4 ms·
It is trivially simple to show that your objection is worthless. 'A thermometer measures temperature when placed in your armpit. But I can take a thermometer an
by gwern 13y ago
It is trivially simple to show that your objection is worthless. 'A thermometer measures temperature when placed in your armpit. But I can take a thermometer and put it in hot water and the temperature goes up! There is no reliable way to measure temperature, therefore, it's not clear the temperature is measuring anything reliably. What, exactly, are you talking about then?'
- mikeash 13y agoReminds me of the surefire way to totally cheat on your tests and always get a great score and never, ever get caught. All you have to do is spend a bunch of time beforehand going over all the material until you understand it.
- Ygg2 13y agoIQ tests are notoriously hard to make. And they don't measure IQ, so much as what the tester considers IQ. An amusing story. There was an IQ comparisson between rural and urban areas and the finding was that people in rural areas do worse on IQ test. But one psychologist thought that this can't be true. So he went back to test and realized that IQ test that were given to rural/urban areas were same. The people from rural areas did worse on the written part of exams, because the rural areas lacked schools and children started learning/reading much later. Once he went back and revised the tests, removing most of test that had to do with reading and generally type of thought that rural children have less encounter. Conclusion - the old IQ test didn't measure intelligence, unless you can increase intelligence by going to a better school, in which case, that isn't intelligence but knowledge.
- rprospero 13y agoThere's another famous, possibly apocryphal, story about an IQ test with the question: Which of the following doesn't belong: a) Basketball b) Polo c) Hockey d) Billiards The answer is obviously Basketball, since it's the only sport that doesn't use a stick to hit anything. Except the answer is obviously Polo, since it's the only one with horses. Except the answer is obviously Hockey, since it's the only one with a puck instead of a ball. Except the answer is obviously Billiards, since it's the only one without a team. In case you're curious about the RIGHT answer, the test found that Canadians and Americans living in colder climates were slightly smarter than average.
- wslh 13y agoFollowing your story I love this scene from Kaspar Hauser: http://www.youtube.com/watch?v=oAnOi0fnxuE http://www.youtube.com/watch?v=oAnOi0fnxuE From the subtitles: The Professor here has come all this way to ask you a question. He wants to see how well you can think... ...and what you have learned in these two years... ...and whether you can think logically. Will you answer him? - Yes! - Good. Kaspar... ...let's pretend that this is a village. In this village live people who tell only the truth. Here is another village. The people here only tell lies. Two paths run from these villages to where you are standing... ...and you are at the crossroads. A man comes along, and you want to know which village he comes from... ...the village of the truthtellers or the village of the liars. Now in order to solve this problem, to solve it logically... ...you have one question, and only one. What is the question? That's too difficult for him, how can he know that. I admit, the question is thorny. If you ask the man whether he comes from the village of truth... ...and he does, then he will say, truthfully, yes. But if he comes from the village of lies, he will lie... ...and also answer yes! Yet there exists one question which will solve the problem. That's much to hard, too complicated. You have one question, Kaspar... ...and only one, to solve this problem of logic. I wouldn't know either. If you can't think of the question... ...then I shall tell you. If you came from the other village... ...would you answer 'no' if I were to ask you whether... ...you came from the liars' village? By means of a double negative the liar is forced to tell the truth. This construction forces him to reveal his identity, you see. That's what I call logic via argument to the truth! Well, I know another question. You do? There is no other question, by the laws of logic. There isn't? But I do know another question. Let us hear it, then! I should ask the man whether he was a tree-frog. The man from the truth village would say: No, I'm not a tree-frog, because he tells the truth. The man from the liars' village would say: Yes, I'm a tree-frog, because he would tell a lie. So I know where he comes from. No, that's not a proper question. That won't do, I can't accept it as a question. That's no logic; logic is deduction, not description. What you've done is describe something, not deduce it. But I understood his question. Understanding is secondary; the reasoning is the thing. In Logic and Mathematics we do not understand things... ...we reason and deduce: I cannot accept that question.
- mikeash 13y ago
- gwern 13y ago> IQ tests are notoriously hard to make. And they don't measure IQ, so much as what the tester considers IQ. They are notoriously hard to make, yes, but psychometricians have put in a ton of effort to make culture-fair tests and investigate claimed biases, so they're pretty good these days. We're a very long way from the WWI Alpha test.
- phaus 13y agoIt's a well established fact that thermometers measure temperature. It doesn't matter if its a person's temperature or an object's, it is still going to measure temperature. It is a highly contested assertion that IQ tests accurately measure I.Q. It's completely possible that if you take a reasonably smart person and make them study for a specific type of I.Q. test, that he will score higher than he would have otherwise, even though his intellectual ability would remain unchanged over a broader spectrum of skills than the test involves. Therefore, it may not be a reliable measurement of I.Q. Your example makes no sense whatsoever, and you should have already known that before you typed it out.
- gwern 13y ago> It is a highly contested assertion that IQ tests accurately measure I.Q. It's completely possible that if you take a reasonably smart person and make them study for a specific type of I.Q. test, that he will score higher than he would have otherwise, even though his intellectual ability would remain unchanged over a broader spectrum of skills than the test involves. Therefore, it may not be a reliable measurement of I.Q. It is not contested in the area in question; psychologists routinely use IQ tests without a qualm, and it is a consensus in the field that they are meaningful. See for example the consensus paper on IQ released in the wake of _The Bell Curve_ controversy or look at more recent review articles like Nisbett's "Intelligence: new findings and theoretical developments". Whatever the layman or politicians may think, the debate is over: if you make a good IQ test, it will estimate accurately general performance on all sorts of cognitive performance. If you destroy the accuracy of a given IQ test by training and then measure performance on the original variety of cognitive performance, this will immediately show up in the factorization and demonstrate how the IQ test's accuracy has been destroyed, and IQ tests developed with different kinds of questions will not show the spurious increase, just like if you took a different thermometer and put it in the person's armpit, it would show a different reading from the thermometer stuck in the mug of hot water. > Your example makes no sense whatsoever, and you should have already known that before you typed it out. My example is exactly analogous to the argument that was made.
- phaus 13y ago