So I got a heated email last night from someone who describes himself as “about as liberal as it gets.” He did not respond to my request to reprint his email, but to summarize, he was taking me to task for my attitude on education reform. This is not an unfamiliar position for me, to be berated about ed reform by people who self-identify as liberal or progressive and in fact use that identity as the source of their anger. His complaint was fairly familiar: by insisting that kids can’t learn, I am cursing them to miserable lives, demonstrating a straightforwardly illiberal worldview, and actually disrespecting the profession of the teachers whom I claim to defend.

Well: it won’t surprise you to hear that I don’t, actually, define my stance as “kids can’t learn.” My position on standardized testing and ed reform is, to simplify

1. Standardized tests can tell us some important things about students. Some degree of testing is important and necessary
2. We have ample ability, with the great precision of inferential statistics, to derive that information while being minimally invasive and hard on students, through appropriate stratified sampling and interpolation of the results. There is no reason to test all the kids all the time. We can test an appropriate sample of students and use the power of inferential statistics to understand the population with remarkable accuracy
3. Standardized testing elides an enormous amount of information about student growth, potential, and intelligence, as many scholars in psychometrics, developmental psychology, and educational testing will readily admit
4. Standardized test scores have strong associations with demographic factors like parent’s income level, race, and parent’s education level, for reasons that are complex and controversial. Standardized test scores, like essentially all educational research, are highly susceptible to confounding and common response variables
5. Test scores for individual students are far more static for most students than most people seem to believe
6. Teachers probably control, on average, a quite small portion of the variance in the test scores of their students, making standardized testing a particularly poor way to assess teacher competence
7. The portion of the variance in student test scores controlled by teachers is likely highly variable itself. There’s probably great diversity in how much individual student scores can be affected by teachers

I expect the empirical foundations of these arguments will continue to be tested and debated for the rest of my lifetime. Most of these, there’s ample empirical confirmation now. Some of them, you’ll have to call educated guesses. For now.

So what I would say is this: under the assumption that standardized test scores are the sole and primary criterion for whether education is worthwhile, then yes, I’m deeply pessimistic about education. But I reject that assumption. There’s plenty of goods that arise from education that have nothing to do with standardized test scores. You’ll forgive my frustration: it is the ed reform crowd that has worked to reduce our definition of what education is and does to raising test scores; they then respond to pessimism about raising test scores in mass by calling us nihilistic about education.

Yes, it’s true: I happen to believe that efforts to raise test scores dramatically and in mass, such as is called for by NCLB, is a fool’s errand that can only end in failure or (more likely) fraud. Such a thing has never been accomplished at anything like the scale of our current efforts, in the history of education. Yes, it’s true: I don’t believe that every individual student has the ability to score highly on these metrics. The insistence that any student can achieve the highest scores on testing instruments is both empirically unfounded and intuitively wrong. And I think most anyone who has worked as an educator for long enough, outside of the halls of private academies and Ivy league universities which have gatekeeping functions that exclude everyone but those most likely to succeed at those tasks, will agree.

Saying that isn’t illiberal. What’s illiberal is reducing our definition of what it means to be an educated, successful human being to those terms. What’s illiberal is believing that the only human beings that deserve material security and comfort are the ones who can build an app in their dorm room, or that the only people who can be creative with software are people who can get into Stanford. What’s illiberal is taking the vast diversity of human productivity and ways of flourishing and declaring only a tiny sliver of it actually productive or important. That’s illiberal.

Our educational “crisis” is actually an economic crisis, a manufactured outrage that insists we focus on teachers and their unions in order to avoid focusing on the breakdown of the social contract, we were told, we would live under. The test scores are telling us something, but it’s not what ed reformers thing. They’re telling us not that teachers have failed, but that we have created a society whose definition of human success is totally flawed, and one that has to change to survive.