June 2nd, 2003
|10:59 pm - Smart bomb|
I estimate that the average IQ of the people on my Friends list is about 125.
People say that IQ is only a measure of how well people can answer intelligence tests. Indeed, I'm not suggesting that if people were to take a commercially-available test tomorrow, people would necessarily score a particular number. I'm envisioning a score based on a hypothetical ideal intelligence test, which tested not just the traditional measures such as logic and inference but all facets of behaviour that is generally considered clever, smart, brilliant, capable or genius-like by modern society. I also envision that this test would take about two weeks to administer, all participants would get (say) eight weeks of practice beforehand and some suitable motivator would be applied (maybe a different one for each person...?) so people really did perform to the best of their ability. Definitely a hypothetical ideal test.
It's also worth pointing out that the figure of 125 is based on a standard deviation of 16 points, which is the standard measure used in the US and on the BBC's televised tests. (UK Mensa traditionally quote figures based on a standard deviation of 24, so the average there would turn out to be 135+.) This means that I think the average person on my Friends list is about one and a half standard deviations above average, or brighter than about 93% of society in general. You're not "one in a million", but you are pretty literally "one in fifteen". I also use "average" very generally; the arithmetic mean and median shouldn't differ much, though I'd guess the mean might be a shade higher than the median.
What does this mean in practice? It means that (almost?) all of you are smarter than most, most of you are more intelligent than pretty considerable swathes of the population, a sizeable minority of you are brighter than everyone they'll meet on the streets most days of the week and there are a few scary (but nice) freaks of nature as outliers who I tentatively estimate at somewhere over IQ 155 - a good three and a half sigmas above the mean, or at least 1-in-4,000 rarity. You know who you are; keep quiet or I'll be forced to explain just why I think you've demonstrated you're so clever. ;-)
Happy 0th birthday to Jonathon Avi Tandy and many congratulations to his wonderful mother, heidi8.
Current Mood: i am so smart, s-m-r-t
Current Music: "MegaMan Elecman OC ReMix" which reminds me of Space Harrier
If Test the Nation is anything to go by, my IQ has jumped 24 points in the last year. If that feat is repeated, in three years time I will have an ABSOLUTELY MASSIVE IQ of 202. Blimey, eh?
Yes, but the way TTN works, you will need to be aged 146 and score 85/70 in order to rack up your 202. You scoring 85/70 I can just about imagine, but you ain't aged 143 right now, so something's got to give along the way.
Interesting post. How did you form the basis of your numbers? Is it gut intuition? Or are you using some sort of reference point?
When I was relatively young, my parents had me take an IQ test and I scored pretty well. IQ is supposed to measure your intelligence relative to others of your age, correct? If that's the case, I suspect that my IQ now would be siginificantly lower than it was then. I'm getting dumber. :-)
How did you form the basis of your numbers? Is it gut intuition? Or are you using some sort of reference point?
Two parts of the former (the middle, I suppose) and one part of the latter. This generic chart
helps a lot - I thought in terms of people being "1 in 4", "1 in 6", "1 in 15", "1 in 50", "1 in 1000" and so forth and looked up how many standard deviations above the mean that made them.
As to how I decided whether someone was "1 in 6" or "1 in 60", a large part of that was based on perceived academic achievements, but very much shunted up or down based on how I regard them and how I think they regard themselves based on what they've said in their LJ because academic achievements' correlation with intelligence, while positive, is some way from 1. For British folk, I can get a starting-point for the order of magnitude of figures based on the proportion of people who go to university and the universities they go to. For overseas folk, it's harder, but I can try to compare people in my mind.IQ is supposed to measure your intelligence relative to others of your age, correct?
Indeed so.If that's the case, I suspect that my IQ now would be siginificantly lower than it was then. I'm getting dumber. :-)
So am I. :-) Seriously! (And this isn't just because I had to look up how to do almost all of glissando
's A-level maths questions and still wasn't able to do all of them...)
IQ is supposed to measure your intelligence relative to others of your age, correct?
Oh. Bugger. There goes that impressive number I got when I was 11... It's the good old "thick as the large print edition of the complete works of Charles Dickens" category for me now - I'm sure I'm getting dumber too. Either there's something in the water or it's a by-product of
university the student union... ;-)
Well, Binet's original definition of IQ was (mental age) / (chronological age) * 100, so someone with a mental age of 17 and a chronological age of 11 would come out with a frankly scary IQ of 154. The scores made on Binet Scales and most similar tests are stated in terms of mental age (M. A.) when a child is described as having a mental age of 8, he or she is able to solve the same test problems as average 8-year-old children. This would suggest to the parent as well as teacher that the child is able to keep in learning with average 8-year-olds, even though he or she might actually be younger or older than 8.
