan to assume the authors of the study are liberal? And bets don't count. I am looking for real argument, not things you invent in your head.
Amodio, Assistant Professor of Psychology. You and Megan should practice the Google thing.
And saying you comment is boring, silly, and trite is not ad hominem. If I said you were not qualified to comment on the topic because you are boring, silly, and trite, that would indeed be ad hominem. And what else was your claim about their political persuasion but knee-jerk? You didn't base it on any real proof, did you? It is by definition knee-jerk, especially when you even had the department wrong. If you look at his academic interests, politics is only a piece of it.
So do a little thinking, find some real facts, use real argumentation, and stop whining. You have already bored the crap out of me.
The facts have a liberal bias?
At one point, biologically, it was probably a good thing to simply pick a story and stick with it. The benefits of doing one thing were greater than the benefits of pursuing multiple courses and frittering away the family unit's energy. Then, the world got complicated.
IIRC, the brains of certain kinds of very intelligent people are NOT more complex. Instead, their brains appear to cull superfluities. So, long story short, "it would be a mistake to conclude that one political orientation was better" hardly seems to the work of "bias".
Charles, it seems that McArdle's sloppiness has banjaxed her once again! Had she known the AFP report was wrong about Amodio's actual job, she might not have so readily jumped to the conclusion (he's a liberal) on which her whole argument is based. A little more time and thought and that "work thing" would have kept her out of trouble. She just doesn't learn.
So a sloppily written blog entry based heavily on a sloppily written news story. And to think the Atlantic was once the publication of Nathaniel Hawthorne and the leading intellectuals of the 19th century.
What she said:
First, I've spent too much time watching groups agree with each other not to think that one's peer group has very large effects on one's beliefs.
What she meant to say:
I've spent too much time watching members of a group agree with each other...
And on and on and on...
I absolutely agree that the "people who disagree will probably look very different than the people who disagree".
(Which magazine was it that was complaining that bloggers needed editors?)
And, by the way, the proper idiom is "different from."
The universe is not here to please you." Yes, but as e. e. cummings said: "nobody looses all of the time". It sounds like a well-conceived study in which the researchers took ample measures to avoid letting their own politics creep into their results. This is absolutely not to say that it is at all above criticism; but such criticisms will need to look at the particulars of what has been done, and what has been claimed.
"It certainly could be true; perhaps conservatives and liberals were programmed from birth." It's important to be careful here not to think that something's having a neurological basis entails that it is in any way innate or 'hard-wired'. I'm not seeing anything in what has been reported about this research that indicates that they are making the latter, stronger claim.
It is possible that there's a sort of geographical confound here along the lines Megan suggested, and I hope it's one that the researchers have already looked at. If they haven't, I'd bet they'd recognize it as the kind of confound to be eliminated on future versions of the experiment. In general, we should expect that overall they've done a pretty good job of eliminating confounds -- _Nature_ is not in the habit of publishing methodologically shoddy work.
Wow, I'm stunned. This whole thread has devolved into harsh criticisms that are fairly incidental to Megan's argument - namely, that one small study asserting deep mental differences between political types is extremely vulnerable to bias and misinterpretation by both the researchers and the reporters, and should be taken with a grain of salt. That's an appropriate stance, methinks, regardless of the piddly sourcing issues people are putting around (a legit gripe, maaaybe, but mostly incidental to the point of the post).
Really, what liberal who's been watching current events over the last few years would disagree with Megan's point, on principle? Haven't you been listening to your own anti-intelligent-design or pro-global-warming arguments? Skepticism of single-study-science is healthy, people. Consensus matters. So wait for the consensus of studies to confirm an unexpected result of a single soft-science study before you leap to its defense.
I confess myself mystified as to the skepticism towards this article. They appear to be saying that political conservatives tend to be privately conservative as well, and similarly for political liberals. Why is this controversial?
James, I am not leaping to any article's defense. I haven't read it. Instead, I am merely asking how Megan could assume she knows anything about the reseacher's political bias. She clearly has not read the original research, and only cited a poorly written news summary of it.
In my world, the way to evaluate research is to begin by actually reading it. Stunning, eh? Then you would start looking at the methodology, the data, the related research, the other papers cited, and so on. I would never, in a million years, assume I know anything about the bias of the authors or the tendency of certain types of research to reflect bias without some sort of evidence to support such an assertion.
