It's true that faster isn't smarter, and I agree that some kids are missed because of this, especially when they don't have a good clinician interpreting their performance, but test developers have to draw a line somewhere...is 30" long enough? 1'? 5'? At some point, it becomes difficult to compare performance across the norm group, if the vast majority of children offer their final response within a certain time period, but one outlier is accurate, but only after a substantially longer time. This is still clinically-useful information, which I would report, and use as the basis for recommending accommodations, but it makes the normative information that should be derived from it less meaningful, if that child's test conditions are that different from those of the standardization sample.
This is why test scores need to be interpreted by a skilled clinician.
