Average IQ Scores Have Risen 30 Points During the Past 100 Years

Did folks back in 1900 really have IQs averaging 70 points?



New Zealand political scientist James Flynn realized back in the 1980s that IQ tests are periodically re-normed to 100 points as average. The re-norming all went in one direction: upward at a rate of about 3 points per decade. This insight meant that person making an average score of 100 in 1965 would likely score just 85 points on today's tests. A new study in the journal Perspectives in Psychological Science analyzes the results of 271 independent samples, totaling almost 4 million participants, from 31 countries. They find that average IQ test scores have increased by 30 points over the past century. From the abstract: 

The Flynn effect (rising intelligence test performance in the general population over time and generations) varies enigmatically across countries and intelligence domains; its substantive meaning and causes remain elusive. This first formal meta-analysis on the topic revealed worldwide IQ gains across more than one century (1909–2013), based on 271 independent samples, totaling almost 4 million participants, from 31 countries. Key findings include that IQ gains vary according to domain (estimated 0.41, 0.30, 0.28, and 0.21 IQ points annually for fluid, spatial, full-scale, and crystallized IQ test performance, respectively), are stronger for adults than children, and have decreased in more recent decades.

Before concluding the people living 100 years ago were nearly all feeble-minded, the press release from the University of Vienna cautions:

The study results showed an average increase of about three IQ points every ten years. But do these results mean that an average IQ test result of 100 points in the present day would translate into an IQ of 130 a century ago? Although gains of about 30 points over a hundred years might suggest so, such an interpretation seems unlikely. Rather than increases in general cognitive ability, these gains are more likely to reflect improvements in specific abilities. "A person with an average IQ score of 100 in the early 20st century might have had quite different capabilities than a person with a seemingly equivalent IQ score of 70 in the present day", explain Jakob Pietschnig and Martin Voracek of the University of Vienna. IQ gains thus appear to be hollow in terms of global cognitive ability changes. Higher IQ test scores are more likely reflective of increasing specialization and better test taking strategies of participants.

The causes for this increase are numerous, including reduced childhood exposure to infectious diseases, better nutrition, more schooling, and vast exposure to more complicated media.

NEXT: Central Allegation in The Hunting Ground Collapses Under Scrutiny

Editor's Note: We invite comments and request that they be civil and on-topic. We do not moderate or assume any responsibility for comments, which are owned by the readers who post them. Comments do not represent the views of or Reason Foundation. We reserve the right to delete any comment for any reason at any time. Report abuses.

  1. I’ll just leave this here.

    1. Several of those questions cannot be answered; percent gain-loss/time? Not specified.
      Other than that, I didn’t see them as beyond what I hope a grammar-school grad should be able to answer.

      1. Another aspect is: most of that stuff is recall of memorized facts and tedious calculation.

        What with smart phones and all, we’re more intelligent than that now.

    2. Eh. Walk into a random 8th grade test today. I’m sure there is a lot of crap you learned but forgot.

      1. Also, I’m sure there’s a lot of Common Core crap you’ll have difficulty interpreting.

        1. True. “Place in the correct order who is the most oppressed groups.” Whaaa?

          1. Paradoxically, if you put down a whit-ish last name and checked the ‘caucasian’ check box, whatever answer you write down is correct.

    3. The really sad thing is that people back then thought all of those were important questions to know the answers to.

      I mean, really why is it important that someone be able to recite from memory which states border the Ohio river or which particular wars a collection of battles were fought in.

      A lot of those questions are indicative of why education has been such a failure because they show that the focus has always been on rote memorization of facts rather than achieving a deeper understanding of the meaning and importance of things

    4. Good demonstration of my point below.

      Who fought the battle of Quebec? Who the fuck cares? Who fought at Normandy?

      1. Normans. Duh!

      2. I’m assuming the French won since Detroit speaks English?

        1. Heh…who won the war between the states, considering Ohio got stuck w Toledo?

      3. Question 2 # ‘When was the War of 1812?’

        1. Africa?

  2. Lead being taken out of fuel…?

    1. Lead taken out of pencils?

      1. Oh yeah, I’ve seen that one batted around before. Makes some sense to me – lead is useful in some instances, being ingested or breathed….not so much.

        1. *puts down lead inhaler*
          Wait, wut?

          BTW: did you get a chance to try hendrick’s?

