Even Eduwonkette has a bad day every once in a while
I’ve come to admire the anonymous edu-blogger Eduwonkette, what with her skillful use of Photoshop, fearless questioning of the high and mighty, and, yes, lavish attention and fun she heaps on us here at Fordham.* But I’ve got to call her out on this morning’s post about New York City’s achievement gap.
My beef isn’t about NYC in particular but her analysis of the achievement gap in general. (An analysis that is strikingly similar to Charles Murray’s, by the way.) She writes:
Proficiency rates, or the percentage of students passing a test, are often used to measure achievement gaps. For example, if 90% of white students passed a test and 65% of black students did, some observers will say that the achievement gap is “25 points.” Proficiency is a misleading and inaccurate way to measure achievement gaps. Primarily, the problem is that we cannot differentiate between students who just made it over the proficiency bar and those who scored well above it. Proficiency rates can increase substantially by moving a small number of kids up a few points—just enough to clear the cut score. But black and Hispanic students may still lag far behind their peers even as their proficiency rates increase…. The most valid way to measure gaps between groups is to compare the test score distributions of the groups. What this means is that we compare average scale scores as well as differences between low-scoring white/Asian and Hispanic/black students (i.e. students scoring at the 10th percentile of their respective groups) and differences between high-scoring students (i.e. students scoring at the 90th percentile of their respective groups).
This is true, as far as it goes, if your goal is to create a world whereby all differences between racial or economic groups disappear. Maybe that’s what some organizations are seeking. But I think that objective is rather naïve and not particularly helpful. As Jay Mathews explains, working toward that outcome leads you to root against the progress of white and affluent students, because every gain they make offsets your attempt to “close the gap.” Mathews says “all children deserve a chance to climb as high as they can,” and surely he’s right.
Eduwonkette’s analysis also stands if “proficiency” has no meaning–if it’s just an arbitrary bar, and a low one at that. And yes, in many states, that’s exactly what it is. But let’s imagine that a state sets a standard for proficiency that actually means something–say, that a student is on track to be college- and workforce-ready by the time he or she graduates high school. Then closing, or at least narrowing, the “achievement gap” at the proficiency level is a worthwhile objective, for it would mean that we are succeeding in getting more students to that real-world standard. Since most white and affluent students are already at that standard (or close to it), the only way to narrow the gap would be to help more poor and minority students become college and work-ready. That’s a goal we can all get behind.
There are plenty of reasons to push back against the Bloomberg/Klein hype, but arguing that closing the “proficiency gap” doesn’t matter isn’t one of them.
* And so, as the senior pooh-ba on this blog (let’s face it, Checker isn’t posting a whole lot these days), I decree that Eduwonkette shall now be included among the “Flyest of the Fly.” (See sidebar.)
Related posts:
You can skip to the end and leave a response. Pinging is currently not allowed.
July 30th, 2008 at 9:54 am
I don’t see how Edwonkette has been saying that closing the achievement gap doesn’t matter. She’s saying that there are valid and invalid ways of measuing that gap. There is an art to “connecting the dots” in order to make meaningful evidence-based judgments (and by the way you should explain that to Liam), and Eduwonkette has demonstrated that NYC is doing that in a dishonest manner.
Maybe BloomKlein are the only educational leaders who use statistics to produce even greater “damned lies.” Or maybe we should next evalaute Rhee and the other true believers who let ideology obscure reality.
July 30th, 2008 at 12:31 pm
Mike-
Why do you think it is a naive and not particularly helpful objective to erase the racial differences in academic achievement? – to paraphrase your words
There is no question that there is a bell curve in terms of student aptitude and we should focus on accelerating students’ learning on all parts of the spectrum. That said, don’t you think it will be troubling if we continue to have a stark difference in the achievement of the average black or latino student and the average white or asian student – using whichever methodology to assess the gap (be it a more in-depth analysis as Eduwonkette describes or the one you think is adequate – comparing profiency rates between groups). Fact is, closing the gap between the races and continuing to raise the achievement of already high performers is not and should not be mutually exclusive (unless we think instruction is a zero sum game). I don’t think it is, even though admittedly NCLB is definitely structured in such a way as to encourage focus on the average performers.
