Thursday, January 08, 2009

Atheists worry more than theists do and agnostics average the highest IQ

++Addition++It's pointed out in the comments that a decade ago nearly three-fourths of the members of the National Academy of Sciences were self-described atheists. Among GSS respondents, Wordsum scores for atheists are relatively clustered at 0-1 correct answers and 9-10 correct answers. The standard deviation value is widest for them. While 12.2% of all respondents who answered the question on the certainty of God's existence scored either 9 or 10 on the Wordsum test, 21.2% of atheists did. Conversely, while only 2.5% of all respondents scored 0 or 1 on the Wordsum test, 10.3% of atheists did. So atheists, more than others, seem to come in two distinct varieties--the hyperintelligent empiricist, and the dumb, vicious social outcast.


In response to the British Humanist Association's ad campaign on London buses, the Inductivist looked at patterns of religious service attendance and life satisfaction in the United Kingdom. He found that those who attend religious services are more satisfied than those who do not.

The GSS allows for a look at professed beliefs in addition to actual religious service attendance. Respondents have also been asked how many days each week they worried about things. Unfortunately, this question was only asked in 1996, a year when respondents' confidence in the existence of God was not inquired about. So I had to settle for a proxy question about respondents' perceptions of the Bible. The average number of days respondents reported feeling worried, by take on the Good Book (N = 959):

Bible is...Days worried
Word of God2.70
Inspired Word2.75
Book of Fables3.24

Click on the graphic for greater resolution. Red shows the percentages of each group who never feel worried. Salmon shows the percentages of each group who feel worried on a daily basis.

The trend among Americans is similar to the trend among Brits. The assertion that religious folks worry too much is simply off the mark. That they remain blithely unaware when they arguably should be worrying might be a more accurate line of attack, but as I often find in exhorting friends who don't pay attention to the news to start doing so, many people do not find getting pent up about things beyond their control an attractive way to spend their time.

The humanists might also point out that the more certain one's belief in God is, the less intelligent that person tends to be. The average IQ by level of confidence in the existence of God by way of converting the mean Wordsum score of whites to an IQ of 100 with a standard deviation of 15 points (N = 6,070):

On God...IQ
Don't believe98.1
No way to find out106.3
Some higher power103.0
Believe sometimes97.0
Believe, but have doubts100.3
Know God exists97.1

Oops, looks like agnostics tend to be more intelligent, on average, than atheists and theists alike! I am reminded (as far as questions about the supernatural are concerned) of Socrates' famous identification of his own superior wisdom in Plato's Apology, where he at least realized his own limitations.

Anyway, just as inviting others to share in your anxiety doesn't constitute much of a sales pitch, neither does informing those who do not share your beliefs that they are stupid.

How should atheists go about spreading their message? The American Humanist Association is trying a different message in DC:

“Why believe in a god?” the ads read, over a picture of a man in a Santa suit. “Just be good for goodness’ sake.”

The GSS shows little difference in volunteering among the three Bible perception categories, although those who believe it to be a book of fables are almost twice as likely to have been picked up by the police (whether or not they were actually guilty of any crime) than those who believe it to be the word of God are.

So maybe the humanist groups' approaches are the best advised, even if they insinuate that becoming a non-believer will make former believers more like actual believers already are.



Anonymous said...

Aside from understanding of Bayesian statistics, this might also reflect social desirability bias: the term 'atheist' is subject of much abuse and bad press in the U.S., where 'agnostic' is not.


As far as the intelligence of atheists, agnostics and theists are concerned, when Nature polled members of the National Academy of Sciences on their religious beliefs in 1998, 72.2% were atheist, 20.8% agnostic, and 7% theist.


BGC said...

The assumption behind this slogan is that atheism is merely the subtraction of a false belief from Christianity, thereby saving time and energy for more important matters - but leaving behind most of the rest of Christianity (such as most of the ethics).

But the many differences between atheists and Christians which have been documented here, and at the Inductivist and GNXP blogs suggests that this isn't correct.

Subtracting belief in God does not leave everything else untouched. At an individual and cultural level subtraction of belief in God makes many differences - mostly over the long term (since both humans and human society have a lot of inertia).