          1. Explains so much

    2. Probably iodine added to salt.

  3. The study results showed an average increase of about three IQ points every ten years. But do these results mean that an average IQ test result of 100 points in the present day would translate into an IQ of 130 a century ago? Although gains of about 30 points over a hundred years might suggest so, such an interpretation seems unlikely. Rather than increases in general cognitive ability, these gains are more likely to reflect improvements in specific abilities.

    So we’re suffering a sort of IQ inflation. Our IQ points (however few we have) are worth less than they were 100 years ago, but we have more of them.

    1. My school teacher once told me that things in Ye Olden Times cost less money because people had less money back then. Which is technically true but also so very wrong.

      1. Strikes me as straight up monetary inflation, applied.

  4. I don’t reject the “average” increase results, but concur with the cautions proposed by Univ of Vienna.

    Basic nutrition and food fortification/enrichment alone, particularly iron, could account for a significant overall increase.

  5. One things for sure — I’m way smarter than the dumbass Thomas Jefferson. I mean, he didn’t even know what Facebook was!

    1. Did too! How do you think he told his slaves apart?

      1. Are you saying that all slaves look alike?

        1. Yours don’t? That would mean you’re both overrationing and failing to start the morning off with a chimney clean.

    2. My standard: Did they have indoor plumbing? I’m obviously smarter than anyone who.shat in a bucket and tossed it out the window.

      1. I do a lot more camping and backpacking as I get older. Will you be my overlord? *brandishes lint-free monocle cloth, beckons for boy scouts to come over*

      2. I’m obviously smarter than anyone who.shat in a bucket and tossed it out the window.

        Install your own plumbing or really only as smart as the Mexican plumber who installed it for you?

        1. Smart enough to know to get a Mexican to do the job.

      3. “[…]I’m obviously smarter than anyone who.shat in a bucket and tossed it out the window.[…]”

        Dunno. At that time, the ‘paper’ may well have been some good reading material.

    3. Well, I had to have Twitter explained to me the other day, so I guess that makes me just like Thomas Jefferson.

  6. these gains are more likely to reflect improvements in specific abilities.

    Texting, for instance?

  7. As the world becomes more “wired”, the cultural biases inherent in IQ tests become less significant.

    *** bites lip ***

  8. So basically better childhood nutrition and reduced incidence of serious childhood illness has lopped the bottom off the IQ curve so that it is no longer bell shaped.

    The middle and top portions have remained relatively static but the poor are now basically equal to everyone else in IQ distribution

    1. Or at least has a different curve. But in any case, that would mean there were a lot more low-IQ people back then. Is there evidence of that?

      I do wonder if pre-modern life was more amenable to those with low IQ: many physical jobs, strict traditional/social rules, etc.

      1. ” Is there evidence of that?”

      2. I think if anything the opposite would be true, nobody paid you to sit around all day watching daytime television. In premodern times you had to figure out a way to survive. Even physical jobs would take some skill and intelligence to do.

        1. Ay and in those days, a person only starved when there was literally no edible matter to be had. Now I hear about folks who get all food-insecure because they are too fucked up to meet with some charitable distributor at a particular time. How it’s too hard to get from one end of town to the other during a three hour window once a week, and so forth. Furthermore, I’ve been through numerous parts of the United States, and everywhere I been I seen a country drowning in a superabundance of unused forageable comestibles that nobody takes the blooming trouble to reach down and pick up out of the ground. I don’t see how a people with motor cars can’t manage to get across town at a particular time in order to be given something to eat when hungry and which will go hungry surrounded by edible things just because they’re too stupid to know how to eat anything that didn’t come out of a cellophane package can be more intelligent than a people that was able to take advantage of almost one hundred percent of available resources to assuage hunger. One might be able to make some arguments that the new man is collectively more intelligent, since he’s able to, as a community, vastly diminish the problems people faced then, but there’s no way it could be extended to saying that an individual today is so vastly more intelligent than an individual then.

    2. Things that can be shown as a normal distribution (aka the dreaded bell curve) are often resistant to changes in the overall distribution. The curve can be shifted left or right on the X-axis, but squashing (narrowing) or stretching (flattening) the curve is extremely difficult, absent major intervention.

      No coincidence, then, that attempts to “level” results must by necessity involve tyranny. Of course, for progs, that’s a feature, not a bug.

  9. Quantifying intelligence cannot be done.

    1. How about ranking?

    2. President Camacho sentences Fist to one night of rehabilitation:

    3. Quantifying intelligence cannot be done.

      I sort of agree with this. I think there are different types of intelligence. Give a one size fits all score can’t capture that. Take my brother for example. He struggled with school, but has an amazing eye for detail and visual memory.

