Sunday, October 23, 2005

Urban Education

This Jonathan Kozol article in Harper's is well-meaning. The education given to urban blacks is a crying shame. But Kozol's article is emotionally manipulative and ignores relevant facts and research.

For example, he spends a lot of time considering the problems faced by various New York City schools:
Dear Mr. Kozol," wrote the eight-year-old, "we do not have the things you have. You have Clean things. We do not have. You have a clean bathroom. We do not have that. You have Parks and we do not have Parks.

You have all the thing and we do not have all the thing. Can you help us?"

The letter, from a child named Alliyah, came in a flit envelope of twenty-seven letters from a class of third-grade children in the Bronx. Other letters that the students in Alliyah's classroom sent me registered some of the same complaints. "We don't have no gardens," "no Music or Art," and "no fun places to play," one child said. "Is there a way to fix this Problem?" Another noted a concern one hears from many children in such overcrowded schools: "We have a gym but it is for lining up. I think it is not fair." Yet another of Alliyah's classmates asked me, with a sweet misspelling, if I knew the way to make her school into a "good" school—"like the other kings have"—and ended with the hope that I would do my best to make it possible for "all the kings" to have good schools.

* * *

In the years before I met Elizabeth, I had visited many other schools in the South Bronx and in one northern district of the Bronx as well. I had made repeated visits to a high school where a stream of water flowed down one of the main stairwells on a rainy afternoon and where green fungus molds were growing in the office where the students went for counseling. A large blue barrel was positioned to collect rain-water coming through the ceiling. In one makeshift elementary school housed in a former skating rink next to a funeral establishment in yet another nearly all-black-and-Hispanic section of the Bronx, class size rose to thirty-four and more; four kindergarten classes and a sixth-grade class were packed into a single room that had no windows. The air was stifling in many rooms, and the children had no place for recess because there was no outdoor playground and no indoor gym.
It goes on and on in that vein. Then Kozol finally discussed the actual levels of spending in the above schools:
The present per-pupil spending level in the New York City schools is $11,700, which may be compared with a per-pupil spending level in excess of $22,000 in the well-to-do suburban district of Manhasset, Long Island. The present New York City level is, indeed, almost exactly what Manhasset spent per pupil eighteen years ago, in 1987, when that sum of money bought a great deal more in services and salaries than it can buy today. In dollars adjusted for inflation, New York City has not yet caught up to where its wealthiest suburbs were a quarter-century ago.
Hold on for a second there. Per-pupil spending has ranged from $8,000 in 1997-98 to $11,700 today. Yes, this is less than the $22,000 per pupil that is spent in one suburban district that Kozol cherry-picked. But how does it compare to the national and New York average?

Pretty well. The most recent statistics I could find show that the national (unadjusted) average per-pupil spending in 2000-01 was $8,859, or nearly three thousand dollars less than the Bronx schools. The New York average in 2000-01 was $11,938.

While this doesn't prove that the level of spending in the Bronx is adequate, it does raise the question: Are these schools really suffering from a lack of funds? Or does the real problem lie elsewhere -- mismanagement, bloated bureaucracies, etc.? Someone who is familiar with the reputation of the New York school system might suspect one of the latter explanations.

Also, as a more general matter, Kozol's habit of arguing by cherry-picked anecdotes isn't very reliable. In "Disparities in Public School District Spending, 1989-1990," the National Center for Education Statistics put out what is still the most recent study that mapped census data nationwide against school district spending levels. According to the executive summary:
More money is spent in districts with the highest percentages of minority students compared to districts with the lowest percentages of minority students ($4,514 versus $3,920). Although minority students in poverty are often viewed as those least served by current systems of public education funding, these findings suggest that while inequalities may remain for students in poverty, they do not appear to be driven by minority status.
There's no reason to think that the financial situation has changed for the worse since 1990, i.e., that minority school districts receive less per-pupil funding compared to non-minority districts.

Back to Kozol:
In Milwaukee, for example, virtually every four-year-old is now enrolled in a preliminary kindergarten program, which amounts to a full year of preschool education, prior to a second kindergarten year for five-year-olds. More commonly in urban neighborhoods, large numbers of low-income children are denied these opportunities and come into their kindergarten year without the minimal social skills that children need in order to participate in class activities and without even such very modest early-learning skills as knowing how to hold a crayon or a pencil, identify perhaps a couple of shapes and colors, or recognize that printed pages go from left to right.
If five-year-olds don't know how to hold a crayon, identify any shapes or colors, etc., they are facing a problem that is much more severe than the lack of universal pre-school, namely, a lack of minimally competent parents. Of course, that might make pre-school all the more necessary in such situations. (But it might not: Some recent research suggests that even kindergarten is not helpful in raising student performance in the long term. See here and here.)

