The obligatory link for any discussion of a report on a research result like that is the article by Peter Norvig, director of research at Google, on how to interpret scientific research.
I would worry about publication bias here, since it's a metastudy and few people would publish such a study showing that, would you know, there's nothing going on...
More food for the poisonous group-think infecting my generational and cultural peers. Why engage with the arguments of people that disagree with you, when they're just stupid and/or evil? Best just to mock them and revel in a comfortable sense of smug superiority. Everybody agrees with you, so your dishonest behavior will be reinforced and rewarded.
I agree wholeheartedly. It's far too easy these days to immerse yourself only in the opinions with which you already agree, and to dismiss anything that is challenging or different.
Here's the thing: if you're going to start looking for challenging opinions, it's best to start in places other than talk radio and television news. Stop dismissing credentials. Stop vilifying academic research. Stop rewarding people who are merely controversial, or who play on petty emotional impulses or thinly-veiled bigotry.
Start rewarding depth of analysis, argument and complexity, because the world is a complicated place, full of colors and shades of gray, and no sound bite can possibly encompass it all.
Err, isn't this the opposite of what you should be doing? Credentials are cached judgements of merit; bypass the cache and judge the merit of the arguments for yourself.
Cached judgments of merit? Give the strained engineering metaphors a rest. This has nothing to do with caching, and has everything to do with filtering out bad information.
For any sufficiently complicated subject, your average person lacks the knowledge, experience or intelligence to weigh arguments solely on their merits. It isn't popular to say here to a bunch of opinionated techies, but credentials are a way of bypassing limitations in our own knowledge.
Said another way: there are subjects that must be learned through experience. I am more inclined to believe the opinions of a World Bank economist than I am to believe the comments of JoeThePlumber123456 in some online libertarian forum.
Stop dismissing credentials. Stop vilifying academic research. Stop rewarding people who are merely controversial, or who play on petty emotional impulses or thinly-veiled bigotry.
These principles may not give consistent guidance with respect to the article under discussion. Isn't that what we're talking about? Maybe we should stop pretending that unobjectionable generalities actually advance the discussion. (Except that one -- it's the only one allowed :-P)
Really? I don't understand why the thesis is even provocative.
Isn't one of the key points of the conservative movement that there's no correlation between smart/stupid/good/evil/judgement/etc and intelligence/iq/g/cognitive-ability/etc? And isn't it a pretty good point? Haven't you seen Forest Gump?
Is it really controversial that the highly educated and intelligent segments of society are massively less conservative than the less educated and less intelligent segments? Come on.
Is there any evidence that the "smart people" make better political decisions? No. Anyone claiming so would be an idiot. But who's claiming that?
"I'd rather be governed by the first 2,000 names in the Boston telephone directory than by the faculty of Harvard." -- William F. Buckley Jr.
> the highly educated and intelligent segments of society are massively less conservative
I don't know what the "intelligent segments" are. People with masters in liberal arts? Let's just pretend earnings power is proxy for smarts.
The top slice of earners skew liberal -- the top 3% or so, who largely owe their earnings power to government granted cartels like the AMA, the bar, and Wall Street banking. They tend to live in urban enclaves where their wealth isolates them from meaningful impacts from social legislation. They aren't personally affected by immigration policy, for example.
The much larger mass of upper middle-class types that cumulatively earn most of the money and run the country (the $150K households scattered all over, not just in the coastal urban centers) skew conservative. Entrepreneurs also skew conservative.
To a certain extent, you have business people who build things of value vs. the heavily government entwined priesthoods. The "mandarin classes", if you will, skew liberal. The merchant and tradesman classes skew conservative.
Educated: studied liberal arts & sciences at any credible college or university.
Conservative: Pat Buchanan. "god and country." Not Milton Friedman. Not Irving Kristol. And not "right-wing", which is an ideology. Conservative might really mean Edmund Burke, but I think the linked article means Pat Buchanan.
Smart: is context dependent and means making the best decisions in a given context. In some forms of human endeavor being smart requires higher than average intelligence, like say being a surgeon. But most aren't especially aided by intelligence. Definitely not politics or business. Too much intelligence can even cause stupidity in some situations.
I'm not sure what the word liberal means to you, but to me, in this context, it's meaningless.
The much larger mass of upper middle-class types that cumulatively earn most of the money and run the country (the $150K households scattered all over, not just in the coastal urban centers) skew conservative. Entrepreneurs also skew conservative.
