Diane Ravitch, “No Child Left Behind”, And The Racial Achievement Gap`s Kryptonite Cause
The Death and Life of the Great American School System: How Testing and Choice Are Undermining Education, the new book by veteran education historian Diane Ravitch, has received a lot of publicity for revealing her newfound doubts about the current conventional wisdom on K-12 education reform. Her summary:
“[The Bush Administration’s No Child Left Behind Act] introduced a new definition of school reform that was applauded by Democrats and Republicans alike. In this new era, school reform was characterized as accountability, high-stakes testing, data-driven decision making, choice, charter schools, privatization, deregulation, merit pay, and competition among schools.”
“It was ironic that a conservative Republican president was responsible for the largest expansion of federal control in the history of American education. It was likewise ironic that Democrats embraced market reforms and other initiatives that traditionally had been favored by Republicans.”
But, as demonstrated by our catastrophic experience with the bipartisan assault on traditional lending standards in the name of equalizing minority homeownership rates, whenever the Republicans and Democrats agree on something, you’d better watch out.
The Bush-Kennedy education consensus that pushed through NCLB launched a decade of sound and fury in public education, signifying, well, not much in terms of its oft-proclaimed goal: narrowing the racial gaps in school achievement. And the Obama Administration has largely gone along.
I will point out here, because no-one else (not even Ravitch) will do so, that this fundamental goal of No Child Left Behind—closing the gap between Non-Asian Minorities [NAMs] and whites/Asians—is wrongheaded.
Thus the current ambition of everybody who is anybody is to take the (roughly) half of the kindergarten population that is minority and raise their performance by (roughly) one standard deviation, while, hopefully, doing nothing to improve the performance of whites and Asians (because if you also improve performance by whites and Asians, then you can’t Close The Gap).
That’s a terrible objective. And it just can’t be achieved in any practical way because the racial achievement gap is based on the racial gap in average IQ. I offer my alternative objective at the end of this article.
Some cure-alls for ethnic disparities, such as “small learning communities,” have fallen out of fashion. But they are always replaced by new crazes, such as tracking down the best teachers and sending them to the slums to somehow work their magic.
“In 2008, the federal government’s education research division issued a report with four recommendations for ‘turning around chronically low-performing schools,’ but the report acknowledged that every one of its recommendations had ‘low’ evidence to support it.'”
That Department of Education document, Turning Around Chronically Low-Performing Schools, admits:
“Unfortunately, the research base on effective strategies for quickly turning around low-performing schools is sparse. The panel did not find any empirical studies that reached the rigor necessary to determine that specific turnaround practices produce significantly better academic outcomes.”
“It seems that the only guaranteed strategy is to change the student population, replacing low-performing students with higher-performing students. … Rather than “leaving no child behind,” this strategy plays a shell game with low-performing students, moving them out and dispersing them, pretending they don’t exist.”
Nonetheless, billionaires such as Bill Gates and Eli Broad have donated enormous sums toward implementing the last decade’s orthodoxy in its various manifestations. As Ravitch acidly observes, this “billionaire boys’ club” of philanthropists ignored the existing infrastructure of alternatives to urban public schools in favor of arrogantly thinking they should reinvent schools from the ground up.
Citing James S. Coleman’s research, Ravitch explains that
“Catholic schools have a wonderful record of educating poor and minority children in the cities. It is a shame that the big foundations have not seen fit to keep Catholic schools alive. Instead, they prefer to create a marketplace of options, even as the marketplace helps to kill off highly successful Catholic schools.”
For example, the Gates Foundation, excited over the chic leftist idea promoted by Bill Ayers and Barack Obama in Chicago in the 1990s that the problem with big city public high schools is that they are too big, spent almost $2 billion toward creating 1,500 new small high schools and “small learning communities“ within public high schools. Gates orated in 2005: “If we keep the [high school] system as it is, millions of children will never get a chance to fulfill their promise because of their zip code, their skin color, or the income of their parents.”
Thus Gates gave a million dollars to Bill Ayers’s brother Rick (who also spent years on the lam) so he could set up “small schools” within Berkeley High School to, in part, take students to Cuba to learn about “social justice.”
