A few weeks ago, I wrote a response to Marc Tucker’s response to Diane Ravicth and Anthony Cody’s comments on Tucker’s proposal for fixing our national accountability system. My response was posted on Anthony Cody’s blog Living in Dialogue. Tucker followed with a response, in which he sort of agrees with my critique:

If [Tucker] believes “…that our test-based accountability system ‘is not only ineffective but harmful,’ he would logically suggest that system be abandoned.  Instead he tries to fix it and the fixes include more tests, more high stakes tests, and more standardized tests.

“Yup, that is what I proposed,” writes Tucker. I was happy to read that because my purpose was to to point out that his proposed fixes would result in “tests, more high stakes tests, and more standardized tests.” Thanks, Marc for the clarification.

But of course, we have different views on standardized tests, and more fundamentally what education means. Marc has much more faith in standardized testing than I do. He writes “I do not think all standardized tests are bad, nor did I ever suggest that I think they are.” I, however, don’t believe standardized tests have much value in improving education and have not seen many good examples of them in the world as he has.

Below is a repost of my original response.

————————

A Response to Marc Tucker’s Response to Diane Ravitch and Anthony Cody on Fixing Our National Accountability System

By Yong Zhao.

I was very impressed with Marc Tucker’s indictment of the test-based accountability system that has been in place for over a decade:

The test-based accountability system now universally mandated in the United States—a system that reflects in every way the blue-collar conception of teaching as an occupation—has had ten years to prove itself. The result is very low teacher morale, plummeting applications to schools of education, the need to recruit too many of our teachers from the lowest levels of high school graduates, a testing regime that has narrowed the curriculum for millions of students to a handful of subjects and a very low level of aspiration. There is no evidence that it is contributing anything to improved student performance, much less the improved performance of the very low income and minority students for which it was in the first instance created.

I was even more impressed with Marc Tucker’s stated philosophy: “But I am an empiricist. I am influenced by theory but most impressed by evidence.” I thought he has given up the idea of test-based accountability.

But then I was puzzled by his action. A reasonable person would assume, if he is a true empiricist, influenced by evidence and the evidence he has is that our test-based accountability system “is not only ineffective but harmful,” he would logically suggest that system be abandoned. Instead he tries to fix it and the fixes include more tests, more high stakes tests, and more standardized tests.

In the recently released report from the National Center for Education and the Economy (NCEE) Fixing Our National Accountability System, NCEE president Marc Tucker, provides a plan to improve America’s accountability system. However worded, the “new” plan does not get away from the old: it is still an accountability system for educators based on test scores of students. Moreover, there will be more tests than under No Child Left Behind (NCLB). The tests would carry high stakes not only for teachers and principals, but also for students, which is not the case under NCLB.

More Tests than NCLB

NCLB requires one annual test in math and reading for grades 3-8, one more during grades 10-12, and 3 science tests (each in elementary, middle, and high school), totaling 16 tests for K-12. States are also required to have a sample of 4th and 8th graders take the low-stakes NAEP.

Tucker’s new plan includes three high stakes tests for all students at 4th, 8th, and 10th grades, but expanded to more subjects: math, English, and “subjects like history, literature, science, social studies, music, and the arts.” Additionally, Tucker wants a first grade “Diagnostic Test” and “every other off year, the state would administer tests in English and mathematics beginning in grade 2, and, starting in middle school, in science too, on a sampling basis.” The total number of tests a child may experience from K-12 could be as many as 30, double what is under NCLB. And even though some are sampling tests, the results are reported (average for schools on public websites) and thus put pressure on the school for producing good results. Since a sample will be used and we don’t know which students will be included, the school is likely to feel the same as if all students were to take the test. This does not count the PISA, in which Tucker wants the Federal government to require all states to participate.

Higher Stakes

Under NCLB and Race to the Top, the standardized tests technically carry high stakes only for teachers and principals, who would be punished or rewarded in some form based on the test scores of their students. The tests do not have prescribed consequences for students. But under Tucker’s plan, “our students would take high stakes tests only three times in their whole school career.” Only three times may not sound much, but as long as the stakes are high, especially for students, it’s more than enough to do the damage. As I have written in my latest book Who’s Afraid of the Big Bad Dragon: Why China Has the Best (and Worst) Education System, China technically has two high stakes tests for students (for entrance to high school at grade 9 and one for college entrance at grade 12) but these two tests turn all educational practices of students, parents, teachers, and schools into test-preparation, resulting in the widely observed damages because as Anthony Cody wrote: “these tests have such huge stakes attached to them, the entire system revolves around them, and students’ lives and family incomes are spent on constant test preparation, in and out of school.”

Moreover, Tucker wants a system to bring even more competition and choice. He suggests that “the state would create a public web site on which would be posted the relevant results of” the tests, much like the MySchool website he admires in Australia. But the Website, allowing parents and the public to compare schools and make school choices based on results of Australia’s national test, has been heavily criticized. Education Minister of New South Wales Adrian Piccoli said that “it puts students under unnecessary stress and is a waste of money,” following Australia federal Minister of Education Christopher Pyne’s complaint about it “being used to name and shame schools.”