I find it interesting that the test groups were students at community colleges and foreign applicants. Wouldn't it be far better to judge "intelligence" from students at a four year university, or those studying higher education? And just students? To me, this paper seems to be written to prove a hypothesis, not to test it.
As a conservative, I concede some negative correlation could be present here. Certainly not a strong one, but perhaps a statistically measurable one.
There are a couple of reasons off the top of my head for this. Our institutions of higher learning are, I think this is not in debate, generally left leaning. Kids that go there are smart enough to go to college, but they are kids, and some of them will be swayed left in that atmosphere, even if they might not have been if not exposed to it. That alone should skew the distribution as the article indicates. There's also the religious aspect, where intellectuals tend to be less religious, and by definition then, less conservative.
I'm not sure what someone would do with this information, other than use it to make statements that would not generally be accurate.
It doesn't really matter how smart someone is, they tend to take on the opinions of their peers unless they are very strong willed. If smart people congregate together, it's likely that they will begin to hold similar opinions. So it's no real surprise if people who are smarter hold similar opinions.
Of course, there's still something to be said about open-mindedness being an attribute of minds capable of handling ambiguity.
When you boil cultural conservatism down to its essence, as separate from the concepts of economic liberty which conservatives lean towards, the basic idea is that there is wisdom in the prevailing traditions and cultural attitudes. The wisdom can defy thoughtful analysis. It's the concept that society is the way it is because a process of darwinian selection has settled on the arrangement and it is dangerous to wantonly mess with this complex system.
I would not be surprised that higher IQ people tend towards a certain arrogance that is dismissive of the idea that cultural traditions and attitudes carry value that defies individual understanding. Away from the cultural conservatism and towards market liberalism you often see the same thing: certain smart people can't cope with idea that there's a complex, self regulating system that defies understanding and should not be messed with.
"It's the concept that society is the way it is because a process of darwinian selection has settled on the arrangement and it is dangerous to wantonly mess with this complex system."
An economist sees a $100 bill on the sidewalk. He neglects to pick it up because it can't possibly be real. If it were, an efficient market dictates that somebody would already have picked it up. A five year old follows behind, picks it up, and exclaims "Look ma, I just found $100." The economist turns around and says, "See? I was right."
Just because Darwinian selection gives better results than any individual decision does not absolve you from making individual decisions. It works because of the presence of mutations: take away the mutations, and evolution ceases to give useful results. You see the same problem in finance: the market is efficient only to the extent that people do not believe it is efficient, because once people start believing that it's efficient, they'll invest in index funds and nobody will actually scout out the information needed to drive good stocks higher and bad stocks lower.
In short, I think that culture is a complex, self-regulating system, and that alone is why it should be messed with.
Important note: the study's definition of 'conservative' does not map directly to 'conservative in domestic US politics'.
Also, it is from a study by a Singapore-based academic, based on a pool of foreign students seeking entry to the US, and community college students (unclear where).
Less-intellectual but traditional-hopeful-and-hard-working conservatives could easily be overrepresented in such a study population.
Yes - their sample selection is a little interesting. As their definition of conservatism included valuing hard work, I wonder if all the study has shown is that 'if you work hard, then you don't have to be as intelligent in order to get into the same university as someone who works less hard'.
The definition of Conservative used is laughable. Religious belief has nothing to do with whether you're a conservative or not, and there's no mention of small government, lower taxes, which are what conservatism is about. The definition seems to be a caricature of America's religious right, but the sample is of foreign students in the US. In other words, the definition is based on a subset of American conservatives, which has no equivalent in many of the countries that the students will have come from.
>"there's no mention of small government, lower taxes, which are what conservatism is about."
I disagree. I believe Conservatism is context-dependent.
In the 1700's, Conservatism would have meant resisting the Democratic revolutions in places like the US and France. That's why Edmund Burke's pieces against the revolution in France are still considered important tomes in Conservative thought.
Modern American Conservatism is weird. It was not meant to preserve the old order, but to rebuild an order that no longer existed. Limited government was thoroughly eradicated by the New Deal, which the Conservative movement was born in opposition to. Thus, it is in some ways a "Revolutionary Conservatism", which is slightly oxymoronic.
