Why are IQ scores so asymmetric?

Factual Questions
21

Sure, but where did you get that number specifically? And someone who is tested at at IQ of 101 is not different in any meaningful way than someone with 100. 68% is 85-115 (or 84-116), all that range is 18%?

Yeah, there is not reason that IQ can’t theoretically be negative. The problem is that way out there you are dealing with 0.00000etc001% of the population. If it so happens that 1 out of nearly 7 billion people has this IQ (and is still alive), who would you compare them to? Extremely low or high IQs are difficult to compare. Marilyn vos Savant claims a bullshit number, 8.53 standard deviations over the mean (again. 99.99999eleventybillion99th percentile). It’s based upon a scaled from childhood score, and I don’t think she’s taken or published a updated figure.

Also, for someone who is nonverbal, you cannot give them tests like the WAIS. The assumption is that the nonverbal tests are roughly equivalent, but you have to convert so it’s not exact anyway.

As not seeing these extremes in public: the very low IQ people will be instituionalized so you won’t see them. The very “high IQ” people have all escaped to the Internet, where it seems 90% of the people are IQ 140 or greater! So all of you have an IQ that is 0.39%. What are the odds!? :smack:

And that Honey Boo Boo show looks atrocious. But don’t assume that a very smart person won’t enjoy it.

22

No, that’s not true, elfkin477. If you were to tell me that, say, 2% or 4% of the population will get a score of exactly 100 on an I.Q. test (so that it would then be true that 49% or 48% are smarter than you and 49% or 48% are less intelligent than you), I could barely believe that, although I suspect that most I.Q. tests distinguish I.Q.'s better than that. I simply don’t believe that there is any I.Q. test so sloppily designed that 50% of the people being tested get scores of 100, leaving only 25% to score better than you and 50% to score worse than you. Give me a citation showing that there is any such I.Q. test.

23

From DSM-IV, courtesy of Wikipedia:

So even at 71, a particularly well adapted individual might not initially strike the observer as mentally disabled. To take an example at random, this person might have difficulty in calculating the change due when tendering $5 for a purchase costing $3.24. Yet the next person in line, with an IQ of 129, might very well not bother work it out either, since it may not be that significant.

It almost seems as if you’d need to draw the person out in order to make a judgment; just ordinary small talk won’t do the job.

24

thelurkinghorror, Marilyn vos Savant was tested back in the 1950’s when she was a child. There were still some people back then who continued to use the older quotient definition of I.Q. instead of the new standard deviation definition of I.Q. She says that she tested at 228 I.Q. The old definition of I.Q. meant that you divided your intellectual age by your chronological age and multiplied by 100. If at age 7 she tested as being as smart as an average 16-year-old, her I.Q. using the old definition would indeed have been 228. But, as you say, using the new definition of the I.Q. that everyone uses today, a 228 I.Q. is simply absurd.

Although it’s theoretically possible for an I.Q. to be above 200 or below 0, there simply haven’t been enough people in the history of mankind to make such a statement about someone’s I.Q. An I.Q. of 0 is 6 and 2/3 standard deviations below the average. An I.Q. of 200 is 6 and 2/3 standard deviations above the average. There have only been something like 100 billion people ever born. I calculated it once and found that this would mean that, comparing every person ever born, the lowest I.Q. would be somewhere between 0 and 5 and the highest I.Q. would be somewhere between 195 and 200. (It’s possible to be more exact, but I don’t remember the precise numbers.) So it’s true that given the present number of humans through all history, there can’t be any negative I.Q.'s or I.Q.'s above 200.

25

No, I know that about why it’s so high, but thanks for the info on the old test, and her situation. I teach psychology classes sometimes. I taught about IQ (briefly, not a 101/psychometric class) and asked about it on a test. A good portion of the students defined it as (mental age/chronological age) despite that. What are 101 teachers doing?

26

In post #22 I wrote:

> I simply don’t believe that there is any I.Q. test so sloppily designed that 50%
> of the people being tested get scores of 100, leaving only 25% to score better
> than you and 50% to score worse than you.

I meant:

> I simply don’t believe that there is any I.Q. test so sloppily designed that 50%
> of the people being tested get scores of 100, leaving only 25% to score better
> than you and 25% to score worse than you.

In post #17 I wrote:

> If you have an I.Q. of 115, you are one standard deviation greater than the
> average. You can look up what standard deviation means, but basically it
> means that you are more intelligent than about two-thirds of the population.

I should have written:

> If you have an I.Q. of 115, you are one standard deviation greater than the
> average. You can look up what standard deviation means, but basically it
> means that you are more intelligent than about 84% of the population.

I apologize for getting that one wrong. What I was badly remembering was that about two-thirds of any ordered set falls between one standard deviation below the average and one standard deviation above the average. This means that about one-sixth (about 16%) of the population is above one standard deviation above the average and about one-sixth (about 16%) is below one standard deviation below the average. So if you have an I.Q. of 115, you are more intelligent than about 84% of the population.

