3
$\begingroup$

I have been watching a lot of Jordan Peterson's videos on YouTube recently. One of his themes is, that IQ is among the best studied metrics, with high predictability, validity and is generally believed to be a very real thing in psychology. He also mentions, that IQ can not be improved or trained.

My question is, whether I could improve my IQ (by that I mean the test score, not general intelligence), by simply studying for it, trying to learn about the different types of questions and how they are generally solved and by simply doing a bunch of tests and memorizing how to solve questions.

I have done a few IQ tests online and at the beginning, I was very confused about everything, like from which direction to read the boxes and stuff like that. After doing a couple and seeing how it is done and reading explanations about the different questions I feel, that I am doing a lot better now.

Similar to that, I was on a job interview once, where they asked riddles, presumably in order to assess IQ, without doing an IQ test.

I knew some of the riddles and simply answered some of them by sheer memory.

Isn't it possible to increase your test score of an IQ test by simply studying? And if you can, what makes the tests really valid? And finally, if IQ was such a high predictor of intelligence, wouldn't most job interviews be geared towards assessing your IQ in a legal manner?

$\endgroup$
7
  • $\begingroup$ What sort of IQ tests are we talking about? The last two I was in were administered by people, with little or nothing to write, although some stuff to draw. I suppose you could try working with geometric puzzles. $\endgroup$ Jan 3, 2019 at 0:00
  • $\begingroup$ I have only done online tests, which I know are not very valid. The ones I did where multiple choice basically, puzzle type tests mostly. But I am asking about the tests, which are considered to be valid. $\endgroup$ Jan 3, 2019 at 10:07
  • $\begingroup$ If you know that the online tests you have done are not valid, have you read anything regarding IQ tests and what makes them valid? Maybe something which makes them valid also makes them unbeatable. $\endgroup$ Dec 16, 2019 at 10:50
  • $\begingroup$ @ChrisRogers yes, I have. There are even tests, that are supposed to not require you to be able to read, purely abstract. But I don't buy it. The puzzle tests with the boxes seem to be part of reputable tests and it absolutely helps to know the mechanisms behind those. There is no way an illiterate tribe member from the amazon is going to perform the same as someone from the west, used to solving pointless problems and having studied the mechanisms of the test. $\endgroup$ Dec 16, 2019 at 10:58
  • 1
    $\begingroup$ See also this previous question on practice effects on the progressive matrices test $\endgroup$ Dec 16, 2019 at 11:46

1 Answer 1

5
$\begingroup$

Short answer:
Yes, but it does not matter

Long answer:
Each question has some inner mechanism that you either know or you don't. Usually you should know how to do it, but you don't know, that you should do it that way. Once you know the right method, next puzzle you would solve much faster - most IQ tests are like this. If you don't know the mechanism, it could be very hard to solve - IQ tests should be like this.
For example this one - once you know, that there are 4 squares all the time, next puzzle of this kind will be much easier. Most puzzles use Subtraction and Addition - once you try this two things, you will solve lots of IQ tests. You are expected to know this methods, and that it will be used in test.
Problem with IQ tests is insufficient number of respondents - since it needs lots of people to create norm.
For example I created test with possible score 0-100, and 50% of people reached 80 points. When you take the test and had 80 score, your IQ should be 100. But when (because of better education for example) most scores begin to be above 80, norm will change and for 100IQ you would have to score 82 or more.

So yes, you can study for it, but when more and more people do it (and knows about using subtraction, 4 squares all the time and so), norm will change and tests would be harder for everyone anyway, since there will be always exactly 50% of people that could reach at least 100IQ. ( = take some online tests, since most people do it, otherwise it would put you into disadvantage.)

Sources:
Flynn effect - raise in IQ due to education and also health (notice not genes).
More about norm
Tests commonly used - in it you can hide math operations without using numbers.

$\endgroup$
4
  • $\begingroup$ So why aren't these methods explained or required to know before you take the test? If they don't tell you these methods, it means that some participants will know them and some won't which will result in decreased validity. After all these tests are supposed to measure intelligence, not knowledge. Do the makers of these tests argue, that it doesn't help knowing how the tests are structured, when it clearly does? $\endgroup$ Dec 16, 2019 at 10:16
  • $\begingroup$ @user1721135 ideally you shouldn't know the method. But that is the problem of IQ tests, because you need thousands of people for norm, but also some of them will talk about your test and "spread the word", so you will have to change your norm or test. From my experience none of this happen (because of money and energy) and 60% of people will have over 100 IQ which is wrong. $\endgroup$
    – Jan Ivan
    Dec 16, 2019 at 10:27
  • $\begingroup$ Welcome to Psychology.SE. Can you please cite some sources of information to back claims made in your answer? From my understanding, IQ tests test for all sorts of measures for intelligence, one being the ability to see patterns and maybe sequences in things. If you have the same pattern/sequence in every question, does that not negate the ability to test this? $\endgroup$ Dec 16, 2019 at 10:54
  • 1
    $\begingroup$ @ChrisRogers Yes, but it is hidden within the question. I'm not talking only about en.wikipedia.org/wiki/Raven%27s_Progressive_Matrices but, it is good example of IQ test, where you can hide common operations without being "too obvious". $\endgroup$
    – Jan Ivan
    Dec 16, 2019 at 11:30

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.