Why I Like Tests
That's right. I am not opposed to testing. Tests are frequently the fastest, most direct way of uncovering what your students have learned. There's no point in cobbling together an "authentic" task if what you're shooting for is a quick check on which kiddos can reliably multiply by fives or spell "democracy."
Rockets, volcanoes and bubble bombs are fun to create and explode, but somewhere in there, shouldn't students be able to describe the chemical reactions that occur when you combine baking soda and vinegar? Isn't that the point of hands-on learning, illustrating knowledge to drive home key points of content?
Ah, you say. That's what reports are for--a real-life task. Well, gee. Isn't a required report just a test, only more prone to excess verbiage and bigger margins? In fact, isn't assessing student work and providing feedback a core responsibility and skill in good teaching? If a test efficiently tells us what we need to know--what to teach next--why wouldn't we use it?
A few months back, I got into a bitter argument with some ardent school reformers over my inclusion of the modifier "inappropriate" before the word "tests" in a policy brief. I fully believed (and still believe) that some tests are appropriate. I don't have any problems with tests that are aligned with content taught in the classroom. And I think that positioning all testing as heartless, damaging, even pointless, has done the real reform movement--meaning the anti-corporate agenda-- genuine harm. There must be room for moderate and nuanced thinking on the value (and dangers) of testing.
Appropriate tests are tightly linked to what was taught, and used to inform instruction, rather than sorting and ranking students or punishing their teachers. Good tests push beyond recall, into application, and focus on the most important disciplinary concepts. Standardized tests should be voluntary, and lead to systemic improvement, rather than penalties and shaming.
Sharing data should be done locally, to fine-tune instruction, not "prove" that some students are more deserving, or bludgeon teachers who choose to work in dysfunctional schools or with kids who got off to a rough start academically. Teachers and school leaders should be rigorously trained in assessment literacy. Good assessment, at its best, will drive good instruction.
Of course, that's the real problem. It's not the tests themselves. It's the way the systemic use of standardized testing and resultant data has been co-opted, to justify the same old winners and losers in the struggle to get ahead, economically. The huge increase in testing hasn't told us a single thing we didn't already know about who holds the cards in the education game, who will take home the biggest piece of the pie. We don't need more data.
Look for the Common Core-porate Assessments (now under construction) to offer "performance tasks" into their assessment toolkits. Look for these tasks to be electronic simulations, machine-delivered and scored. Look for a limited number of these, as they're difficult to construct and challenging to score with precise psychometric accuracy. But expect them to be advertised as the Next Big Thing, a dazzling 21st century addition to boring old standardized tests.
There may come a time when we all long for the good old days of teacher-made tests and #2 pencils.