# How is IQ caluated at the lower ends of the scale? **URL:** \ **Category:** Factual Questions\ **Created:** [December 22, 2009, 8:20am UTC](https://boards.straightdope.com/t/how-is-iq-caluated-at-the-lower-ends-of-the-scale/522181 "2009-12-22T08:20:48Z")\ **Posts on this page:** 15\ **Page:** 4
**Author:** ![Spectre\_of\_Pithecanthropus](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/spectre_of_pithecanthropus/32/12343_2.png) [@Spectre\_of\_Pithecanthropus](https://boards.straightdope.com/u/Spectre_of_Pithecanthropus)\ **Post date:** [December 28, 2009, 1:40am UTC](https://boards.straightdope.com/t/how-is-iq-caluated-at-the-lower-ends-of-the-scale/522181/61 "2009-12-28T01:40:39Z")
> [@constanze](#): > > Aren’t a lot of IQ scores calculated via the SAT tests that many Americans take, and where the scores can be looked up? Unless each US school in each state does really test each 4th grader with an official IQ test, and then release this confidential data to the Press, I don’t see how they get the scores for Presidents. But then, the US has a very different attitude towards data protection. To the best of my knowledge, Americans’ IQ scores and detailed academic histories are supposed to be kept confidential; for example your university can only confirm that you attended, and that you earned such and such degrees. I can’t believe that they would reveal the results of an intelligence test (which by no means all university students undergo). I’m not sure about SAT tests, but what you suggest is somewhat true about the GRE and SAT. A given total score (quantitative + verbal) is said to correlate fairly well to a given IQ range, and this is not too surprising since the test is essentially an evaluation of your ability to quickly acquire, comprehend, and apply information. Unlike a traditional IQ test, however, it is possible to significantly improve your performance on the GRE or SAT by concentrated study and practice. However, this improvement may not be terribly significant in IQ correlation. If your weak subject is math, you may be able to achieve a very gratifying improvement in your quantitative score, say on the order of 100 points. But when you look at the estimated IQ correlation, you’ll find the change is not that significant. The results of all these tests are supposed to be kept confidential too, of course, but when it comes to prominent people private information does have an unfortunate way of getting out. I can’t blame famous actors and other celebrities for such extravagances as private islands and walled estates, since our society affords them privacy in no other way. ---
**Author:** ![Spectre\_of\_Pithecanthropus](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/spectre_of_pithecanthropus/32/12343_2.png) [@Spectre\_of\_Pithecanthropus](https://boards.straightdope.com/u/Spectre_of_Pithecanthropus)\ **Post date:** [December 28, 2009, 1:42am UTC](https://boards.straightdope.com/t/how-is-iq-caluated-at-the-lower-ends-of-the-scale/522181/62 "2009-12-28T01:42:59Z")
ETA: I should mention that Mensa, the high I.Q. society, used to accept GRE scores above a certain level, up to 1996 or so. At that time the test was changed, and they stopped doing so. ---
**Author:** ![Spectre\_of\_Pithecanthropus](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/spectre_of_pithecanthropus/32/12343_2.png) [@Spectre\_of\_Pithecanthropus](https://boards.straightdope.com/u/Spectre_of_Pithecanthropus)\ **Post date:** [December 28, 2009, 3:18am UTC](https://boards.straightdope.com/t/how-is-iq-caluated-at-the-lower-ends-of-the-scale/522181/63 "2009-12-28T03:18:30Z")
> [@Spectre\_of\_Pithecanthropus](#): > > To the best of my knowledge, Americans’ IQ scores and detailed academic histories are supposed to be kept confidential; for example your university can only confirm that you attended, and that you earned such and such degrees. ETA: Of course, they will send a detailed transcript of your grades to someone–usually another school that you are applying to–but that only happens at your request. ---
