oh and....predictive reliability is also used inside each test. Any subtest or subscale on any given test should strongly predict the scores betweent themselves.
So a child with a very hi verbal score would be expected to have a very hi perceptual score. Or a child with a very hi writing fluency score would be expected to have a very high mathematics score.
When a child's score within a test fluctuate more than a standard deviation (15%ile points), one needs to consider "specific learning disabilities" and get the child more thoroughly evaluated.
My son had as much as a 91%ile point difference between subtests!!
An iq score shouldn't really be drawn from such a test. The test should just be used diagnostically in this type of situation.
For my son there was severe asynchronous development, so severe that he has specific learning disabilities. His doctors say it will pull together over time and with interventions to help him overcome the LDs.
Of course, the school psychologist says the scores indicate an average iq. ARGH!!!!!
but I digress....quickly pull me back!!! chuckle
