That's interesting speculation. I'll add some more speculation. Maybe the schools don't get norm-referenced scores, and can only report % correct (vs %ile). Not so sure JHU could use school grade-level administration to generate norms, if the schools are also using it as a GT selection test, because then it wouldn't be a representative sampling. Unless the schools try to develop local norms? Or use arbitrary cut scores? Or just pick the top 10% of test-takers?

...pronounced like the long vowel and first letter of the alphabet...