So if I'm reading the norms correctly, your DC was already right up in the 99th %ile range on the fall reading and language tests, which is probably why his scores fell slightly over the course of the year. MAP has item lockout, which prevents the same items from being re-administered within a certain time frame. There aren't that many very high level items in the 2-5 pool. I would imagine that he exhausted the pool early in the process, so the adaptive item selection started tossing him slightly easier items, as they were the highest ones remaining. If the school re-set the item pool on the winter reading retake, that might have allowed him to go back up at that point, and then fall again in spring, on the same item pool issue. His SRI/lexile puts him above the 75th %ile for 12th graders, which tends to support the test ceiling hypothesis. He did make significant gains as assessed by that measure.
I wouldn't say that the data definitively show that he has lost ground in reading and language.
Minorly off topic rant: that's just bleeping stupid. I can understand an assessment having a ceiling, especially one like the MAP that is more for tracking growth over time instead of ability at a given time. But how hard would it be to generate an error when the test stops giving good results. Error: questions out of range is more useful than a ceiling that artificially and silently moves down.
Anyway, back on topic.