Requiring each and every state to participate in the PISA would cost the nation millions of dollars, but more important put states in a competition that will further damage America’s education as I have written in a series of blog posts How Does the PISA Put the World at Risk.

The Cause of the Tucker Inconsistency

Why does one who condemns test-based accountability system so much want more test-based accountability? The inconsistency exemplified by Marc Tucker does not make sense to me at all. Yet it is widespread so it must make sense in some way. I try to put myself in the shoes of Tucker and other similarly minded people and learned the chain of reasoning underlying their inconsistency:

Premise #1: Education quality matters to individual and national prosperity.

Premise #2: Education is a top-down process through which students are instilled the prescribed content and skills (curriculum) deemed universally valuable by some sort of authority.

Premise #3: Teachers and schools are responsible for the quality of education, i.e., instilling in students the prescribed knowledge and skills.

Premise #4: How well students master the prescribed knowledge and content is measured by tests.

Conclusion #1: Thus test scores measure the quality of education, and thus the capacity for individuals and nations to be economically prosperous.

Conclusion #2: American students have lower test scores on some international tests, thus American schools offer a lower quality education than countries with higher test scores.

Conclusion #3: Therefore, American teachers must be less effective than their counterparts in other countries.

Conclusion #4: Therefore, to prepare Americans to succeed in the global economy, American teachers and schools must be held accountable for improving the quality of education, which is to raise test scores (Tucker’s goal: “the only acceptable target for the United States is to be among the top ten performers in the world” [I assume top 10 on the PISA league table]).

Conclusion #5: Hence we must improve the test-based accountability system, which then leads to higher quality education, which then leads to economic prosperity.

Bait and Switch

Marc Tucker’s objection to Anthony Cody’s questioning his assertion that “the economic future of our students will only be guaranteed if we educate them better” is a standard bait-and-switch tactic, playing with the afore-mentioned logic. It starts with the premises. Education is a term that has a positive connotation, but in practice it has many different, sometimes, contradictory, incarnations, in the same way the word “democracy” is used in reality. For example, some of the worst dictatorial countries claim to be democratic. Thus whether education matters to the prosperity of individuals and nations depends entirely on what it means.

If it means imposition of prescribed knowledge and skills as Tucker seems to suggest, the question would be if the prescription is what the society needs because only when the prescription matches the needs can this sort of education bring prosperity for the individual and nations. Then a follow up question is whether the prescribed knowledge and skills can be accurately measured by tests, whatever quality they may have. Only when test scores accurately reflect the level of master of skills and knowledge can they be used as indicators of the quality of this flavor of education.

My interpretation of Cody’s argument is that Tucker’s version of education does not have much to do with the economic success of America and its people in the future. Tucker uses prominent economists and their theories to defend his argument that “for both individuals and nations, people’s stock of education and skills is a major determinant of economic growth.” But again he did a bait-and-switch with imposing his version of education and skills on the economists. Paul Romer, the Nobel laureate Tucker borrows his ammunition against Cody and Ravitch, emphasizes on the ability to innovate, to create new combinations of resources as driving force for economic growth in developed countries. According to Romer:

Leading countries like the United States, Canada, and the members of the European Union cannot stay ahead merely by adopting ideas developed elsewhere. They must offer strong incentives for discovering new ideas at home…the one safe measure governments have used to great advantage has been subsidies for education to increase the supply of talented young scientists and engineers. They are the basic input into the discovery process, the fuel that fires the innovation engine. No one can know where newly trained young people will end up working, but nations that are willing to educate more of them and let them follow their instincts can be confident that they will accomplish amazing things.

It is hard to believe that Romer’s creative talents come from the education Tucker endorses. Past international test scores have been a poor indicator of nation’s economic performances, as Keith Baker and a number of others have shown. More recently, Google has determined that “G.P.A.’s are worthless as a criteria for hiring, and test scores are worthless. … We found that they don’t predict anything.”

When economies change, as Tucker notes, so fast and on a global scale, it has become even more difficult to predict the skills and knowledge that matters in the future. But one thing seems to be clear. Even if Americans are equipped with the same skills and knowledge as Chinese and Indians, America’s favorite competitors, Americans won’t have an economic advantage simply because it costs much less for these countries to develop the same skills. So more of the same skills and knowledge won’t work, neither will the same education. America does not need a quantitatively better education, it needs a different kind of education.

There are of course other problems with Tucker’s chain of reasoning; for example, are American teachers truly worse educators than their counterparts in other countries? Again it depends on the definition of education. Is education about test scores? Or is it about cultivating diverse, creative, passionate, and curious innovators and entrepreneurs?

Tucker has much faith in this plan. “We know this form of accountability will work because it is already working at a national scale in the countries that are outperforming us.” Even if Tucker were right, America will at best outperform the top performing country—China. But is that what we want? My answer is NO and my reasons are in my book Who’s Afraid of the Big Bad Dragon: Why China Has the Best (and Worst) Education System in the World.