What's Wrong With Teaching to the Test? (Part 1)

This is not a defense of standardized testing.


Any teacher who teaches only to the test, or any district that tries limiting its teachers to doing so, is surely misguided. But the occasional misguided or lazy teacher doesn't mean the tests are themselves the problem.

That also doesn't mean the tests are unimpeachable. As a classroom teacher, I am all too aware of how the current NCLB-engendered reliance on standardized tests can pervert a solid, well-designed curriculum. (I'm somewhat immune, because I teach government and economics in Arizona, where we don't have a state exam in social studies, but raising scores drives everything we are doing in our school improvement process, and the folks teaching English and Math to sophomores are directly in the firing line.)

But the fact that an almost exclusive reliance on standardized testing can skew what happens in classrooms, doesn't mean it necessarily skews what's going on.

It's kinda like hearing people quote the Bible as saying "money is the root of all evil," when Saint Paul really wrote, "For the love of money is the root of all kinds of evil." Close, sure, but not really the same thing, is it?

Now I don't claim to be smart enough to understand the whole testing debate, but from a classroom teacher's perspective, it seems like everyone ought to relax and start with the things reasonable people can agree on:

Standardized tests stink as a way to evaluate students. Very little we do in life is going to involve filling in bubble sheets. Many students who can give a great oral presentation, make pertinent observations in class and write insightfully given adequate time, just don't do well on tests focused on recall and recitation. And by definition, "standardization" means the test cannot be adjusted to reflect geographic, socioeconomic or other differences.

The tests are getting better, and the shift to common core standards should move us up Bloom's ladder in ways that open up performance possibilities for those students mentioned above. We use the TAP system for evaluating teachers in my school -- most schools will soon use TAP or something similar -- and the shift to common core is starting to make sense to our faculty. To the extent that the state exams (AIMS in Arizona) assess basic academic skills, skills that apply across the curriculum, they're not so bad.

Standardized tests also stink as a way to evaluate teachers. Not everything teachers do is measurable right away, and teacher quality is not necessarily causally related to improved student learning. Without going all Kumbaya on you, we are forming minds and bodies, and some of the seeds a good teacher plants may not germinate for years. Besides, a teacher can teach the hell out of something, and a particular student still not get it on test day.

Anybody got a better idea than tests? You know, a workable one? It'd be great to have a corps of trained evaluators fan out across the country to conduct one-on-one interviews with every student and teacher, providing inter-rater reliability on a grand scale. But who's going to pay for it? And whom are you going to hire? We already lose too many teachers to administration and non-education careers; are we also going to pull out thousands more to staff our evaluation corps?

We need a little Solutionism around here. We're educators, which means we're at least a little smart, and a little educated, so we ought to try finding a solution accepting the level of standardization that's unavoidable because we're a big country with a lot of students, while still preserving the kind of human-scale assessment that any decent teacher applies to her students. (I'm stealing the idea of solutionism -- "the belief that, together, science and humanity can solve anything" -- so go to http://www.draftfcb.com/work-detail.aspx?work=453 for details on the original idea.)

See, I'm not defending standardized testing, I'm just not sure there's much in the way of an alternative.