Some problems with the definition of intellectual disability and their implications

There is a degree of consistency in how ID is defined, that it is:
• A significant impairment in intellectual functioning;
• A significant impairment in social/adaptive functioning;
• All occurring before adulthood.
A significant impairment in intellectual functioning is further defined as a measured IQ below a stated cutoff point usually 70 (or two standard deviations below the mean).

This would seem to assume that everybody has a true level of intellectual ability that is measurable with sufficient accurately (about 4 to 5 points) to make stating an IQ cutoff point in the definition a reasonable thing to do. What I do is to look at how valid these assumptions are.

I will present the work I have done on the accuracy to which we can measure low IQ: I have done a meta analysis of the stability of measured low IQ, indicating that the 95% confidence interval for stability is 12.5 IQ points. I have shown that there is a major flaw effect in the WISC-III and WISC-IV. I review studies that have compared the major gold standard IQ tests and have found that the WAIS measures consistently higher than the WISC and Stanford Binet. I also note we do not currently know how the Flynn effect is effecting IQ scores sufficiently well for us to correct for it. Taking all these factors into account it looks as though we can only measure low IQ to an accuracy of about 15 points either side of the measured IQ, which is much greater that the 4 to 5 points assumed by the writers of the definitions. This will result in individuals being misdiagnosed, which will result in some individuals not getting the service they require, being given a stigmatizing label or, in the extreme of somebody appealing a death penalty in the US on the grounds of having an intellectual disability, being wrongly executed. We need to accept that measured IQ is only a rough indicator of true intellectual ability in the low range, that intellectual disability is not something that can be precisely defined in terms of measurable variables and we need to develop a series of qualitative definitions that meet the needs of various stakeholders.