This tends to break down once people pass a certain age, though, because while average 17-year-olds might be able to solve harder problems than average 16-year-olds, it doesn't follow that average 20-year-olds should necessarily be able to solve harder problems than average 19-year-olds. (Or does it? I'm not so clear here. Where and when do people tend to reach their potential? Do they ever?)
More about mental age
is available from the Google cache.
Testing people's IQs, especially at the impressive levels, is quite a controversial subject. This is one of many articles
on testing people who are many standard deviations above average; there's more at the Uncommonly Difficult IQ Tests
site, alongside much about the scary politics of ultra-high-IQ societies and their members. The old saw about "the lower the stakes, the more insidious the politics" rings true here.
One name which crops up in terms of ultra-high-IQ is that of Richard Rosner. Yes, that Richard Rosner
. Not that he's a bad loser
, or anything.
Going back to ultra-high IQs; scarily, I can think of a few people for whom this could
be an issue - though, happily, I don't think any of them want it to be one. I am very
happy that I may be the only person on my Friends list who cares about this. :-D
IQ is supposed to measure your intelligence relative to others of your age, correct
Yes, but if you were to read (and believe) Herrnstein and Murray's work then the answer is not exactly. IQ tests (according to them) measure G which is a persons inate intelligence/potential. Meaning that although the tests are different depending on your age because of what the tests measure there should be little difference between your score when you are 5 versus when you are 55.
"Here are six conclusions regarding tests of cognative ability, drawn from the classical tradition, that are by now beyond significant technical dispute:
-- (Herrnstein/Murray 1996 page 22-23)
- There is such a thing as a general factor of cognitive ability on which human beings differ
- All standardized tests of academic aptitude or achievement measure this general factor to some degree, but IQ tests expressly designed for that purpose measure it most accurately
- IQ Scores match, to a first degree, whatever it is that people mean when they use the work intelligent or smart in ordinary language.
- IQ scores are stable, although not perfectly so, over much of a person's life.
- Properly administered IQ tests are not demonstrably biased against social, economic, ethnic, or racial groups.
- Cognative ability is substantiall heritable, apparently no less than 40 percent and no more than 80%."
Now I think that this is psuedo science at its best/worst, but that is another discussion altogether. doooook
would love their work.
On a very vaguely related note, d00d. Birthdays count the UNIX (*cough cough right cough*) way! I never realized it until now, but yes, the baby's day of birth, their first birthday, is their 0th. cool.
Jonathon will be a whole milliyear old by now :-)
I was given an IQ test when I was ickle, and even my parents don't know what the results were. They said they didn't want to know. Interesting, I think.
Haha, I think it'd be funny to make you explain. It could be the new meme. :p
And like songmonk
, I definitely think I've gotten dumber. Mostly since going to uni, I believe.
|Date:||June 2nd, 2003 07:00 pm (UTC)|| |
Heh. I feel the same way about University. It zapped the majority of my creative impulses and made me quite lazy. On the bright side, it (indirectly, of course) introduced me to poker, which is a good thing (I think?)
Definitely a good thing. Uni introduced me to other people who are up very late, and to late-night instant messenging. And we see where that's gotten me...
Mmmm, insomnia. :D
Haha, I think it'd be funny to make you explain.
I'm sayin' nuthin'. I definitely think I've gotten dumber. Mostly since going to uni, I believe.
I wonder if it actually is dumber or just less disciplined? I definitely suspect the latter in my case at least and in many cases in general. Definitely happy to forget how to do things, so long as I still feel capable of picking up new things just as quickly as ever before. Now I'm not sure that the things that I can do now which I couldn't have done five years ago come under what people would generally regard as intelligence, but it's still progress of a sort.
Insomnia: are you familiar with this Faithless song
? Not the sort of thing I generally like, but this particular one has stuck with me for years.
Am definitely less disciplined than I was back in high school. Rather a pity, that.
I've never heard of this song before. I'd like to hear a recording of it now that I've seen the lyrics though. This one by Chicago
is more my style.
