Gaming high stakes tests

by David Safier

Those of you who haven't taught may not have thought about the most rational way for a teacher to maximize the number of students who pass high stakes tests.

Here's the thinking teacher's guide to getting the highest passing rate on AIMS (or any other state test):

1. Don't waste time working with your top students. They can pass those tests in their sleep.

2. It's not "cost effective" in any rational cost-benefit analysis to waste much time with the lowest students either. Even if you raise their scores a few points, it probably won't be enough for them to pass AIMS by the time test day rolls around.

3. Invest your time in students who are close to the pass-fail line. If they get two or three more answers right, they pass — just barely, but they pass. If they blow a few answers, they fall just short of the cutoff.

Cynical, right? No teacher with a conscience would ignore the top and bottom students and drill-and-kill, test and retest the middle students within an inch of their lives. And yet, the pass/fail incentives built into high stakes tests are set up for that kind of behavior, and if teachers want to maximize their student pass rate, it's pretty hard to resist spending extra time teaching to the almost passing and barely passing while letting the top and bottom kids fend for themselves.

"Value added" assessment makes this kind of behavior somewhat less attractive. That's when students' scores are measured for improvement, not just the passing rate. If a low student makes over a year's progress in your class, for instance, that's more important than if he/she makes it all the way to that magic "pass" number. But when you look at published numbers on school comparisons, the pass/no pass percentages are up there in neon lights. Even with "value added" thrown in, passing rates are the bottom line.