Evolution vs. wealth

March 26, 2012 • 5:16 am

From Calamities of Nature comes this unprofessionally drawn but most enlightening graph of the relationship between national wealth (gross domestic product) and belief in evolution, with each dot representing a country (dots from similar regions have the same color). Click to enlarge:

The points are not all independent, of course, because geographically contiguous countries have similar religions (ergo similar feelings about evolution) and similar degrees of wealth.  Still, it shows what I’ve mentioned several times before: northern Europe has high national wealth and high belief in evolution; eastern Europe has the opposite, even though many of those countries were once Communist, and the poorer you are, the less likely you are to accept evolution.

The positive relationship suggests that countries that are wealthier, and whose inhabitants are doing better, have less impetus to be religious, and hence less need to reject evolution.

Most striking to Americans is our status as an outlier: our acceptance of evolution is much lower than our GDP would indicate.  This is probably the result of other factors that make American society, while wealthy, more socially dysfunctional: greater income inquality, lack of social support like national health care, and so on (see Greg Paul’s work on the strong negative relationship between societal “success” and religiosity).

I’m not a sociologist, of course, so my take is at best superficial, but more and more data show that religion takes hold when society fails to fulfill certain fundamental needs of its members, making them  less secure and more likely to grab for the supernatural.  While the U.S. is a wealthy country, it is very low on Greg Paul’s “successful society scale,” and we have inordinately high levels of income inequality, also a sign of social insecurity.

I’m not a Marxist, but data support Karl’s famous statement, “Religion is the sigh of the oppressed creature, the heart of a heartless world, just as it is the spirit of a spiritless situation. It is the opium of the people.”

73 thoughts on “Evolution vs. wealth

  1. A weakish positive relationship is evident, but the curve fit to the data is arbitrary and rhetorical.

      1. There are (at least) two conclusions that can be drawn from the figure, and one of them is the general positive relationship. The curve fit is a dishonest and rhetorically biased depiction of that relationship. It’s not irrelevant.

  2. To be honest, the use of a complex curve fitting ( y= A(1-B/x) ; y= evolution belief ; x=GDP [with adjustments]) seems unnecessary. I’d have looked at the data as fitting two (sub-)parallel lines, one linking the US and Turkey, the other being the regression for all the other cited countries. Without access to the original data, I can’t demonstrate that this would produce a better fit (define “better” ; I’m thinking sum-of-squares-of-errors, but I can’t remember the normal statistics notation), but I think it would probably produce a more normal distribution of errors, and probably a better fit overall.
    Culturally linking Turkey and America … hmmm, provocative.
    The selsction of countries seems a little … eclectic. Probably determined by the data in the Science paper – I don’t have access to it. I note that the other “Western Asia” country, Cyprus, plots squarely in the main band in contrast to Turkey (which has de facto control of half of the island).
    It would be interesting to compare South Korea (where protestant Christian evangelical nuttiness is strong) with Japan’s placement in the main band.
    Hertzsprung and Russell, where are you when we need you?

    1. It is a showcase curve.

      – Anything else than a linear relationship would promote certain parts of the curve. Here the low GDP/capita data and its errors dominates the fit.

      – A multi-factor analysis, say a PCA, would equalize linear and so forth factors.

      – It is an unphysical fit. You can’t have zero GDP/capita. I suspect the artist’s dodge is that no one survives on that, but that has nothing to do with the major part of the data dependence.

      The HR evolution idea is perceptive I think! Hasn’t Rosling noted that nations can either promote wealth (or Gini index) before social medicine (most western nations) or social medicine before wealth (China) and end up with a healthy population both ways? HR diagrams have such degeneracies as well, I believe.

      1. I think I see how you got from my comment on Hertzsprung and Russell (on stellar colour vs absolute luminosity) through (stellar) evolution pathways, to different ways of achieving a healthy population. But I’m not entirely sure that I get you. Could you elaborate a bit, please.

  3. Jerry,

    A minor point – GDP is a flow (gross domestic product per year) whereas wealth is stock. This graph shows that higher income countries have higher acceptance of evolution, although presumably the same is true is you used wealth.

