I was driving into work today listening to Steve Inskeep on NPR finish up a report on presidential candidates (focus Mike Huckabee). And what transitional music do they play in between that report and the next? The first two notes– and only the first two notes– of the initial guitar lick of Depeche Mode’s “Your Own Personal Jesus.” I laughed all the way to school.

It got me thinking about the school improvement plan we were handed yesterday at our faculty meeting (I might post about this fascinating document later), a key component of which was addressing “weakness in background knowledge at all grade levels.” I can’t help but reflect that is only my odd, twisted, totally unique combination of “background knowledge” that let me in on an NPR auditory joke that I would be surprised if anyone else actually comprehended.

So a widely-administered test on understanding this joke, I’m sure you’d agree, would be unfair. And yet isn’t this exactly what we are doing in our listening and reading comprehension passages on a standardized exam?

E.D. Hirsch’s work here, and related articles in the same American Educator, have perhaps the most well-sourced argument I have seen for reading comprehension being tied almost completely to background knowledge. There’s also the old related standby that the “background knowledge” required to pass a standardized reading exam inevitably is reflective of socioeconomic status.

But I think I’m going one step further here. I heard NPR’s joke NOT only because I’ve heard some Depeche Mode, and NOT only because I happen to listen to NPR regularly, BUT because I personally happen to have some musical sense. My “background knowledge” is tied intimately not only to my SES, but to that pattern of exposure, internalization, predilection and talent that everyone has, and never repeats, like snowflakes.

Can we test the reading ability of a snowflake? Sure. (Give it a book about phases of matter and watch the magic.) But can we do it properly with a sole arbitrary reading exam that we administer to approximately 2,800,000 kids in New York State alone? We need different data than that. And we need better.