IQ Dilemma: Inconvenient Correlations, Uncomfortable Data

I was having a discussion with another blogger, Staffan in his blog Staffan’s Personality Blog. His blog was recommended to me by a human biodiversity (HBD) proponent, I believe it was JayMan in a comment in HBD Chick’s blog. I commented on several of Staffan’s posts, but the longest discussion was in his post about studies on intelligence, religion and ideology.

I pointed out the numerous studies that show a correlation between liberalism and above average IQ. Staffan is unconvinced. He doesn’t like Kanazawa and so thinks criticisms of one researcher disproves all other research by other researchers, a line of reason that I don’t follow. He seems to refuse to take the other researchers seriously, even though the criticisms toward Kanazawa don’t apply toward them or their  research. It comes down to Staffan believing all psychological research is biased toward liberalism and so can be dismissed nearly out of hand.

I’ve criticized this criticism before. The conservative bias about the existence of a liberal bias seems to fall apart under scrutiny, especially the claim about psychology. But maybe I’m missing something.

There are two sets of data that are obvious and yet don’t quite connect for people like Staffan.

Most psychologists are liberal. Most scientists, most academics, most well educated people in general identify as liberals, express liberal views and support liberal policies; in particular social liberalism that most closely correlates to the liberal-minded traits such as openness. However, it doesn’t occur to Staffan to wonder why psychology attracts liberals in the first place or even why higher education in general might not just attract but encourage a liberal mindset.

Liberals, along with having above average IQs, unsurprisingly tend to be above average in education and wealth. To remove the liberal part from the equation, most people who are above average in education and wealth unsurprisingly tend to have above average IQs.

This could be explained simply from a perspective of poverty.

Poor people tend to be socially conservative and so easily swayed by the rhetoric of conservative politics in the US, and because of this not as many poor people self-identify as liberal even among the subset of the poor that tends to vote for Democrats (yes, around a third of Democrats self-identify as conservative and, surprising to some, only around a third self-identify as liberal). Poverty and poverty-related factors have been proven to inhibit cognitive development which is shown in IQ tests, and lower IQ leads to lower rates of higher education. Wealth and wealth-related factors have been proven to increase cognitive development which is shown in IQ tests, and higher IQ leads to higher rates of higher education.

All of this is well known. It’s proven in numerous studies and sets of data, both correlationally and causally.

It would be strange if the average liberal who is wealthier and more well-educated (than both the average American and the average conservative) turned out to not also have an above average IQ. That would be one of the strangest discoveries in all of social science research. What would make liberals somehow different from every other comparable group? Considering so much research shows a correlation between liberalism and higher IQs that fits the expected pattern, it would require massive alternative data and careful analysis to explain this bizarre phenomenon, if it were to exist. No such data or analysis is offered by Staffan.

Conservatives love to point out that poor blacks have lower IQs. Yet they suddenly become righteous when it is pointed out that poor conservatives also tend to have lower IQs. Conservative political correctness police are no better than their liberal counterparts. To get at the real point, poverty sucks which is something liberals have been saying for generations. But it isn’t to argue that liberals have any reason to be proud simply for being among a privileged demographic that has experienced less poverty.

The other factor that correlates to lower IQ is authoritarianism. As much research has shown, conservatives and right-wingers in the US show higher rates of authoritarianism whereas liberals and leftists in the US show lower rates of authoritarianism (and I suppose the same would be found in other liberal democracies). Once again, going by the known data, it would be logical to conclude that American conservatives with a higher average rate of authoritarianism than American liberals would have a lower average IQ than American liberals. In different societies, different correlations would be found. For example, communists in authoritarian communist countries unsurprisingly show higher rates of authoritarianism; and if it were studied, these authoritarian communists would probably have lower IQs. It would be surprising to find some other result.

As poverty sucks, so does authoritarianism. Like a good liberal, I’d love to end both of these plagues upon humanity.

