Reforming the Schools
We Must Take Charge: Our Schools and Our Future.
by Chester E. Finn, Jr.
The Free Press. 365 pp. $22.95.
There are essentially three kinds of remedies for the widely acknowledged shortcomings of our schools. (I do not consider simply spending more money on schools a remedy.) The first, usually proposed by people whose information comes from intensive classroom visits or who themselves have been teachers, is that the ethos of the classroom and the techniques of the teacher must be altered, and that this can be done if only teachers and principals would better understand how learning occurs. A good example of this approach is Theodore Sizer’s book, Horace’s Compromise. There are many others. Out of such fine grained, up-close examinations of teaching and learning we get a vivid picture of those classrooms that work and those that do not, and ideas for how to turn the latter into the former.
The second kind of remedy, often proposed by economists and political scientists and based on aggregate data about school achievement, proceeds from the argument that the organization of the school, and especially of the public-school system, is faulty. Schools that succeed are those that empower principals, compete for students, and resist the suffocating constraints of a central bureaucracy. Politics, Markets, and Schools, by John Chubb and Terry Moe, is the most recent and best-documented book propounding this thesis, which revolves around the idea of choice. But the idea itself has been around at least since Milton Friedman suggested it quite some time ago; many of the data with which to support the case for choice have been gathered by James Coleman and his colleagues.
The third remedy, proposed chiefly by past and present government officials, is based on the view that the rules by which teachers are accredited, curricula approved, schools judged, and students graduated are faulty. To change behavior we should change the rules. For many years the preferred changes were those that told teachers how and what to teach; of late they have come to focus on what students ought to know. The new book by Chester E. Finn, Jr. follows largely in this tradition. Its title, We Must Take Charge, conveys its message: accountability.
It is easy to see why these competing perspectives are held by those who hold them. Teachers, and those with a lot of first-hand knowledge about teaching, want to liberate instructors and encourage them to enter into a more constructive partnership with their students. Though not necessarily opposed to changes in structure or rules, they are suspicious because in the past most such plans have been the work of naive outsiders whose misguided efforts have put unrealistic burdens on overworked teachers. By contrast, economists and political scientists believe that any reform which relies on persuading poor teachers to become gifted ones, while it may succeed in isolated cases, is doomed to failure because it neglects the incentives operating on teachers and the institutions within which they work. As for government officials, they know that they cannot motivate teachers or manage classrooms, and suspect that it is utopian to think that any broad-scale plan for parental choice will be adopted by more than a fraction of the thousands of school systems in the country. The best that can be done is to have rules specifying outcomes: requirements for graduation or tests of student achievement.
Finn, who was an Assistant Secretary of Education under William J. Bennett and who has been a key adviser to the new Secretary, Lamar Alexander, has produced the best case we are likely to find for the accountability strategy. Apart from the fact that accountability may be easier to achieve than either revitalization or reorganization, Finn’s case rests on a powerful insight: though Americans think that in general our schools are doing poorly, they also think that the schools their own children are attending (or, in the case of teachers, the schools in which they are working) are doing just fine. If that is true, then it would be foolish to expect parents to demand broad structural changes or teachers to embrace new classroom methods.
Despite the countless tests, polls, books, articles, and television specials about our failing students, Finn believes that most parents are still, as he puts it, “asleep at the wheel.” They are victims of the Lake Wobegon Effect: any given child and any given school is above average. In 1988, American thirteen-year-olds took part in an international achievement test. Though our children got the lowest scores in mathematics, in response to a background question they said that they thought they were doing the best. In other countries, students did much better than the Americans but did not believe they were doing as well as they should. This was not an isolated instance. Harold Stevenson of the University of Michigan has repeatedly found that American students think they are among the best students in math (though in fact they are among the worst), whereas Japanese and Taiwanese pupils do not think they are among the best (though in fact they are).
Parents hold the same distorted perspective. Over 90 percent of the mothers surveyed in one study thought their children’s schools were doing an excellent or good job, twice the fraction of Chinese and Japanese mothers with this view.
The reason for this schizophrenia, Finn suggests, is that for decades American children have been bringing home report cards showing good grades even though on national and international achievement tests they are doing poorly. Why do parents not notice the contradiction? For one thing, the grim test data are typically reported for the nation as a whole, or sometimes for a state or region, but not for individual schools. Though SAT scores are falling and U.S. students are flunking international math and geography tests, these numbers can be ignored by parents on the assumption that they apply to other people’s schools and other people’s children. For another thing, there are complex methodological factors that cause all of the six major national achievement tests to place almost everybody above average. According to one survey, not one of the 50 states was below the national average in elementary-school achievement levels, and 90 percent of local school districts were above average. Small wonder that most parents and children think that things are fine when in fact they are not.
