The Study on Intelligence and Religion That You Weren’t Supposed to Hear About

March 5, 2013
Yoda - the beautiful union of intelligence and spirituality (and possibly also dyslexia).

Yoda – the beautiful union of intelligence and spirituality (and possibly also dyslexia).

Numerous studies have consistently arrived at one and the same conclusion: religious people are less intelligent than atheists. So maybe we should accept this fact and move on. Nah, just kidding. On the contrary, when behavioural scientists all come up with the same result it’s time to get suspicious. The normal thing in this field is a variety of results.

So, now that we’re all duly suspicious, the next step is critical scrutiny. I’ll just look at some of the major studies in recent time, but it should give you an idea of the situation.

Nyborg

In 2008Danish psychologist Helmut Nyborg conducted a study that showed that atheists scored an average of 1.95 IQ points higher than agnostics, 3.82 points higher than liberal persuasions, and 5.89 IQ points higher than dogmatic persuasions. This may sound impressive but Nyborg’s sample was made up of children age 12-17. It is well known that both intelligence and personality are under much larger environmental influence during childhood than in adulthood. Nyborg himself noted that religion declines under this period but failed to see how this makes the study flawed. Whatever the environmental factor is – peer pressure, rebellion etc – it decreases with time. This sort of short-term environmental influence goes for intelligence as well. It becomes stable once you reach young adulthood. So young people make a horrible sample for making general conclusions of how religion and intelligence might be related.

Lynn

The same year British psychologist Richard Lynn (together with Nyborg) conducted a similar study. He simply compared the national averages of intelligence and rates of disbelief in God in 137 nations, making up 95 percent of the world’s population. Lynn found a very high correlation of 0.6 between these two variables. While this design appears to give the final answer, it is in fact measuring widely different countries which makes it very problematic. On such problem is that if we look at the big picture most nations in the world have an IQ close to 100 and then we have the Muslim nations that average around 80-85. So a large part of Lynn’s findings is about Islam, not religion in general.  If we restrict this sample to Western nations the correlation drops to 0.42. Lynn also admits that the surveys on religious disbelief may have problems with low response rates, being representative of the population etc. It remains an open question how much further the correlation would drop if we could adjust for this factors but we can’t so it really isn’t more than guesswork.

Kanazawa

The third major study in recent years was conducted by Japanese psychologist Satoshi Kanazawa, who of course corrected the persistent flaw of using children in his samples. Well, kind of. He used participants aged 18-28 to measure religiousness but then took the same participants IQs at age 12-17 so it’s not much of an improvement. His sample was also the 75 percent remains of an original sample which leaves the question of how representative it was open. Kanazawa also made another mistake in how he measured religiousness. Rather than using a multidimensional test to separate quality and quantity he merely had people answer how religious they were on a scale from 1 to 5.  A fundamentalist will no doubt claim to be “very religious” and score a 5. If his intelligence is low (and there is something to suggest this, more on that later on) then that would contribute to the relation between religion and IQ by suggesting that a non-fundamentalist is essentially the same kind of person but with a slightly weaker faith. There are various ways a person can be religious but Kanazawa suggests this is a matter of degrees and in doing so everyone gets to be stupid on behalf of the fundamentalist. Furthermore he uses a simple vocabulary test as a proxy for IQ. Given all these problems, Kanazawa finds that atheists average around 103 in IQ and the very religious at 97. With all the crap he has been pulling I suspect even his fans were a little disappointed with such a meagre result.

Gary J. Lewis?

Although these three studies have been very publicized and quoted, there was a study conducted by psychologist Gary J. Lewis and colleagues at the University of Edinburgh in 2011 that never got much attention. This study actually used an adult sample (mean age 48), measured several aspects of religion and featured a composite measure of intelligence. It did have some limitations in that the sample was mainly White and only concerned itself with Christianity. But it should provide a pretty good picture of the relation between Christianity and intelligence among White Americans.  An especially interesting feature was that it controlled for the personality trait Openness to experience, relevant to both intelligence and religiousness.  So what did they find? Well, they did actually find that religious people are less intelligent, but the results may still come as a surprise. Here is the gist of it from the article,

 The results indicated that intelligence is significantly negatively associated with five of the six measures of religious belief, confirming previous work (Bertsch & Pesta, 2009; Kanazawa, 2010; Lynn et al.,2009; Nyborg, 2009). Certainty about these findings is enhanced by the fact that, in the present study, these relationships to intelligence remained after we controlled for both openness and education, two potentially confounding factors. It should be noted, however, that the effect sizes were small for all associations with intelligence, the largest being the modest intelligence–fundamentalism link (β=−.13), with all other associations estimated at less than β=−.10.