**Author:** ![I\_Love\_Me\_Vol.I](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/i_love_me_vol.i/32/521_2.png) [@I\_Love\_Me\_Vol.I](https://boards.straightdope.com/u/I_Love_Me_Vol.I)\ **Post date:** [December 28, 2009, 5:15am UTC](https://boards.straightdope.com/t/how-is-iq-caluated-at-the-lower-ends-of-the-scale/522181/64 "2009-12-28T05:15:55Z")
> [@Chief\_Pedant](#): > > Actually, I think intelligent people do consider (formal) intelligence testing a good indicator of intelligence. Sure; there are variations on trying to define intelligence but if we’re talking g or “smart guy” or whatever, it’s a pretty good indicator. > > It’s a much poorer indicator for things like social skills, initiative, character, dependency…on and on…and those things (depending on circumstance) can be equally or more important in overall “success” in modern society. I agree with this–especially the second paragraph. It is a standard, unfunny, joke response of mine whenever these sorts of threads come up to merely answer something like: 198 11.5 These numbers are exaggerated, facetious claims to certain measurements that are often of great concern to many people. The truth is neither of my numbers are that high, yet they are rather large. I just don’t care to say what they actually are. Another truth, which I find rather odd, is that both of my measurements have been shrinking slightly over the last 10 years or so. However, in this post I’m only talking about IQ’s, and the fact that while mine is unexpectedly high (for such a spacey, and occasionally dim-witted person), I have realized more and more over time that there are so many so-called “intellectual abilities” that I’m simply not very good at. I don’t know what that means; fortunately, I don’t really care. ---
**Author:** ![I\_Love\_Me\_Vol.I](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/i_love_me_vol.i/32/521_2.png) [@I\_Love\_Me\_Vol.I](https://boards.straightdope.com/u/I_Love_Me_Vol.I)\ **Post date:** [December 28, 2009, 5:19am UTC](https://boards.straightdope.com/t/how-is-iq-caluated-at-the-lower-ends-of-the-scale/522181/65 "2009-12-28T05:19:01Z")
> [@Dinsdale](#): > > I believe intelligence is an imprecise term, and can legitimately encompass much more than what IQ tests measure. The same person can be a genius in some respects, and a fucking idiot in others. That’s all I meant. Yes. This is me. That is what I was talking about in my previous post. And it’s kind of a living Hell. I would prefer to be just one or the other… none of this confusing mish-mash. Blaggghhhh! ---
**Author:** ![Spectre\_of\_Pithecanthropus](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/spectre_of_pithecanthropus/32/12343_2.png) [@Spectre\_of\_Pithecanthropus](https://boards.straightdope.com/u/Spectre_of_Pithecanthropus)\ **Post date:** [December 29, 2009, 2:08am UTC](https://boards.straightdope.com/t/how-is-iq-caluated-at-the-lower-ends-of-the-scale/522181/66 "2009-12-29T02:08:23Z")
> [@I\_Love\_Me\_Vol.I](#): > > However, in this post I’m only talking about IQ’s, and the fact that while mine is unexpectedly high (for such a spacey, and occasionally dim-witted person), I have realized more and more over time that there are so many so-called “intellectual abilities” that I’m simply not very good at. It’s a reflection of the fact that intelligence is a poorly understood grab-bag of mental acuities. While I think it’s fair to say in general it’s the ability to receive and process information, and then produce a useful result, it’s obvious that there are many different facets of this. Two people might take the same IQ test and get the same score, but there can be lots of variations in how the correct answers were distributed. For instance, I never do well on the questions where you have to pick out a recognizable picture or geometric pattern from a patchwork-quilt-like jumble, but I do fairly well on the tests generally. ---
**Author:** ![Spectre\_of\_Pithecanthropus](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/spectre_of_pithecanthropus/32/12343_2.png) [@Spectre\_of\_Pithecanthropus](https://boards.straightdope.com/u/Spectre_of_Pithecanthropus)\ **Post date:** [December 29, 2009, 2:36am UTC](https://boards.straightdope.com/t/how-is-iq-caluated-at-the-lower-ends-of-the-scale/522181/67 "2009-12-29T02:36:59Z")
> [@Pyper](#): > > While that’s true (a person with an I.Q. of 50 will probably never master multiplication, for instance), I think people generally underestimate mentally retarded individuals’ ability to learn. I started working a teenager with moderate mental retardation last year. Since then he has learned how to make simple meals, bathe