You give Megan far too much credit. She says nothing of substance as to why such research is subject to bias. She merely assumes that it is, though she waves her hands a little about, "people who disagree will probably look very different from the people who agree." (A point that several people have correctly noted is barely understandable.) She doesn't know about the data used, or the methodologies. At the end of the day, Megan's response is incredibly simplistic. "Oh, it must be by a liberal." Such a response has no value at all.
I am reminded once again Megan is the blogger who dismissed the Lancet studies of the civilian deaths in Iraq when she clearly had not even read at least one of the articles. Sorry, the world of research doesn't work that way, where critics try to chime in without having read the material in question.
A note to any Atlantic editors who might be reading. It is time to raise the bar here, because this blogger reflects very badly on what is an otherwise fine institution.
So Charles, where, exactly, does Megan make a claim about the researcher's politics?
She mentions Murtagh and she mentions the Yahoo story. From there she makes a general point about such studies.
Anyway, in the goal of advancing actual discussion, the study involved this (from the LA Times version):
Each participant was wired to an electroencephalograph that recorded activity in the anterior cingulate cortex, the part of the brain that detects conflicts between a habitual tendency (pressing a key) and a more appropriate response (not pressing the key). Liberals had more brain activity and made fewer mistakes than conservatives when they saw a W, researchers said. Liberals and conservatives were equally accurate in recognizing M.
What exactly is being tested here, anyway? Perhaps liberals play more video games than conservatives. Or perhaps they are more vigilant to the evil of the letter W.
Hi Henry,
Please don't make me do basic work for you.
The coverage of the story indicates that the research found that liberals tend to be more flexible thinkers.
Megan says, "So when I read about research that has confirmed that a group the researchers themselves belong to is smarter, or more flexible, or just generally groovier, than some other group of folks, I can't help but reach for the salt shaker."
Ergo, Megan has concluded that the researchers are liberal.
Personally, I'm still trying to jibe this post with the one below chiding Chait for being too rhetorically dismissive of an expert in a field he's not a professor of.
Megan, a question. Why is ok for you to question the validity of a study from a field you have zero knowledge of, but a trained journalist (whose book you didn't effing read but agreed to discuss nonetheless) rhetorically dismissing someone who doesn't agree with him has failed journalism itself?
As has repeatedly been mentioned by many, you don't even bother to google before writing. You don't even bother to proofread your posts, as evidenced by the misspellings, garbled syntax, and plain nonsensical constructions. By any standard, you fail as a journalist.
And yet, you trust your own biases more than those you "see" in a study you haven't read.
You really are a perfect pundit. I'm sure we'll be seeing you on Lou Dobbs soon. Just remember, reality wants you out of work, so keep ignoring it.
Why is a study with a sample size of 43 taken seriously by anybody?
To Charles:
And when a comment poster is defending a study that has confirmed that a group [he himself] belong[s] to is smarter, or more flexible, or just generally groovier, he sometimes calls anyone who questions the study a "wingnut".
Charles Giacometti - Please don't make me do basic work for you.
Please don't make me a party to your assumptions.
As has repeatedly been mentioned by many, you don't even bother to google before writing. You don't even bother to proofread your posts, as evidenced by the misspellings, garbled syntax, and plain nonsensical constructions. By any standard, you fail as a journalist.
*************************************************
Actually, she's several steps above Yglesias in grammar and misspellings
"I first heard rumor of this "General Betray Us" ad plan Friday night and it sounded dumb. And when I saw it this kornng it looked dumb"
kornng???? Does he type with his knuckles?
Charles--
generally, it's a safe assumption that faculty (especially in the social sciences/humanities) are liberal. There was a study done on this, as it happens:
http://www.ratio.se/pdf/wp/dk_ls_diverse.pdf
That said, Megan's arguments rest mainly on potential methodological issues with the research-- sampling bias, y'know. The researcher's own possible biases are only brought up to suggest why we shouldn't accept the results uncritically; they're not a reason to reject the research entirely, and she doesn't present them as such.
One last thing-- it's a lot easier to convince folks of your reasoning if you aren't constantly hurling insults/nit-picking their grammar to score rhetorical points. Just a thought.
WSS, if you are suggesting that I defended the study at any point, you are not a very good reader. But if this kind of nonsensical response helps you feel better, I am very happy for you. Don't expect me to answer in the future though.
25Hour, same point about reading more carefully. Where did I nitpick an opponent's grammar? I made a correction to someone's wording--someone who agreed with me. If I made an error in language, I wouldn't burst into tears if someone corrected me. If you think I am hurling insults when I (correctly) point out someone is being trite or knee-jerk, then you are hypersensitive and/or determined to just disagree. That is your problem and not mine.
