Did Achivement Gaps Grow in Arne Duncan’s Chicago?
Ezra Klein and Dana Goldstein observe that black-white achievement gaps, as measured on the National Assessment of Education Progress, generally increased during Arne Duncan’s tenure as head of Chicago Public Schools. I looked this up via the useful TUDA site and it’s true. At the same time, the evidence is overwhelmingly good. Of the five sets of metrics available (4th grade reading, 4th grade math, 8th grade reading, 8th grade math, and 8th grade reading) black scores improved on four metrics. And Hispanic scores improved on all five. The trouble is that the small white minority (about 10 percent) in Chicago public schools also improved and sometimes showed a larger increase than did black test scores.
Here, for example, is 4th grade math where you see that racial achievement gaps grew. On the other hand, scores for minority students went up . . they just went up more for white students:
I don’t think that’s necessarily the worst thing in the world. You see the same pattern for 8th grade math:
By contrast, in 4th grade reading you saw the same general upward trend but also a slight narrowing of the white/nonwhite gap:
In 8th grade reading, the whites stagnated while Hispanics improved so that gap narrowed. But black scores got worse. Since Hispanics improved by more than blacks declined, this constituted an overall improvement in average scores:
In terms of writing, 8th grade scores went up across the board in Chicago between 2002 and 2007, and racial gaps narrowed:
To me, this makes Arne Duncan look pretty good. It also exposes some conceptual problems with efforts to narrow the achievement gap. It would have been politically difficult for Duncan to somehow try to implement policies that were designed to prevent white students from improving their performance. Nor does it seems like deliberately thwarting the efforts of the highest-performing groups purely in order to close gaps makes much sense as a policy. I think about the most you can ask of a city superintendent is that achievement broadly increases — including for poor students and minorities — during his tenure. A federal policy maker, by contrast, has the ability to not only back policies that enhance achievement but also to back policies that substantially increase the volume of resources available to high poverty schools and, therefore, will plausibly have some gap-narrowing bite.
41 Responses to “Did Achivement Gaps Grow in Arne Duncan’s Chicago?”
Register
RSSLogin
Topic Cloud
ArchivesBlog RollAbout Matt YglesiasContact Me
Donate
December 17th, 2008 at 1:54 pm
Is the difference between a score of 211 and 212 the “same” as the difference between 212 and 213? These scales usually are only met to be interpreted as ordinal (more is better) and not as cardinal.
December 17th, 2008 at 2:02 pm
Looks like reading scores were down for blacks, regardless of any comparative measure. Isn’t that bad? I’d also like to know what those asterisks in the graphs are referring to.
December 17th, 2008 at 2:21 pm
I agree the best way is to allocate new funding disproportionately to low achievers. Then over time over-achieving Asians will be highly underfunded and without say funding for computers or smaller class sizes they will do just as well as the rest of the poor.
And best of all it will make going to expensive New York private schools even more rewarding (relatively). Those damned high-achieving Asians will never be a threat to us upper class folks again.
December 17th, 2008 at 2:41 pm
According to http://www.cps.k12.il.us/ataglance.html
the Chicago public schols are only about 8% white but 47% black. My guess is that the standard deviaiton for blacks is greater than whites. and makes the year to year change statistically insignificant.
December 17th, 2008 at 2:44 pm
I have brought up Somerby before and I do not want to sound like an apostle of his; but these tests do not inform us on Duncan. One does not have to be cynical to ask questions-did the Board create a mandate to teach the test at the expense of other curricula? Why the growing disparity between blacks/ hispanics and the rest? Four years is a short table that may be statistically worthless-what are the results for other years? One could go on but I think I made the point. To be sure, I am not saying that Duncan is unqualified- just that the data provided is tenuous.
December 17th, 2008 at 3:23 pm
Not to be too picky, but isn’t this just a general instance of the case that we should focus on improving the lot of the worst-off rather than reducing inequality per se? (It might even be worth counting the interests of those who aren’t worst-off as being worth something, rather than nothing or a negative amount.)
This sort of point applies beyond education to the economy as well.
December 17th, 2008 at 4:05 pm
Just speculating here, as I don’t know Chicago well, but perhaps there are demographic changes within the white population that are leading to higher achievement scores. What might be happening is that the city’s white population includes a diminishing proportion of working-class ethnics in the neighborhoods and an increasing proportion of affluent Stuff White People Like types in downtown condominiums.
December 17th, 2008 at 4:12 pm
Part of the difference in the test scores for black kids may be explained with an exodus of the black middle and working class out of Chicago. Leaving relatively poorer black kids behind.
There has been a large drop in the overall Chicago public school population since 2000 and it is almost entirely attributable to black students leaving the city and the system.
The number of hispanic students has increased, while the relatively small number of white students has stayed about the same.
December 17th, 2008 at 4:12 pm
Theoretically, there should be a point at which the advantaged group begin reaching diminished returns (with respect to teaching effort) in their progress while the disadvantaged groups do not. That would result in reduced disparity. It would also indicate that you are beginning to approach the limit of just how good a job you can do teaching.