Why do we tolerate a testing system that misleads us? The answer, as is so often the case with governmental failure, is that it is not in the interests of any of the key players to have an honest system. Many professional educators are deeply opposed to standardized tests because they violate the child-centered philosophy of education: there should be no fixed standards for achievement, there should only be variable, relative standards tailored to the needs of the individual child. (I am reminded of the comparable view in penology: there should be no fixed punishments for a crime, only sentences tailored to the needs, character, and rehabilitative potential of each offender.) Naderite advocacy groups insistently argue that standardized tests are culturally biased, though in fact the best are not. Elected officials reinforce these theories for their own reasons. They do not want test results reported for individual schools, districts, or states because such data might be used to compare schools, districts, and states in ways that would be embarrassing for their constituents and thus for themselves.
As a result, when the federal government created the National Assessment of Educational Progress (NAEP) in 1969, Congress ensured that it would publish results only for the nation as a whole and for regions within the nation. When Secretary of Education Bennett tried in 1987 to remedy this defect, Congress again ensured that the NAEP would not be used to compare children, classrooms, schools, or school districts. Efforts to improve testing so that it could be used to close the gap between low actual achievement and high perceived achievement were met, in Finn’s words, by a determined effort to shoot the messenger.
Given the absence of clear educational goals and individually reported tests to measure progress toward those goals, Finn is not surprised to find that proposals from the first group of reformers—the teacher-centered, classroom-oriented ones—make so little headway. When Maine offered extra money and freedom from burdensome rules to ten schools, not one applied. When Massachusetts provided funds for seven schools to restructure their programs, only one applied. When New York City School Chancellor Joseph Fernandez asked each school to tell him what it would do if it had the freedom to innovate, only one in six responded, and half of those responses were unsatisfactory. Theodore Sizer, the most skilled and determined of the classroom-oriented reformers, has built up a network of like-minded school officials called the Coalition of Essential Schools, but he has faced exactly the same problem—resistance to change from within.
In general, Finn makes quite clear that the central issue is not money. In real (that is, inflation-adjusted) dollars, per-pupil spending on public education has tripled since the mid-1950’s and doubled since the mid-1960’s. It rose by nearly 30 percent in the 1980’s alone, and the average teacher salary rose by 27 percent in the same decade. By 1990-91, the average public-school teacher was earning (on a twelve-month basis) the equivalent of $44,000 per year. At the same time the number of pupils in the classroom has gone down. Between 1961 and 1986, median classroom size fell in the elementary schools from 30 to 24 and in secondary schools from 27 to 22. In 1966 the median number of pupils per teacher per day was 130; in 1986 it was 105.
Yet despite the fact that we are spending more money than ever before, we are not getting better results. Part of the reason, now so well-established by scholarship that only an officer of the National Education Association could deny it, is that there is no relationship between spending and outcomes, at least over the range of variation that we observe in this country’s schools. Another part is that a lot of the increased spending never finds its way to the classroom but is siphoned off by school-district bureaucracies. For example, only one-third of the $6,100 per student that New York City spends on education reaches the classroom; half the money goes to pay for citywide management, and another one-sixth goes to school management.
Finn wants to reform education by emphasizing outputs rather than inputs. He wants the schools to become accountable to parents and the public. Accountability rests on a tripod: there must be clear goals, reliable information on how to get to those goals, and strong incentives to attain them. In his view, none of the three legs is strong enough.
Though Finn makes many recommendations, the ones that are developed at greatest length concern the first two legs of the tripod. He wants the national government to set clear goals for learning, and to ensure that reliable tests exist that will measure the progress of individual students and schools. As I write, a National Council on Education Standards and Testing, co-chaired by Governors Carroll Campbell of South Carolina and Roy Romer of Colorado, is at work on these matters. Finn is a member. It will be an uphill fight, given the resistance of so many professional educators to either goals or testing (and given the fact that Keith Geiger, president of the National Education Association, is also a member of the Council).
Skepticism about national goals and tests is not confined to educators or the political Left, however. Many non-educators and political conservatives also have worries. Finn acknowledges this and is at pains to rebut many of the objections from all sides. He notes, correctly, that the claim that tests are biased is generally wrong (although, curiously, he does not marshal the impressive body of studies that demonstrates this). He admits that many teachers will “teach the tests,” but observes, wisely, that if the tests are any good this would be a good thing—or at least a better thing than letting them teach materials with little bearing on cognitive achievement. He acknowledges that standardized multiple-choice tests often measure memory more than they do learning, but points out that essay questions can be a large part of national tests, as they already are in England, France, Germany, Japan, and many other nations, without any obviously harmful effect on local educational levels. Hundreds of thousands of American high-school students are also given essay questions in the New York Regents exams and the private Advanced Placement tests.
