Some questions about IQ tests - and no, it's not anything to do with G&T.....

[meandmyflyingmachine]

I have seen various references to IQs recently, and they are given as a single number (eg 131). Is this how the number is given by the person doing the test, or do they say there is a margin of error. Does the test discriminate between, for example, 130 and 131? And if you were to do a different test, would that come out with the same number?

I know absolutely nothing about these tests really. My gut feeling is that they must surely be a fairly broad brush, but that isn't actually based on any knowledge , hence my questions. Anybody able to inform me?

I'm talking off the top of my head as it is a long long time since I looked at this sort of stuff but no, you wouldn't expect to get the same result everytime you took the test, assuming it is a different test. There would be a variation although the results should be in the same ball park. You shouldn't get say, 100, which is deemed average on one test, and 130 or 70 on another. If you got 131 on a test I would have thought it was reasonable to get anything from 125 to 135 in another test. (That is just my opinion by the way I don't know what the official margin of error should be but certainly I have had scores in that range over the years). We can vary a lot when taking the tests depending on how we are feeling, how much time we are given or where and when we are sitting the test. Also people tend to get better scores as they get older as their knowledge and experience grows.

The thing about these tests is that whilst they try really hard to test your ability to think it is almost impossible to devise a test that isn't based on some sort of learnt knowledge. For example, you need to know how to calculate the maths questions or at the extreme how to read the questions in the first place. There are cross cultural differences in knowledge too. You also need different tests for extremes of intelligence as people with a low IQ simply wouldn't cope with the test for the majority of the population and the highly intelligent wouldn't be sufficiently tested to get a proper result at the other end of the scale.

There is a big debate on the usefulness of IQ tests. They give an idea of your ability to think and reason but there are so many variables that they aren't foolproof. Others argue about how one test can't measure the different types of intelligence.

I am beginning to be running out of info on this but I hope that helps.

There are a few widely used IQ tests (summary), and they each have different rules about how they are supposed to be interpreted. They come with standard deviations which are supposed to indicate the confidence intervals (not sure if you use +/- 1.96*std deviations or what). Try Google to learn more.

See I told you somebody would know better than me.

I was just looking at Wikipedia (searched on IQ) which give quite a good, if long, summary.

I thought you're explanation was better, actually, Niecie. But I guess the basic answer to me&flying machine is definitely "Yes".