himself, make his own bed, do his own laundry, and operate the TV and DVD player. No one had ever made an effort to teach him these things before because they assumed it would be pointless. I read a newspaper article years ago about how some workers with IQs in the 50 range managed to hold down very responsible jobs like driving a truck. These folks managed to compensate for their inability to read in a variety of ways. I don’t remember the details, but I think above average social skills played a part. Most clinical definitions of retardation now combine the aspects of raw intelligence with social and adaptive skills, which makes me think that some of the people discussed would not have met the clinical definition. The article pointed out that, apart from any obvious physical symptom (e.g. Down syndrome), you could talk to an adult with an IQ in that range and not notice anything unusual for quite some time, at least if the conversation was just about lightweight, everyday stuff. ---
**Author:** ![AboutAsWeirdAsYouCanGet](https://avatars.discourse-cdn.com/v4/letter/a/22d042/32.png) [@AboutAsWeirdAsYouCanGet](https://boards.straightdope.com/u/AboutAsWeirdAsYouCanGet)\ **Post date:** [December 29, 2009, 5:48am UTC](https://boards.straightdope.com/t/how-is-iq-caluated-at-the-lower-ends-of-the-scale/522181/68 "2009-12-29T05:48:20Z")
> [@](#): > > Most clinical definitions of retardation now combine the aspects of raw intelligence with social and adaptive skills, which makes me think that some of the people discussed would not have met the clinical definition. The article pointed out that, apart from any obvious physical symptom (e.g. Down syndrome), you could talk to an adult with an IQ in that range and not notice anything unusual for quite some time, at least if the conversation was just about lightweight, everyday stuff. Yes. Meaning an inner city kid or a rural kid who is low IQ wouldn’t nessarily be MR unless they have adaptive skills defiects. It is possible to have adaptive skills defects with normal IQ as well. (people with cerebal palsy, Asperger’s, learning disabilites etc) ---
**Author:** ![Scruff](https://avatars.discourse-cdn.com/v4/letter/s/6f9a4e/32.png) [@Scruff](https://boards.straightdope.com/u/Scruff)\ **Post date:** [December 29, 2009, 6:20am UTC](https://boards.straightdope.com/t/how-is-iq-caluated-at-the-lower-ends-of-the-scale/522181/69 "2009-12-29T06:20:39Z")
> [@Spectre\_of\_Pithecanthropus](#): > > ETA: I should mention that Mensa, the high I.Q. society, used to accept GRE scores above a certain level, up to 1996 or so. At that time the test was changed, and they stopped doing so. Dude, if you can get 1996 on your GRE, you deserve to be in Mensa.😛 ---
**Author:** ![The\_Bith\_Shuffle](https://avatars.discourse-cdn.com/v4/letter/t/ce73a5/32.png) [@The\_Bith\_Shuffle](https://boards.straightdope.com/u/The_Bith_Shuffle)\ **Post date:** [December 29, 2009, 6:45am UTC](https://boards.straightdope.com/t/how-is-iq-caluated-at-the-lower-ends-of-the-scale/522181/70 "2009-12-29T06:45:01Z")
One thing I’ve been curious about for a long time is the relationship between IQ testing, the normal distribution, and randomness. You see, the normal distribution is meant to model the distribution of a quantity that depends on a variety of random factors. Since psychologists tend to think that intelligence is the result of a wide array of genetic factors (rather than just a few genes), they figure this explains the normal distribution of intelligence. But since the normal distribution is the distribution of quantities depending on various random factors, it also is the distribution of scores based on a bunch of random guesses. If I asked twenty thousand people to answer A or B 500 times (and only A or B was right or wrong each time), the distribution of how many correct answers a person got would also be a normal distribution. Someone who scores 150 on an IQ test is then very unlikely to be of average intelligence. On the other hand, if a bunch of average-IQ people take IQ tests and are forced to guess on many problems, there will be a small number who get impressive IQ scores through luck. I don’t know exactly how the relationship between random-guessing and actual intelligence affects the study of intelligence and the use of IQ scores, but I imagine that there could be multiple significant effects. Or maybe there could be no effects at all; I could just be clueless. ---