    Also, I agree with the previous post that both USA and are Turkey are probably outliers, which is not a surprise given Turkey’s Islamic religiosity.

    — Chris

    1. Didn’t Kemal Pasha try hard to get rid of said religiosity? I heard till recently Turkey used to have a law prohibiting the use of religious items in political manifestos of parties, and perhaps it still does. Wouldn’t that be signs of a very aggressive secular society?

    2. both USA and are Turkey are probably outliers, which is not a surprise given Turkey’s Islamic religiosity.

      Not having visited there, I wonder how much of Turkey is actually strongly religious as opposed to having a lot of public display, given that high amounts of public display are required by Islam. As Ahannāsmi points out, Turkey was founded as an explicitly secular state, with significant legislative effort going into trying to prevent religion from becoming a significant force in the politics of the country. This has been under sustained attack from both religiously inspired forces and politicians wanting to use religion to attract support. (As I understand it, from press reporting.)
      Another country not considered in the original posting, which would be interesting to see, is Israel. Having worked there, the bulk of the people don’t seem to be particularly religiously devoted. but their politicians seem if anything to be even more religious nutjobs than the modal American politician.

  4. The sad thing is, I bet that large segments of the population would prefer to be “right with God” than to be successful as a society. Much the same as the Islamic theocracies would.

  5. When examining the United States, it would be useful to break the country down regionally. You would most likely find that the northeast and west coasts are where the greatest wealth distribution occurs, an where there is much higher acceptance of evolution among the citizenry, while in the south and Midwest there is greater poverty, less education, and a rejection of rational science in favor of mysticism and superstition.

  6. The points are not all independent, of course, because geographically contiguous countries have similar religions (ergo similar feelings about evolution) and similar degrees of wealth.

    I would say “tend to have” — as a contrary data point, Canada has slightly lower per capita GDP than the US, but much higher acceptance of evolution (59% by one survey), putting Canada somewhere near Austria in that graph, and well away from its contiguous neighbour.

  7. I wonder what made the author choose “A(1 – B/x)” as the formula.
    It seems pretty obvious to me that such a curve is an extremely poor fit for most countries with GDP < 20000$. A simple linear equation fits the data a lot better.

    Rough estimation gives the following coefficients:

    BiE = 0.42 + 6e-6 * GDP
    (That would be 42% + 0.0006% * GDP for those who think in percents, not in ratios)

    1. The curve fitter has made the assumption that the curve should asymptotically approach some value when GDP per capita becomes large. That value is A and should be constrained to be less than 100%. The formula is obviously simplistic and gives nonsense (negative percentage) when GDP per capita is less than B. General inferences such as “the USA is an outlier” and “the data indicate a strong positive correlation between GDP per capita and belief in evolution” are justified without the curve fitting.

  8. If income insecurity is supposed to lead to greater religiosity, then why in America has church attendance been dropping most rapidly among those in the lower income brackets? It seems you should see an inverse relationship between wealth and religiosity, if the argument is true.

    1. Not income insecurity, but insecurity/security generally, i.e., levels of social dysfunction/social well-being: e.g., homicide rates, alcoholism, divorce, degree of economic inequality, access to health care, education, etc, etc.

      1. But this is an inappropriate seperation. “General” social insecurity and disfunction somehow explains greater religiosity, but actual individual insecurity and disfunction don’t?

        1. I do not understand why you think this distinction is inappropriate: if everyone were to share your view social science would be limited to idiographic approaches.

          I am also perplexed by your latter clause – general levels of social well-being/social dysfunction are the result of individual, actual levels.

          1. The indices of social well-being you cite (homicide rates, alcoholism, divorce, degree of economic inequality, access to health care, education, etc, etc.) are higher among low income earners. Yet, instead of producing more religiousity as Greg Paul would contend, we’ve been progressively witnessing less.

            There is thus something wrong with Paul’s claim that “social dysfunction” leads to higher religiousity.