Not sure it's the sort of thing you normally go for, but there should be a link to a .mp3 of it third link down from this page
Looks to me like that page sends you huge, huge amounts of data, probably the vast majority of it pop-up related, but if it gets you mp3s, it may be worth a dabble...
I was given an IQ test when I was ickle by an edukayshunal trick cyclist. My parents did know what the results were, but
they didn't even mention the test to me for another decade and my mum only revealed the score with great reluctance.
This was probably a good thing.
|Date:||June 2nd, 2003 08:41 pm (UTC)|| |
When I said I am getting dumber, I meant relative to my peers, not to myself at a younger age.
I was being somewhat tongue-in-cheek, but there was some degree of seriousness. I do feel that I was more "outstanding" when I was younger. Now I don't even feel like the smartest of my friends. Granted, some of that has to do with the people that I associate with now vs. then (I know some pretty bright people), but relative to the population as a whole (of my age group), I feel like I have slid back in the pack.
I have the method for doing that first maths question, if you're interested! It was interesting to see how you worked through some of the questions - I got the same answers but in slightly different ways. On a side note, when you did Trig, did you ever use the "All Stations To Cambridge" rule? I just wondered if we were unique in using that because we are in Cambridge.
I refused on principle to do the TTN quiz on the BBC - I forget what my principle was now, probably that I'd be embarrassed if I got a lousy score *grin*.
I did it in the first year and got 67/70, which translated into either 130 or 133 - can't remember which. I decided that was good enough for me and have retired from the IQ test taking game.
Hadn't heard that mnemonic; only the Native American chief Sohcahtoa or the equivalent sentence 'Tom's Old Aunt Sat On Her Coat And Hat'.
We used Sohcahtoa too, but also the mnemonic "Some Old Horses Can Always Hide Their Old Age".
Seems to be especially fertile ground for making good mnemonics, more so given that Sohcahtoa himself is extremely memorable.
SOHCAHTOA here too, though I always liked "Some People Have Curly Black Hair To Push Back", which uses the terms Perpendicular for Opposite and Base for Adjacent.
I had completely forgotten about sohcahtoa, but then I went and did arty stuff at A' level which involved me creating the interesting (but now entirely useless) mnemonic of AGIAT-REP-TEC-EX which tells you all of the stages of Stanislavsky's 'Method'
I think that A is 'As if' and G is 'Given Circumstances' but the rest is a mystery.
Rivers of Yorkshire: SUNWAC! (Swale, Ure, Nidd, Wharfe, Aire, Calder.)
I do like your ducks icon. For some reason, LJ displays different icons on the reply-to page and the comment page, so it's showing your ducks icon as it's your default. Anyway, welcome - you now know a full three fourths of the ukgameshows "thang"...
For SOHCAHTOA, we use
"The Old American Sat On His Car And Hiccoughed"
All Stations To Camridge is used for showing in which quadrants the trigonometric values are positive (e.g. 0 - 90 = All, 90 - 180 = Sin, 180 - 270 = Tan and 270 - 360 = Cos) The rule works anticlockwise around the little diagram.
Native American? Never heard that. We were taught that Sohcahtoa "was a volcano" - like Krakatoa - which had the side-effect of my doodling billowing smoke and lava flowing out of any triangles drawn in my rough working, but never mind.
Yep - please forward the solution to Q1 forthwith.
Trig: we did not.
It is likely that there are errors along the way, but I think this is the method used for solving equations like these (we did a similar example in class).
Solve log (base 3) x - 3 log (base x) 3 + 2 = 0
y = log (base x) 3
(x^y) = 3
log (base 3) (x^y) = log (base 3) 3
y log (base 3) x = 1
y = 1/(log (base 3) x)
Therefore (log (base 3) x) - 3/(log (base 3) x) + 2 = 0
((log (base 3) x) ^2) - 3 + 2log (base 3) x = 0
let z = log (base 3) x
(z^2) + 2z - 3 = 0
(z + 3) (z - 1) = 0
Therefore z=-3 or z=1
log (base3) x = -3
log (base3) x = 1
Hmmm, not sure if i've done it entirely right, but that's the vague method anyway. Sorry for spamming your LJ with this, and i hope the notation is all correct.
Now 'tis time for me to learn about out the dear little Ottomans...and I don't mean the furniture...
when you did Trig, did you ever use the "All Stations To Cambridge" rule?