**Author:** ![Kelby](https://avatars.discourse-cdn.com/v4/letter/k/a587f6/32.png) [@Kelby](https://boards.straightdope.com/u/Kelby)\ **Post date:** [December 29, 2009, 7:02pm UTC](https://boards.straightdope.com/t/how-is-iq-caluated-at-the-lower-ends-of-the-scale/522181/71 "2009-12-29T19:02:00Z")
> [@Spectre\_of\_Pithecanthropus](#): > > I read a newspaper article years ago about how some workers with IQs in the 50 range managed to hold down very responsible jobs like driving a truck. These folks managed to compensate for their inability to read in a variety of ways. I don’t remember the details, but I think above average social skills played a part. Most clinical definitions of retardation now combine the aspects of raw intelligence with social and adaptive skills, which makes me think that some of the people discussed would not have met the clinical definition. The article pointed out that, apart from any obvious physical symptom (e.g. Down syndrome), you could talk to an adult with an IQ in that range and not notice anything unusual for quite some time, at least if the conversation was just about lightweight, everyday stuff. I’d like to see that article. The individuals I’ve encountered who are functioning the 50’s IQ range are almost always quickly distinguishable from normal peers. Their expressive speech would be clearly lacking in content and complexity. I’m also not aware of anyone that low functioning ever obtaining a driver’s license. In fact, most of these folks are considred disabled and receive SSI benefits. They are employable, but usually in assembly-line type jobs that are highly structured and closely supervised. ---
**Author:** ![Kelby](https://avatars.discourse-cdn.com/v4/letter/k/a587f6/32.png) [@Kelby](https://boards.straightdope.com/u/Kelby)\ **Post date:** [December 29, 2009, 7:09pm UTC](https://boards.straightdope.com/t/how-is-iq-caluated-at-the-lower-ends-of-the-scale/522181/72 "2009-12-29T19:09:26Z")
> [@The\_Bith\_Shuffle](#): > > One thing I’ve been curious about for a long time is the relationship between IQ testing, the normal distribution, and randomness. You see, the normal distribution is meant to model the distribution of a quantity that depends on a variety of random factors. Since psychologists tend to think that intelligence is the result of a wide array of genetic factors (rather than just a few genes), they figure this explains the normal distribution of intelligence. > > But since the normal distribution is the distribution of quantities depending on various random factors, it also is the distribution of scores based on a bunch of random guesses. If I asked twenty thousand people to answer A or B 500 times (and only A or B was right or wrong each time), the distribution of how many correct answers a person got would also be a normal distribution. > > Someone who scores 150 on an IQ test is then very unlikely to be of average intelligence. **On the other hand, if a bunch of average-IQ people take IQ tests and are forced to guess on many problems, there will be a small number who get impressive IQ scores through luck.** > I don’t know exactly how the relationship between random-guessing and actual intelligence affects the study of intelligence and the use of IQ scores, but I imagine that there could be multiple significant effects. Or maybe there could be no effects at all; I could just be clueless. There are very few subtests on an individually administered scale that can be influenced by luck. Most tasks require a constructed response. ---
**Author:** ![Spectre\_of\_Pithecanthropus](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/spectre_of_pithecanthropus/32/12343_2.png) [@Spectre\_of\_Pithecanthropus](https://boards.straightdope.com/u/Spectre_of_Pithecanthropus)\ **Post date:** [December 29, 2009, 7:35pm UTC](https://boards.straightdope.com/t/how-is-iq-caluated-at-the-lower-ends-of-the-scale/522181/73 "2009-12-29T19:35:14Z")
> [@The\_Bith\_Shuffle](#): > > I don’t know exactly how the relationship between random-guessing and actual intelligence affects the study of intelligence and the use of IQ scores, but I imagine that there could be multiple significant effects. Or maybe there could be no effects at all; I could just be clueless. For the GRE and SAT there’s a wrong-answer penalty of, IIRC, a quarter point; this should theoretically neutralize any productive value in guessing. I don’t know if the same is true of typical IQ tests, but I would assume so. Conceivably, by sheer luck one could still guess most of the answers correctly and achieve a high score, but here I think we’re veering into the realm of typewriters, monkeys, and Shakespeare. ---