In 2009, Gates admitted that he’d largely wasted his donations on this small schools boondoggle.
Why didn’t the press and the think tanks point out clearly to Gates that he was being ripped off by charlatans?
Because, as Ravitch points out, the Gates Foundation had bought off most of the self-proclaimed experts—handing out $57 million to “advocacy groups” in 2005 alone:
“Never before was there a foundation that gave grants to almost every major think tank and advocacy group in the field of education, leaving almost no one willing to criticize its vast power and unchecked influence.”
(“Almost no one” is right. Personally, I’ve been criticizing the damage being done to children by the Gates Foundation for years. But, then, I’m not important enough to be bribed by Bill Gates, so why listen to me?)
Gates has now given up on “small learning communities” as so 2000s. Now he has jumped aboard the educational fad of the 2010s: Blame Teachers!
“It is amazing how big a difference a great teacher makes versus an ineffective one. Research shows that there is only half as much variation in student achievement between schools as there is among classrooms in the same school. If you want your child to get the best education possible, it is actually more important to get him assigned to a great teacher than to a great school.”
This unfalsifiable line of reasoning—“If students succeeded, it was the teacher who did it. If students got low scores, it was the teacher’s fault”—has swept wonkdom.
According to Ravitch:
“So, depending on which economist or statistician one preferred, the achievement gap between races, ethnic groups, and income groups could be closed in three years (Sanders), four years (Gordon, Kane, and Staiger), or five years (Hanushek and Rivkin). “
“Over a short period of time, this assertion became an urban myth among journalists and policy wonks in Washington, something that that ‘everyone knew.’ This particular urban myth fed a fantasy that schools serving poor children might be able to construct a teaching corps made up exclusively of superstar teachers, the ones who produced large gains year after year.”
The hot new idea embraced by the Obama Administration and the Gates Foundation is to develop statistical techniques to find effective teachers, so that they can be taken out of white schools and sent to black and Hispanic schools.
It’s crucial to keep in mind that these influential papers are not studies of actual success stories of school districts that closed the racial gaps. Nobody has done that. Instead, they are merely mathematical projections of what might happen if all else remained equal. The authors are just assuming that the effect seen in one year of a good teacher over a bad teacher can be multiplied by any number of years.
For example, Gordon, Kaine, and Staiger write:
“Therefore, if the effects were to accumulate, having a top-quartile teacher rather than a bottom-quartile teacher four years in a row would be enough to close the black-white test score gap.”
But that turns out to be a big “If.” Ravitch notes:
“The fact was that the theory had never been demonstrated anywhere. No school or school district or state anywhere in the nation had ever proved the theory correct. Nowhere was there a real-life demonstration in which a district had identified the top quintile of teachers, assigned low-performing students to their classes, and improved the test scores of low-performing students so dramatically in three, four or five years that the black-white test score gap closed.”
Ending the black-white disparity has been the Holy Grail of education reform since LBJ. Considering all the rewards that would befall any educator who could achieve it, you might assume that absence of evidence after all these decades is evidence of absence.
And, theory alone suggests we should be skeptical that the effects of star teachers would “accumulate” for three, four, or five years in a row. That’s due to one of the most famous concepts in economics: diminishing marginal returns.
Consider a hypothetical example from a different kind of teaching: golf instruction.
These days, I have the money and time to only play golf about twice a year. I now average about 40 strokes per round worse than the superstars of the game.
But imagine that I somehow convinced the swing coaches of Tiger Woods, Phil Mickelson, Pádraig Harrington, and Jim Furyk to drop their most famous clients and instead each work with me intensively for one year in succession.
Assume that during the first of these four years, Tiger’s new ex-coach Hank Haney helps me cut 10 strokes off my average score, from 108 to 98. Does that mean I would therefore be on track over four years to cut 40 strokes, all the way down to 68, and thus challenge my teachers’ former pupils for the green coat at the 2014 Masters?
Of course not.
That’s almost as mindless as saying that if I then got a fifth year of world-class golf instruction, I’d be averaging 58 strokes per round and winning every pro tournament by 20 strokes.