Staffan’s response, once again, is that psychologists are biased liberals and so all the massive amounts of research on authoritarianism can be dismissed as biased. This verges on denialism, but Staffan doesn’t quite go down that road. He does seem to hold some basic respect for science. The problem he faces is that he can’t square all of the data with his preferred assumptions and conclusions. In our discussion, he has yet to comment on the studies I presented to him that weren’t referred to in his post. The debate can’t go on until he does.

This kind of debate can and does just go around in circles.

The interesting part to me isn’t whether the correlation exists. It is becoming increasingly convincing that we are beyond that point. We have good evidence that it exists. What interests me is why and how it exists. What might be the causation behind the correlation? Or is their any direct causal link at all? Can it be explained away by confounding factors and circumstantial conditions? Maybe it is just an artifact of the data, but considering the data is large and growing: How did such an artifact arise so consistently and pervasively in this data from our society, consistent and pervasive across decades of data sets collected by numerous researchers and other data gatherers, some of whom gathered data without expectation of later researchers finding this particular correlation?

The point being that we won’t find anything of worth if we stop at mere dismissal of the correlation. If one wishes to disprove that correlation, then more power to them. But no one yet has disproven it, much less fully explained it. There are hypotheses that have been presented to explain the correlation and these hypotheses are falsifiable. Go forth and falsify them, if you can.

This is where my frustration stops me short for I don’t know how to continue this kind of discussion. It doesn’t seem to go anywhere and frustration begins to bring out the worst in me, and I already feel like I’m being mean to Staffan who is a perfectly nice guy as far as I can tell. I don’t want to judge Staffan or people like him. I want to explore ideas and data with intellectual curiosity and hopefully intellectual humility.

Maybe I’m wrong about significant parts of my analysis. That wouldn’t surprise me. But the debate-ending response I receive just doesn’t satisfy me. Claims of liberal bias are unfair and ultimately unhelpful; I would go so far as to call them unfounded to any great extent. Each study must be taken on its own terms and not waved away with one’s hand. Is that asking too much?

I could go in more depth with my analysis, but I’ve already explained my view in Staffan’s blog. Instead of restating everything I wrote in the comments section of Staffan’s post, I’ll just offer the comments themselves below (which do repeat some of what I wrote above).

What point I’m trying to make?

I was just trying to describe the confusion of ideologies in the US. I was also trying to show how labels are in some ways arbitrary. It isn’t the label that defines the person but the person who defines the label. Liberal and conservative meant something quite different when my grandparents first was eligible to vote which isn’t that long ago in the big picture.

Liberalism used to be a very large and inclusive category. Old school Republicans like Eisenhower and Nixon would speak positively of liberalism. These days it has become a very narrow term. I don’t know if it has narrowed in definition, but it has narrowed in demographics.

Liberal as a label isn’t comparable to conservative as a label. Conservatism has become an amorphous category. If given a forced choice, most Americans will identify as conservative. But if asked about particular issues, most Americans will more often state support for liberal issues. This is what is called symbolic conservatism. So, conservatism includes both those that support the most radical of right-wing politics and those who support standard liberal politics.

As for those who identify as liberals, they are more comparable to the demographics of libertarians. Both groups are a smaller portion of the population. Both tend to be above average in wealth and education. It would be surprising, going by these demographic facts, if they weren’t above average in iQ. Wealthier, well-educated people in general tend to have higher IQ for a host of reasons as I’ve pointed out, many of which are environmental. Also, wealthier, well-educated people have more opportunities such as legacies that allow them to go to the best ivory League schools. Furthermore, both liberals and libertarians tend to be socially liberal and measure high on liberal-minded traits such as openness to experience.

So, maybe it isn’t that liberals and libertarians are inherently smarter in that these labels don’t magically confer intelligence. It’s just that because of present societal conditions the wealthier, well-educated demographic tends to identify as either liberal or libertarian. I guess you could call this an artefact, but I doubt it is because of wishful thinking.

Is there a reason you’d think people who are above average in wealth and education wouldn’t also be above average in IQ?

