I feel incredibly lucky to be at a place where standardized testing is not the norm. While a lot of schools have lost time to days of testing, and teachers, parents and students have questioned the purpose and validity of the tests, we’ve plugged along, doing what we do. That’s not to say that tests, even standardized ones, don’t infiltrate our lives. For us, the SAT and ACT loom large. When our aggregate scores show a decline, we worry. Is there something we can do to improve those scores? Have we changed something in the way we teach x that might have inadvertently impacted those scores? We resist teaching to the test, but we can’t help but be influenced by tests.
Assessment is a good thing sometimes, but it’s clear to me that we’re relying too heavily as a society on what certain assessments tell us. And even when assessments tell us something, we’re reluctant to do the hard work (and spend the money) to fix the problem. The issue of underperforming schools is a complex issue, and what a lot of research tells us is that we need to fix the problems outside of schools–poverty, drugs, poor parenting skills, absent parents, etc.–in order to fix the problems in schools. Telling those students over and over again that they’re not performing at grade level according to a test (whose validity is questionable at best) does not help things.
I’m a big fan of data, but it seems to me that we’re relying on too narrow a slice of data to deal with a complex issue such as learning, which is tied up with a lot of things. I’m a bigger fan of humanistic interpretations of data that look through a variety of lenses. Instead of just saying, here are the numbers, we should be saying, what do these numbers really mean? And why? The why is hugely important, and I don’t see too many people who are supporters of heavy testing asking that question. Ever.
When we look at data at our school, the first question we ask is why? Why does it look like this? What are some possible reasons for this? Is this something we can do something about? Or is it something outside of our control? And sometimes it goes the other way. We have a problem and we think, hmm, can we get some data that might tell us something? It’s always in context, and I think that that’s what’s often missing from testing these days. There’s no context.