Going back to purple cow’s OP:

> For example, a person with a 125 IQ is smart but i don’t think noticeably so in
> an everyday conversation. But a person with a 75 IQ is in the realm of Forest
> Gump, slow, unable to grasp complex ideas, etc. Go out to 150 and the person
> is very smart, probably nocticeably so even in a short span of time. But at 50
> the person has trouble taking care of him or herself.

I think this is how a person with an I.Q. of 125 looks at the world, especially if they can manage to spend most of their time with other people of around that I.Q. To them a person with an I.Q. of 125 isn’t really noticeably different from other people. A person with an I.Q. of 100 is a little slower, but they will have difficulty at times grasping some more difficult ideas that the person of I.Q. 125 grasps. A person with an I.Q. of 75 will seem impossibly slow and annoying. The person with an I.Q. of 125 will have trouble interacting with them at all and will try to ignore them as much as possible. A person with an I.Q. of 150 will be clearly smarter, but the person with an I.Q. of 125 will generally enjoy interacting with him, as long as the person with the 150 I.Q. doesn’t seem to be bored.

On the other hand, the person with an I.Q. of 150 will spend most of his time interacting with people with I.Q.'s no higher than 125. He will occasionally get tired of his interactions with someone with an I.Q. of 125. He realizes that the person of I.Q. 125 is smarter than average, but he tires of that person thinking that he is as smart as he is. It’s easier actually to interact with someone of I.Q. 100. That person doesn’t think of himself as anything but average. He doesn’t feel the need to pretend that he is almost as smart as the person with an I.Q. of 150. The person with an I.Q. of 150 will possibly enjoy interacting with someone with an I.Q. of 75. Here, at last, is someone who makes no pretense of being smart. They can be nice to each other and not try to compare themselves.

27

Is there any sort of rough equivalence between the old and new system, based on current demograhics? On an unofficial test I got a 147, and I dismissed it as obviously inaccurate, but I wonder if would be more reasonable in the old system. I was in my early teens when I took the test.

28

I had the same sorta problem. Gotta convert centimeters to inches.:smiley:

29

As I understand, the Flynn effect - a long-term rise in overall IQ scores - is real. So while the scores may have at one time been centered about 100, they no longer are. So, to the OP, we would expect to encounter more people with an IQ of 100 + x than 100 - x.

One cause postulated for the Flynn effect is people’s increased familiarity today with testing environments.

So when you meet a person with an IQ of 100, their counterpart 80 years ago might have been assigned a score of 85 or 90. They were just as capable of functioning in society as a person today with a 100 IQ , but the fact that they’d never seen a puzzle or a word game made them look on paper like Forrest Gump today.

30

I’m not sure what you mean by “the old system and new system.” If you’re talking about the quotient way of calculating I.Q. and the standard deviation way of calculating I.Q., no, there’s no easy equivalence between them, unless you count the fact that an average I.Q. on both of them is 100. When did you take this test that you scored 147 on? It’s unlikely that you took a test where the score was calculated on the quotient system rather than on the standard deviation system unless you’re a lot older than I expect. Marilyn vos Savant is probably older than you think and she only took a test using the quotient system when she was a child.

Part of the problem with trying to find some equivalence between the old system and the new system is that you expect to see changes in the tested I.Q. as a child grows up in the old system. In the new system, on the other hand, the I.Q. is generally supposed to remain the same as a child grows up. (Whether I.Q. truly stays the same throughout one’s life is actually questionable, but the new system more or less assumes that it does.) In the new system, if you’re better than 999 out of 1,000 other 10-year-olds when you’re 10 but not better than 9,999 out of 10,000 other 10-year-olds at that age, it’s expected that you’ll be better than 999 out of 1,000 other 15-year-olds when you’re 12 but not better than 9,999 out of 10,000 other 15-year-olds at that age. In the old system, if you test at intellectual age 15 when you’re 10 years old (which gives you an I.Q. of 150 at that age), it’s not expected that you’ll necessarily test at an intellectual age of 18 when you’re 12 years old (which is also an I.Q. of 150). There’s no requirement in the old system that the proportion of people with any particular I.Q. score remains the same over time.

To be more specific, in the new system it’s not quite assumed that I.Q.'s don’t change over time as a child ages. To be exact, it’s assumed that the proportion of people with any given I.Q. (in a given age group) remains the same over time as that group grows up. The requirement is actually that the distribution of I.Q.'s on a normal curve remains the same over time. It’s theoretically possible that I.Q.'s of people could simply jump up and down randomly on the normal curve as they age. This wouldn’t violate the conditions of the new system. It would merely make the idea of I.Q. pointless.

