IQ tests
IQ tests
First off - I don't know how to use a wiki properly, let alone your fancy talk page messaging system. Please excuse any errors. Secondly - I noticed you said you didn't buy into iq tests to determine intelligence potential on the High IQ society talk page. I was wondering what you meant by that; or is it just a 'fuck IQ tests' kind of statement?
Hi, you did just great.
IQ tests are remarkably unable to look at intelligence, cause we dont' really have any way to measure that. They look at your vocab - if you grew up on a home that is not English, or grew up deaf, or grew up speaking BEV, you will score lower on that section, no matter what your intelligence is. They rate how well you can solve "logic" problems, but the logic is a very western way of looking at logic problems. Native Americans as one very particular example, are known to score badly on that part, cause they don not learn the same methods of solving problems that western americans do.
IQ tests have no way at looking at how you put peices of the world together to come up with unique solutions; they have no way of examining how you use so called "common sense"; they have no real way of assessing your creativity - a vital part of intelligence; they have no way of compensating for bad/good test takers.
And again, the very premise they start with is a logical "fail". it's very unclear what they are defining as "intelligence", cause no one really knows what "intelligence" is.

Well, we do know what "intelligence" is, just that there's multiple definitions for it and ways of measuring it, and no one method that's actually reliable and capable of giving us this one reproducible answer. Your actual mind, the thing that displays intelligence (as an absolute, not as some higher end of the scale), has many different attributes all with different strengths. When these attributes are skewed or exaggerated in one direction or another we interpret it as something like mental illness (Aspergers, for example). It's not unlike building the stats of an RPG character - you only have a certain amount of EXP to spend and you can't max out everything.
The main problem with IQ tests is that, despite what people think, practicising does improve your score. Whereas if you wanted an actual objective test you'd want to get as consistent a score as possible regardless of how many times someone took that test or ones similar to it. There are only so many different types of problems you can give people before the recognise all the methods needed to successfully complete an IQ test and score highly - this is your spatial awareness one, this is the one where you convert dots and shapes to numbers, this is how you look for the "primer" to break a code, and so on. Its the same reason I can do those logic grids you get in Puzzler really quickly; I've done 3 or 4 by thinking, now I can regognise the procedure required, I haven't necessarily become more logical, though.
Certainly, I don't think you can simply do a test and prove that you're clever. That has to come from long-term interactions with others (should you actually care about what others think) who can assess you in multiple situations instinctively and spot what you're good at. As Stephen Hawking said; people who brag about their IQs are losers.
See, I could be wrong, but i've never seen a definition of intelligence that wasn't "beauty esque, I know it when i see it". I often ask of people - you know a good salesman is clearly intelligent, cause they can size you up, then figure out what motivates you and sell directly to that motivation. That's not some easy thing. But you're never going to see anything like that on a test - cause how would you test for it. One of the most intelligent men I know of, would fail any IQ test, cause he left school at 16 (typical "mom died" or "dad was drunk" or whatever makes you be 'the man of the house') and took a job as a janitor for the school. It's a state job, so he kept it, and still has it at 50 or so. He doesn't read well, and has had very little "world experience", doen't know about Iraq or how far the sun is. But when you present a problem to him, you see his mind at work. When you do show him something (the one I did was how far the sun is from the Earth. and in a way that shocked me, he took that info and started talking about just how hot the sun must have to be, and then how do we survive that much intensity... from his own thinking, not with math or science, just thought), he can instantly use it in his world. This guy would not do well on any "IQ" test we've yet to devise.
Deaf people are another measure. Because of the nature of the language they use, English or any written language can be a serious puzzle for them. Their own vocab in sign are 1000-5000 words, cause signed languages aren't really built on new signs, but on how you use existing sings to fill new concepts. How are they going to rate well on a test.
Puzzle solving, another marker for intelligence, is a measure of how you take your existing knowledge gained by experience or "books/being told", and manipulate them to make novel experiences. But to do this, you have to have a set of world experiences that actually relate to what you are solving. in a very very very stereotypical example, a black kid who is asked "how can you survive and not jump a gang", who comes up with 100 workable ideas to do that is no less intelligent than the guy who can stack up to blocks and tie a string across them to make a sled - they just have different experiences. and right now, IQ tests don't test the black kid's (again, stereotypical of me just to overstate a point) prior knowledge, but they do test the privileged kid's.
And i'd not thought about practice, but you are right. You see the types of problems and get shown the answer once, you can use that to figure out others.