      1. Agreed.

        Dipshits a century ago who had to ‘make their mark’ in order to buy/sell cattle could stick their finger in the air and predict the weather have been replaced by college-educated dipshits that can type advance calculations into computers and consistently fail to predict the weather a century from now.

        Not that things haven’t gotten any better but if you said there was a huge degree of confirmation bias and/or cargo cult, I wouldn’t be surprised.

      2. The numbers seem to vaguely mean something or other. Folks rated upwards of 120 seem discernably different to average and below average types, whereas I can’t really detect an reliable and consistent difference among anyone in the range of 70 to 120, something I’ve heard reported by many other people in the higher brackets; whereas, folks in the narrow high average range seem universally to claim to be able to easily recognise a difference between people of high-average and low-average intelligence, though I am suspicious of whether they really can tell such a difference from performance or if there isn’t some social sign that identifies some people as less intelligent than others outside of function. There’s also that at the higher end, intelligence becomes wildly more variable in character. The functions of someone of 100 IQ are very similar to those of another person of 100 IQ, but two people both rated at 140 will generally be extremely different in function, and by the very nature of it it’s sair likely that IQ scores are decreasingly meaningful the more one deviates from average. But don’t listen to me. I don’t know dick.

    4. I’m doing it right now.

  10. Couldn’t it be that the tests are being made easier? Zod knows they’re dumbing down plenty of other things.

    1. +30

    2. They re-centered SAT scores to raise the “average score”.

      SAT is no longer an acronym for anything because “aptitude” and then the more PC “assessment” were determined to be damaging to self-esteem or racist or something.

    3. Or that a hundred years ago 30 points of IQ were being used for essential skills that aren’t tested for, while today people can survive who have almost no practical survival or problem solving skills and so can dedicate more IQ points to things that are represented on intelligence tests.

  11. Somewhat related, very interesting: Why Arabs Hate Reading

    Politically correct academics insist that one kind of writing is as easy to read as another, as Haeri does in the quotation above. Yet they frequently acknowledge, sometimes in the same breath, as Haeri also does, that readers of consonantal scripts like Arabic must figure out the message before they can read it. This forces them either to pretend, absurdly, that all messages are uniformly challenging?in other words, that grasping a passage from “Finnegans Wake” is no harder than grasping a shopping list?or, more often, to avoid the question altogether, as Olson and Haeri both do.

    Because if one message can be harder to grasp than another?as is clearly the case?then reading that message in consonantal writing like Arabic, which relies on readers grasping it first, is harder by definition. In practice, also by definition, such writing will not only discourage reading, it will also inevitably favor messages that are simple and familiar over those that are complex and challenging (which might explain why the yardstick of literacy in the Arab world is rote memorization of the Quran).

    In other words, consonantal writing by definition is a recipe for poor literacy.

    1. Reading is of this world, infidel.

    2. Thanks, papaya. Now I need to look up “consonantal” writing.

      1. There are no vowels so you have to infer the difference between “cat” and “cut” from context.

        1. Well a more apt description of “consonantal” writing might be “retarded” writing. I mean why not just spread feces on the wall and try to glean meanings from the context.

        2. Brett L|6.1.15 @ 3:39PM|#
          “There are no vowels so you have to infer the difference between “cat” and “cut” from context.”

          Sort of like Hebrew, no?

          1. Yes, the author mentions Hebrew but doesn’t bother to answer the obvious question of whether Israelis hate reading too.

            He also doesn’t mention that both Arabic and Hebrew writing are outfitted with perfectly usable marks to indicate vowels. The “sacred texts” omit them but I have seen them used in everday writing.

            1. Hebrew writing is a special case, a consonantal script for a dead language that was brought back to life by European Zionists for use in Israel, where alphabetic script is also commonly used.

              Emphasis added.

      2. It’s explained at the link, which is basically a medium-length blog post.

    3. I wonder if reading Chinese requires figuring out the message beforehand as well.

      1. If it’s anything like Japanese (which uses the same characters + hiragana and katakana), no.

      2. No, because there is no more ambiguity in written Chinese than in, say, English.

        1. The main difference is that you can’t “sound out” Chinese, as the characters are full words which combine to create other words. As such, literacy involves increasing lexicon, while learning phonemes is part of speaking. In English reading, literacy is built from the phoneme to the lexicon to the context.

          Japanese is a hybrid, having both the lexigrams (borrowed from Chinese), and the phonemic writing in the kana system. Literacy is much like English.

