No Profit In Ending Poverty

Here's an extremely sobering, depressing, thought-provoking interview with the creator of the acclaimed HBO series The Wire. David Simon was a crime reporter for twelve years with The Baltimore Sun before turning to a career in television. Here's an excerpt from the interview, conducted by the inimitable Bill Moyers.

DAVID SIMON: The people most affected . . . are black and brown and poor. It's the abandoned inner cores of our urban areas. . . . (E)conomically, we don't need those people. The American economy doesn't need them. So, as long as they stay in their ghettos, and they only kill each other, we're willing to pay a police presence to keep them out of our America. And to let them fight over scraps, which is what the drug war, effectively, is. . . (S)ince we basically have become a market-based culture and it's what we know, and it's what's led us to this sad denouement, I think we're going to follow market-based logic, right to the bitter end.

BILL MOYERS: Which says?

DAVID SIMON: If you don't need 'em, why extend yourself? Why seriously assess what you're doing to your poorest and most vulnerable citizens? There's no profit to be had in doing anything other than marginalizing them and discarding them.

Simon's solution?

I would decriminalize drugs in a heartbeat. I would put all the interdiction money, all the incarceration money, all the enforcement money, all of the pretrial, all the prep, all of that cash, I would hurl it, as fast as I could, into drug treatment and job training and jobs programs. I would rather turn these neighborhoods inward with jobs programs. Even if it was the equivalent of the urban CCC, if it was New Deal-type logic, it would be doing less damage than creating a war syndrome, where we're basically treating our underclass. The drug war is a war on the underclass now. That's all it is. It has no other meaning.

What Would Replace "Grade Level"?

I think common sense says that any single measure that claims to assess something is always enhanced by a different kind of measure that attempts to corroborate its claims.

So if I want to know if my kid knows something about the Revolutionary War, I might give her a multiple-choice test. But I'd also want her write a 3-page paper on the cause of the war, give a presentation on the Battle of Bunker Hill, and write and act in a skit about George Washington. At the end of the unit, I'd want her to select items she worked on and place them in her portfolio and then write a meta-cognitive summary of what she learned, the challenges she faced in learning them, and how she overcame them.

Duh, right? Most good teachers do these sorts of things all the time. All of these assessments/measures focus on the question of what my kid knows. But each produces different information in different ways. And each involves different skills.

So go tell that to your state DOE. What will they say? Something like, "These classroom-based assessments are very nice, but they're certainly not reliable. We can't possibly accept your judgement about what students in your classroom know and can do."

THAT'S the problem.

Of course you can't accept the teacher's judgment if punitive high stakes are associated with the assessment, e.g., the teacher getting fired (thanks NCLB!) or the school getting shut down (thanks again, NCLB!) So there's an incentive (thanks Campbell's Law!) to cook the books and make things seem what they aren't.

It doesn't have to be this way. Ultimately, what we're really concerned about is (1) what do kids know? and (2) what can kids do? A single measure (usually a norm-referenced, multiple-choice test) that tells me if my kid is "at grade level" does not tell me what my kid knows and what my kid can do. It tells me if my kid's score is the same as her peers, below her peers, or above her peers. In short, it tells me zippety-doo-dah.

Multiple measures (such as the ones I mentioned above) are the evidence we need to answer the questions (1) what do kids know? and (2) what can kids do? These measures reveal nothing about "grade level," i.e., where these kids are "supposed" to be in relation to each other. Rather, these measures give a very real sense of where these kids ARE. Once we know where they are, we can help them get to the next place. How they get there and when they get there is an open question. But in the best circumstances, getting there is kind of fun. Anyone remember teaching and learning is supposed to be fun? It's different for each kid and for each teacher. It's what learning is all about, and what makes teaching a thrill.