            1. No. It operates at a societal level. Societies that have high levels of social welfare, have lower levels of religiosity; and societies with higher levels of social insecurity/social dysfunction have higher levels of religious belief.

              1. So, it operates at the societal level, but not at the individual level? And here I thought societies were aggregates of individuals, not a kind of seperate entity.

              2. There is such a thing as society. Individual persons only exist as members of a society.

    2. Maybe it’s just not a two-way street. The *existing* religious could lessen as wealth increases but is may also lessen for different reasons as wealth decreases. I’m pretty sure new uptake has nothing to do with increasing wealth or poverty.

      I’ve said this here before in the discussion about some stats that related education to religion: a statistic can only point to the existence of a relationship, it cannot say anything about the nature of the relationship, ie the causes. In this case The wealthy may leave as easily because of a clash of values as for enlightened philosophical reasons while the poorer may leave simply out loosing faith due to hard times. Stats like this are pretty much useless without getting to grips with the actual reasons.

  9. The figure for Canada would be a useful comparator for the US.
    The questuion posed identifies only part of the problem. Those who accepted evolution for the purpose of the survey may have meant (to parahrase Star Trek) “Evolution, but not evolution as we know it”! It would be interesting to have comparitive figures for those who accept the scientific theory of evolution (Ie. with no reference to a “prime mover” or supernatural interventions. I feel that the US would turn out even worse.

  10. I think that USA should be represented as federation of states rather as one state. US states might be very diverse so using decoupled statistics instead of aggregates it might show that in US are some spots close to Scandinavia and other states might find themselves on the opposite side of spectrum closer to Turkey. I would find it more interesting. Is NY more like France, Spain or Poland etc?

    1. Second this recommendation. I’ve done similar state-by-state replots at least one on other such graph, and found the curve for the US states individually shifted a lot closer to the curve than the aggregated point.

      Contrariwise, based on the last time I tried such a thing, I’d expect the curve for the US states would still be on the low side.

    2. Interesting you picked NY state. Like California, the presence of a large city (or in CA’s case, cities) skews the political landscape. NY and CA are reliably “blue,” but that occurs only in the urban areas. The rest of each state is as “red” as area in the south. In the Central Valley of CA, I often passed those mega churches one sees more often in the South. (I have not traveled the interior of NY yet.) I would suspect, that parts of the state would be per capita on the poor side, (not much money in dairy farming) but the predominant belief system is unknown.

  11. In order to get an accurate picture of the USA, you would need to break te country down into roughly 4 regions: the north eastern seaboard, the south, the middle states, and the pacific coast. The areas of highest median wealth will be located on the eastern seaboard and the pacific coast, and the acknowledgment of evolution highest.

  12. It seems to me that epistemology must be a problem for persons who both subscribe to evolution and are atheists.

    Evolution does not select for knowledge of truth but for appropriate behavior. The two are not necessarily related.

    For the sake of argument , it may well be that a lion is actually a white snake-like creature but humans alive today descended from ancestors who perceived lions as brown and cat-like. These humans would survive and others with a more accurate perception die out if the latter’s response to lions were to try and pet them.

    All that is necessary to survive is not the truth about lions but an appropriate response to counter the treat posed by lions.

    1. You’re question-begging, Phosphy — how do you know what the “truth” actually is? If the situation you describe actually was the case, how could you independently know that evolution had selected the “wrong” answer?

        1. Right, so you admit that you can’t distinguish such situation from where you have direct access to “truth”. Are you just saying you hope that you can tell what is true, even though you can’t know whether you can?

            1. And do what, precisely? Wish there was a god that somehow magically ensured our perceptions and beliefs were veridical?

              I seriously don’t understand your point. You’ve conceded that you have no alternative way of determining truth, so your position is equivalent to someone saying “But we really could be in the Matrix!” Sure, but so what, since as you admit, we’d never know if that were the case.

              1. In their discussion on biological information at :


                the Stanford Encylopedia of Philosophy makes the observation that

                “paradigm cases of semantic information are the products of an intelligence”

                but you reject this with regard to living organisms hence its is you who are in the matrix not me.

