Education - Mayoral control is NOT the answer

cross posted from Daily Kos

I was not a supporter of the selection of Arne Duncan as Secretary of Education, but I was willing to withhold judgment, to see where he would attempt to take the nation in education policy. I thought perhaps policy might be made in the White House, with him serving as the public face. I was wrong. Duncan is attempting to drive education in ways that will destructive. Many of the policies he is pushing demonstrate his fundamental lack of understanding.

Today I will briefly explore the issue of mayoral control of big city school systems. Remember, such is Duncan's experience in years in Chicago. He started as an assistant to Paul Vallas in a system controlled directly by Mayor Richie Daley, succeeding him in that position for a number of years before being tapped by his basketball buddy, the new President, to head our national educational efforts.

In this exploration I am going to rely on an op ed in yesterdays New York Times entitled Mayor Bloomberg’s Crib Sheet, by Diane Ravitch.

Diane Ravitch is currently a research professor of education at New York University. One of her books is The Great School Wars: New York City, 1805-1973. Trained as an educational historian, she served as Assistant Secretary of Education and Counselor to Secretary of Education Lamar Alexander from 1991 to 1993, where she was responsible for the Office of Educational Research and Improvement in the U.S. Department of Education. As Assistant Secretary, she led the federal effort to promote the creation of state and national academic standards. She holds positions as a senior researcher simultaneously at Hoover and at Brookings. She has become an outspoken critic of No Child Left Behind. While I do not always agree with her, I consider her a friend. And before we start with her op ed, I have to put you on notice: she is NOT a fan of Duncan, having recently described him as "Margaret Spellings in drag."


Ravitch begins by noting Duncan's call for mayors to take control of the nation's school and of his pointing at New York City as an example. She then writes

Actually, the record on mayoral control of schools is unimpressive. Eleven big-city school districts take part in the federal test called the National Assessment of Educational Progress. Two of the lowest-performing cities — Chicago and Cleveland — have mayoral control. The two highest-performing cities — Austin, Tex., and Charlotte, N.C. — do not.

Stop for a moment, remember that Chicago has had mayoral control of its schools since Paul Vallas was put in place by Daley in 1995, with Duncan succeeding him in 2001. The National Assessment of Educational Progress (NAEP) is considered the best single, independent measure of school performance we have. Let me quote from the linked Wikipedia article to provide a bit of context:

NAEP conducts assessments periodically in mathematics, reading, writing, science, and other areas.[1] New assessments in world history and in foreign language are anticipated in 2012.[2]
NAEP is administered by the National Center for Education Statistics (NCES), a division of the US Department of Education.
Since NAEP assessments are administered uniformly to all participating students using the same test booklets and identical procedures across the nation, NAEP results serve as a common metric for all states and selected urban districts that take the assessment. The assessment stays essentially the same from year to year, with only carefully documented changes. NAEP reports all results at the national level and provides state results for some assessments. On a trial basis, NAEP is releasing the results for a number of large urban districts.
NAEP results are based on representative samples of students at grades 4, 8, and 12 for the main assessments, or samples of students at ages 9, 13, or 17 years for the long-term trend assessments. These grades and ages were chosen because they represent critical junctures in academic achievement. NAEP provides data on subject-matter achievement, instructional experiences, and school environment for populations of students (e.g., all fourth-graders) and groups within those populations (e.g., female students, Hispanic students). NAEP does not provide scores for individual students or schools, although state NAEP can report results for selected large urban districts.

Educational researchers consider the main NAEP the best single indicator of educational performance over time. There is somewhat less confidence in the accuracy of what is known as state NAEP, especially since participation became mandatory in 2001 with the passage of No Child Left Behind. State NAEP scores provide a check on claims by states for improvement on their own state assessments, those state assessments being used to determine Adequate Yearly Progress under NCLB.

Two quick comments about what NAEP has shown before we return to Ravitch. First, an examination of NAEP scores completely destroys the idea of any Texas miracle in education during the 6 years G. W. Bush was governor - and remember, it was that Texas miracle that was used to sell the nation on NCLB. The Nation's Report Card, as NAEP is sometimes described, showed no improvement for Texas in the 1990s, and has shown little improvement in the 6+ years since NCLB went into effect. Second, Duncan spent 7 years in charge of Chicago schools in a system of mayoral control that predated him by another 6 years. Vallas was cited by Clinton for raising test scores, but (a) the scores that were raised were a selective set of Illinois tests, not consistent across all of the state tests, and the city showed little progress on NAEP. As we return of Ravitch remember her point - that of the 11 urban districts participating in NAEP for separate scoring, the two lowest scoring were under mayoral control while the two highest were not. And Chicago, after 13+ years of mayoral control, including more than 7 under Duncan, was at the bottom.

