They aren't all that stable at any age, and yes, less so in early childhood. It was just the specific number I was wondering about. It could be right, though, and I certainly agree that IQ testing preschoolers is weird. It doesn't seem as though it should be beyond the wit of humankind to educate each person appropriately...
An interesting thing is that the nature of a normal (or nearly normal) distribution makes it almost inevitable that something shocking like that number is going to arise. If you're drawing a line towards the right hand end of a bell curve, then by the nature of the thing, a high proportion of those who are to the right of the line are *close* to the line. So then even a small test-retest difference will lead to a high proportion of those who landed to the right on a test not landing to the right on a retest. And this is true irrespective of exactly where you put the line - there will always be some people who are clearly to the right of the line, and a lot more who are close to it. Which just illustrates the stupidity of using such a line to make macro differences to individuals.
Of course, although the UK doesn't go in for IQ testing children, and GT programmes of the kind common in the US are rare and getting rarer (all state funding for GT provision was recently withdrawn), we don't have a system to hold up as an alternative. The comprehensive school system which was supposed to address precisely this does not effectively do what it was supposed to do. So we have various selective schools, i.e. instead of trying to get into a GT programme you're trying to get into a school where all children are what might be called GT elsewhere, which is even more rigid. At least for C that'll be at 13 not 3 (interestingly, his rather academic prep school was at pains to point out that they were not academically selective at entry, although this didn't stop me worrying at the time!).
I'm not sure what the right answer is, honestly. Probably "make teaching a really prestigious career" is a large part of it.
An interesting thing is that the nature of a normal (or nearly normal) distribution makes it almost inevitable that something shocking like that number is going to arise. If you're drawing a line towards the right hand end of a bell curve, then by the nature of the thing, a high proportion of those who are to the right of the line are *close* to the line. So then even a small test-retest difference will lead to a high proportion of those who landed to the right on a test not landing to the right on a retest. And this is true irrespective of exactly where you put the line - there will always be some people who are clearly to the right of the line, and a lot more who are close to it. Which just illustrates the stupidity of using such a line to make macro differences to individuals.
Of course, although the UK doesn't go in for IQ testing children, and GT programmes of the kind common in the US are rare and getting rarer (all state funding for GT provision was recently withdrawn), we don't have a system to hold up as an alternative. The comprehensive school system which was supposed to address precisely this does not effectively do what it was supposed to do. So we have various selective schools, i.e. instead of trying to get into a GT programme you're trying to get into a school where all children are what might be called GT elsewhere, which is even more rigid. At least for C that'll be at 13 not 3 (interestingly, his rather academic prep school was at pains to point out that they were not academically selective at entry, although this didn't stop me worrying at the time!).
I'm not sure what the right answer is, honestly. Probably "make teaching a really prestigious career" is a large part of it.
Reply
Leave a comment