As to your attempt to justify Megan's wholly unsupported accustation of bias on the part of the researchers, you pointing out the general demographics of the political leanings of academics says absolutely zero about the particular accusation she is making wholly without basis at this particular researcher. Consider the logic only for a moment. Because some academics lean one way politically, I can immediately assume this particular academic also does? If you think so, you understand zero about logic, argumentation, and debate. And if my reply offends you, too bad. Unless you can make better points, don't expect me to answer you.
Look, this is very simple. Your little blogging heroine here Megan is not very bright. She wrote about something that she is absolutely clueless about, and in the process, revealed both her ignorance and her biases. If that makes you mad at me, I really can't help you.
Not in a million years would I have expected that the Atlantic Monthly web site would be so full of readers with such a fundamental lack of reasoning skills, but I guess this is what happens you add a blogger like this.
I agree with 25hour that the nitpicking about spelling, etc. is just ridiculous. It's a BLOG, people. It's not a book manuscript.
There is a question here as to what different people have in mind by taking the results "with a grain of salt". If it just means being motivated to take a close look at what the researchers did, then that's great. But a lot of people seem to be taking it as an invitation to discount the study _without_ taking such a close look, which is... lame. And some people seem to be taking a look, but not really having much of a clue about what it would mean to really do so; Joseph Hertzlinger's comment, for example, reveals that he doesn't actually know much about the relevant methodological norms here, and for that matter, that he fails to understand that _Nature_ isn't going to publish an article with a trivial statistics error in it.
Also, speaking as a liberal academic: it is far from crazy to take as an operating assumption that the authors of the research in question are likely to be on the liberal side of the dial. What would be crazy would be to simply reject their findings out of hand on that basis. And it's not clear to me that Megan is doing any such thing.
So, Philosopher, you would be comfortable with a blogger writing a lazy "critique" of your research paper--without having read it--that concludes that your liberal political bias colored the results of your work?
Like I just said: I'd be fine if their identification of my political orientation motivated them to take a close, hard look at what I'd said. The only part of Megan's post that is actually anything like a _critique_ was the proposed geographical confound, which (as, again, I already said) seems to me to be entirely plausible.
The Nature Neuroscience article itself costs money, but the supplementary information is free! It includes the procedure.
Discuss.
Looking at the procedure, I spy one big potential source of selection bias. This appears to be a study much like those I participated in when I was in college - a sample is achieved by giving psych students class credit for participating, or others a small amount of money for their time. While this is fine for many purposes, you're selecting from a group of probably youthful and/or poor people in a large, super-liberal city, and that can have bias effects on certain types of experiments. I can imagine that people who self-identify as conservative under those circumstances are much more likely to be stubborn in their political beliefs than the liberals around them, or who come from a conservative milieu. Really, you'd have to be a rare specimen indeed to be a cash-strapped college student in New York with solid conservative beliefs, and if that's their core sample...then at most, it's just talking about the mental composition of conservatives who live in the vicinity of NYU. Wierdos, in other words.
But that's not to say the study is wrong. It's just that like all behavioral studies, one unexpected result means approximately zip. When you're talking about inferring the workings of the mind from behavioral cues, controlling for every possible bias is just about impossible. You need a much larger set of studies, all pointing in the same direction, to have any probative value.
And Charles, concerning Megan's predilection to assume NYU academics are liberals, sure. She does that, like most conservatives do, and also like most liberals assume that deeply religious people are conservative. Which is to say, it's a decent guess but not a fair assumption. But it's not incredibly pertinent to the point of her post either. You're digressing pretty massively to tar her reputation on a flippant mistake she made that has little to do with the substantive point she's making, which if you look, is actually a bit nuanced.
You say you're not debating her main point, but why hijack the thread to address something so incidental? Even well-written blogs can be sloppy with snark from time to time, and it's nothing to feel aggrieved over unless it reaches Hugh-Hewitt-on-Islam levels. Some of us actually want to talk about the substance of her post, and presumably that's why the comment section is here. Instead those of us talking about the study have to discuss over a zillion angry back-and-forth posts about an aside.
That being said, I'm glad you've chosen the righteous path of un-anonymity. You stand behind what you say, and that counts. Unless your name is George or something.
We would also have to see the distribution on the political spectrum. With n=42, there are probably 8-10 conservatives in the group, if that. I doubt that means much.