The tricky part, as always, is determining the reasons for the correlation and whether there is a causal link between the factors or to some other factor.

“I do actually believe people with higher education and income have higher intelligence. I just haven’t seen credible statistics that says liberals are smarter.”

The data I’ve seen shows that liberals tend to be more well educated and wealthier than the average American. College professors, college students and the college educated are disproportionately liberal. It is also known that higher education correlates to higher IQ. So, it would be bizarre if most average liberals (or libertarian) got their above average education without being above average in IQ. That would be counterintuitive and contrary to the pattern fitting other well educated groups.

There are many lower class conservatives, but because of the rhetoric since the Reagan era the lower classes have mostly stopped identifying as liberals at all. They might call themselves independents, moderates or progressives, just usually not liberal because that label has become stigmatized in the mind of the average American. As such, there would appear to be a lot less distance between the average liberal and the liberal elite than there is between the average conservative and the conservative elite.

This can be seen in the parties. Back during the Reagan administration, many moderates, independents, liberals and union members voted for Reagan and identified as Republican. It was a really broad party back then, as it had been in the decades before that as well. Surprise, surprise, the Republicans at that time had a higher average IQ than did the Democrats. Today, however, Democrats have a higher average IQ. Also, today, the Republican Party is no longer a big tent party that is inclusive of anyone other than conservatives. Why did the Republican’s average IQ go down as their rates of conservatism went up?

There is something interesting about Democrats these days. The Democratic Party is now broader than the Republican Party, even as those who identify as liberal have narrowed. Only about a third of Democrats identify as liberal, another third as moderate and the rest conservative. Minorities have lower average IQs and are more socially conservative than most Americans and at the same time they tend to vote Democrat. This mean that the non-conservative non-minority Democrats must have very high average IQs to make up the difference and still maintain the higher average IQ than Republicans.

I don’t know entirely what that might mean. I’m not a Democrat and don’t feel any need to defend them. But the data seems to confirm that something of relevance is going on.

“But those knowledge quizzes Pew do are always won by people who vote Republican. Not to say that’s the final word but there is reason to be suspicious because their is a media bias.”

There are several distinctions to be made.

Democrats aren’t the same as liberals since there are nearly as many self-identified conservatives in the Democratic Party as there are self-identified liberals. Democrats include the most well educated and the least well educated, but the least well educated Democrats are also the ones least likely to identify as liberal and more likely to identify as conservative.

If you look at those political knowledge quizzes, you’ll see that groups that are known to be extremely liberal do very well on them. For example, the audiences of Jon Stewart and Stephen Colbert are always at the top of the most informed about politics.

If you look at other data, you’ll find that Republicans and the audiences of conservative shows often show high rates of being informed. The interesting part, though, is that they also show high rates of being misinformed. This is called the smart idiot effect. They tend to watch tons of political shows and they know a lot of information, but they unfortunately aren’t good at discerning between what is true and what is false. So, it isn’t just that there are both Republicans who are very well informed and others who are very misinformed. Rather, Republicans are more misinformed precisely to the degree they are informed. That is mind-blowing!

You don’t find this same high rate of smart idiot effect among liberal groups. this might relate as well that the most liberal demographic (i.e., the youngest demographic) get more of their info from alternative media than any other demographic. Part of this difference is because conservatives are a relatively older demographic who rely more on the established mainstream media. The younger generation is way more socially and fiscally liberal on most issues, way more critical of capitalism and way more supportive of socialism. Furthermore, the younger generation along with higher rates of alternative media consumption have higher rates of college education.

I’m not arguing low IQ liberals don’t exist. Besides, if it becomes a label that more people identify with beyond those who are above average in education and wealth, then the average IQ of liberals would probably decrease some or quite a bit (while the rate of the smart idiot effect might increase). Broaden the demographics behind a label and the IQ range will likewise broaden, specifically in this case among the less well educated lower classes and minorities.