Note the level of these correlations. It’s a well-known fact that variables of this kind always correlate with each other so small correlations are to be taken with a pinch of salt. Here is one take on correlation levels from David Buss & Randy Larsen’s textbook (which I highly recommend) Personality Psychology: Domains of Knowledge About Human Nature (2010),

Although what is considered large or small depends on many factors, social scientists have adopted a general convention. Correlations around .10 are considered small; those around .30 are considered medium; and those around .50 or greater are considered large (Cohen & Cohen, 1975).

So, four of six measures of religiousness showed less than small correlations, a level you might call negligible. Spirituality was unrelated to intelligence, and, perhaps most surprising, fundamentalism (of the Christian variety that is) showed only a small correlation of -0.13.

We Are All Stupid in Our Own Way (Except Me)

Now, the higher quality of this study means that the correlation can be taken more seriously. But at the same time the result indicate that not even the fundamentalists are very unintelligent. This may seem puzzling but fundamentalists are mainly considered stupid because the deviate from the norm.  Consider for instance the mainstream delusions of Optimist Bias. This refers to how ordinary people have an optimism which goes against common sense. Typically the majority of all people believe that they have a lower than average risk of becoming alcohol dependent, catch an STD or have a heart attack. This rosy outlook is of course mathematically impossible but those who engage in this sort of thinking are not in general considered stupid, most likely because they are in majority.

And What About Openness?

I almost forgot the interesting trait known as openness which is known to be related to intelligence as well as a liberal attitude. It was actually slightly related to two measures of religiousness, spirituality and mindfulness, while unrelated to the others, with the exception of fundamentalism which it was slightly inversely linked to (-0.12) – which is pretty much what you’d expect. Another interesting finding is that fundamentalism was unrelated to both sex and gender, contradicting the notion of this person as a middle-aged or older man.

Hopefully, there will be more high-quality studies like this one to put the religious-people-are-so-stupid research to rest. The interactions of intelligence, religiousness and personality are very interesting and deserve a better fate than to be skewed and dumbed down for ideological purposes.

Advertisements

Solving a Mystery for Satoshi Kanazawa

February 14, 2013

Over at the Big Think, psychologist Satoshi Kanazawa lists what he considers the unsolved mysteries of evolutionary psychology. One of these mysteries is the question of why people with many siblings have many children. According to Kanazawa this doesn’t make any evolutionary sense at all,

This is because people with many siblings have the option of investing in their younger siblings and increasing their reproductive success by doing so.  Humans are just as genetically related to their full siblings as they are to their own biological children; both share half their genes.

I think the solution to this mystery can be found in viewing extraversion and introversion as evolutionary strategies. There is plenty of research showing that extraverts are more sexually active than introverts. This would lead to them having more children (we’re dealing with an evolutionary time scale so we can ignore the effect of contraception), and since it’s also a highly inheritable trait it would make a person with many siblings more likely to have extraverted parents and for that reason he is likely to have inherited their extraversion and have many children himself.

But if so, why aren’t we all extraverts – if it’s enables us to spread our genes so well? To understand this I think we need to look at differences in this trait between people living in environments with varying degrees of resources. In good times with plenty of food available, extraverts would propagate and spread their genes by having many children but a relatively small parental investment. But sooner or later hard times would come and then the introverted strategy of few children and high parental investment would pay off. If there is a scarcity of food, giving all you can spare to one child rather than sharing it between 4-5 children becomes a winning strategy. That way, hard times would ensure that genes for introversion would survive.

Now, it’s a fact that different regions vary in resources. People who have originated in a cold climate, or at least in a region with long cold winters should, if this theory is correct, have a higher level of introversion than people from warmer regions. We are all familiar with the stereotypes of the introverted northerners and extraverted southerners. Is there any truth in it?

Looking at extraversion scores from Richard Lynn’s study (1995) of 36 (I omitted Iceland because it has a microscopic population) nations there is actually a bit of a pattern supporting the theory.

The average level of extraversion for all countries was 18.4. But those who originated in a cold climate (Nordic or Central Asian) averaged at 17.2, while those originating in a warm climate averaged at 20.0, with intermediary countries averaging at 17.9.

Extraversion scores by country and climatic origin.

Extraversion scores by country and climatic origin.

I’m probably not the first person to have this idea but since Kanazawa insists that it’s a mystery and this seems to be a plausible explanation, I thought I should share it.


%d bloggers like this: