Education Myths: What Special Interest Groups Want You to Believe About Our Schools — And Why It Isn't So, by Jay P. Greene. Rowman & Littlefield, 2005, 267 pages.

---

Lies Your Teachers Are Telling You

by Gary Jason

Jay P. Greene, senior fellow in the Economic Research Office of the Manhattan Institute, has published widely on education topics. His research has been cited four times in Supreme Court decisions. This book brings together his views on the state of contemporary American K–12 education. His ideas are clear and compelling, backed up by extensive research.

He considers 18 myths about education — views that are both pervasive and demonstrably false. He groups these myths into four categories: myths about resources, myths about outcomes, myths about accountability, and myths about choice.

He starts with six myths about resources. First is the view that public schools perform badly because they are underfunded. This is a common refrain among modern liberals, not to mention teachers' unions, but Greene notes that the average person accepts it as well. He demolishes the myth — in part by showing that national inflation-adjusted per pupil spending has increased nearly linearly every year from 1945 (when it was only $1,214) to 2001 (when it hit $8,745), while the NAEP (National Assessment of Educational Progress) scores stayed essentially flat from 1971 to 1999.

The NAEP is the most reliable and representative national measure of scores we have. During the period when it has been used (1971 to the present), per pupil funding went from $4,479 to $8,745, while average scores increased only an insignificant three to five points out of 500 possible. High-school graduation rates from 1971 to 1999 were nearly flat. They went from 75.6% down to 72.5%.

Greene attributes the prevalence of the "underfunding" myth to people's reluctance to believe that more spending has been useless, and the fact that during the period for which outcomes can be reliably measured, funding rose only slowly, while the number of students dropped. People therefore didn't notice the per-pupil rise. The prominence of this myth may also be attributed, as Greene maintains, to well-publicized anecdotal accounts of funding problems.

There is also the myth that schools are doing poorly because of a rising tide of developmentally disabled ("special ed") children. Greene refutes this myth by noting that while the number of kids enrolled in special ed classes has mushroomed (from 8.3% in 1976–77, to 13.3% in 2000–01 — an increase of over 50% in less than 25 years), this has been because of changes in diagnostic practices, not to an increase in actual disabilities. For example, during the same period the number of mentally retarded kids dropped from 961,000 to 599,000. Greene suggests that the drop was caused by better prenatal care and improved safety devices, but doesn't mention a more plausible explanation: abortion. He does provide a good deal of evidence that funding incentives are important to the increase in students in special ed: "Not only is the Special Ed Myth false, it is the reverse of the truth: special education is not draining school budgets, it's inflating them" (34).

The next myth he explores is the claim that schools are doing poorly because of social problems (broken homes, poverty, poor parenting, or the like). Greene and his co-researcher Greg Forster addressed this hypothesis by developing a "teachability index" of 16 social indicators, establishing it as a reliable predictor of student success. They then examined each state's level of achievement compared to what would be predicted by its teachability index score. If social problems in a student population trump the type of schooling which that population receives, you would expect not to see much variability in student achievement among states with similar teachability indexes. But you do. This means that student achievement has to do with something more than the social advantages the students have. It has to do with the sort of education they are given. Greene urges four school strategies that have been shown to improve student achievement: accountability programs, school choice, early intervention for struggling students, and instruction based upon specific sets of skills and factual knowledge.

Greene then addresses the theory that small class sizes produce dramatic gains in student achievement. There are some data supporting the idea that lowering class size improves performance, but as Greene shows (in a probing evaluation), they are not powerful. The gains are small and the costs quite large. For instance, when California adopted a small class program, the number of teachers went up from 62,226 to 91,112 in just three years (you can see why teachers' unions push this myth so aggressively), without any commensurate increase in educational quality. And again, we have nationwide statistics that undercut the theory: from 1970 to 2001, the average number of students per teacher dropped by 29% (from an average class size of 22.3 down to 15.9), while achievement scores and graduation rates remained flat.

Next is the myth that professional credentials and experience make a big difference in teacher effectiveness. Here again Greene does an excellent survey of the literature. The evidence shows that possessing a teaching credential doesn't result in any significant increase in teaching effectiveness, and neither does possessing an advanced degree in education — though he does note that one study has found that having a master's degree in the academic subject one teaches results in a significant improvement of performance among high-school teachers. In other words, what is taught in ed schools is of dubious value. As Greene nicely puts it:

The main focus in education courses is the study of pedagogical theory, and it is possible that the formal study of pedagogy might not contribute substantially to the teacher's ability to actually teach. It is certainly true in many other fields that the study of theory, although important to scholars, does not contribute much to professional performance. Economic theory is not the main focus in schools of business, legal philosophy is even less important in law schools, and biological theory (which includes, for example, debates over evolution) is either peripheral or totally absent in medical schools. (70)

The evidence also indicates that teaching experience improves effectiveness only for the first few years of a teacher's career, and is associated with a decline in effectiveness later on — probably because of the role of tenure in deterring teachers from striving for better performance.

Then there is the myth that teachers are paid badly. Greene notes the obvious: teachers work far fewer hours than other professionals. They work only nine months a year, have many more holidays, and work about 7.3 hours per day. Factor this in, and the average teacher salary of $44,600 (in 2001) works out to a full-time equivalent salary of $65,440. The common reply is that teachers do a lot of work offsite, such as grading papers and "keeping up with the field" (as if that were similar to what a science professor at a research university or medical doctor has to do). But most other professionals do offsite work as well.

If we look at the 2002 Department of Labor figures on hourly compensation, teachers do quite well indeed: elementary school teachers earn $30.75 per hour and high-school teachers earn $31.01. Teaching compares well with other professions: architecture ($26.64); biology ($28.07); civil engineering ($29.45); mechanical engineering ($29.46); physics ($32.86); computer science ($32.86); electronic engineering ($34.97); and even dentistry ($35.51). And this doesn't include the notoriously generous benefits that teachers receive — medical insurance, retirement programs that are defined benefit (pension) rather than defined contribution plans, and the king of all perks, tenure.

Turning to myths about outcomes, Greene starts with the idea that schools are performing worse than they used to. He argues, surprisingly, that test scores show that students are performing roughly as well now as at any time over the past 30 years. For instance, twelfth-grade scores on the NAEP have been essentially flat from 1971 to 1999. So have the percentages of students scoring at the highest levels of proficiency. ACT and SAT scores were roughly stable over this period, and graduation rates stayed about the same, or declined only slightly. Fine — but I cannot refrain from emphasizing the fact that Greene's figures don't address the period before 1971. Even if schools are doing as well now as they did in 1971, how does their performance compare to that of the 1960s? Or the 1940s?

But to continue . . . Greene considers the myth that almost all kids graduate from high school. This makes a most enlightening chapter. The official figure, put out by the U.S. Department of Education in the year 2000, puts the high-school graduation rate at 86.5%, with many states claiming even higher rates. Greene reviews various methods of estimating graduation rates, and shows them to be flawed — for example, the National Center for Education Statistics (NCES) method that was used to come up with the official U.S. graduation figure excludes from the sampling anyone who is incarcerated (even though incarcerated persons are often dropouts), and counts anyone who gets a GED as having graduated (even though a GED is not really comparable to a real diploma). Greene's more accurate method involves dividing the number of diplomas actually awarded in the year 2000–01 by the average of the enrollments of this twelfth-grade cohort in its eighth, ninth, and tenth grades (modified to reflect an actual slight increase in the high school population). His method yields a more realistic figure of 71%. This is disquieting enough, but when he does the breakdown by ethnicity, the results are horrific: only 51% of black and 52% of Latino students graduated in 2001.

Greene debunks another bogus notion: that the primary reason for low college enrollment figures for minority students is a lack of financial aid. He shows that this claim is based on an unrealistic notion of what counts as "college-ready." If you believe (as the NCES does) that "college-ready" just means having graduated with a 2.7 GPA, then it will indeed appear that qualified students are not attending college. But if you believe (as Greene does) that "college-ready" students are only those who have graduated with a regular diploma, taken the minimum courses required for admission by a college, and demonstrated basic literacy, then nearly all college-ready students wind up entering college. That is true of minority students as well as white students.

The third group of myths is about accountability. Greene first considers the prevalent idea that the results of "high-stakes" tests are unreliable because of extensive cheating, student stress, and teachers "teaching to the test." By "high-stakes" tests he means tests with substantial consequences for poor performance, consequences such as students not being allowed to graduate, or schools receiving less funding. He again reviews the evidence, and shows the flaws in some highly touted studies. His own, properly designed study (conducted with Marcus Winters and Greg Forster) shows that there are strong correlations between high-stakes and low-stakes test results. This correlation would be unlikely if high-stakes tests had more cheating, coaching, or stress-induced failure than low-stakes tests (such as the SAT-9). He makes an excellent point regarding the buzz phrase "teach to the test." Teaching to the test is bad, if what you mean by that is coaching students with the correct answers (i.e., cheating by the instructor) or forcing students to memorize unimportant facts. But teaching to the test is good if you mean changing curricula and methodology to make all teachers focus more on the skills that the public (which pays for the whole educational system) deems important.

Greene critiques the similar misperception that high school exit exams increase dropout rates. He begins by noting that employers are becoming increasingly dismissive of the worth of a high-school diploma. In response, some states have instituted high-school exit exams, which in turn have caused concern that dropout rates will soar. But a study he conducted with Marcus Winters found that "changes in graduation rates in states that adopt exit exams are not significantly different from the normal fluctuations in graduation rates that are present in states that do not adopt such exams" (131).

This is surprising, and he considers three possible explanations. First, exit exams might be so easy as to affect almost no students. Second, the fact that students can take these exams repeatedly (and get additional instruction to help pass them), means that even students with poor academic skills can get through by chance. Third, the number of kids who are weeded out may be offset by the number of kids whom the schools are forced to help. Greene argues that the third explanation is the most adequate: exams do weed some people out, but they also encourage some students who would otherwise drop out to work harder, and some schools to give them the requisite remedial education. The tests take diplomas away from unqualified students and put them in the hands of qualified students . . . what a revolutionary concept!

Greene reviews the common lament of school administrators that accountability programs, such as the admittedly flawed No Child Left Behind Act (NCLB), are unduly burdensome. (Translation: give us the money, but don't require that we test students to verify the results). Numerous "experts" testify that instituting NCLB's mandates will require expensive reforms. But Greene wryly notes that "No field other than education makes important financial decisions solely based on claims made by experts . . . without demanding empirical evidence to back up those claims. The reason for this is simple: expert professionals often have a large financial or political interest in the recommendations they make" (137). I would add that no other field has lower quality "experts" than education.

The real costs of accountability programs, Greene argues, are only the costs of test preparation and administration, not the cost of preparing students for the tests — after all, that is what the education budget is for to begin with. He cites one careful study (by Harvard economist Caroline Hoxby) that shows the cost of testing as at most 0.35% of the education budget, even under the most onerous testing regimen. This is quite modest, especially when one considers the increase in quality that testing can produce.

The discussion of the fourth group of myths, those regarding choice in education, is the most valuable in the book, because consumer choice is the precondition for all other reforms (such as testing). Accountability is great, but unless parents can remove their children from lousy schools, it must be of limited effect. Greene's chapter 13, on the issue of choice, is perhaps the most useful of all. It targets the pernicious myth that there is no evidence that vouchers work.

Vouchers have been introduced in various places in the United States. They have been around for a century in Maine and Vermont. Arizona, Florida, Ohio, and Pennsylvania have launched publicly funded voucher programs, as has the city of Milwaukee (where 15% of children now attend private voucher schools). Philanthropists have set up privately funded voucher programs in Charlotte, Dayton, New York, Washington, and other cities. Numerous studies have been conducted on these limited voucher experiments. In addition, vouchers have been introduced in a number of foreign countries, including Denmark, New Zealand, and Sweden, although strangely, Greene doesn't appear to regard their experiences as useful in discussing the American system. (I will return to this point later.)