I cited the one book by Ravitch because writing it provided her with probably more knowledge about the history of schools in New York City than anyone else in the country. Diane was trained as an educational historian, and IIRC, her dissertation was supervised by perhaps the greatest historian of education we have had, Lawrence Cremins. While I will sometimes disagree with the conclusions she draws, she is a solid researcher on educational history. When evidence proves her previous ideas to be inaccurate, she will acknowledge and correct them, as she is doing in the book on which she is currently working.

Duncan recently came to New York to urge renewal of the state statute, passed in 2002, that gives the mayor control of the schools in New York City. That law expires at the end of this school year. Ravitch points out two key things to know about NYC schools

1. Mayoral control is nothing new: "From 1873 to 1969, the mayor appointed every single member of the Board of Education. The era of decentralization from 1969 to 2002 was an aberration, because the mayor had only two appointees on a seven-member board."

2. The control over schools Bloomberg currently has is unrivaled in the city's history, with previous mayors respecting the independence of school board members they appointed. By contrast, "The present version of the board, the Panel on Education Policy, serves at the pleasure of the mayor and rubber-stamps the policies and spending practices of the Department of Education, which is run by Mayor Bloomberg and Schools Chancellor Joel Klein."

Let me deviate from Ravitch a bit. One of the ironies of mayoral control has been the pattern of appointing people to run schools who really lack the background as professional educators one might expect. I teach in Maryland. A superintendent must meet certain qualifications in order to head one of our 24 school divisions (23 counties and the City of Baltimore). One of those requirements is a doctorate in education, although that State Superintendent can waive some of the requirements (and Superintendent Nancy Grasmick, who herself started as a teacher, has done so). Ever since Seattle experienced some success with hiring a non-educator to run their schools, mayors and governors have somehow thought such an approach was the solution to the seemingly intractable problems of urban education. But retired Maj. Gen. John Stanford was sui generis, and the success he had in Seattle has not been duplicated by similar appointments, whether of Generals (Julius Becton in DC), former Governors (Roy Romer in Los Angeles), financial managers (Paul Vallas in Chicago, Philadelphia and New Orleans), or lawyers (Duncan in Chicago and Joel Klein in New York). [Michelle Rhee in DC did spend several years in a classroom with Teach for America, during which time by her own admission she was a lousy teacher until near the end of her second year. Her subsequent experience was running The New Teacher Project, a non-profit that was one of many spinoffs from the TFA family. Her highest degree is a Masters in Public Policy]

Ravitch - and remember her background and her responsibilities in the US Dept of Education - examines the claims of supporter of the Bloomberg-Klein regime of spectacular improvements. They argue for approval without change of the current law. She quotes Sec. Duncan

I’m looking at the data here in front of me,” he said while in New York. “Graduation rates are up. Test scores are up ... By every measure, that’s real progress.”

Except that claim is unsupported by independent measures:

On the federal National Assessment of Educational Progress — widely acknowledged as the gold standard of the testing industry — New York City showed almost no academic improvement between 2003, when the mayor’s reforms were introduced, and 2007. There were no significant gains for New York City’s students — black, Hispanic, white, Asian or lower-income — in fourth-grade reading, eighth-grade reading or eighth-grade mathematics. In fourth-grade math, pupils showed significant gains (although the validity of this is suspect because an unusually large proportion — 25 percent — of students were given extra time and help). The federal test reported no narrowing of the achievement gap between white students and minority students.

When supporters of the Klein regime try to point to scores on state tests, which have improved, Ravitch responds:

indeed, the state scores have soared in recent years, not only in the city but also across New York state However, the statewide scores on the N.A.E.P. are as flat as New York City’s. Our state tests are, unfortunately, exemplars of grade inflation.

She also points out how other measures, such as graduation rates reported by the city schools, do not indicate improvement:

The city says the rate climbed to 62 percent from 53 percent between 2003 and 2007; the state’s Department of Education, which uses a different formula, says the city’s rose to 52 percent, from 44 percent. Either way, the city’s graduation rate is no better than that of Mississippi, which spends about a third of what New York City spends per pupil.

Moreover, the city’s graduation rates have been pumped up with a variety of dubious means, like “credit recovery,” in which students who fail a course can get full credit if they agree to take a three-day makeup program or turn in an independent project. In addition, the city counts as graduates the students who dropped out and obtained a graduate-equivalency degree.