Related to party politics, consider geographic regions and areas. Conservatives are disproportionately found in the South and liberals disproportionately in the North. It was through the Southern Strategy that the Republican Party took over the South. The average IQ in the North is higher than the average IQ in the South. Or look at this in terms of rural and urban. The rural areas tend to be more conservative and have lower average iQ and the opposite for urban areas.

If you’re interested to know why and on what basis I make the above analysis, here are a variety of things that inform my views:

“Data from the National Longitudinal Study of Adolescent Health (Add Health) support Kanazawa’s hypothesis. Young adults who subjectively identify themselves as “very liberal” have an average IQ of 106 during adolescence while those who identify themselves as “very conservative” have an average IQ of 95 during adolescence.”
“Most (62%) identify themselves as liberal… most highly educated group (49% have a college degree or more)… Liberals are second only to Enterprisers in following news about government and public affairs most of the time (60%). Liberals’ use of the internet to get news is the highest among all groups (37%).”–and-reality

I wanted to clarify one thing. I don’t think we are fundamentally disagreeing. Like you, I have great interest in personality, moreso than politics. I’m not an ideologue, but I do see politics similar to culture as a useful lense to explore psychology.

I don’t care about liberalism in and of itself. And the so-called liberal elite often irritate me. I agree that the average liberal, the average conservative, and other relatively more privileged people have no reason for being proud of achieving more than those with fewer opportunities, resources and advantages that come from upper class social capital.

Besides, I’m more interested in general liberal-mindedness, upon which all social democracy is dependent upon. Even the average modern conservative is more liberal-minded than the average conservative of past centuries. Liberal-mindedness does correlate to liberalism but not perfectly or absolutely.

My father is a conservative, especially fiscally conservative, with libertarian tendencies. He worked in the business world where he had a successful career, but ended up being a professor. Two things stand out about him, as relevant to this discussion. I’m sure his IQ is well above average. And he has great capacity for liberal-mindedness, most certainly above the national average.

I must give credit to my father for my own liberal-mindedness. My parents were going through a liberal phase of their life when they raised me, although they have never identified as liberal. Even more significant, they raised me in one of the most liberal Christian churches in the US, Unity Church which is New Thought. My father says my mother even was pro-choice when she was younger, but se denies it now. I have to at the same time also give my parets credit for my conservative streak.

People are complex, way more complex than political ideologies and religious theologies. Still, within that complexity, patterns can be detected.

“Those studies and articles don’t say much.”

They say no more nor less than any other scientific data. Dismissing them out of hand isn’t helpful. Kanazawa’s study is just one among many. I even purposely quoted that article about his study because it pointed out that his data was confirmed by other data.

“Would a psychologist (95 percent of which identify as liberals) pick up on that and make a study on it?”

Yes, most academics, scientists and other well educated people identify as liberal. And yet you oddly find it surprising that well educated people have above average IQs. Ask yourself why, in the first place, conservatives are on average less well educated, specifically in the context of higher education having an extremely strong correlation to higher IQ.

I have nothing against Gary Lewis’ study. However, as far as I can tell, it only dealt with certain factors of religion and not conservative vs liberal.

Many religious people are liberal, especially greater social liberalism among the religious who are middle class and upper class. Also, even though in the US there are a lot of poor religious people, those who attend church most often and are most involved in religious activities the most tend to be those who are wealthier and hence who are more well educated. So, there is a very mixed demographics in religion that isn’t found within, for example, contemporary American liberalism.

Gary Lewis’ study doesn’t speak to this greater complexity of the relationships between religion, ideological labels, socio-economic status, and cognitive development. I would also add the issue of psychological traits. The real issue I’m trying to disentangle here has to do with this larger context of factors.

In the US, liberals unlike conservatives are disproportionately found among the middle to upper classes which correlates to above average education and IQ. If you just look at middle to upper conservatives, you would expect to also find above average education and IQ. Ditto for any similar wealthier group. You’d have to explain why what applies to all other wealthier people doesn’t apply to contemporary American liberals who tend to be wealthier.