You'll never get consensus on what "conservative" means; conservatives themselves will never agree completely, just as liberals will never agree on liberalism, Marxists on Marxism, etc. It's a political question. If we accept the paper's definition for the sake of argument, then maybe we can have an interesting discussion about it.
Of course, it's a given that the definition and the means of measurement determined the results, so the definition used in the paper is extremely important to interpreting the results. Whether the definition is correct, however, isn't important.
To look at this politically, the more educated an American, from high school dropout through college graduate, the more likely they are to vote Republican-- a pattern that flips dramatically at the very top. College grads are more likely to be GOP but folks with graduate degrees are fairly strongly Democratic. Even in the Democratic wave in 2008 this general pattern held true-- Obama only beat McCain by 2 points among college graduates, and it was the first time a modern Democrat candidate beat a Republican candidate among college graduates.
Something I always look at in these kinds of studies is what the sample group is. There's a statistical phenomenon called "explaining away" which often causes spurious correlations. For example: measure the athletic and intellectual abilities of American college students. You'll find a fairly strong negative correlation, even though in the general population there's no correlation or a weak positive correlation.
Why? Admission to college has two possible (but non-exclusive) explanations, athleticism and intellect. These explanations "compete" to explain the fact that a given student got in. Roughly speaking, the presence in the general population of unintelligent non-athletes contributes a positive term to the (near-zero) correlation. These people are removed from the student population, so the total correlation is lower, and hence negative.
Now, this experiment studied student applicants to American universities. If both intelligence and social conservatism in some way cause people to apply to American colleges, a negative correlation between these traits will be induced in the pool of applicants. Off the top of my head, one such cause might be that people with higher social status are more likely to apply to university, and high social status correlates positively with increased acceptance of the social order. I find this rather more plausible than the easy conclusion that "conservatives are dumb", even though it's a more complex causal chain.
33 comments
[ 2.3 ms ] story [ 89.9 ms ] threadhttp://norvig.com/experiment-design.html
Checking this blog entry point by point against Norvig's checklist would be good mental exercise.
Here's the thing: if you're going to start looking for challenging opinions, it's best to start in places other than talk radio and television news. Stop dismissing credentials. Stop vilifying academic research. Stop rewarding people who are merely controversial, or who play on petty emotional impulses or thinly-veiled bigotry.
Start rewarding depth of analysis, argument and complexity, because the world is a complicated place, full of colors and shades of gray, and no sound bite can possibly encompass it all.
Err, isn't this the opposite of what you should be doing? Credentials are cached judgements of merit; bypass the cache and judge the merit of the arguments for yourself.
As for the study, I find it silly.
For any sufficiently complicated subject, your average person lacks the knowledge, experience or intelligence to weigh arguments solely on their merits. It isn't popular to say here to a bunch of opinionated techies, but credentials are a way of bypassing limitations in our own knowledge.
Said another way: there are subjects that must be learned through experience. I am more inclined to believe the opinions of a World Bank economist than I am to believe the comments of JoeThePlumber123456 in some online libertarian forum.
These principles may not give consistent guidance with respect to the article under discussion. Isn't that what we're talking about? Maybe we should stop pretending that unobjectionable generalities actually advance the discussion. (Except that one -- it's the only one allowed :-P)
Isn't one of the key points of the conservative movement that there's no correlation between smart/stupid/good/evil/judgement/etc and intelligence/iq/g/cognitive-ability/etc? And isn't it a pretty good point? Haven't you seen Forest Gump?
Is it really controversial that the highly educated and intelligent segments of society are massively less conservative than the less educated and less intelligent segments? Come on.
Is there any evidence that the "smart people" make better political decisions? No. Anyone claiming so would be an idiot. But who's claiming that?
"I'd rather be governed by the first 2,000 names in the Boston telephone directory than by the faculty of Harvard." -- William F. Buckley Jr.
I don't know what the "intelligent segments" are. People with masters in liberal arts? Let's just pretend earnings power is proxy for smarts.
The top slice of earners skew liberal -- the top 3% or so, who largely owe their earnings power to government granted cartels like the AMA, the bar, and Wall Street banking. They tend to live in urban enclaves where their wealth isolates them from meaningful impacts from social legislation. They aren't personally affected by immigration policy, for example.
The much larger mass of upper middle-class types that cumulatively earn most of the money and run the country (the $150K households scattered all over, not just in the coastal urban centers) skew conservative. Entrepreneurs also skew conservative.