**Author:** ![Irishman](https://avatars.discourse-cdn.com/v4/letter/i/b487fb/32.png) [@Irishman](https://boards.straightdope.com/u/Irishman)\ **Post date:** [December 30, 2009, 12:31am UTC](https://boards.straightdope.com/t/how-is-iq-caluated-at-the-lower-ends-of-the-scale/522181/74 "2009-12-30T00:31:01Z")
**Mr. Excellent** said: > [@](#): > > With respect, this isn’t so. Most modern IQ tests assess memory, sure - but they also assess analytical ability. (One very basic way they do this, though far from the only one, is through analogies - “Black is to white as up is to \_\_\_”, and so forth.) I can’t remember what the opposite of up is, but I know it’s that way / ! 😉 **electronbee** said: > [@](#): > > Hasn’t everyone in the US been tested in elementary school? I remember my classmates and I taking tests back in the 80’s. And, IIRC, my HS psych teacher told me we could go to the DOE once we were over 18 and get the result. There’s a fair amount of standardized testing (and moreso now than when I was a munchkin), but I don’t recall IQ being one of those. But I might just not remember. **Silverstreak Wonder** said: > [@](#): > > Excellent, well my test was long ago, but even then I do recall those awful anal-things you mention and tried after to find out how in the devil does one study for them? They would be fine if there was one correct answer, but in almost every one there are like 3 of 4 right answers and it is just someones opinion which one is “Better”. Ever think that maybe those did have just one correct answer, that those were the questions that you weren’t smart enough to get? Seriously, how do they sort the 140 IQ from the 160 IQ if they don’t have questions at the top end? If they don’t have questions that even the 160 folk can’t get right? > [@](#): > > So I have always called those opinion questions. Can anyone explain why they would have many correct answers? I think maybe you missed something in the question. > [@](#): > > How does one study for this test, is there some trick to it? Get smarter. 😉 Seriously, I bought one of those IQ test books and took several of them. My score did slightly improve from the first to the last, largely because of familiarity with a couple of the puzzle types. The first time it was puzzling, but I studied it to figure how it worked, and on subsequent tests I was more easily able to pick out the pattern from that kind of puzzle because I knew how it was put together and how the first one worked. But my score didn’t dramatically improve across the set - something like 125 to 132 across 5 or 6 tests. And IIRC there were one or two puzzles I couldn’t figure out even with the correct answer provided. One can “study” by encountering lots of different puzzles and familiarizing oneself with the way the puzzles are built, with the kinds of patterns that are commonly used, etc. Practice improves your abilities, whether it’s having a large vocabulary (which helps dramatically on the SAT), or studying spatial relationships so one can visually “move” a 3-D image in your mind and keep track of the faces. Learn exotic number sequences, such as prime numbers, Fibonacci, whatever. Ultimately, though, the tests are timed to put additional pressure and measure how fast you recognize the patterns, not just that you do figure them out. **Silverstreak Wonder** said: > [@](#): > > In the examples on the test there was ALWAYS just one correct answer and 3 or 4 bad answers I’d never pick. On the test there would almost always be 2 or 3 correct answers that fit, so do you pick the largest one as “Better” or what? And just who decides what is the best answer anyway? I think a true test would involve writing a sentence or two to explain the choice, and not having multiple right answers and someones opinion. Then the persons actual logic could be seen and judged. A better way could be that the answers would have explanations and then you’d pick the one with the clearest too, there are many ways to