    4. SF’d link

      1. Should work fine.

    5. To call it “consonantal writing” is kind of misleading. It’s entirely unalphabetic. Arabic follows very different theory of writing than the alphabetic system. Spelling in Arabic is nothing like spelling in an alphabetic system. One thinks about the relation between writing and speech in a very different way. It’s hard to emphasise the difference enough. It’s not a small thing. Each word root is a series of three consonants. Specific applications of the underlying abstract concept are determined by varying the word-shape, which is to say the pattern of vowels and additional consonants stuck in between and around the three consonants of the root. The Asiatic writing systems are fit to this scheme and there’d actually be a loss in efficiency in writing an Asiatic language in an alphabetic system. The Asiatic systems also permit a fairly standardised writing system to be used for a single language which represents with equal fidelity many dialect variants. The problem with Arabic isn’t the writing system; it’s the language itself. I know many languages and Arabic offhand doesn’t strike me as particularly good at any angle of communication and in many respects seems decidedly inferior to other languages. There seems to be a sort of stultifying genius to it in general as well.

  12. Private Joe Bauers would like a word with you.

  13. I don’t know. People seem generally dumber than 100 years ago and much dumber than 200 years ago based on their writing and what they will believe.

    1. What dumb writing are you reading from the “200 years ago collection”?

      1. see what I mean?

    2. Jokes aside, only the smartest damn people knew how to read and write back in the day.

      Now with compulsory education, everyone gets taught. It’s like the driving analogy I made a few weeks back. If EVERYONE HAD to play lacrosse, you’d have a lot of shitty lacrosse players milling around the field, interfering with the few that took it seriously.

      1. That’s true for most of the world. For the English-speakers, 99% had basic reading skills by 1800. And this was before compulsory education. They probably wouldn’t be able to read a contract, but they could read “Beef – five cents a pound.” If you told them to write that phrase only 30% could probably do it.

        Thanks to compulsory education, over 70% of the English-speaking population can write “Beef – five cents a pound” and almost 95% can read it.

      2. Wouldn’t the measure be that for those who were educated way back when, versus those who are now, what would their letters to one another look like? Who were the deeper thinkers, the better philosophizers, and so on?

        If I read Solomon from 3,000 years ago, I realize he may have been the wisest man who ever lived. When I read Marcus Aurelius, Cicero, or Thomas Jefferson, I think that they exhibit intelligence far beyond our own. I think we modern humans confuse technology with intelligence.

        1. I think we modern humans confuse technology with intelligence.

          We do.

    3. I think that dumb people had a habit or dying young back in the old days. So, school children as a whole may have had lower scores, but a lot of idiots auto-darwinated by early adulthood.

    4. as evidenced by this post?

      You should go back and research the lead up to the Spanish American war if you really believe that. You’ll find that there is nothing new under the sun.

  14. Or maybe IQ tests are mostly bullshit.

    1. “Think, man, think! What does the clown who wrote this problem believe the correct answer is?”

      1. I see you went to college. “What does the professor believe? I’ll just right something that confirms his biases and get an A for sure.”

        1. *That’s* intelligence.

          More seriously, I meant this with regard to conventional IQ tests.

          For instance, sometimes more than once pattern could apply; and the take is in the position of choosing “the most likely” rather than “an arguably correct” answer.

          1. *one*

            *** gets coffee ***

        2. Which is a distinct departure from the Socratic school where philosophy was studied through argument and debate.

    2. Yep – I know plenty of guys who never took to school, but can tear down an engine, rebuild it, and plunk it into a car. Or my friend who dropped out of HS, drives a tow truck – makes more $$$ than I do – and is a world traveler, having been in all the best brothels of Asia.

    3. nah

  15. But Derp is expanding exponentially…

    1. But peak derp is still a myth.

  16. I’m no expert on IQ tests but wouldn’t 70 points be borderline retarded?

    1. Oh yes.

    2. That’s kinda where the wheels come off, for me.

      I just have a really hard time believing that the average person back in 1920 was borderline retarded by current standards.

      I can go with some marginal increases in intelligence due to better nutrition, but that conclusion makes me question the premises.

      1. Read that paper I linked below.

      2. I was only half joking with my iq inflation comment above. This seems to claiming that the thirty point rise has been calculated after any inflationary effects. 100 years isn’t like ancient history. Some of us might be and have been acquainted with people who were around back then. Did they seem borderline retarded?

  17. I wonder how present-day people would score on IQ tests from a century ago.
    We could probably dig some up and find out. Surely there are records.