And I do agree that it is a reasonable (and accurate) goal of the education system to get every student to a proficiency level such that he/she can become a successful adult (whatever that means, but for the sake of argument, let’s say that means an individual who is workforce ready and can make a decent income, is of good character, and can participate as an engaged and reasonably knowledgeable citizen in a democratic society). Again, I just don’t see how hoisting our high flyers necessarily conflicts with trying to close the academic performance gap between races?
August 1st, 2008 at 3:19 pm
That said, don’t you think it will be troubling if we continue to have a stark difference in the achievement of the average black or latino student and the average white or asian student – using whichever methodology to assess the gap (be it a more in-depth analysis as Eduwonkette describes or the one you think is adequate – comparing profiency rates between groups).
People really need to break free of their creationist thinking on this matter. How troubling are other aspects of human biodiversity? Isn’t it troubling that so few (none) East Asian sprinters win the 100 m at international track and field events? Is it so troubling that Jews are so over-represented as Nobel Laureates. Human Bio-Diversity is the hand that evolution has dealt us and there is nothing troubling about celebrating diversity, or for that matter, recognizing reality.
Fact is, closing the gap between the races and continuing to raise the achievement of already high performers is not and should not be mutually exclusive
You’ll have an easier time closing the stature gap that is part of gender diversity than you will in closing the racial achievement gap because we already know of ways to increase a person’s height through pharmaceutical intervention. To be fair here, if you think that closing the gap is possible then you should have no trouble in digging up solid research demonstrating that it is possible, after all, this has been studied for over a century now, and the question has had burning urgency since the 1960s, so surely you must be tripping over evidence of successful methods and trials.
When you write “Is not and should not be” you’re combining descriptive and normative statements. Normative statements can often be unbound from reality, as yours is, but descriptive statements are either correct or incorrect, and yours is incorrect. No one can pronounce statements on what “should be” as being incorrect, but to close the gap between the races and continuing to raise the achievement of already high performers runs into the problem of no one having any success with closing the gaps. Everything that has been tried has failed. Having to rely on unimagined future interventions in order to reconcile your two statements is a very high risk gambit that has no relation to the reality we deal with and is more a statement of faith than it is a statement of reason.
Again, I just don’t see how hoisting our high flyers necessarily conflicts with trying to close the academic performance gap between races?
Overlap is not equality. There are high achieving Blacks and Hispanics and there are low achieving Whites and Asians. The problem is that the populations don’t have the same distributions. So, when the high flyers are allowed to reach their full potential we’re going to see a widening of the achievement gap for more students in the White and Asian populations will excel compared to students from the Black and Hispanic populations. The most successful strategy developed thus far in the battle to close the achievement gap is to restrain high achievers, so one can’t, using strategies known to work, reconcile allowing high achievers to fly and making progress on the achievement gap.
August 1st, 2008 at 3:44 pm
But let’s imagine that a state sets a standard for proficiency that actually means something—say, that a student is on track to be college- and workforce-ready by the time he or she graduates high school. Then closing, or at least narrowing, the “achievement gap” at the proficiency level is a worthwhile objective, for it would mean that we are succeeding in getting more students to that real-world standard.
I don’t disagree but for this scenario to play out a few things need to happen.
1.) The mission of the education sector has to be reoriented – right now Job #1 is closing the gap. That mission takes precedence over any other. Any efforts that actually result in an increase in the gap are looked upon with disfavor. It’s career suicide to boast of being responsible for getting all students over a proficiency bar but by doing so increasing the achievement gap. Observers who note the increased achievement gap will characterize your efforts as failure at best, and racism at worst.
2.) One of the most powerful incentives administrators and politicians have for lying and obfuscating is that, even when caught doing so, this alternative is better than acknowledging the achievement gap and their inability to close it. When faced with an inability to achieve desired results THE POLICY becomes obfuscation.
3.) Abandon the Lysenkoist premise of NCLB. Denying reality is a surefire way of wasting resources and causing harm to innocents.
So, by removing penalties for failure to close the Gap we can institute actual, non-gamed, proficiency standards. We can institute policies, procedures, and pedagogies that work to further the goal of getting most students past a basic level of proficiency and any consequences that result take a back seat in importance, in other words, let’s get our priorities straight.
May 14th, 2009 at 6:16 pm
I don’t usually comment on blog posts… but this was a good post.. keep up the good work