improve that kind of a test. But that’s the point. To a 160 IQ person (or maybe the hypothetical 180 IQ person), the actual answer is just as obvious as the simple example questions are. If you’re not able to deduce the logic, then the test is doing its job - filtering the high level performers. If they provided the explanations and you picked from those, they wouldn’t be testing your cognitive ability, they would be testing your ability to judge someone else’s reasoning. Those questions serve exactly the purpose they’re supposed to. Just because _you_ can’t figure out the right answer doesn’t mean there isn’t a right answer. And as for why they don’t tell you the answer, if they told you what the reasoning was (after the fact), then that kind of puzzle would likely cease to be useful. Because once you see the reasoning used in one, it is easier to duplicate that reasoning than to create that reasoning for yourself. **constanze** said: > [@](#): > > To those who have answered here knowing about IQ: My impression was that the kind of intelligence that a high IQ measured was the ability to see patterns quicker than normal people. Or is this a misconception of Hollywood? Because that (and not photographic memory) is what a lot of “genius” people often seem to display - they say “But don’t you see that ------?” to an ordinary person, who only does after explantions. That may be Hollywood shorthand for how to convey someone is a genius (he figures out these complicated things and does so quickly - the “Elementary, Watson” approach to conveying intelligence). But I think that is actually a reasonable depiction of what IQ tests are testing. The cognitive abilities being addressed are various forms of reasoning: logical (deduction and induction), spatial (this shape looks like this from one side, what does it look like from the other?), comparison and contrast (analogies, this is to this as that is to \_\_, etc). A genius is someone who can rapidly spot whatever pattern is in question. It’s not just the ability to recognize the pattern, but the ease or speed with which it occurs. **constanze** said: > [@](#): > > I don’t see how they get the scores for Presidents. Those scores have to be determine through some type of evaluation process from their extant writings and historical record. Here is an interesting article. > **[Presidential IQ's](https://cseweb.ucsd.edu//~gary/iq.html)** Note that there was an interent urban legend circulating about Presidential IQs that was a hoax, as cited in that article and on snopes. > **[Presidential IQ Hoax](https://www.snopes.com/fact-check/presidential-iq/)** > > According to a study by the Lovenstein Institute, President Bush has the lowest IQ of all presidents of past 50 years. The first site above includes a table of computed scores for the first 42 Presidents. Interesting that Jefferson scores highest (3.1 on “Intellectual Brilliance”, with Kennedy in #2 at 1.8), while Harding scores lowest at -2.0 His IQ scores still fall 107.8 to 139.9, while Jefferson’s are 145.5 to 160.0. Make of that what you will. ---
**Author:** ![hibernicus](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/hibernicus/32/2802_2.png) [@hibernicus](https://boards.straightdope.com/u/hibernicus)\ **Post date:** [December 30, 2009, 6:45pm UTC](https://boards.straightdope.com/t/how-is-iq-caluated-at-the-lower-ends-of-the-scale/522181/75 "2009-12-30T18:45:50Z")
> [@Silverstreak\_Wonder](#): > > Dinsdale and others, I can tell you why someone who may be all wrong about views and their thinking can test so high in IQ tests, the reason is they are not just testing intelligence at all, but MEMORY. Remember he said he need not take notes in class to do well? That isn’t being smart at all, it is having a photographic memory. > > Yes, you can make a living and seem smart with just that. There are some jobs that is all you need in fact, teaching, foreign language interpreter being a couple of good examples. Absolutely not. If memory were all that were required to be an interpreter, machines would be able to translate effectively between languages. They are not. Your remark suggests that you seriously underestimate the differences between languages, and the skills of the interpreter. [Previous page](https://boards.straightdope.com/t/how-is-iq-caluated-at-the-lower-ends-of-the-scale/522181.md?page=3)