Perhaps the most egregious passage is this:
Perhaps in order to deflect these recognitions, or to soften them somewhat, many people, even while they do nor doubt the benefit of making very large investments in the education of their own children, somehow — paradoxical as it may seem — appear to be attracted to the argument that money may not really matter that much at all. No matter with what regularity such doubts about the worth of spending money on a child's education are advanced, it is obvious that those who have the money, and who spend it lavishly to benefit their own kids, do not do it for no reason. Yet shockingly large numbers of well-educated and sophisticated people whom I talk with nowadays dismiss such challenges with a surprising ease. "Is the answer really to throw money into these dysfunctional and failing schools?" I'm often asked. "Don't we have some better ways to make them `work'?" The question is posed in a variety of forms. "Yes, of course, it's not a perfectly fair system as it stands. But money alone is surely not the sole response. The values of the parents and the kids themselves must have a role in this as well you know, housing, health conditions, social factors." "Other factors"—a term of overall reprieve one often hears—"have got to be considered, too." These latter points are obviously true but always seem to have the odd effect of substituting things we know we cannot change in the short run for obvious solutions like cutting class size and constructing new school buildings or providing universal preschool that we actually could put in place right now if we were so inclined.
There's quite a bit of scholarly literature on the effects of school spending on student performance. In fact, the Coleman Report ("Equality of Educational Opportunity") -- one of the biggest social science studies in history -- famously found that school spending didn't have much relation to student performance. Far more important were the characteristics of the student body as a whole. As the Coleman Report found, "the social composition of the student body is more highly related to achievement, independent of the student's own social background, than is any school factor."

There are lots of other studies -- some showing modest increases in performance from increased spending and some showing that increasing spending after a certain point has little or no effect. The most that one can say is that there are conflicting studies on the issue. And it is certainly not the sort of question that can be resolved by rhetoric to the effect that rich people "do not [send their kids to expensive schools] for no reason." Rich people buy lots of different things for their kids, but that doesn't guarantee that any one thing has an effect on academic performance.

Kozol ends the paragraph by saying that "cutting class size" is one of the no-brainer ideas that we "could put in place" if we wanted. Well, not so fast. Some research suggests that broad measures to reduce class size actually end up harming inner-city schools. For example, Jepsen and Rivkin found that when California tried to reduce class sizes, this meant hiring many more teachers, some of whom were less qualified or experienced. Plus, some inner-city teachers ended up moving to suburban schools to fill the new vacancies there. As a result, inner-city schools in California had fewer good teachers than ever, thereby shortchanging their students on the one factor that probably makes the most difference.

Finally, Kozol spends a good deal of time describing a "scripted teaching system" used in some inner-city schools. The passage is written with a sneering tone towards the very notion of a scripted and regimented program. Kozol concludes this passage by saying that such programs are "desperation strategies that come out of the acceptance of inequity," and that "[i]f we did not have a deeply segregated system in which more experienced instructors teach the children of the privileged and the least experienced are sent to teach the children of minorities, these practices would not be needed and could not be so convincingly defended."

What Kozol doesn't mention -- though it appears in a New York Times article that he links to -- is that "many of the schools' scores have climbed steadily since the program was introduced." (Perhaps that is what he means by his suggestion that the program is "convincingly defended.")

So Kozol's complaint here boils down to something like this: Because of de facto segregation, teachers are forced to rely on teaching methods that actually work, rather than having the freedom to experiment with less successful methods. This is a bizarre complaint.



Blogger P. G. said...

On the expenditures per student thing... uh, but... costs are higher in New York city! You can't compare the expenditure in NYC to the national average where that national average includes almost all communities where the prevailing wages (due to cost of living) are much lower, the cost of land is much lower, etc.

3:10 PM  
Blogger P. G. said...

oh, and as for the scripted learning thing, if, as Kozol reports, it relies on Skinnerian approaches, then this is a serious problem regardless of test scores. B.F. Skinner and his acolytes were only interested in changing behavior, not imparting knowledge. (See e.g. "Beyond Freedom and Dignity.") With such a theoretical background, I doubt that test scores (which would measure, perhaps, rote application of techniques and/or memorization of facts) would attest to the effectiveness of the teaching method for producing *understanding* rather than just spouting of data.

3:28 PM  
Blogger Stuart Buck said...

On expenditures -- yes, NYC is more expensive. Still, an extra $3,000 per student per year translates to an extra $60,000 per classroom (assuming 20 students per class). Assume that the teachers are paid $30,000 more than the national average (a tidy sum). Then that still leaves $30,000 per classroom per year to go towards extra building expenses or whatnot. If a school has 40 classes, that's an extra $1.2 million per year. Nothing to sneeze at, even if NYC does cost more. (These are just guesses on my part, of course, but the point is that $3,000 per student could go a long way.)

In my limited understanding, a scripted program for teaching reading would have a strict regimen for teaching phonics, various words, etc., etc. I don't know that it makes any sense to say that one reading program emphasizes "data" over "understanding," or vice versa.

8:50 PM  
Blogger P. G. said...

Fair enough on the money, although I think there are a few other expenses to take into account when one is dealing with poor and/or urban kids: increased reliance on subsidized school lunches, increased reliance on school buses, etc.

In terms of the scripted learning thing, my point is that there's already a real question about pointing to improved test results as evidence for increased learning. (Teachers who "teach to the test," ability to remember and repeat information but not critically consider or synthesize it, etc.) That question can only be made more nagging when the teaching is scripted, which certainly raises the suspicion that the script is modeled to the test as well as making one worry about the lack of an option for teachers to provide context for the material, adapt it to different students, etc. The question becomes impossible to ignore when one learns that the scripting is modeled after the theories of a pyschologist who professed complete unconcern with the contents of people's heads.

9:00 AM  

Post a Comment

Subscribe to Post Comments [Atom]

<< Home