Not On The Test

Why We Need to Debunk "Grade Level"

As we have seen from the Obama administration and so-called "progressives," NCLB is going to be tweaked, not fundamentally altered or discarded. I'd argue that it's being tweaked and not discarded because there is a very strong sense that there is a magical thing called "grade level." This magical thing called "grade level" is (1) very real, (2) can be measured empirically with a fine degree of validity and reliability, i.e., it really tells us something useful and is beyond repute, and (3) lots of low-income minorities are not at "grade level" and is therefore cause for concern, as it is our sacred duty to get them to this magical place.

But, as I have been discussing, we can show that "grade level" (1) is a phantasm, (2) cannot be measured accurately or reliably and does not yield any kind of useful information whatsoever, and (3) is therefore meaningless when we talk about the academic achievement of low-income minorities.

This is a foundational critique that, if successful, will raise the following questions:

1) If "grade level" is a phantasm and does not accurately measure what students know and can do, what are other means by which we can better understand what students know and can do?

2) If it's meaningless to say that low-income minorities are not "at grade level," then what is a meaningful way to talk about the disparity that exists between low-income kids and their more affluent peers?

If we get lots of folks asking these questions, then there's an opening for discussion of alternatives. But if folks are not asking them, then they still uncritically and unquestionably accept that "grade level" is real and will, therefore, always be caught in a box. They will design more assessments -- maybe even some pretty good ones -- but these assessments will all be for the purpose of determining if kids are at "grade level" or not. Ergo, we are still where we are now.

With all due respect to the work that everyone -- including me -- has tried to do on debunking NCLB, we have clearly not achieved our goals. So that's why I'm suggesting this tactic.

Is "Grade Level" a Load of Hooey?

At the heart of the "achievement gap" is the contention that lots of low-income minority kids are "not at grade level" and are often said to be several "grade levels" behind.

But what do we mean by "grade level"? Grade level is the score of the average child in a particular grade on a norm-based test. But, by definition, 50% of all children are always below grade level. When using standardized, norm-based tests, you always guarantee that half of the students taking the test are below grade level. So when we say that low-income blacks, for example, are not at grade level, aren't we overlooking the rather obvious fact that LOTS of kids -- in fact, HALF of all kids by definition -- are below grade level?

Further, aren't we overlooking the fact that standardized tests are rather poor measures of what students know and can do? And as a colleague of mine reminded me recently, "Most kids need family and adult support to become readers. And as we know, many kids don't have that support. Thus the need for schools --in loco parentis." So aren't we also overlooking the fact that low-income minorities often don't have this kind of family support, so their being "behind" is not all that surprising?

So why not accept that lots of low-income minority kids do not read at the level of their white, affluent peers and -- instead of pathologizing them for this and then handing them a dumbed-down, See Spot Run curriculum with lots o worksheets -- work with them from where they are and at their own pace? In other words, why not just accept that all kids learn differently and at different paces? Would this just be too ridiculously practical? Instead of giving them the dumbed-down curriculum, you give them all the support and encouragements and structure they need without framing their development as "behind" or "slow" or "impaired." It's just where it's at. They are where they are. Don't sacrifice PE and art and music so they can do more phonics drills. Give them a broad-based experience of schooling that still makes it fun and interesting. They may not read Moby Dick -- ever. But they may not want to read Moby Dick, even if they could. (Confession - I tried to read it once and gave up because I found it lethally boring.) Am I missing something here?

In the end, here's what I think this would accomplish. We'd reframe the "achievement gap" and replace it with what we know to be true of all kids (actually, all people): a continuum. We know that ability varies greatly on everything, and that some kids are simply better at reading than others. There's nothing wrong with this, in the same way that there's nothing wrong with the fact that some kids are better communicators than others or better dancers or better weavers or better at computer games than others. Reading is tricky, though, because it's seen as so foundational, and there's a belief (probably romantic) that while it's OK for kids to be better at some things and not others, ALL kids have to be equally good at reading. Maybe they just aren't? And, since reading is so heavily affected by socioeconomic factors, it only makes sense that affluent kids will be slightly better at it than low-income kids.

Given all of this, we'd see growth measures in place for each student and completely get rid of norm-based standards and measures and only measure students in relation to their own growth and development.