31

To be precise, although the Flynn effect shows that people are smarter (or at least do better on I.Q. tests) today than they did before, it doesn’t mean that the average I.Q. has risen. What the Flynn effect says is that if a group of, say, 10-year-olds take an I.Q. test today that was given to a group of 10-year-olds ten years ago, you’ll expect that kids today will score an average of 103 instead of am average of 100 as those ten years ago did. Their scores on that test will not be generally quoted as averaging at 103 though. Their scores on that ten-year-old test will have three points subtracted from them, so the average will still be 100. The rule is that the average always stays at 100 no matter what people actually do on the test. In fact, it’s unlikely that the children will be given an old test. A new test will be created and the scores will be calculated so that the average is 100.

32

BigT, what was this unofficial test where you got a score of 147? Was it on the Internet? Was it in some magazine? Most of those unofficial I.Q. tests are hopelessly inaccurate. They often rate reasonably smart people quite a bit higher than they really are. They like to make people feel good about themselves, and assigning them an I…Q. that’s higher than their true one is one easy way to do it, since most people have no easy way to check their I.Q. You may indeed have a 147 I.Q., or you may have fallen for one of these worthless tests.

33

When I was a kid, I’d play the vidya game ‘Leisure Suit Larry’.

The game had a sort of adult only protection scheme that involved the player volunteering their age, and then answering a series of questions based on that age.

Needless to say at age 12 I didnt know much about the culture of adults from age 18-30. It was much easier answering questions of my parents ages.

Easiest was to just enter age 99(the oldest the game permitted). Eighty year old adult trivia had permeated even the society of children.

34

Trivia contests are not I.Q. tests. They are tests of memory. Sometimes they are tests of quick recall.

35

Yeah, but if you play today, you’ll have to answer questions about Lee Majors and Iran-Contra. I remember just memorizing the answers as a kid.

Or until I realized I could just push Alt-X to skip it.

36

That wasn’t quite the point I was making. Cultural exposure and experience trump aptitude here. You could teach a caveman to drive a car. Profoundly illiterate people can learn traffic signs.

Just for example: “What year was the Empire State Building erected?” would have four dates. “Who was Cher married to?” Sonny Bono, Frank Zappa, Marvin Gaye, John Lennon?

As you can imagine, Al Lowe probably knew as much pertaining to the marital status of Ella Fitzgerald as 12 year old me knew about Cher. The older questions were less sophisticated.

But anyway, years ago I took some aptitude tests and my immediate thought was: I could have studied for this. Parts of it anyway. Most of it even. There is an element of memory in testing.

I had an anomaly in my results. The math portion I did very poorly at, and my English skills were significantly better, such that they had me take another test, a non verbal one. The Raven test or some variation.

I realized I could have studied for even that. While the more advanced pattern matching had many permutations, I could have ran drills to prepare. In fact, my first thought when sitting down was this: “There is probably only a limited number of variations in print.”

Anyway, I felt pretty fuzzy headed, and thought I did poorly. I’ve recently figured out I should be avoiding gluten products, and I’m feeling much more sharp these days. I’d like to try again.

In the end though, I did fairly well, and received some sort of derived IQ score of 119, plus or minus 10 points.

37

FuzzyOgre, I’m still not sure why you brought this up in reply to my post about the Flynn effect. I had presumed that you were trying to say that someone your age at that point (12 years old) would, on average, have an I.Q. which was 87 * (3/10) = 26.1 points higher than someone who was 99 years old. The Flynn effect is about I.Q., not about memory, quick recall, cultural exposure, or experience. The points you make in your second post may be true, but I don’t understand why they’re relevant to what I wrote.

38

IQ is, in theory, an ordinal/ranking measurement and an IQ point doesn’t correspond with any “quantum of smartness”. For example, if the stupidest person who ever lived suffers brain damage and gets stupider, his IQ would stay the same because his position on the (theoretic) IQ chart represents the fact that he’s the stupidest person and not exactly how stupid that is.

39

And one of the big questions of our time is whether or not the Flynn Effect means that humanity is getting smarter on average. I’ve wondered how well I would do studying at Oxford in 1350 if I had a time machine that I could use right now. Would the Flynn Effect mean that I would be worlds ahead of my fellow students and would run all over them and make top of the class because I’m just so dazzlingly smart based on the standards of the time and can pick up Greek conjugations with only five hours of studying while my fellow medievally-raised students need 50 hours and still get worse grades than me?

40

You’d definitely know it there, although it would be hard to quantify. I mean, there are people I’ve worked with in the past who, while otherwise intelligent and capable, didn’t have that extra spark or insight that the smarter ones had.

And there have been the people who were on the other side of that divide- they’re the ones who immediately pick up on what I am explaining, and it’s implications, even though it might not have been immediately obvious to me, and took some work to puzzle out.

I think professionally, someone’s intelligence is usually displayed by how they comprehend not only the subject at hand, but any implications and expected next steps that logically follow from it.