This ideological pattern of demographics in the US may not fit the pattern found in other countries. Maybe liberals elsewhere don’t tend to be above average wealth, education and IQ. That would then require an understanding of what liberalism is in different contexts and whether there is any general cross-national pattern at all to be found between ideologies and other factors.

I feel wary about generalizing too much. A lot of research I’m familiar with has come from studies done in the US, but American demographics and ideologies aren’t representative of the rest of the world. For example, research has shown that there is a greater correlation with authoritarianism among American conservatives. This isn’t to say that there is anything inherently authoritarian about conservatism in general, but that the issues movement conservatism has emphasized in America has drawn into the ranks those who rate higher on authoritarianism. Nonetheless, there are conservative traditions that are explicitly non-authoritarian.

As I’ve pointed out, I have a conservative streak. The type of conservatism that I’m fond of has more to do with agrarian traditionalism, specifically as described by Wendell Berry. The problem with American conservatism is that it doesn’t tend to be very traditional. A distinction is that mainstream American conservatives tend to be very supportive of capitalism even when it is destructive of traditional values such as a place-based sense of community and localized kinship social order.

I don’t know about liberalism, but I know there is a similar thing with data about left-wing politics. In the US, left-wingers tend to rate low on authoritarianism. But in communist countries, left-wingers have measured high on authoritarianism. And I’d be unsurprised if left-wingers in authoritarian communist countries had lower average IQs than left-wingers in non-authoritarian countries.

I think you are seeking to criticize from the wrong angle. Religiosity can’t be used as a proxy for conservatism nor liberalism a proxy for atheism. You need studies that specifically include all of these factors and seek to discern the causal links.

“How much they say is a matter of quality and it’s clear that Kanazawa did several things that reduce quality – the age of the interviewees, the one-dimensional measure of religiousness, measuring intelligence with vocabulary.”

I offered this link which is of an article written by a conservative:

From that link, you will find the link to the study itself:

In the blog, the author explained about the quality of the study:

“Yes, they used different types of intelligence tests; verbal and spatial. Yes, they corrected for socioeconomic background. Their replication was in the UK and USA.”

Along with other caveats, he offers this important detail about the study:

“It is important to emphasize that the authors do not posit an independent direct causal connection between low I.Q. and more reactionary attitudes towards race and homosexuality. Rather, they start out with a model where low cognitive ability people are drawn (or remain in) to conservative orientation, and this is further correlated with these specific racial and sexual attitudes. Like almost all psychology you can’t get the causation airtight (if you are a hardcore Humean you could probably say this for everything), but the correlation is suggestive in light of political and psychological models.”

T’his is basically along the lines of what I’ve been saying. Correlation between higher IQ and liberalism (or just liberals in certain countries) is not the same thing as causation. But the correlation still remains and appears to have good studies backing it up.

I offered another blog about yet another study. Here are the links to both:

As the blogger notes:

“The paper describes a meta-analysis based on data from three studies that employed the same set of psychological measures. Twenty-two of these measures were selected, drawn from four domains: personality, social attitudes, values, and social norms. While the paper finds strong support for the hypothesis that low cognitive ability is associated with high conservatism it doesn’t make any statements about causality.”

Once again, causality is the tricky part. Nonetheless, it is irrelevant for the point I’ve tried to make in my comments. I honestly admit to not knowing for sure why this particular correlation continually is found in so many different studies. I think many people get upset because they read into this an argument for causation, but that doesn’t seem to be what most researchers are proposing at this point.

A third link I offered is of an article that appears to refer to the same study as referred to in the first link above:

From that article, a critic points out the correlational nature of such research:

“Hodson and Busseri’s explanation of their findings is reasonable, Nosek said, but it is correlational. That means the researchers didn’t conclusively prove that the low intelligence caused the later prejudice. To do that, you’d have to somehow randomly assign otherwise identical people to be smart or dumb, liberal or conservative. Those sorts of studies obviously aren’t possible.”

That could be seen as a criticism, but from my point of view the correlation alone is interesting. Notice that even a critic like this admits that the correlation was demonstrated by the study. His criticism is that there is a lot of complexity involved and also other similar correlations might be found with other extreme ideologies.