Leaving aside questions of truth - it seems that long-term atheism allows a person/ society to be more individualistic/ selfish, more this-worldly/ short-termist, less dutiful/ more hedonistic and more careerist/ less devoted to transcendental values.

In Human Accomplishment, Charles Murray concludes that the highest level of achievement probably requires a level of dedication which can only be supported by a society with an active valuation of the transcendental values (eg virtue, truth and beauty). To be sustained in the long term, these values may require belief in God.

However, secularization has prevailed for many decades because religions are often hostile to science and economics, and the most scientific and wealthy individuals and societies have generaly been those that became most powerful. But the long term demographic trends, and perhaps the morally enervating effect of excessive individualism, hedonism and short-termism, will probably lead to a progressive decline in secularism.

The future will probably be about a choice between religions, specifically a choice between revealed religions; not about a choice between god and atheism.

Audacious Epigone said...


Any chance you'd look at an editorial in Medical Hypotheses along these lines? I'd love to see you flush these ideas out even further.

Yes, religiosity and fecundity are strongly linked, both at the individual level and at the national level. I have an upcoming post looking at TFR and data on acceptance of an atheistic leader from the WVS.

Tangentially, in his interview with GNXP's Razib, Charles Murray mentioned that as he has aged, he has become progressively more theistic. I wonder if this has been in part a result of his scholarly work.

Anonymous said...

Interesting contrast with Barna's US survey showing that atheists and agnostics have the lowest divorce rate (21%) versus born again Christians (27%), other Christians (21%) and Jews (30%).

To quote:

"Barna's results verified findings of earlier polls: that conservative Protestant Christians, on average, have the highest divorce rate, while mainline Christians have a much lower rate. They found some new information as well: that atheists and agnostics have the lowest divorce rate of all. "

Worry more, divorce less?

Audacious Epigone said...


It's not really a contrast. Atheists tend to have higher average IQ than most religiously affiliated people (Jews and Episcopalians excepted), and divorce is inversely correlated with intelligence. But agnostics and especially atheists are less likely than the religiously affiliated to ever get married. Consequently, they are more likely to live alone, even though when they do marry (like most marriages between people with higher IQ), it tends to work.

William Hone, Jr. said...

But as the "Atheists in Foxholes" memorial celebrated at the following website demonstrates, the claim that there are No Atheists in Foxholes is simply not true:

BSR said...

In U.S. and much of the world, there is a strong disapproval of atheism & agnosticism. This, and the fact that atheist & agnostic fractions reported are usually small, care has to be taken in drawing conclusions. My guess is, if there was enough respectability for atheism as it is for other classifications (like sexual orientation, gender, race/ethnicity, language), the number of atheists/agnostics would be much higher.

Damian said...

Atheists are agnostics (and most admit to this, as well). As Richard Carrier has pointed out, it's not really possible to be one and not the other, depending on which conception of god that we are considering (which is very important). Agnostics are, after all, still without theism (a-theismos), so this is more a game of semantics than anything else.

And I would suggest that anyone that has thought about this issue — even for a moment — would admit that we are strictly, and philosophically, without knowledge (a-gnostikos) if we are considering "god" in the abstract — i.e. a "first cause" — and not as described in many of the worlds religious texts.

The question appears to apply to all gods, so it is eminently reasonable to expect that intelligent atheists would give the second answer.

So, these statistics neither surprise, nor impress me, because most (agnostic) atheists that I have known would have answered that there is "no way to find out" (particularly as the question isn't specific).

Oops, indeed.

Audacious Epigone said...


The dictionary definition of atheism is the "doctrine or belief that there is no God." The second definition is "Disbelief in the existence of a supreme being or beings." The GSS response that God does not exist captures that much better than the "there is no way to find out" response does. The latter is better fits the dictionary definition of agnosticism: "an intellectual doctrine or attitude affirming the uncertainty of all claims to ultimate knowledge."

The distribution for the atheist response category is wide though, with one-fifth scoring either a 9 or 10 on the Wordsum vocab test (indicating an IQ of 120+).