    (Why is this even a hard question? Just go find some old IQ tests and take them… )

    Actually what I suspect is that IQ test answers tend to evolve. Certain kinds of knowledge start off in elite communities and gradually become mainstream. This is what causes the average test scores to rise. I.e. 70 years ago, hardly anyone knew what DNA was, now everyone does. What was considered elite knowledge 100 years ago is common knowledge today and vice versa.

    I wouldn’t be too shocked if people taking 100 year old IQ tests scored poorly on them because they would probably have all sorts of quesitons about things like pneumatic pumps and events in the Taft administration.

    1. IQ test aren’t tests of knowledge. Or at least shouldn’t be. They do sometimes require some cultural knowledge, though. E.g. the one question I remember from the one I took in second grade was a drawing of a man standing outside, in the morning or afternoon. They asked me what was wrong with the picture. I said: “His shadow is going the wrong way (pointing towards the sun).” They clicked the stopwatch and wrote down how long it took me to see that. (Basically instantly. *COUGH*) But that test wouldn’t work with, say, some pygmy just pulled out of the jungle who had never before seen a representational drawing with perspective.

      However, part of this increase might be that people are more used to taking tests of all sorts…?

    2. Not very well, but old IQ tests unfortunately tended to be mostly knowledge based.

      I found an old book with one in it when I was in the military and took the test. I only scored a 105, on modern IQ tests I typically score in the 130 – 140 range

      1. Check out the big brain on Brett Rasilio! You’re a smart muther fucker.

      2. Did you get a trophy for the modern test?

        1. Lol no

          Sadly the actual value of a high IQ is massively overrated.

          It is more akin to having a 14 inch dick, a nice conversation point but it really doesn’t help you out that much in the real world

    3. This is what causes the average test scores to rise. I.e. 70 years ago, hardly anyone knew what DNA was, now everyone does. What was considered elite knowledge 100 years ago is common knowledge today and vice versa.

      But the biggest gains were ‘fluid’ and the smallest were ‘crystallized.’

    4. I wouldn’t be too shocked if people taking 100 year old IQ tests scored poorly on them because they would probably have all sorts of quesitons about things like pneumatic pumps and events in the Taft administration.

      I distinctly remember the term youngblat meaning a hot dog in the same way a squib was a firecracker.

      Now, a squib is a wizard-born person who can’t do magic and nobody has any clue what a youngblat is (even test administrators).

  18. Ooohh baby, you are so talented. And they are so dumb.

    1. +1 Gosh darn it, Mr. Lemarr, you use your tongue prettier than a $20 whore

  19. 30 IQ points in 100 years. I think that is +3 sigma. I believe that +3 sigma is the 99.87th perventile, so it looks like Flynn is telling me that the average person today has more brainpowet than 99.87% of the population in 1915.

    Will someone please tell me how my interpretation of Flynn’s findings is wrong before I laugh myself into an aneurysm.

    1. Well, “percentile”.

      1. Perventile. Interesting.

    2. 2 sigma

      1. Thanks for that. I feel better now.

  20. How does this explain the performance of the Central Intelligence Agency over the past several decades?

    1. They’ve gotten us to accept being spied on… and even demand it.

      I’d say that was certainly a clever move, if not smart.

  21. So IQ scores have gone up while global temperature has gone up? There really is nothing that global warming can’t do.

  22. Average IQ Scores Have Risen 30 Points During the Past 100 Years

    You don’t have to thank me all at once.

  23. Have these always been written tests? If so, how could they possibly test the intelligence of the people who couldn’t read them? Illiteracy and stupidity are two different things.

    We’ve now reached the point in literacy where people can write but cannot read. For example, Congress allegedly writes several thousand pages of legislation but Congress hasn’t actually read any of it.

  24. The increasing scores aren’t surprising. IQ tests, which supposedely measure innate ability only, also measure a lot of confounders: education, test taking strategies, etc. and are unable to untangle where innate ability ends and the confounders begin. Better nutrition may also play a part but not nearly to the extent seen here.

  25. I’m a Brick! – Ralphie Wiggum

  26. There’s also this I’ve observed: the more times the person administering the test has administered it before, the easier it is for the subject to score higher. Not only do people administering IQ tests now have much more experience doing so than people administering them 100 years ago did, but I am certain that administration methods are much more standardised, and the subtlest differences in apparently superficial niceties may have a considerable effect on the subject’s performance.

    1. Of course, people back then were a bunch of dopes, so it may be as expected.

Please to post comments

Comments are closed.