This whole approach can be expanded; testing need not mean legions of students using Number 2 pencils to fill in little boxes. Indeed, the effect of some recent efforts at expanded testing has been quite dramatic. Barbara Lerner has reported (“Good News About American Education,” COMMENTARY, March 1991) that the use of minimum-competency tests beginning in the 1970’s has led to dramatic gains, especially among black pupils. She terms this the only truly successful educational reform in recent decades.
Goals and tests have indisputable advantages. The problem for many people, myself included, is whether there should be national goals and national tests at all. Finn’s experiences in Washington have left him more optimistic than I am about what can be accomplished by the federal government or a National Council on Education Standards and Testing. Not that I am a libertarian, or a reflexive advocate of minimalist government; I am simply a person whose service on national commissions and studies of politics have been sobering experiences.
I am also the president of the American Political Science Association (APSA), and in that capacity I am kept informed by the Association’s staff of the National Council’s efforts to produce goals. I have learned that initially the Council did not seem inclined to include civics and government among the standards. Needless to say, the APSA has endeavored to correct that oversight. I am pleased by the prospect but suspicious that other organizations, less deserving than the APSA, will also be successful in lobbying the Council. I assume that neither the Flat Earth Society nor the Ku Klux Klan will prevail, but who will? For possibilities, consider the individuals and groups that persuaded New York State Education Commissioner Thomas Sobol to embrace the loonier aspects of multiculturalism. Or consider the examples given by Finn himself: a California statute demands that students acquire a true comprehension of the rights and duties of American citizenship, “including kindness toward domestic pets,” and a California commission, undeterred by the barbs thrown its way by Doonesbury, urges the teaching of “self-esteem” as required by “new-age thinking.” Do not dismiss these as the excesses of Lotus Land; California may lead the nation, but it is not different from it.
Now, it is quite possible to design national tests that will assess the intellectual skills considered vital by virtually everybody. The minimum-competency tests evaluated by Barbara Lerner are an example. But is it possible to do that, and only that, if you entrust the task to a broadly representative national commission that must obtain congressional assent to its recommendations? Finn seems to think so; I hope he is right, but I doubt it. Instead, I envisage the triumph of the same sorts of interest-group politicians and lathered ideologues who now shape “policy” on almost every social issue, from teenage pregnancy to abortion funding, and, at the state level, on educational policy itself.
This is why I prefer parental choice to national accountability. It is only one leg of Finn’s tripod, but it is the least dangerous one: parents deciding what constitutes a good education and voting with their feet and their pocketbooks to enforce those decisions. Some schools will respond by giving out reliable test scores, some will respond by blathering about self-esteem, and some, doing neither, will go out of business. Tests will proliferate, some under public auspices, some under private ones. States and districts will differ about which tests, if any, they use or require. We will have a chance to learn what works and what does not.
In several places Finn endorses parental choice and refers approvingly to the book by Chubb and Moe, but his heart seems to be in test-driven accountability. He wants to have local choice and national goals and national tests. So, I judge, would the Bush administration, since it has formed a council on goals and testing but not one on choice. But there is an unseen political bias in this evenhanded approach: goals (if stated generally enough) can be adopted, tests (if made politically palatable) can possibly be adopted, but choice is a far more difficult enterprise and one, therefore, likely to be either traded away or left at the starting post.
Let me make clear what does not worry me: I do not worry that national goals will fail to include my own pet subjects, be they spelling, geography, or pride in nation. Nor do I really worry that national goals will make it easier for some local high school to teach black nationalism or Iroquois political theory. I worry that the political capital expended on goals and tests will leave the vital third leg of the tripod, incentives, too weak to carry its important share of the load. And I also worry that our nation is too big and too diverse to enable us to spell out wisely any educational requirements beyond competency in the essential skills of reading, writing, and arithmetic. Our national goal should not be to get Mary and Johnny to read The Federalist Papers, Mark Twain, or Herman Melville, but to get them to read anything—seriously, frequently, and with comprehension. We should care that our students write, but I care little what they write about, so long as they write a lot, write under criticism, and have to make corrections.
When former Secretary of Education Bennett set forth the curriculum for his ideal “James Madison High School,” he was asking the nation to imagine an educational world which, right now, cannot exist and, for many pupils, should not exist. If, beyond basic competency in the three R’s, some of our schools were to teach carpentry, cooking, or auto mechanics, I would be pleased—so long as they produced excellent carpenters, cooks, and mechanics. I would not even mind if they omitted civics. Aristotle remarked that politics cannot be taught to the young; I would only add, certainly not by some of the people I see trying to do it. But please do not report to the American Political Science Association that I said this.