I'll have to bookmark that tidbit on sign language. I'm currently sketching out my little NaNoWriMo adventure and I'm sure language and thought are going to take up the best part of a chapter. In fact, with a new Stephen Fry documentary on language, a few wikipedia articles on colour perception lingering in the background and a few other things it might even take up a massive wad of the 50k words. :S
NaNoWriMo? Yeah, i started the Fry last night. I'm an addict to anything with language. then you add fry? and i'm all over that shit! but i fell asleep, so we'll see.
True signed languages (as opposed to signed english, or signed french) are amazing. You map your language differently. I get it intellectually, but i can't really accomplish it to save my life. :-)
- reduplication - If you sign something, like "to learn", one time, it means one thing, and two times means something more. In learn, once means "I learned something", and two or repeting means "i am in the process of learning this". I teach sometimes, vs., I'm always teaching.
- arching - there is a concept of arching that if i make one action / word in one place, it means a specific thing, event, time, etc. So pointing once, means "he", but if i arch, it means "many, always, all of them", so pointing to my right, then drawing across more to my right means "they". Signing night at my wrist means "tonite" or "night" in general. Signing it at my elbow and moving downword means "every night" or "at night the sun sets", kind of "always". (arching is cool, I've read 2 studies on it, but they were at the time, beyond my understanding of asl).
- direction "verbs". So most (?) any (?) verb can be signed with direction in mind. If I sign the word "meet" and draw my finger from you to me, when making the sing, it means "you meet me". But if i sign the same sign starting from me to you, it means "I meet you". almost any verb that would take an indirect object (ask to, walk to, give to) you can drop the subject and indirect object by the direction you make the sign.
These things sorta blow me away as english speaker and linguist, cause it makes signing a far more dynamic language than English.
I'm really loving the works on Linguistics of Sign and how it helps us understand language at its core.
Course- all that said, when I sign it's like "Me umm ummm ummm WANT umm ummm ummm GO umm umm umm Store", which is about as grammatical in ASL as it sounds in english. ;-)

I am writing some bad RW fan-fiction as a piss-poor excuse to write an extended essay about skepticism and rationalism. I have to reach 50,000 words over the course of Novemeber. I'll devote a couple of pages to thought and language, but it's something I need to research. The way sign language can add an extra layer of complexity is something I can't afford to ignore if I want to do that topic justice.
That is the classic objection to IQ testing, but it is widely acknowledged now. It's getting a bit old, but I think the APA's response to Herrnstein and Murray's neo-eugenicist drivel is still a good starting point. I like to think of IQ like BMI -- it has its place but is easily and often abused.
I think as long as you know what you are looking for, and what you are testing for, you're right it can "have it's place"
no, i take that back. Seriously, if a test is flawed, then what good is it. and if it is not flawed, what is the point of saying "you are better at doing this one specific kind of test than someone else". Does it really mean you are more intelligent, and if so, what does that even mean in the real world? Will someone who is intelligent be of more value than someone who is not?
I took IQ tests when i was in primary school, and was way way over the top (obscenely, which is emotionally why I initially distrusted them cause i'm simply not that smart). Yet i could not spell or do math. And when i took them at 16, i was on the lower end. (cause again, i can't spell, do math, or intuit grammar). So what did it gain me, one way or the other? What did that test gain for the school? Did they put me in the right classes? Was I smarter when i was 5, then when i was 16?
I mean, what are they for? Other than giving you bragging rights over someone "dumb", or give a school the right to put some kid in a lower class or some kid in a higher class (which is just bad bad bad form), what do they serve?
Other than "how fun, i got all those questions right".

In my experience, IQ gives your parents an excuse to yell at you for "not fulfilling your potential"
not that i EVER want this implemented, but i do somedays wish i had a facebookesquw "like" button. "like ty's statement!!!"