"With n=42, there are probably 8-10 conservatives in the group, if that. I doubt that means much." That would be something for the statistical analysis to determine. As a general (and, really, nearly-absolute) rule, no one's rough eyeball sense of how many subjects is enough is very reliable.
I agree with James that it would be nice to know more about their subject recruitment. The Nature piece itself is just a "Brief Communication", so it's pretty sparse about such things.
I wonder how self-described libertarians would have fared? I'd really like to know what team I should bet on if someone ever decides to hold a Tetris tournament between ideological groups.
IMO, the Yahoo! and LA Times articles both sucked. I would have liked more basic info on the study, rather than paragraph after paragraph of outsiders like UC Berkeley's Frank Sulloway letting us all know what conclusions should be drawn from it (the broadest ones possible, of course).
There were 43 students in this study; something we learned from the "mushy" Yahoo! piece, but not from the supposedly "much better" LA Times article. I wonder what the breakdown of liberals to conservatives was? Obviously, it wasn't even. In a study that small, a couple of imperceptive dolts on one side or the other could really skew the data. Were the students asked to identify themselves as liberal or conservative before or after the experiment? If after, I wonder how many conservatives they managed to snag in a random sampling of 43 NYU and UCLA undergrads? If before, what's to prevent students from lying and then deliberately scoring poorly, to make "the other side" look bad?
So many questions, yet all we get in these mainstream articles is the unquestioned assertion that a study of 43 students essentially playing a video game constitutes scientific proof that liberals are fluid, dynamic thinkers who constantly recognize new patterns in human affairs (completely analogous to "M" and "W" flashing by on a computer screen, of course), and radically adjust their world views accordingly, while conservatives are...um...how do we put this tactfully?...steadfast and loyal (traits that are also inextricably tied to the inability to recognize inverted characters flashing by on a computer screen, of course).
If after, I wonder how many conservatives they managed to snag in a random sampling of 43 NYU and UCLA undergrads? If before, what's to prevent students from lying and then deliberately scoring poorly, to make "the other side" look bad?
Note: I wrote this before seeing Kwyjibo's post of the procedures from the Nature Neuroscience article.
...then at most, it's just talking about the mental composition of conservatives who live in the vicinity of NYU. Wierdos, in other words.
No, it was all college students, according to the Yahoo! article.
Charles Giacometti wrote: In my world, the way to evaluate research is to begin by actually reading it. Stunning, eh? Then you would start looking at the methodology, the data, the related research, the other papers cited, and so on. I would never, in a million years, assume I know anything about the bias of the authors or the tendency of certain types of research to reflect bias without some sort of evidence to support such an assertion.
So in other words, a person must be an expert in any given field in order to have even a general opinion about research conducted in that field, or the typical qualities of research that come out of that field?
Hope you didn't complement the chef tonight, unless you were a cook yourself and spent a good deal of time reviewing the recipe, ingredients, and technique. After all, that's the only way to know for sure whether you're being served a good meal, right?
You give Megan far too much credit. She says nothing of substance as to why such research is subject to bias. She merely assumes that it is, though she waves her hands a little about, "people who disagree will probably look very different from the people who agree." (A point that several people have correctly noted is barely understandable.) She doesn't know about the data used, or the methodologies. At the end of the day, Megan's response is incredibly simplistic. "Oh, it must be by a liberal." Such a response has no value at all.
Hold the phone -- did you just take up a bunch of stuff out of thin air based only loosely on the original post, ascribe it to Ms. McArdle in total, and then explain why the methodology of exactly that kind of approach has no value? My head is spinning.
I am reminded once again Megan is the blogger who dismissed the Lancet studies of the civilian deaths in Iraq when she clearly had not even read at least one of the articles. Sorry, the world of research doesn't work that way, where critics try to chime in without having read the material in question.
No, Megan is the blogger who posted several good arguments why the methodology in the Lancet study had some non-trivial plausibility issues, thereby making it very difficult to accept the claims made without a much more thorough defense of how those claims were possible. Her interlocutors generally responded to that with, "Well, you weren't there, so we should take the study at face value for now," which gave the Laugh Test a twenty-minute abdominal workout.
In any case, don't try to foist this kind of discredit-by-alternate-history crap here; it's beneath someone who thinks himself intelligent enough to claim --
A note to any Atlantic editors who might be reading. It is time to raise the bar here, because this blogger reflects very badly on what is an otherwise fine institution.