                As I have said before the statement “In the beginning God…..” is an axiom. Without it life is incoherent.
                No truth, no free will,no law,no morality.

                There can be no intellectually satisfied atheist.

              2. As I have said before the statement “In the beginning God…..” is an axiom. Without it life is incoherent.
                No truth, no free will,no law,no morality.

                But you have also conceded that this actually could be the state of affairs, and you couldn’t tell differently. How can you be “intellectually satisfied” if you can’t know that your belief is actually true?

                How do you avoid the dilemma that you think arises for a selection-based naturalized epistemology?

              3. “But you have also conceded that this actually could be the state of affairs, and you couldn’t tell differently. How can you be “intellectually satisfied” if you can’t know that your belief is actually true”?

                I have never conceded that this dilemma could actually be the state of affairs as I have never conceded that syntax for prescriptive semantic information could arise from the laws of physics and chemistry. This has never been demonstrated and I do not accept that it could be done. You belief that it can be done and in fact did occur. The dilemma, if it exists therefore, is yours.

                But even if it were demonstrated that syntax for prescriptive semantic information could arise from the laws of physics and chemistry merely finding this out neither removes the dilemma you will face on the incoherence of concepts such as knowledge of truth, free will, morality, law etc without the God axiom or the fact that a transcendent intellect is a logical explanation for living organisms.

                So we both live by faith each with a different outcome.

                The one (atheism) leads to incoherence and disorder – nihilism – the other (theism) to coherence and order.

                That’s why we are instructed to have a “reasonable” faith.

              4. finding this out neither removes the dilemma you will face on the incoherence of concepts such as knowledge of truth, free will, morality, law etc without the God axiom

                Again, Phosphy, you’re just question-begging, as you haven’t demonstrated that knowledge of truth, free will, morality, law, etc. are in fact coherent concepts. Do that first, and then we can talk. But until then, you’re just assuming a wished-for conclusion, then arguing from it.

              5. You ask too much of my simple intellect and I have no intentions of ” staring into the abyss” so we’ll leave it there.

              6. Tell you what, Tulse, live as though knowledge of truth,free will, morality and law are not coherent concepts for one month, start with your bank, and then let’s continue the discussion.

  13. A much simpler, more obvious explanation for such a trend would be that the whole science effort behind evolution is an expensive affair and the wealthier countries are more likely to have more people doing it. Media is also more pervasive in wealthier countries so what is learned in science is more easily distributed. School textbooks are also more regularly updated. Etc.

    It is also worth noting the number of countries which are higher on the evolution scale than wealthier countries.

    The US as an outlier can easily be explained by Prof Coyne’s unique ability over there to get all the religious people’s backs up. 😉

    1. If you compare South African schools you are likely to see the opposite of this graph. The wealthy private schools tend to be religious schools with names like Bishops and St Cyprians and I’m quite sure the uptake of evolution is much higher in those schools than in poor rural schools which have no formal religious structures.

    2. It is also worth noting the number of countries which are higher on the evolution scale than wealthier countries.

      Good point. Most pertinent observation.
      The specious fit curve is not even the worst aspect of this plot.
      Unweighted per capita GDP is grossly misleading to begin with.
      But a vertical quantile-by-quantile inspection reveals more troubling questions regarding the Acceptance of Evolution scores:
      – Why do poorhouse Romania, not-too-well-off Croatia, emergent Poland and Slovakia score within the same vigintile (5%-bandwidth) as wealthy Austria?
      – Why does the Swiss average score slightly below arch-Catholic Malta and not-notoriously-Scandinavian, not-notoriously-well-off Portugal?
      – Hungary, the Czech Republic, Italy, and the Netherlands all score well within the same vigintile bandwidth. Etc.
      Anyone moderately acquainted with the histories, societies and cultures of these countries will propose a number of answers. They won’t be so simple, and they won’t be so linear.

      The question of factors correlated with, and possibly causal to, the level of acceptance of Evolutionary Theory in a given society is a serious one, requiring serious data and deserving serious treatment. Anecdotal seat-of-the-pants statistics like this one are not helpful.