Let me step back for a moment. First, remember the requirement of NCLB to participate in NAEP. This was required precisely to serve as a check on state's manipulating their own tests to "show improvement." One can establish a first year cut score (the raw score which represents passing) to show a low pass rate, then lower the cut score to show" "improvement" even if the raw scores have not changed. I experienced that in the one year I taught middle school in Virginia. The year before I arrived our school had a 58% pass rate on middle school American History. The year I was there, with the other two teachers being first year teachers and me being new to the curriculum, our pass rate was 81%, which seems to be a spectacular improvement. Except that the cut scores were changed to have a more acceptable passing rate - if we had restated the previous year's scores according to the new pass rate, it would have been about 71-72% - we improved, but not that much. And of course, we were comparing two different cohorts of students.

The manipulation of graduation rates is a well-known phenomenon. We saw it in Texas during the tenure of Gov. Bush, especially in Houston under Rod Paige. Students would be held back, sometimes more than once, in th grade (because the Texas tests were in 10th grade), until they dropped out, then they would not be listed as a drop out if you could get them to say they might go eventually for a GED, instead being listed as transferring to an alternative educational program. All this was in this professional literature in work by Walt Haney BEFORE NCLB was passed into law near the beginning of Bush's first term.

Let's return to Ravitch. She notes that the NY figures do not include as dropouts those listed as discharged during their hs years:

Some discharges are legitimate, like students who moved to another school district. But many others are so-called push-outs, students who were ejected from school even though they had a legal right to be there, often because their grades and test scores were bringing down their schools’ averages. The Department of Education refuses to disclose how many students are in each of these categories. We do know, however, that more than one-fifth of the members of the class of 2007, or 18,524 students, were discharged and not counted as dropouts.

One point to bear in mind is that Ravitch is not totally opposed to some level of mayoral involvement in the governance of schools. She is opposed to the model one sees in NYC, in which there is no oversight of the actions taken by the mayor and his designee, and hence no public participation in s school governance. She is willing to have the mayor appoint the members of the Board of Education for fixed terms,

Candidates for the board should be evaluated by a blue-ribbon panel so that no mayor can stack it with friends. That board should appoint the chancellor, and his or her first responsibility must be to the children and their schools, not to the mayor.

What a remarkable idea - the head of the school system has as first responsibility the children. If one returns to the history of Paul Vallas, for example, one finds him finishing second in 2001 (to Blagojevich and ahead of Burris) for the Democratic nominee for Governor of Illinois, has since considered running again in 2010 and has announced that he plans to run this year for the Cook County Board as a Republican. Reasonable people might well question his dedication to the children that should have been his primary responsibility.

Ravitch believes that school boards need to make their decision in public, subject to public scrutiny. She further advocates for some level of parental control, writing

Local school boards composed of parent leaders should oversee the schools in their districts, although they should not have any financial authority.

She wants independent auditing to evaluate claims of improvements in test scores and graduation rates. The current New York law has none of these features. Instead all power resides within the hands of a chancellor / ceo, Joel Klein, who is answerable only to the mayor. So far that model has not proven successful, and yet that is what Duncan wants to propagate across the nation, perhaps because that is his own personal experience, an experience which has not shown positive results.

If our schools are truly public schools, they should be answerable to the public. Their governance should be democratic. The model of mayoral control, especially as implemented in New York City, meets neither of these criteria. By itself that should be sufficient reason to reject that model of governance. The model is further undercut by the lack of success that can be demonstrated by independent evaluation, not only in New York, but also in Chicago under the leadership of Duncan and of his predecessor.

Let me offer the final paragraph penned by Ravitch in this piece:

Not every school problem can be solved by changes in governance. But to establish accountability, transparency and the legitimacy that comes with public participation, the Legislature should act promptly to restore public oversight of public education. As we all learned in civics class, checks and balances are vital to democracy.

checks and balances are vital to democracy - we have just escaped from an 8 year administration that did not believe it should be subject to checks and balances, and we came close to destroying our economy and our international standing as a result of actions taken without such checks and balances. If nothing else, we should have learned that no public function can be trusted to people who are not subject to checks and balances. Our public schools should be preparing our children not only to be employed, but to be participating citizens in a representative liberal democracy. The model of governance advocated by Duncan is opposed to that. By itself that should be sufficient reason to reject it. And it does not work, as both his experience in Chicago and the tenure of Joel Klein in New York demonstrate.

Peace.