It has its place if the scope is correct. In the case of IQ, like BMI, the scope is for an average over a population. The objections are only really valid for an individual because any one person can override the entire point of an IQ test. Taken over the course of hundreds, if not thousands of people, and split by various demographics, they can give you meaningful information because those little bumps that are so objectionable in an individual case are ironed out.
Consider the analogy of anecdotal evidence. A single anecdote is meaningless, as is applying an IQ test to one person. Build them up in a controlled manner, however, and you have something you can use.
Maybe i'm missing something hugely obvious. "used to do what".
Say Denver kids are less intelligent than NY kids? Say today's kids are less intelligent than yesterday's kids? Unless it's used individually (which as you said, and i lived, is highly problematic), I can't see where these numbers really help.
I must be blind about it, but what kind of info does it give us, other than how good we are at making and taking tests?

Well, it's been used to study the Flynn Effect, whereby IQ's gradually increase over time. It's been shown that certain racial demographics have lower IQs but probably owing to social rather than genetic factors. And they're made over many observations, so like a medical trial saying that homeopathy is bunk while paracetamol isn't so much, it's definitely a real effect. If the tests were random and useless, such effects wouldn't be seen at all.
Of course, the Flynn effect is most interesting in that certain US states can't execute people if they're retarded - i.e., below a certain IQ. Now, putting aside the fact that one individual's performance isn't indicative of actual intelligence (because we have a hard-and-fast law using this as its foundation, we can "assume" it's real), the Flynn effect means that the level needed to meet such "retarded" criteria changes over time. So a person killed 30-40 years ago might be considered under the retard threshold to be killed today.
Serious question for all of you then.
Do you think we are getting more intelligent as a culture over time? or that our experiences are greater, more numerous (and our exposure to reading and, you know, test taking) is greater and therefore we appear more intelligent on these tests.
If you took 1000 fetuses from 1000 years ago, and planted them in women today, would they be "less intelligent".
(or am I misunderstanding the flynn effect).

It's probable more of a cultural thing than an outright genetic thing, but like off this nature vs nurture crap the answer is undoubtedly both. If you did take humans from 1000 years ago and made them grow up today, I'm sure you'd probably see a statistical tendency for them to be less intelligent, but not by the amount you'd expect it to drop if they were actually 1000 years ago.
wiki did mention an increase in nutrition, especially in utero, so I guess i can buy that. But i just have a hard time buying that we are becoming more intelligent, and that it isn't simply that we have significant more experience at our hands.
I mean, how much mental activity could you do, when you didn't read, and you spent all day tending your fields. The potential might be there, but you aren't using your brain so you aren't pushing it to learn more.

Well, if you accept evolution you must accept a genetic basis for intelligence. Secondly, if you accept evolutionary principles, you'll be aware that there's no way to stop it working. Only if a) we reproduce exactly through cloning and b) do not die before we reproduce, would evolution be stopped. So while we've come down from the physical trees and climbed up the tree of intelligence we should continue to do so as long as there's descent with modification and a driving force to select for higher intelligence. That local factors in development outstrip this effect by an order of magnitude at least doesn't automatically make it negligible or non-existent, though. So while our wider experiences, better diet and improved access to culture and technology will improve intelligence significantly (i.e., we could pluck a baby from the bronze age and I'm pretty sure they'd be up to speed with our computer technology just as fast as any modern child) there will most certainly be a small genetic factor visible on a statistical level.
Usually the most accepted idea of nature-vs-nurture is that genetics enable you to do it, but your environment allows you to reach that potential. I think it's slightly more complex than that, though, and certainly isn't a case that you wouldn't be merely limited by genes. They determine not so much how far we can go, but how easy we find it to get there. You can have the best potential genes on the planet but unless your parents instill a work ethic into you, you're not going to Oxford or Cambridge. Similarly, you can come from "poor stock" but for only a marginally greater amount of effort you could equal or better anyone from "good stock". Even then, it's important not to make a rule out of this or platonify it. Variation is what drives evolution, so we're undoubtedly going to find a high amount of variation in intelligence's genetic factors - thus manifesting in the anecdotes of "oh, Joe Blogs had dumb parents and grand parents but he ended up with a PhD!" and "Sally Sue had everything, but still flunked out and failed everything!" - actually, even then that's an cartoonish exaggeration, genetic variance may be great, but it's also subtle.
The thing is, while both nature and nurture might improve our intelligence, the real interesting thing is whether they conspire and feed back on each other. Do we improve intelligence selection criteria at the genetic level because of our more nurtured abilities?
Yeah, I get the evo thing, but i remain dubious that evo happens in 100 years. At least in this case, where there's no real environmental pull to cause us to evolve. Evo requires a pressure, and i don't see that. we are not selecting for the more intelligent - sighs - if only cause statistically more intelligent / successful people tend to have LESS kids.
"The thing is, while both nature and nurture might improve our intelligence, the real interesting thing is whether they conspire and feed back on each other. Do we improve intelligence selection criteria at the genetic level because of our more nurtured abilities?" -- Ohh, ahhh... ;-) So you are saying that it's possibly (likely?) that the biology the actual genetics can/might/does changes because of non genetic things like nutrition or the opposite say - malnutrition, poision in the water, etc?

