Raising and Educating a Deaf Child

International experts answer your questions about the choices, controversies, and decisions faced by the parents and educators of deaf and hard-of-hearing children.

Question from Greg D., Longmeadow, MA

Have you heard of discrepancies between the SAT 9 vs SAT10? I am seeing deaf children who made nice gains in the classroom, but the SAT 10 scores show little or no gain over the SAT 9 given 3 yrs earlier.

Question from Greg D., Longmeadow, MA. Posted April 26, 2010.
Response from Sen Qi - Gallaudet University

I don’t recall that they did equating studies between SAT-9 and SAT-10 for the deaf and hard of hearing students.  Short of having a student sample who takes both two versions of the tests, we have no way of knowing how these two measures differ. But the question at hand is actually about using classroom progress as benchmark to compare the to standard tests.

The logic goes like this: if the students have gains in school work, they are supposed to have gains in the SAT scores as well.  There is no gain from SAT-9 to SAT-10, therefore the two versions of the SAT must be different, presumably SAT-10 is more difficult.

The above reasoning is fallable in several fronts.

1. Some standardized tests are notoriously misaligned with classroom teaching. So to avoid comparing oranges and bananas, the teacher should check if the contents of the tests and the contents he/she taught in the classroom are indeed matched.

2. If the contents of the tests are pretty much what they taught in the school, then there is another reason why the students don’t have score gains. Perhaps they are not used to the standard test settings, and have not received coaching on taking such tests since 3 years ago.

3. I am not sure which test level of SAT-9 and SAT-10 these students took. If they took lower level for SAT-9 and higher level for SAT-10, they may have less gains because the scale is not necessarily linear when they take different tests, although they claim that SAT has a vertical scale acoss test levels.  It’s always dangerous to strech too much.  For example, if you ask a 10the grader to take a 2rd grade tests, he will get a scale score, but the score’s validity and reliability will be questionable.

4. Of course, there is always a possibility that the two version of the tests have true differences.  In fact the main reasons for the test publisher to release new versions has been to keep up the changes in curricula and student populations. The new version is supposed to fit the new curricula and new generation of students better.

All in all, if I were a teacher, I would trust my classroom tests better than standard tests, especially if what I want to know is whether my students learned learned something from me, instead of how my students compared to the national norm.

I feel that I need to add one more point. The best way for a teacher to have his/her students gain on standard tests is to teach the tests, although there is no guarantee for success.