Similarly, by the logic of this latest schooling theory, if we gave blacks and Hispanics better teachers for not three, four, or five years, but instead six, eight, or ten years, then they’d be scoring twice as high as whites and Asians!
It’s easy for contemporary Americans to understand why the best imaginable teaching won’t erase the Superstar-Duffer gap in golf. Yet, apparently, it’s very difficult for most intellectuals to grasp why returns might diminish in attempts to close the white-Non-Asian Minority gap in school achievement.
That’s because the racial gap in academic achievement, and its unmentionable IQ cause, works on American intellectuals, liberal and “conservative”, like Kryptonite on Superman. It deprives them of all their powers, leaving them helpless as babies.
If you proclaim enthusiastically that sending the best teachers into the slums would close the Diversity Disparity, it serves as a sign that you aren’t one of those horrible heretics who thinks that genetics might play a role. It’s a way of saying: “Don’t do to me what you did to James Watson. I believe, I believe!”
The question we should be asking about this latest fad: why would we want to send the best teachers to teach the worst students?
In all non-politicized forms of instruction, you never hear anyone say anything analogous. Everybody simply assumes that the best teachers should work with the best students—that it is reasonable that Butch Harmon coaches Phil Mickelson rather me, that Phil Jackson coach Michael Jordan and Kobe Bryant rather than a junior high school basketball team. For 2400 years, it has been assumed that it was a good thing for all concerned that Plato had Aristotle as a pupil rather than some random dolt.
Similarly, superstar economics professor and liberal pundit Paul Krugman is never criticized for being a professor at Princeton. You don’t hear anyone say Krugman should quit Princeton and go teach at Passaic County Community College where he could be doing more good.
On the other hand, Ravitch strikes me as overstating her case that (in the words of her title) Testing and Choice Are Undermining Education. While they aren’t panaceas, I don’t see much proof in her book that high stakes testing and charter schools are particularly hurting public education, either.
I haven’t noticed that public schools did a worse job in the 2000s than in the 1990s. They almost certainly are doing better with the students they have to work with than in the 1970s, the dark period in U.S. K-12 education history that Charles Murray and Richard Herrnstein described in The Bell Curve as “The Great Decline”.
For example, in third party tests, the white-black racial gap has been slightly narrowing on the federal National Assessment of Educational Progress tests and slightly widening on the SAT. In other words, despite all the hullabaloo, the results are about the same.
What about charter schools? Ravitch emphasizes that by skimming off the inner city’s most diligent students, the much-celebrated Knowledge is Power Program chain of 60 charter schools [KIPP] leaves the nearby neighborhood public schools with even worse student bodies. Furthermore, KIPP can’t be mass-produced
KIPP provides a Marine Corps boot camp-like environment that appears to do some good for a small, self-selected segment of the hardest-working young people, just as Parris Island does some young men good. But mass-producing hundreds of Parris Islands wouldn’t solve our social problems. (“The many, the blasé, the Marines”—not much of a slogan!) Similarly, KIPP can’t ever become large enough, while staying true to itself, to move the needle much.
Yet, because they give “the few, the proud” a chance to get away from ghetto culture, I can’t oppose KIPP schools.
Similarly, we do need better statistics on school and teacher quality. But, as Ravitch demonstrates, we shouldn’t expect them to be the miracle cures that Bill Gates, Barack Obama, and Malcolm Gladwell presume.
As I’ve been pointing out since 1995, the traditional method of judging the job done by schools from their students’ test scores is self-evidently silly. Because scores are IQ-dominated, you’re mostly just measuring how smart the students were before they even enrolled.
Amazingly, over the last couple of years, my long-standing suggestion that we should instead measure “value added”—how much test scores go up relative to how smart the kids were when they started—has actually become fashionable. The Obama Administration is spending huge sums on capturing this kind of data via its Race to the Top slush fund.
Ravitch, however, contends that value-added statistics are too new, too iffy, and too manipulable to use in determining teacher pay and employment. Nobody knows yet, for instance, if a teacher would tend to average a higher value-added score if she takes on last year’s best or worst students. But when there’s salary on the line, teachers will quickly figure out how to game the system.