      As an exercise for the statistically inclined reader, I suggest an analysis of the “swiss” dataset that comes with every standard installation of R. It represents fertility and socio-economic data from just the French-speaking part of Switzerland, district by district, in 1888/89. The numbers tell a cautionary tale.

      1. Richard Dawkin’s site has the actual study:


        Both Austria and Bulgaria have a large “not sure” segment. Many Austrian kids end regular school at the age of 14 and go to trade school and vocational training. They probably never hear about evolution.

        The Austrian version of the World Values Survey is done by the catholic faculty of the University of Vienna. Those are qualitative, personal interviews. There might be some selection bias and observer-expectancy effect, if the interviews are done by prospective priests and religious instruction teachers.

        1. The Austrian school system alone cannot be faulted. It is, by and large, not too dissimilar to the one the Swiss have, and has only recently failed to evolve in the same direction as Germany.
          A more recent (2008) representative (n=1520) poll offers some different insights, which may produce different explanations:
          “How do Austrians think about Evolution?”
          (sorry, in German only)

          A few salient questions and answers:
          – humans and apes have common ancestors: Yes 80%, No 12%, Don’t know 8%;
          – how did living organisms and biological species survive cataclysmic events?
          (multiple responses possible)
          — through individual adaptation: 72%
          — formation of new species: 40%
          — through particular resilience: 20%
          — they didn’t survive at all, most organisms and species were exterminated: 11%;
          – Evolution is the result of random events: Yes 45%, No 41%, Don’t know 15%
          – Evolution does not suffice to explain the origins of life: Yes 51%, No 30%, Don’t know 19%.

          I think quite a few amusing problems and serious contradictions are self-evident.
          If readers are interested and Jerry agrees, I’d be glad to translate the entire study.

          1. Hm, I would have a lot of trouble answering many of those questions, as they seem ambiguous to me. For example, I’d tend to say “No” to the last two, since natural selection is not random, and evolution doesn’t explain abiogenesis, but I tend to think I’m “supposed” to answer “Yes” if I want to register agreeing that evolution is the most supported theory for the appearance of life we see today. On the other hand, I tend to over-think these sorts of questions, which is probably not very common.

            Of course, it is also possible that many of those who will state that they agree that evolution is the best explanation assert it dogmatically, because they were told it was so, and don’t actually understand the basics of the theory.

  14. Good information. Thanks for verifying something I suspected was true. The relationship between security and religion is not nearly as strong as between education and religion.

  15. So the obvious question is how does the US differ economically from the other wealthy nations? One obvious answer is the progressiveness of the US tax system. Most other wealthy nations have extremely regressive tax systems compared to the US. Those nations depend largely on VAT and similar regressive tax structures for most of their revenue, while the US depends on progressive income taxation much more heavily. The overall percentage of taxes paid by wealthy people in those nations is a small fraction of the overall percentage of taxes paid by wealthy people in the US.

  16. “even though many of those countries were once Communist, and the poorer you are, the less likely you are to accept evolution.”

    Communist regimes disliked Darwinism. Evolutionary biologists were considered to be untrustworthy by the Marxist-Leninst establishment because of the obvious reactionary implications of evolutionary theory.

    You are also conflating lack of belief in Evolution with religiosity.

    Only 16% of Estonians and 19% of Czechs believe in God according to Eurobaromater 2005, which makes them the most atheistic nations of Europe. 80% of the Polish believe in a God, but all three nations don’t differ on acceptance of Evolution at all.

    38% of Icelandics believe in God and Iceland has the highest acceptance of Evolution.

    The marxist notion that being determines consciousness doesn’t work at all. The different levels of religiosity in economically similar post-communist countries can be explained by the level of oppression against religion under the respective communist regimes.

    What makes the USA exceptional is that there is an active and well-organized anti-Darwinian propaganda effort that counters the efforts of the education system.

    P.S.: It is also a false assumption to believe that people who don’t accept Evolution must believe in theistic creation. A lot of people don’t care or haven’t even thought about the origin of species.

