The results are crazy for advanced kids -- post high-school math and reading levels in elementary for WJ. Same issue for WIAT although not sure the levels are quite as excessively elevated.
Just a reminder, this is because age/grade-equivalents reflect only that your student obtained the same raw score on the test as was obtained by the 50th %ile of the referenced age/grade on this same test, NOT that they are functioning at that level. They are not as elevated on the WIAT only because the reference group tops out at 12th grade.
If you want to know the instructional level of a student, these are not the appropriate instruments. The WJ and WIAT, like most norm-referenced standardized measures of cognition or achievement, are intended for comparing a student to their normative age/grade peers, which is why the most valuable (and most accurate) piece of information is the age-referenced standard score.
As to which is better for classification purposes, I would agree that either works. You just need to be aware of the limitations and strengths of each. I think I have mentioned elsewhere that the WIAT has item sets by grade for certain subtests (Reading Comprehension, Oral Reading Fluency), and a couple of subtests aren't available until grade 1 (or age six). The item set restrictions mean that some very skilled students will hit the test ceiling, which may underestimate their achievement in that area. OTOH, the WJ Broad achievement composites (BReading, BMath, BWritten Language) all include fluency measures, all of which are timed pencil-and-paper tasks, which can unduly penalize small children with age-appropriate fine motor skills, but advanced academic skills. The WIAT fluency subtests are not included in the area scores, and the reading fluency subtest is motor-free, unlike the WJ. Reading Comprehension and Written Expression on the WIAT are also deeper measures than the equivalents on the WJ, which, I suppose, may mean that some children score artificially high on the WJ in those skills.