I'm saying that to claim it doesn't happen is a bigger stretch. But the effect will be small, and only visible at the population level. But the sheer number of confounding factors make it difficult to fully quantify. Undoubtedly I think we will be selecting for more intelligence, just not in the obvious ways you might expect. Idiocracy was an interesting concept, but not really a documentary, I think.
Ugh, Idiocracy, don't even get me started on that.
Oh, i wasn't thinking it was a documentary. It is statically provable that the more successful (again, not "intelligent", as i have a super hard time quantifying that term) individuals have less (even 0) kids. My own dyslexia will be out of the gene pool, but so will my smarts.
What isn't considered is that just cause someone hasn't had a historical connection to things that allow him or her material success (like a good education) doesn't mean their genes are not equally "smart" when given the same chance.
But again, i'm a big believer in nurture more than nature.

The thing is, while both nature and nurture might improve our intelligence, the real interesting thing is whether they conspire and feed back on each other. Do we improve intelligence selection criteria at the genetic level because of our more nurtured abilities? -- I made a comment to this, and it got eaten (or i didn't hit SAVE... which is more likely).
So you are saying that it's possible/likely/proven that what happens to an individual in the here an now, can/does actually change their genetics? (ie., nutrition or malnutirion; exposure to environmental pollutants, etc)

It's not the only factor, but I'd put my money on nutrition and health care. People are getting taller as well, most likely thanks to that.
Ok, see that part i can accept and understand. I know we see it in menstrual changes in girls (age of onset), and age we live to be. So it could well make you smarter. make your brain fire better. Conversely, the chemicals around us, and lack of variety in food, and light at night make us more prone to depression, autism, etc.
I'm just really stumbling at accepting any idea that we can evolve in 100 years, without serious pressure on it.

Yeah, that's essentially what I was just typing except worded better.
I think we might all be operationalising intelligence a little bit differently. The way I think about is as the processing speed of your brain. Imagine it is the megahertz/gigahertz and ram capability of your brain (a distinctly cognitive outlook). It correlates strongly with things like reaction time or inspection time tasks (how quickly you can react to stimulus or decide which of two lines is longer) probably because the faster your nerves can 'transmit' the faster your brain can 'think'. Most authors would define it as how quickly a person can learn and operationalise new information - there might be small differences between definitions but not great big fatal ones. Obviously intelligence is different to education. Intelligence is fluid, it can be applied to ANY field, whereas education is crystallised it can only be applied to specific fields. That's why tests like Ravens or Catells are considered more intelligence loaded, they contain very little variance due to prior learning - you have to learn how to do the test as you go along.
There are problems - Retesting effects to my mind are the most obvious and dire (it's where I do my research). However, the proof always lies in the pudding, what can you do with IQ scores? How about predict work performance to a correlation of .6? How about the BEST single predictor of work performance; in the absence of previous work performance data?
The flynn effect might actually be because people are getting smarter through environmental effects. Don't tell Schlafly, his head might explode.