/sci/ - Science & Math

>>6538452
Solve an IQ test. Then learn from your mistakes and see what you did wrong. Repeat 100 times. Congratulations, the tests now state that your IQ went up by 50 points.

IQ tests are like BMI. They're not completely worthless but they're far from accurate and can easily be done incorrectly, as is usually the case when attempting to measure something in an incredibly indirect manner.

>>6538452

Take two equally intelligent individuals

>one is motivated to do well, interested in the test, has prior test taking experience

the other

>has little to no testing experience and no reason to apply effort or care about the test

why are people oblivious to the fact these two people will have very different IQ test results?

>>6538463
Proved that effort is more valuable than natural intelligence. So I guess it did work.

>>6538485
take two individuals with no prior test taking experience

they each take the test once.

IQ becomes valid again.

>>6538452
Hilarious pic, I lost it

>>6538463

It is true that IQ tests can be practiced for. However, this defeats the purpose of the test.

As I recall, there is usually a limit on having to take several months off in between tests for the result to be considered valid. I suppose this would also apply to any practice in doing IQ tests. Of course you can't really supervise that in most cases, but then you're only cheating yourself because it's not like you win much by having a paper stating you have an IQ higher than it actually is.

>>6538452
Can we stop with these fucking retarded posts? Where the hell are the janitors and/or mods?

>>6538452
> I think the only thing you'll get out of IQ-tests is the persons ability to solve IQ-tests.

The first law of psychometrics is : A subjects response measures only the subjects response to the test.
So, yeah, pretty much.

Linear, one dimensional scales are for measuring linear, one dimentional things.

IQ is just plain stupid.

>>6538624
If you want good threads then make them you whiny faggot.

>>6538636
Y-you, too.

>>6538628
IQ isn't linear. Obviously yours is too low to understand what the word "linear" means.

>>6538628
>what are IQ subtests?
>what does linear mean?
>what is a normal distribution?

>>6538463
>take a gun to a boxing match
>win instantly

Wow, you can manipulate the results by cheating? Cool story, bro. It is explicitly written in the instructions of every IQ test that you're not supposed to take the same test twice. And you're not being told the correct answers either.

>>6538519
This.

Do no-IQ people think that every person gets the same result the first time and then they go higher as they try harder?

There are people that simply are better at doing them without experience, how's that supposed to work then?

>>6538934
>It is explicitly written in the instructions of every IQ test that you're not supposed to take the same test twice.
Nobody said it has to be the exact same test, you can simply learn to better solve the types of problems they give by practice. Any objection to doing so only proves that the system is flaud.

If IQ tests do not measure intelligence... then surely the average IQ of prison inmates and PhD students will be about the same, right?

>>6539012
Intelligent people take an iq test and get a good score and are satisfied and stop. While stupid people take the test score poorly and are upset and practice the test until they score higher. Then they come here bragging about muh 150 iqs... look how much I practiced i even googled the answers and retook it 6 times! Don't lie I know you do it.

While the rest of us are getting on with our life working on important stuff.

IQ tests have certain predicative value in judging the future academic and work success of children.

Still there's interesting effects like Flynn effects that say that the mean IQ score in 1982 would be 121 if it were judged by the same criteria as those in 1952. I don't know if the effect continues as strongly but there's certainly alot of aspects about IQ that we really don't know much about. There's no one region of problem solving in the brain, and although genes have been found that are linked to increase of certain receptors increasing IQ scores it's still very much a question what general intelligence actually corresponds to in the brain.

>>6539016
Noone has ever argued that random IQ test online are particularly good. Or that cheating on a test is still valid.

IQ measures a very specifc ability. It does not measure creativity for example, wich is an extremely useful ability to have and surely a characteristic of great genius. Also, people think differently, i know people who are smarter than me and i have already helped then solving some problems just because i could think in a different way.

Every IQ test I've seen has at least one section which requires you to guess the next element in a sequence based on a finite prefix. Anyone with a modicum of mathematical understanding will tell you that such a question is illdefined. There are infinitely many sequences which satisfy the criteria, and the IQ test does not specify on which grounds the one unique answer should be selected from an infinite number of candidates.

So whatever it is that an IQ test measures, it is not mathematical intelligence. It seems to me to be more of a social intelligence - you are graded on your ability to infer which sequence the test writer had in mind, which one seems "natural" to the examiner. Which would explain why high IQ results are correlated with employment prospects. It's easy to nail an interview if you can accurately guess what your potential employer is thinking.