In my (thankfully) brief career as a teacher, this provided perhaps the single most memorable lesson. Fourth formers at an all-boys school, double maths first thing on a Monday morning. Teach 'em which bits are positive where on the circle. At the end of the exposition (involving a cross and a pointer) bring out the mnemonic "Auntie's Scones Taste..." and let the lads finish it themselves. Bingo, one lesson they'll never forget.
Which scale are you using? On one scale, 160 is top 1% but on another 160 is "practically most intelligent person on the planet". My understanding is that the general media and Mensa themselves quote the higher-valued scale where 148 is top 2%, except for some reason the BBC test uses the lower scale where it is 134.
By the way, I was pleased to see I (TRELLIS) was mentioned on the seven-letter word-fit puzzle in the latest WPC test.
I took a trickcyclist test when I was a child, and the Mensa test at 19, and again about three years ago, and every time got a figure on the "145=genius, 148=Mensa entry" scale. (Not telling but I am eligible for membership and not just a scrape-through.) The TTN test gave me quite a low figure first time round, and I was irritated by that. (This time I was on a plane when it was broadcast and haven't bothered looking for the web version.) Realising that they're scored on different scales came a bit slowly to me (doh) but by then I was so CBA about the whole thing I forgot to even remember my score (although I did score higher than DH, who says Anne Robinson put him off).
OK, here I'm going to contradict myself, but go off on a long- but quietly- held rant. It is known that the Cattell tests (the ones with "148 = Mensa entry") have standard deviation of 24. (Source 1
, source 2
, source 3
and so forth.) Yet "two standard deviations above average" does not translate directly to "top 2%" - it translates to "top 2.3%". If Mensa really do use "two standard deviations above average" as the qualification target, then Mensa isn't a "1 in 50" society, it is a "1 in 44" society, which is a slightly significant difference... at least, in my book.
There's an interesting quote from a Mensa Bulletin article by Dr. Abbie Salny, (American) Mensa's supervisory psychologist
(which is also "source 1" above):
For Mensa, an applicant must achieve a score at the 98th percentile on a standardized, supervised intelligence test or equivalent. The 98th percentile is two standard deviations* above the mean (rounded off).
That's a seriously non-trivial piece of rounding-off in my view. The difference between 2 standard deviations and ~2.06 standard deviations, between "1 in 50" and "1 in 44", is about 11% of Mensa's eligible membership. I think this is definitely a cynical business move on Mensa's part. (I've also wondered whether Mensa-supervised IQ tests ever do reveal people to be one or two points short of the Mensa level - surely it can't be in Mensa's interest to turn people away? Maybe if it'll get them to buy another book or two and have another test then it's a possible money-spinner, but...)
Tr**is Pe**ry frothing-at-mouth mode disengaged.
(which is also "source 1" above)
"source 1" as in "source 2", obviously. Just keeping you on your toes.
I rather fear that when Mensa published this book
, especially if they were also responsible for the accompanying synopsis, they destroyed their own credibility as an IQ testing organisation, and have very nearly destroyed the credibility of IQ testing in general. Ok, so you won't actually be more intelligent, but everyone will think you are.
And that's pretty much the Mensa credo, asfarasI can judge.
Very good. Confirms all my prejudices about the crew.
That's slightly unfair; I do think there are good reasons for joining, particularly if you want to meet more people (I attended a Mensa meeting as a guest, once, and found the folk there entirely personable) or if you know they have particular Special Interest Groups of particular interest. However, LiveJournal coupled with a little (dare I say it) selectivity would probably do those things at least very nearly as well (in some cases rather better, based on numerical advantage alone), is a lot cheaper and has flaws which are a lot less odious.
Definitely the one with standard deviation of 16. I go on about this at considerable length in the third paragraph and even talk about this in terms of standard deviations from the mean and proportions of the population; pay at-ten-tion, double-0-Trellis. (Oh, and even a result that is four standard deviations above the mean, so a 164 on the SD 16 scale, is "only" a 1-in-30,000 outlier - so "most intelligent 2,000 in the UK", rather than "practically the most intelligent person on the planet".)
Binet's scale, the original, had SD of 16, which is why most of the US (and the BBC test, as far as I can tell) use it. Wechsler's had SD of 15, Cattell's had SD of 24 - not sure why. (The jiggery_pokery Smartness Index, which I have just made up, has mean of 396 and standard deviation of 843, so lots of people in my Friends list have four-digit scores - and a negative score is also quite possible.)
