Affirmative Action for School Reformers?

Affirmative action is a policy that does not poll well these days. Large majorities typically reject the idea that hiring or school admissions should take into account the discrimination that African-Americans, women, and other groups face. Most people will argue that we should have a strictly merit-based system with people evaluated on their records.

It might seem like a good idea that people stand or fall on their track record, but that doesn't seem to fit the United States today. It's certainly hard to maintain that standing in my profession of economics bears any relationship to performance. Nearly everyone in a top policymaking position today was among the group of economists who could not see the $8 trillion housing bubble that wrecked the economy when it collapsed

Economics is clearly not the only field where failure seems the best track forward. The top ranks of the so-called school reform movement are packed with people who have prospered in spite of failure.

To see this point, let's start at the top. Last fall, Thomas Friedman devoted one of his columns to the failure of the US education system. The frame is a trip to Chicago where he met with its new mayor, Rahm Emanuel and several business executives, all of whom complained about how the city's schools were not producing graduates who were qualified to fill the jobs that were being created. (There is good reason to question their assertions.)

Friedman bemoaned the fact that the kids in the Chicago school system couldn't get the skills needed to work in a modern factory. He told readers Mayor Emanuel has plans to fix the system, but it would take time.

Those wondering what sort of dunderhead left Chicago's schools in such awful shape need look no further than the Secretary of Education, Arne Duncan. Duncan headed up Chicago's school system from June of 2001 until he left to join the Obama administration in January of 2009. If Chicago's schools really are as bad as Friedman claims, then Duncan would certainly have to take much of the blame.

Duncan is not the only school reformer who seems to have done well in spite of a mixed track record. Washington's former school Chancellor Michelle Rhee seems to be doing quite well for herself as head of Students First, an organization that she founded after leaving her position with the Washington school system.

Rhee had taken open pleasure in firing teachers who she considered incompetent, largely based on student test scores. Her performance as chancellor had been the main issue in the incumbent mayor's re-election campaign in 2010. When the mayor lost, she resigned to set up her new organization, which claims to be advising school systems around the country.

Shortly after leaving her position as chancellor, USA Today found serious evidence of cheating on the city's standardized tests. Rhee responded angrily when questioned about the evidence, claiming that it was a plot to undermine school reform. In Rhee's defense, Washington Post columnist Jay Mathews commented that Rhee probably had not had much chance to give the city's test security serious thought.

While this comment was intended as a defense of Rhee by a supporter, this assessment is in fact incredibly damning. There had already been a series of prominent cheating scandals in school systems around the country by the time that Rhee took her job in DC. There is obviously serious motivation to cheat on tests where the results can either lead to large bonuses or firing. If Rhee had not implemented strong measures to protect against cheating, then clearly she was in way over her head.

The third high-profile example of a school reformer who does not seem to be making the grade is New York City Chancellor Dennis M. Walcott. Mr. Walcott apparently thought it was a clever idea to make public measures of teacher value added by school class. This measure is supposed to show how much a set of students improved their test scores under a teacher, after controlling for various demographic characteristics.

The problem with these scores is that they are highly erratic, with even the same teachers having enormous variations in their scores from year to year. Some of this may reflect the fact that teachers, like everyone else, have good years and bad years; but much of the variation is undoubtedly due to random factors that are outside of the teachers' control.

In fact, it turns out that there is enormous variation in results even for the same teacher in the same year, in some cases teaching the same subject. This graph matches the results for teachers who taught the same subject to two different grades (e.g. fifth grade and sixth grade math). There is almost no correlation in results, with many teachers scoring near the top in one class and near the bottom in the other class.

Chancellor Walcott made this information available with the idea of increasing the accountability of teachers, but the lack of accountability is on his part. How many parents will be outraged that their child was in a class where the teacher ranked near the bottom by this value-added measure, when it turned out that by the same measure the teacher had consistently been an outstanding performer in other classes?

If Walcott was being responsible, he would have made a point of informing parents and others of the limitations of these tests, which have long been known to serious scholars. But providing information does not appear to have been the purpose of this exercise. Mayor Bloomberg and other advocates of school reform are trumpeting the release of this data as another victory for the school reform movement over incompetent teachers and teachers' unions.

The rich and powerful have lined up firmly on the side of school reform, which is defined as a system where teachers lack job security and standardized testing becomes all important. Having well-educated students seems to be secondary in this story. And those who do the dirty work – in Chicago; Washington, DC; New York City; and elsewhere – can count on handsome rewards, even if their work doesn't pass muster by most standards.

This article is not covered by Creative Commons policy and may not be republished without permission.