>>6538452
IQ tests are vague measurements of intelligence which have never been developed to tease apart the upper end, where the quotient might be meaningless anyway. They are mainly developed so when running DIFFERENT tests on subjects, they can make sure there isn't a large effect of intelligence, which is why the tests are designed to discriminate coursely between retards, average people, and low geniuses. It is also used to identify learning disorders, so bad performance in one area can be compared within subject to their baseline. This is pretty much the attitude of every tester, cognitive scientist, and person working in psychometrics I have ever met. And btw, they ain't no Einsteins neither. If you believe these are precise tools which based on deep psychological principles distinguish between a 140 and a 145, you are fooling yourself. Things in that range are so narrow, and usually based on a single question or a few seconds in faster performance. Speaking as someone who was tested for dyslexia multiple times whose IQ ranges 135-145, which really means nothing except that I'm somewhere in that 95-98 bracket, maybe. Also, who cares?

>>6538452
>IQ tests are completely accurate.

Find me one legitimate psychologist who says that.

>>6539154
While there does exist an infinite number of possibilities to continue the row of elements, there is usually one obvious pattern one can observe. For example, for the row of numbers 1 2 3 ... the most obvious pattern would be 4, though different logical assesments indeed can bring you to a different answer which will be no less correct, yet far less obvious. The ability to choose the "obvious" answer from a list of given possibilities, however, does not indicate a higher level of intelligence than seeing multiple possibilities, quite the contrary in fact. An intelligence that is only able to choose an answer from a finite list is the lowest level of possible intelligence (with the second level being able to solve a problem without said list, the third providinge multiple and original answers and the fourth and final level being able to see far beyond the setting, the stage and given problems and focusing on something outside the comprehention of the first three levels).

>>6539171
>IQ tests are vague measurements of intelligence which have never been developed to tease apart the upper end, where the quotient might be meaningless anyway.

>believing in threshold myth

IQ is only important when comparing the different races, in which case it's one of the most important metrics to exist.

I DARE someone to find something wrong with the logic behind these studies. I fucking DARE you.

>>6539736

>I DARE someone to find something wrong with the logic behind these studies

Sure. "Black" is not a race.

>>6539736
I've heard the difference between blacks and whites in IQ has decreased in recent years to be around half a standard deviation. It's still pretty large but hey, it's a progress.

Also
>Sample: 7
lel

>>6539736
>that sample size

>>6538452
They are generally constructed to test one skills in logic and related intelligence, and doesn't really tell you much about ones ability to abstract or be creative.

I think aptitude tests are more useful.

>>6539744
I wouldn't bet on that finding. Remember that there is enormous peer pressure to claim that the difference has been reduced. There are many studies and outlier studies pop up once in a while. The gullible can then just cheery pick the outlier studies in his preferred direction of effect size (smaller or larger).

>>6539743
>"Black" is not a race.
According to the government and pretty much every institution, "black" is a race.

>>6539750

>2800
>1300

seems pretty significant to me. The others we can safely throw out but the black/white divide is just striking.

>>6539813
So, in other words, people who want to deny the usefulness of IQ tests, look to redefine intelligence to include creativity and abstract though, instead of the traditional definition of logical thinking and critical analysis ability, and then declare that because the IQ test does not properly gauge these other abilities, it does not measure intelligence. Do they imagine some stigma exists against valuable intellectual abilities that cannot be called "intelligence"? Wouldn't they still have a problem if we renamed "IQ" to "logical thinking quotient"? It sounds to me like an excuse, and not a reason.

>>6538452
You're probably retarded but you should understand the difference between linear systems and traits and non-linear ones.

Mental functions don't operate as in a 1+ 1 = 2 model, or 130 is whatever percent smarter than 100. Mental functions interact in remarkably complex ways, and aren't reductive to some test.

Srsly yo science is fucked its trying to be some kinda math but that aint how it works. Watch this video to better understand how retarded you are.

http://www.youtube.com/watch?v=_njf8jwEGRo

>>6540135
This, so very much.

>>6540069
not in my country.

sage goes in all fields

http://www.nytimes.com/2004/12/12/magazine/12QUESTIONS.html?_r=0

What is your I.Q.?

Stephen Hawking: I have no idea. People who boast about their I.Q. are losers