WPC test: I note with a little amusement that the summary page has pushed back the estimate for the release of top scores again. First it was "Sunday", then it was "late Monday", now it reads "Monday evening...scores still being tabulated; top scores tonight are very unlikely." The poor, hard-working dears.
The definition of Mensa is "top 2%". The two S.D. part appears to have been given as an approximation (hence the "rounded off" quote).
That said, I wouldn't be surprised if they actually gave offers on a 1-in-44 basis given that those on the borderine would probably pass on a retest.
In reply to ericlendl, Mensa has destroyed its credibility many many times over since inception (even since the 1950s) and publishing a book on IQ tests is hardly going to change that.
Well, 'sall relative; when I say 155, I think of that as being the 1-in-4,000 level, as discussed. (Or, rather, I have defined my figural scale so that 155 corresponds to 1-in-4,000.) OK, 1-in-4,000 isn't one-in-a-million, but it's still pretty damn rare - it's "one in a big school every few years".
I also noted that their estimate was "somewhere over IQ 155", not least because I would find it difficult to tell whether they were 1 in 4,000, or 1 in 14,000, or 1 in 40,000, or so on. Not that expert, dude. :-)
"1 in 4,000" roughly translates to "best result in a particular year out of an entire county", for the sizes of counties we have in this country, or "one of the best students in their field in the one of the best universities in the country". It's nice to think that 1 in 4,000 might well be a significant underestimate for some of 'em, though. (I am starting to consider 125 as being a not-too-generous estimate for the average, too.)
Care to compare numbers in your experience?
*reads your LJ, enjoys your turn of phrase on things like "the fabled naked ace bluff of legend"*
Right, you're British. The British (except the "Test The Nation" thing, for what it's worth) normally quote IQs with a standard deviation of 24, not a standard deviation of 16. So when I say "three and a half standard deviations above average", or 155, that works out as 184 in British money.
Your 195 puts you about four standard deviations above average, which is about 1 in 30,000. I'm happy enough to believe this; after all, someone's got to be. By comparison, your average poker home game having IQ over 175 is a bit over three standard deviations in average, so you're suggesting the average player is about 1 in 1,000. I can believe that, too.
Seeing you're a chess player, in chess terms, "1 in 30,000" would put you in roughly the top 200 in Scotland. Looking at Chess Scotland, 100th place is a FIDE rating of 2070. So if you're something like 1800, +/- some margin, then you're probably something like top 200 in Scotland, so again "1 in 30,000". This ties in with your most recent comment about being "good" and what "good" means; the chess rating system, as accurate or non-accurate as it is, and the IQ scale, do give a definite scale to measure against.
Have you ever encountered David Levy? Scottish chess IM from a few decades back, made a famous bet about computer chess, plays a fair bit of poker himself. Mostly seven stud hi-lo, I think, though. Possibly a dumb question because he and his game are based in London whereas you're up in Edinburgh, but the Internet makes the world a much smaller place, so it's not impossible.
Sir Clive Sinclair is known to be several standard deviations above the average also but apparently cannot count to five on a poker table to save his life. When he got hammered on the first couple of series of "Late Night Poker", those were apparently some of his better days...
Mmm, interesting. I think your experiences do put you at a point where I can neither comprehend nor compare. All my information comes from the Uncommonly Difficult IQ Tests
site; the Titan Test
is apparently still being scored and may be an IQ test with a higher ceiling than the ones you have taken in the past. Certainly there are some interesting questions on there if nothing else. (There are links to a few other tests which may also stretch you at the site.)Thirty thousand isn't very many. It's an average sized town.
To be the single outstanding person in an average sized town is... well, it impresses me at least!Do you have any kind of distribution I could take a butcher's at?
The Uncommonly Difficult IQ Tests site also has the Generic IQ chart
to which I referred and from which I have been heavily drawing upon in my comparisons.And good is 2450+
I'm curious to know why you pick that figure in particular, though it's certainly as good as any. I could understand a feeling that "good" was a shifting goal, always a little out of your reach.
Unrelated questions: 1) where were you when the UK team for the World Puzzle Championship needed you most? :-) This year's team has been selected already, but if you enjoy the field, if the challenges involved are sufficiently to your taste, then perhaps you should consider trying out for next year's team. Details as I get them, but the past few years' qualifying tests are still online
2) How's tonight's poker (because I presume you've been online this late to fleece drunk folk operating on PDT!) been going?