Debating Michelle Rhee
by Richard Kahlenberg
Earlier this week, I reviewed in Slate Magazine a new biography of former D.C. Schools Chancellor Michelle Rhee called The Bee Eater, written by former USA Today editorial writer Richard Whitmire. In my review, I noted that Rhee is wildly popular among members of the elite media, even though there is very little evidence to back up Rhee’s claim that low quality teachers, and their union protectors, are bigger impediments to equal educational opportunity than poverty and segregation. I was critical of Whitmire’s highly favorable treatment of Rhee’s tenure.
In the old days, authors took their lumps when their books were reviewed, but now they routinely respond, which is a healthy development. Whitmire, who is a good and clear writer, responded to my review on his Bee Eater blog. He makes some interesting points and raises some good questions, which I’ll try to respond to here.
The two central questions posed by Whitmire’s book are: Why do D.C. schools perform so much worse than many other big city school districts? And was Rhee’s the right approach to improving the schools?
Why Do D.C. Schools Perform Poorly?
In comparing school districts, virtually every education researcher will point to poverty levels and segregation as the key factors in why certain districts do better than others. Districts are made up of schools, and schools which have high levels of poverty consistently perform far worse than those with lower levels of poverty. These differences outstrip the much-discussed achievement gap between the U.S. and other countries. For example, on the recently-released 2009 Programme for International Student Assessment (PISA), Stephen Krashen, professor emeritus at the University of Southern California, notes that 15-year-old Americans in low poverty schools (those with less than 10 percent of students eligible for free or reduced price lunch) scored 551 on reading, higher than the overall average of any participating country. By contrast, students in schools with more than 75% of students from low-income families, scored 446 on average, second to last among the 34 nations in the Organization of Economic Cooperation and Development (OECD).
The biggest reason that D.C. scores poorly overall is that it has higher numbers of students coming from low-income backgrounds and attending high poverty schools than most other American school districts. But Whitmire then narrows the question to ask, if we take poverty and race off the table, why do poor black students in D.C. perform two years behind poor black students in places like New York?
For Whitmire the answer is obvious: teacher quality. He says that “experts” inside and outside of the D.C. school system estimated that two-thirds of teachers needed to be fired when Rhee came in. He provides no explanation for how this astonishing figure was arrived at, nor does he provide a comparable figure for the percentage of teachers he considers sub-par in New York City. Whatever one thinks of Rhee’s IMPACT teacher evaluation program, it didn’t exist when she first came on so it can’t be the source of the two-thirds figure.
Whitmire spent many hours visiting Washington D.C. public schools – something I have not done – and so it should count for something when he indicates that based on his visits he found “no reason to dispute” the two-thirds figure. But I would be far more convinced if Whitmire’s observations and interviews were backed up by published research. I don’t think the evidence presented by Whitmire supports his declaration that Rhee “fired too few teachers, not too many.”
Whitmire is correct that teacher quality is very important, but other factors are also relevant – such as the standards and curriculum in a district, the availability of support services and pre-K programs, access to extended learning time, levels of segregation, class size in the early grades, the quality of professional development provided to teachers, and on and on. According to Diane Ravitch’s analysis of research, teacher quality, by itself, accounts for between 10 and 20 percent of overall achievement outcomes.
Was Rhee’s the right approach to reform?
Let’s assume, however, that Whitmire is right and the central reason that D.C. performs worse than systems like New York City's is that it has a lower quality of teacher on average. Does that vindicate Rhee’s approach?
Hardly. As I pointed out in the Slate review, there are good ways and bad ways to get rid of poorly performing teachers. Peer review programs can weed out low performers in a way that is considered fair and doesn’t demoralize large segments of the teaching population. Likewise, merit pay can be structured in a way that encourages, rather than discourages, the sharing of good teaching ideas. Rhee didn't go about her program of upgrading teacher quality in the right ways.
Moreover, once a school district removes poor performers and institutes a credible system of rewarding high performers, there is still the question of how to connect the best teachers with the disadvantaged students who need them most. In D.C., there is evidence that using Rhee’s own measure of teacher effectiveness – the IMPACT analysis – the strongest teachers are in the most affluent schools. According to the Washington Post, the District’s most affluent ward has four times as many “highly effective” teachers as the poorest ward.
Rhee instituted a good and progressive program to provide the biggest bonuses to highly effective teachers who go into higher poverty schools. But there are limits to this approach. Research finds that teachers care more about working conditions than salary, which is why bonuses have to be very large – as much as 40% of salary – to effectively keep great teachers in high poverty schools for sustained periods of time.
A more cost-effective way to connect strong teachers and low-income students is to integrate schools by economic status. That’s part of why I wish Rhee had followed through on her good and productive efforts to attract more middle-class families to the public schools with a conscious strategy of school integration that would reassure low-income families that the middle-class influx would help, not hurt, them. It’s telling that in a recent interview with Seyward Darby of The New Republic, Rhee declined to spell out a position in favor of integration. Integration is “a very tricky situation,” Rhee told Darby. “StudentsFirst is not at this point going to take a policy stance on the issue.”
In his blogpost, Whitmire suggests that Rhee had to be tough on teachers because “Michelle Lite,” which he defines as “Rhee’s reforms with more cooperation and smiles,” has not produced “Rhee-like gains” in other districts. But as I noted in the Slate review, recent research by Alan Ginsburg, the former Director of Policy and Program Studies at the U.S. Department of Education, finds that in fact the two prior superintendents in D.C., Paul Vance, and Clifford Janey, produced “Rhee-like gains” without all the fireworks. What does that do to Whitmire’s unsubstantiated theory that “Part of the resentment against Rhee in some parts of D.C. appears rooted in the fact that it took a Korean American to actually improve schools—after a long string of black schools chiefs produced no improvements”?