I was thinking of a way of getting at a place of agreement between us. In my previous comment, the last quote was of a critic. He made a good point about low IQs maybe being correlated to extreme ideologies in general, whether right-wing or left-wing. That seems a key point to my mind.

From what I can tell, both of us agree that there is a distinction between being conservative and being right-wing. In the US and similar countries, however, conservatism and right-wing ideologies have become conflated. But the same thing hasn’t happened as much in recent history with liberalism and left-wing ideologies because Cold War fear-mongering has caused liberals to disown and distance themselves from left-wingers, although this might be changing now with new criticisms against capitalism arising in the mainstream again.

Going by this, the reason lower IQ would be correlated to conservatism is because conservatism has become correlated to right-wing ideologies. So, it might actually be right-wing ideologies that is forming the correlative bridge between lower IQs and conservatism, and hence no direct or even indirect causal link may exist between them.

What extreme ideologies may signify is simply authoritarianism. In the US, conservatives on average measure higher on authoritarianism than liberals. This is yet another one of those correlations with no certain causal link. At the same time, liberals and left-wingers in the US measure low on authoritarianism, but the opposite is found in authoritarian left-wing countries. The key component seems to be when a particular ideological movement becomes conflated with authoritarianism and hence conflated with extremist ideologies.

We’d need to look at countries where right-wing ideologies and fiscal conservatism don’t dominate. In an authoritarian fiscally liberal left-wing country, I suspect that the minority of counter-cultural ‘conservatives’ would have above average IQs. Maybe it could be as simple as low IQ people in general just like to fit in with the dominant ideology of their society or are less likely to think outside of the dominant ideology of their society.

At present, the right-wing ‘conservative ideology of capitalism dominates nearly all of Western civilization and most of the rest of the world through globalization. So, one would expect to find the high IQ people disproportionately opposing or standing outside of this dominant ideology. It could be a mere historically contingent condition of the ideological spectrum.

Does that make sense?

2 thoughts on “IQ Dilemma: Inconvenient Correlations, Uncomfortable Data

  1. My hypothesis would be easy enough to test.

    First, determine which countries have low poverty rates (along with low rates of poverty-related social problems: economic inequality, unemployment, malnutrition, polltion, violent crimes, etc), low authoritarian rates, and no major political parties or movements that are strongly influenced by extremist ideologies. Basically, what is needed are the most economically and politically moderate countries in the world with well functioning social democracies. The most optimal countries would have effective public education systems that fairly and equally educate the entire population which would allow every individual the opportunity to fulfill their furthest cognitive development.

    Second, determine the specific demographic that is both most politically liberal and most liberal-minded. This would be the most hardcore liberals who most strongly and clearly embody liberalism which, depending on the political culture of a country, may or may not correspond to self-identified liberalism. Determine the same for conservative-minded conservatives.

    Third, determine the average IQ of liberal-minded liberals and conservative-minded conservatives. Within these demographics, determine whether average IQ increases, decreases or remains the same between liberals who are more politically liberal and liberal-minded and those who are less. Determine the same with conservatives.

    If a study followed this plan, it could be determined the causal link or lack thereof between liberalism and higher averag IQs and between conservatives and lower average IQa. This would determine if the higher average IQs of American liberals is merely an artifact of their lower rates of poverty, authoritarianism and extremism. If that turned out to not be the case, further studies would necessitate controlling against these confounding factors in order to determine the actual causal factor(s).

    Obviously, more research is needed. Staffan would have been correct to state this, but insted he sought to dismiss the research already done. We need to build on that research and not bury or heads in the sand because some think it is politically incorrect to point out some demographics have higher and some lower average IQs.

    If we discovered that liberalism somehow causally encourages and promotes greater cognitive development, then maybe we as a society should encourage and promote liberalism. Or if some other factor turns out to be key, we can encourage and promote it. Progress sometimes comes by facing uncomfortable areas of our own humanity.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s