Also, I have heard many stories of HG+ kids doing poorly on the CogAT - it seems better suited to ID MG kids, but misses many HG+ kids for some reason.
My guess would be that it is due to the test being multiple choice and relying so heavily on fast processing after grade 2, when it is timed, and convergent thinking.
For example, the verbal classification section gives the kids a list of words and asks them to select from a group of choices how these words are the same. A child who comes up with some unique manner in which they are related cannot explain further why that fits since it is multiple choice. A deeper thinker may see too many possibilities.
I'm honestly not even sure on iding MG kids with the CogAT. I'm sure that there are some gifted kids (MG or otherwise) who do well on it, but I'm not sure that it is good for testing giftedness in general. Just b/c the scores on the CogAT correlate fairly well with IQ scores for children who fall within 1 SD of the mean, doesn't mean that the scores correlate well with IQ for kids who fall further from the mean.
There are, admittedly, studies of the validity of the CogAT as compared to the WISC-III and the WJ Cog. which show a reasonable correlation btwn these different tests, again for average kids. However, I don't believe that there have been any studies done of the correlation btwn scores on the CogAT and an IQ test for gifted children or for children with intellectual disabilities.