You really don't want to start that -- the first and easiest way they can improve the quality of the site is to bar commenters who actively lower the bar by grinding ideological axes against the blogger, and then where would you be?
Of course. I was implying that someone else could do it, since I'm too dumb.
On the old blog, this would have been done hours ago.
Joseph Hertzlinger's comment, for example, reveals that he doesn't actually know much about the relevant methodological norms here, and for that matter, that he fails to understand that _Nature_ isn't going to publish an article with a trivial statistics error in it.
I'm not deaf; I'm ignoring you.
I'm perfectly well aware of the propensity of social scientists to publish papers based on small samples and think they should be embarrassed by it.
Charles should instead write notes to the editors of Science and Nature, explaining that journals of general interest aren't in a million years useful to anyone.
Megan asserts:
"My other problem is that the west has a long history of research on race and IQ, and gender and IQ, and so forth, that generally finds--quel surprise!--that the dominant group is genetically superior."
No, that's pretty close to the opposite of the truth. For example, the first person to seriously argue, using data, that women were just as smart as men was pioneering IQ researcher Cyril Burt way back in 1912. Ever since the 1930s, IQ test designers have simply assumed that the sexes are equal in average IQ and thrown out questions that would lead to gender differences in average IQ.
A general rule of thumb is that since people are allowed to get away with making up lies about IQ researchers, so most of the bad things you hear about psychometricians are lies.
"Wow, I'm stunned. This whole thread has devolved into harsh criticisms that are fairly incidental to Megan's argument"
That seems to be a fairly good description as to what has happened across the board in comments to this blog. Megan seems to has stepped on a hornet's nest of leftists who seem intent on discrediting her. Many are making points that contribute to a well rounded discussions of the issues. But I find myself increasingly wading through comments that are picking nits with minor points, comments that attribute to Megan positions that she does not hold, and comments that entirely misconstrue what Megan is saying. If you don't understand what I am talking about, you might be part of the problem.
That seems to be a fairly good description as to what has happened across the board in comments to this blog. Megan seems to has stepped on a hornet's nest of leftists who seem intent on discrediting her.
The difference between Megan's blog here and on her own site is not (necessarily) that commenters here are dumber than the ones who were there; the problem is that there is a core group of people here who don't want to argue in good faith, but want to "win" instead.
Good faith requires that when one's opponent in a debate makes an argument that can be interpreted more than one way, one does not assume that one's opponent is intending the dumber of the two interpretations, or the less moral of the two interpretations.
Not surprisingly, partisans are more likely to argue to win than to argue in good faith. Megan's out of place here because she's not a partisan, whereas some of these commenters are.
To follow up on my previous point: unfortunately, there's a blog that I know for a fact is run by bad faith arguers -- who also happen to be, collectively, not smart enough to run a lemonade stand -- that seems to be funneling some commenters over here.
Anony-mouse writes, "So in other words, a person must be an expert in any given field in order to have even a general opinion about research conducted in that field, or the typical qualities of research that come out of that field?"
Not even close, mr./ms. mouse. It is far simpler than that. I expect people to actually read something they comment on. Pretty simple, eh? And I am quite certain that Megan never read at least one of the Lancet articles, but that didn't stop her from superficially attempting to criticize it.
I am not saying I agree or disagree with Megan. I am not saying I agree or disagree with this study or the Lancet study. I am not saying that you have to be Albert Einstein or Alfred E. Neuman. I am simply saying that if you want to weigh in on a research paper, you actually have to have read it.
That is an incredibly low bar. If we can't reach that, then the quality of public dialogue indeed has sunk very, very low.
You can pick each other's nits and make ad hominem arguments all day, but Megan's original post speaks to an important point about the nature of science and the conflict between Kuhn and Popper.
Scientists like to believe that they are exposing the truths of the universe through direct experiment and falsifiable hypotheses. The reality is that because 1)they find evidence where they look; and 2) they look where the current paradigm tells them to, they spend most of their effort reinforcing the dominant paradigm. (That is until the Kuhnian 'paradigm shift' comes.)
I think Megan's point is that these guys found what they were looking for -- liberals are more flexible thinkers. It's not so much that the statement is wrong, but more that it's of no value.
However, I think that such a criticism of the paper is misguided. First, it was a note, not a full blown research paper, and as such its whole point was not to establish the cognitive differences between liberals and conservatives, but rather link already established (in the literature) cognitive differences to a physiological process. Megan's issue should be with the previously published literature cited in this one.
