Ask teachers or administrators what they think about high-stakes testing. You’ll usually get the same answers:
“We hate it! We worry every year about AYP. Our instruction suffers. We can never tell what scores will be because we have a new group of students every year. These tests are not particularly reliable [or not particularly useful for our particular population]. It doesn’t make sense to evaluate teachers based on students’ once-yearly, high-stakes exams,” and so on. If you are a teacher, no doubt you’ve said all this and much more!
High-level educational administrators seem to love high-stakes standardized testing. They love to compare our national scores with scores from Japan or Sweden (fo0lishly: how can you compare a small country with a relatively homogenous population with wildly variable populations across a huge nation like America?). Politicians love high-stakes testing too. In this way, they feel sure their expensive education budgets are properly spent.
Seems like there’s an unbridgeable gap between what decision-makers want and what educators want.
Can we find some middle ground?
Maybe it is not as hard as it seems. One good step would be to rely on formative tests rather than last year’s test scores. This way, we could more accurately track and assess student progress as we go through the year, fine-tuning our instruction to actual student progress as we go along. Administrators can keep legislators apprised of student scores throughout the year, rather than just relying on a high-stakes test score once a year.
Now that many states have moved to the Common Core, we have some wonderful built-in assessments which are not just high-stakes standardized tests. The project results can be posted on school websites for local communities and politicians at all levels to examine.
There may be other good ways to find middle ground, but implementing even just these two suggestions could create more reliable assessment to satisfy politicians as well as freeing us to better teach our students.