December 17th, 2008 at 5:43 pm
I’m always wary of judging the success of a school system based on a couple of years of test scores. Bush’s first Sect of Ed was responsible for the “Houston Miracle”, which was later found out to be lots of smoke and mirrors. Of course, congress had already approved NCLB, which it was patterned after.
December 17th, 2008 at 5:44 pm
If Mr. Duncan was doing such a great job with the public schools that numerous affluent white parents who were paying to send their kids to private schools decided to send their kids to public schools, then the racial gap would increase.
December 17th, 2008 at 5:49 pm
In the Chicago Public School system, like most urban public school systems, you have basically three types of white students: ones who get into special programs, ones who live in the handful of mostly white neighborhoods, and ones from broken homes where mom can’t afford anything better.
If you were to improve the overall quality and discipline of the school system, there’s a huge untapped reservoir of potential white students, and they would tend to do better than the broken home white kids who are there now.
December 17th, 2008 at 5:52 pm
not convinced about the achievement gap, then look at chicago’s NAEP scores — CPS ranks low compared to other big cities on the TUDA comparison of NAEP scores.
December 17th, 2008 at 5:54 pm
The surest way to reduce the racial gap is to hit the white and Asian kids over the head with the a ballpeen hammer.
Wouldn’t it make more sense to junk the goal of reducing racial gaps in favor of a goal of helping each individual student attain more of his or her individual potential?
December 17th, 2008 at 5:59 pm
These changes, over a short time, are all within the standard error of measurement on these tests and really tell you precisely nothing. These aren’t the scores from sporting events – they are products created by deeply fallible measures. Small changes (within five points + or -) mean just about nothing.
December 17th, 2008 at 6:18 pm
These are NAEP scores. You might want to do a little research to find out how many students took this test/ Were they ALL students in Chicago, or just a select few?
December 17th, 2008 at 6:47 pm
Well, no. The charts include indicators of statistical significance — they’re the little stars (or lack of stars) next to the numbers. For example, the increase in the 8th grading writing scale score of black students from 128 in 2002 to 138 in 2007 was statistically significant. The increase in scores for white students from 165 to 170 was not.
December 17th, 2008 at 6:51 pm
Let’s stop to think about the dysfunctional incentives generated by the Bush-Kennedy NCLB mandate to narrow racial gaps. If you improved the Chicago School System overall, one effect would be to lure in to the public schools more of the children of affluent, well-educated whites who currently send their kids to private schools or move to the suburbs when their kids reach five. Making the Chicago public schools good enough overall to attract these kids would increase the black-white gap by bringing in more smart white kids. So, you better not do that!
That’s a pretty stupid disincentive to build into policy, but since nobody thinks intelligently about intelligence (or you might end up like James D. Watson), we get all sorts of stupid disincentives built into the system.
December 17th, 2008 at 6:57 pm
Conversely, here’s a surefire way to _reduce_ the white-black gap in public schools: lower the quality of the school system so drastically that the only white kids left in it are ones whose parents are too poor or drugged up or whatever to haul their kids out of it. For example, I believe that La Griffe du Lion estimated that the average IQ of white students in the Baltimore public schools was 86, which goes a long way toward closing the racial gap!
A much less destructive goal for public schools than worrying about racial gap is trying to add the most value to each student individually, aggregated across all individuals.
December 17th, 2008 at 8:21 pm
Steve Sailer: I’m pretty sure that NCLB concerns itself with absolute achievement scores by racial/ethnic subgroups (according to “proficiency” thresholds set by the state), not with comparisons between the subgroups.
I welcome being corrected if I’m wrong.
December 17th, 2008 at 8:43 pm
Test scores are a bullshit indicator of educational progress.
December 17th, 2008 at 8:44 pm
disclaimer: Steve Sailer is a racist megalomaniac hack.
that said, in this specific instance, he actually is right.
December 18th, 2008 at 9:35 am
RE: “I don’t think that’s necessarily the worst thing in the world.”
Of course, it’s not a bad thing.
We want the gap to close, sure. But, not by bringing whites down a notch. We want all outcomes to increase (now we can argue on how to measure, but this is what is available.)
Again, We want all outcomes to increase. Period.
December 18th, 2008 at 12:27 pm
My son attended elementary school in a DC suburban school district that promised they would get all children reading at grade level. In practice this meant they intended to get all children reading below grade level up to grade level and all children reading above level down to grade level. We moved to a different county.
March 2nd, 2009 at 4:53 am
levitraVery interesting site. Hope it will always be alive!
March 14th, 2009 at 5:11 am
Excellent site. It was pleasant to me.
xanax
March 22nd, 2009 at 10:24 am
buy viagra online
Excellent site. It was pleasant to me.
April 2nd, 2009 at 5:09 am
If you have to do it, you might as well do it right
buy cheap viagra
April 2nd, 2009 at 12:24 pm
So why would you say that, I mean does it really matter?
April 3rd, 2009 at 4:06 am
It is the coolest site,keep so!
cheap brand pfizer viagra
April 9th, 2009 at 5:25 am
Excellent site, It was pleasant to me. viagra