A more proper criticism is that measuring microcurrents on the scalp is about two steps removed from phrenology in terms of understanding how the brain functions.
Another critical question would be how the lines in Figure 1b were drawn? We have a spectrum of political orientation from -5 to +5, but then just two lines representing lib/cons. Are those simple averages? Where do the 0's go? Can we see ranges as well?
And as others have pointed out only 7 of 43 identify themselves as conservative, to the right of zero, which can have profound effects on the regression.
Finally, it would be nice to see similar graphs and p-values for the same participants scored for age, sex, standardized intelligence score, and self-reported happiness. It could go along way into showing that these results were not based off another indicating variable.
"So when I read about research that has confirmed that a group the researchers themselves belong to is smarter, or more flexible, or just generally groovier, than some other group of folks, I can't help but reach for the salt shaker."
Adam Smith called it the "vanity of the philosopher". A couple of McArdle's buds from the econ department at George Mason University used that as a title for their book:
http://www.amazon.com/Vanity-Philosopher-Hierarchy-Post-Classical-Economics/dp/0472114964/ref=sr_1_1/002-6676040-2563255?ie=UTF8&s=books&qid=1189521126&sr=8-1
Oh, argh.
"I'm perfectly well aware of the propensity of social scientists to publish papers based on small samples and think they should be embarrassed by it." Not nearly as embarrassed as you should be, in being so willing to hold forth as if you had half a clue as to what you're talking about. (If you have an actual, like, _argument_ to make, based on real knowledge of methodology, statistics, etc., please do make it. But so far you sound like someone who is simply uninformed and stubborn.)
Taking Jim's comments in reverse: I'm certain that they've either already looked at variables like those; or have good reasons to think aren't confoundsd that need considering (e.g., if there have already been studies using this ERP paradigm looking at, say, reported happiness differences, and found nothing like this, then the proposed happiness confound isn't a plausible one); or they are planning to do so in the continuations of the study. I would be very surprised if they didn't at least check gender; and there's probably not enough age difference in the sample for it to be doing anything. (Is there already an established correlation between IQ scores and declared political orientation, btw?)
A number of ERP methods are extremely well-established, and your attempt to assimilate it to phrenology is nothing but rank ignorance. I mean, do you think you have a better grasp of what is or isn't a fundamentally sound method here than the folks at _Nature: Neuroscience_? Really? If there are objections to be made to this research, _this_ sort of broad-brush tarring isn't very likely to be it.
The important Popperian point here is not: did the researchers look for something they expected/wanted to find? But rather: did the researchers take appropriate steps so that, if what they wanted/expected to find wasn't really the case, could their expectation be falsified? The answer here is "yes". Popper is totally on their side here. (And Kuhn is simply besides the point; we're firmly inside the boundaries of what Kuhn would call "normal" science here.)
Philosopher -
I admit the 'phrenology' comment was baiting (and cribbed from a more knowledgeable source than I who was criticizing brain blood flow experiments, which I realize are an entirely different thing.)
I realize the tools of neuroscientists are severly limited by ethical considerations, so they must squeeze as much analysis out as they can out of what they have. But ultimately these are just microcurrents on the scalp, and there must be a point were the conclusions become too tenuous to hold. The fact that other neuroscientists accept these methods may have much to do with the professional need of cognitive neuroscientists to differentiate themselves from psychologists.
In terms of confounding variables, I am unfamiliar with publishing research notes. From what you say I assume we are to trust that they checked for the and didn't note it. Sounds good.
But on the Popperian issue, I'm not quite with you. Quoting the paper, "To test the hypothesis that political liberalism (versus conservatism)
would be associated with greater conflict-related ACC activity . . .[description of study]" Seems like a simple falsifiable hypothesis to me. But the last line of the paper gives away their nefarious game: "More broadly, this research demonstrates how integration across multiple levels of analysis can begin to elucidate how abstract, seemingly ineffable constructs, such as ideology, are reflected in the human brain." That's the real goal.
We have no idea how long the search for a physiological artifact of ideology has been sought, and how many other studies came back with no results.
The ONLY 2008 Swiss Rolex site that offers. replica rolex NEXT DAY DELIVERY via COD or Credit Card payment! [url=http://replica.makeforum.org]replica rolex[/url].
Play Blackjack Online. play blackjack Play blackjack for free or for money - [url=http://www.xanga.com/play_blackjack]play blackjack[/url].