To a certain extent, you have business people who build things of value vs. the heavily government entwined priesthoods. The "mandarin classes", if you will, skew liberal. The merchant and tradesman classes skew conservative.
Intelligent: capable of graduating from MIT.
Educated: studied liberal arts & sciences at any credible college or university.
Conservative: Pat Buchanan. "god and country." Not Milton Friedman. Not Irving Kristol. And not "right-wing", which is an ideology. Conservative might really mean Edmund Burke, but I think the linked article means Pat Buchanan.
Smart: is context dependent and means making the best decisions in a given context. In some forms of human endeavor being smart requires higher than average intelligence, like say being a surgeon. But most aren't especially aided by intelligence. Definitely not politics or business. Too much intelligence can even cause stupidity in some situations.
I'm not sure what the word liberal means to you, but to me, in this context, it's meaningless.
Citations, data please?
I wonder if age and conservatism are positively correlated.
I wonder how many significant lurking variables are ignored in the study.
There are a couple of reasons off the top of my head for this. Our institutions of higher learning are, I think this is not in debate, generally left leaning. Kids that go there are smart enough to go to college, but they are kids, and some of them will be swayed left in that atmosphere, even if they might not have been if not exposed to it. That alone should skew the distribution as the article indicates. There's also the religious aspect, where intellectuals tend to be less religious, and by definition then, less conservative.
I'm not sure what someone would do with this information, other than use it to make statements that would not generally be accurate.
Of course, there's still something to be said about open-mindedness being an attribute of minds capable of handling ambiguity.
I would not be surprised that higher IQ people tend towards a certain arrogance that is dismissive of the idea that cultural traditions and attitudes carry value that defies individual understanding. Away from the cultural conservatism and towards market liberalism you often see the same thing: certain smart people can't cope with idea that there's a complex, self regulating system that defies understanding and should not be messed with.
An economist sees a $100 bill on the sidewalk. He neglects to pick it up because it can't possibly be real. If it were, an efficient market dictates that somebody would already have picked it up. A five year old follows behind, picks it up, and exclaims "Look ma, I just found $100." The economist turns around and says, "See? I was right."
Just because Darwinian selection gives better results than any individual decision does not absolve you from making individual decisions. It works because of the presence of mutations: take away the mutations, and evolution ceases to give useful results. You see the same problem in finance: the market is efficient only to the extent that people do not believe it is efficient, because once people start believing that it's efficient, they'll invest in index funds and nobody will actually scout out the information needed to drive good stocks higher and bad stocks lower.
In short, I think that culture is a complex, self-regulating system, and that alone is why it should be messed with.
Also, it is from a study by a Singapore-based academic, based on a pool of foreign students seeking entry to the US, and community college students (unclear where).
Less-intellectual but traditional-hopeful-and-hard-working conservatives could easily be overrepresented in such a study population.
I disagree. I believe Conservatism is context-dependent.
In the 1700's, Conservatism would have meant resisting the Democratic revolutions in places like the US and France. That's why Edmund Burke's pieces against the revolution in France are still considered important tomes in Conservative thought.
Modern American Conservatism is weird. It was not meant to preserve the old order, but to rebuild an order that no longer existed. Limited government was thoroughly eradicated by the New Deal, which the Conservative movement was born in opposition to. Thus, it is in some ways a "Revolutionary Conservatism", which is slightly oxymoronic.
Of course, it's a given that the definition and the means of measurement determined the results, so the definition used in the paper is extremely important to interpreting the results. Whether the definition is correct, however, isn't important.
http://elections.nytimes.com/2008/results/president/national...
Why? Admission to college has two possible (but non-exclusive) explanations, athleticism and intellect. These explanations "compete" to explain the fact that a given student got in. Roughly speaking, the presence in the general population of unintelligent non-athletes contributes a positive term to the (near-zero) correlation. These people are removed from the student population, so the total correlation is lower, and hence negative.
Now, this experiment studied student applicants to American universities. If both intelligence and social conservatism in some way cause people to apply to American colleges, a negative correlation between these traits will be induced in the pool of applicants. Off the top of my head, one such cause might be that people with higher social status are more likely to apply to university, and high social status correlates positively with increased acceptance of the social order. I find this rather more plausible than the easy conclusion that "conservatives are dumb", even though it's a more complex causal chain.