  17. If the USA is exluded then Turkey why not? The curve would be much more flat without Turkey.

    I really love that Hungary accepts evolution the most in the eastern europan group, but there are forces fighting against this, we have quite a lot of religiuous lunatics in the current government. Even if most of them would not be seen as a radical in the USA, but I really hope we never sink that deep.

  18. I think that the opening remark that the graph is ‘unprofessionally drawn’ is a little off the mark. It has obviously been produced without computer graphics and drawn by hand. It is very clear and the hand writing is very neat and legible.

    1. I agree: I was unfair in my characterization. I should have said, “Not drawn by computer”!

  19. Oil-rich Arabian states shower their citizenry with lavish public spending, including free healthcare, housing, electricity, subsidized gasoline, basic income etc. Yet religion and creationism are strong in these places.

    This is another case in point against the marxist analysis.

    Furthermore the frequently made disctinction between a singular “European welfare and universal healthcare model” and “american free market exceptionalism” is a false one. Healthcare and welfare differs greatly from one european country to the next and the US system seems to me to be more part of the norm rather than an exception when you actually look at the details of what is covered and under which conditions, etc.

    Medicare and social security are generous, cover the most vulnerable part of society, yet american retirees are highly religious for sure.

    The american tax system is actually more progressive than what we have in most of Europe. The US relies heavily on property taxes whereas Europe uses a regressive VAT to raise government income.

    1. I think it’s a simple bread and circuses stratagy so the masses don’t revolt against the totalitarian regime that, if they didn’t do these things, would probably be over-thrown quite quickly.

  20. WEIT: Come for the articles, stay for the systematic dismantling of the descriptive statistics used in a comic that’s usually about talking dogs.

  21. I think that the US is an oulier in terms of GDP and religiosity because of an accident of geography. I remember in my American history class when our instructor began by saying that the single greatest reason for the success of the United States is geography: what a magnificent piece of real estate.
    What I’m saying is that the US would have been a super power no matter what Americans believed in or didn’t believe in. If a hundred million chimpanzees managed to consolidate the whole of that land under one flag then you’d have ended up with a super power of chimpanzees. Instead, we have a super power of Americans. The United States can get away with believing in nonsense on a grand scale and still being rich because of geographical luck which no other country even approaches. If the United States was the size of Ireland, we’d see a very different outcome.

    1. Surely geography has been advantageous to the US; those two oceans were mighty helpful during the twentieth century. Your case is still very much overstated. When Europeans showed up on this piece of real estate, they were not opposed by a superpower.

  22. “I’m not a sociologist, of course, so my take is at best superficial, but more and more data show that religion takes hold when society fails to fulfill certain fundamental needs of its members, making them less secure and more likely to grab for the supernatural.”

    Well said.

    This of course is why your religious right was so vocal in opposing healthcare reform, whining throughout the debate that the Obama administration was trying to replace the churches by providing for the poor.

    And they were technically right!

    Although there was no planned attempt to weaken the influence of the churches over Americans, that is a very likely byproduct of providing basic necessities to your citizenry.

    Christofascists know that a wealthy America with a strong social safety net is a mortal danger to their religion, and by extension their political power.

    That is why they’re so strongly opposed to government social programs.

    And that is why I consider religion, leading the opposition to every attempt at human progress, to be literally anti-human.

  23. Or for the US, check out Colin Woodard’s book “American Nations”, on the US’s enduring regional cultures and their colorful and intertwined history. He identifies Yankeedom, New Netherland, the Midlands, Tidewater, Greater Appalachia, the Deep South, the Far West, El Norte, and the Left Coast, along with a few on the periphery.

    He interprets much of US history as a contest between Yankeedom and the Deep South, with whatever allies they could make. This contest once led to a major war, the US Civil War.

    Turning to Calamities of Nature – Holy Hypocrites, I find:

    Yankeedom searches for “god” the least, closely followed by the Midlands, the Far West, the Left Coast, and El Norte.

    The Deep South and Greater Appalachia search for “god” the most.

Leave a Reply