There’s a lesson here. Innumerate commentators, which is to say essentially all pundits and politicians, don’t realize that statistics don’t necessarily tell you what you think they do.
Consider sports statistics. One of the most famous numbers in American sports history is Wilt Chamberlain’s average of 50 points scored per NBA game in 1961-62 (including 100 in a single game). Nobody else has ever averaged 40. He also led the league in rebounding, the second most prominent statistic, with an astonishing 26 per game. So Wilt must have been, at minimum, the best player in the league that year, right?
Wrong. Scoring 50 points per game is more impressive to naïve stats fans today than it was to Wilt’s rivals and own teammates at the time. He lost the Most Valuable Player balloting in a landslide to Bill Russell, who averaged only 19 points per game. Wilt received just nine first place votes from his colleagues compared to Russell’s 51. Wilt loved winning statistical championships, but Russell loved winning team titles. (His Celtics won 11 in 13 years.)
If we had had today’s more sophisticated basketball statistics back in 1962, they might have been able to demonstrate quantitatively what was apparent to the players at the time: that Russell was better at helping his team win than Wilt was. It took a couple of generations to invent the requisite measures.
Moral from the study of sports statistics: it’s easy to come up with useful measures of small questions, but it requires long, hard work to devise valid overall summative ranking systems of who is better than whom—even though that’s what everybody craves.
Bill James, for instance, began revolutionizing baseball statistics in 1975. But it took him until 2000 to come up with his Win Shares for inclusive ranking of players.
And rating teachers is less like rating players and more like rating baseball team managers—an even larger challenge, one that baseball statistic fanatics have made only fitful progress in quantifying.
There’s another problem. While value-added ratings of teachers may be a good idea, there is simply no chance that they will be implemented honestly in Obama’s America (or George W. Bush’s, for that matter).
Coleman noticed this in his 1966 study of schools: black teachers averaged more years of formal education, but that didn’t have any correlations with student achievement. The one thing that did matter, besides students’ backgrounds, was the IQ of their teachers. But to avoid hurting black teachers, he left this crucial finding out of the Coleman Report.
Tellingly, and ominously, Arne Duncan, the Obama Administration’s Education Secretary who is promoting “value-added” statistics, has also recently announced a new Civil Rights crusade to demonize school districts whose policies appear to have disparate impact on protected minorities.
How are Duncan’s two contradictory campaigns going to be reconciled?
The same way that everything else is rigged in Obama/ Bush’s America: more quotas and more lies.
Ravitch ends her book with a call for a return to the quality neighborhood public schools she attended in Houston in the 1940s:
“Going to school is not the same as going shopping. Parents should not be burdened with locating a suitable school for their child. They should be able to take their child to the neighborhood public school as a matter of course …”
But in 2010, this is simply unrealistic. Massive demographic change, induced by federal immigration policy, has meant that the central problem with the neighborhood school is, typically, the neighborhood. As the real estate agents say when talking about “good schools,” “Location, location, location.”
We’re not supposed to talk about the downside of diversity, so the quality of chatter about schooling is low. Ravitch’s book is well above average, but it would have been even better if she didn’t have to tiptoe around this central fact.
So what goal do I propose instead of Closing The Gap?
My goal, instead, would be to raise the average performance of all racial groups by half a standard deviation.
In other words, both goals are intended to improve the national average by half a standard deviation—but the Gates-Obama-Bush-Kennedy consensus wants to do it entirely by raising the scores of the minority half.
Which objective sounds more achievable?
Mine, obviously, for two reasons:
- Diminishing marginal returns: a one standard deviation improvement is not merely twice as hard to accomplish as a half-standard deviation performance, it’s much harder.
- Real improvements tend to better everybody’s performance. For example, I can drive a golf ball farther off the tee than I could 15 years ago because driver technology has significantly improved. (Clubheads are approaching the size of toasters, so you can now take a wild swipe at the ball without fear of whiffing). But then, Phil Mickelson can also hit the ball farther, too. So the pro-hacker gap in driving distance hasn’t closed.
In summary: my aim is both more achievable, more fair, and more sensible than the Gates-Obama-Bush-Kennedy consensus.
And therefore, of course, it’s also much more unmentionable.