Major media outlets, such as the New York Times, Time magazine, and the Washington Post, say that the evidence about these programs is inconclusive. Greene, in a detailed literature review, strongly argues that it is not in the least inconclusive. He rightly focuses on random assignment studies, which are statistically the most reliable. (In these studies, subjects are assigned to the experimental or control group randomly. This helps guarantee that the two groups are matched, with confounding variables ruled out.) There have been eight random assignment studies of vouchers, and all but one have shown statistically significant gains on test scores for students in voucher programs. The one exception showed a gain in test scores, but it wasn't statistically significant. Every voucher program proved tremendously popular with parents. All this, from programs that cost half what public schools spend per child. None of the American voucher programs is pro rata; none gives parents a share of public taxes equal to what is spent per capita by public schools.

Greene discredits the common perception that private schools are elite, expensive institutions that can dump low-performing students. U.S Department of Education figures for the year 2000 show that the average private school charges $4,689 per year tuition; religious private schools charge only $4,063, with Catholic schools charging only $3,236. Public schools — or, I should say, the taxpayers — spend an average of $8,032 per pupil. Nor does the research support the claim that most private schools are very selective, or that they expel a lot of students. Greene doesn't make the point, but it seems clear that the possibility of expulsion tends to concentrate the students' minds. Indeed, many private schools — especially Catholic ones — wind up taking in and dealing with students expelled by public schools.

Private schools are often accused of hurting public schools by draining them of all the decent students. Greene grants that this claim is plausible, when considered a priori, but argues that the a posteriori evidence belies it. His review of studies of voucher programs suggests that they actually tend to improve public schools. The reason is clear if we think dynamically rather than statically: competition forces public schools to improve the quality of their service out of fear that they will lose some of their students, and thus some of their government funding. Hoxby's study of the Milwaukee program found that public schools that were exposed to greater voucher competition made significantly greater test gains than those that were less exposed. Charter schools force similar improvement, as does the kind of school districting that makes it easy for people to choose their public school.

By another close review of the evidence, Greene dispatches the view that private schools don't and won't serve disabled kids. The evidence shows that private schools do accept disabled students and (especially in the case of religious schools) educate them for far less money. The parents of disabled students tend to be more satisfied with private schools, especially because these schools are better at protecting their children from bullying, and at teaching them good behavior.

Greene likewise tears down the myth that private schools don't teach tolerance as well as public schools. Studies show that they do a better job of inculcating such civic virtues as tolerance and volunteerism. Why? Well, first, private schools teach everything better. Second, it appears that students who are enabled to exercise a better sense of self-identity — as they often are in private schools — are less threatened by people who are different. Third, private schools are typically smaller and have more student and parental involvement, which encourages more civic spirit than large, bureaucratic schools manage to do.

The last myth that Greene takes up is the notion that private schools are more racially segregated than public schools, and that vouchers will therefore make segregation worse. This is a myth fomented by hysteria from teachers' unions and civil rights organizations. Here again the evidence argues to the contrary — though there are fewer relevant studies on this matter. For one thing, most measures of "racial diversity" are flawed. Greene's own study (based on the 1992 National Education Longitudinal Study) showed that racially segregated classrooms were more prevalent in public schools than private ones. This whole subject is tricky, however. A school can be "integrated" as a whole, but there can still be segregation within it; and after decades of busing, people have already done a good deal of self-segregation.

Greene concludes that the various myths he debunks are all part of a "mega-myth," namely, "that education is different from other policy areas in that the types of incentives that normally shape human behavior do not shape educational behavior" (218). We think that rewarding good behavior and penalizing bad behavior works in parenting, and in every kind of business — but not in K–12 education! He reasonably urges us to begin to think in more economically realistic ways.

Greene's work is a testament to the usefulness of libertarian and conservative thinktanks. The academy is thoroughly dominated by leftists. Most humanities and social science departments are inhabited almost exclusively by people of the Left, with the possible exception of economics departments, and even those are changing rapidly. Add to that the creation of new departments, such as "labor studies," funded by labor unions (including teachers' unions), and the difficulties in doing research that reflects both sides of certain issues, or even addresses certain issues, become truly formidable. Thinktanks are a desperately needed counterbalance to a politically biased academy. Indeed, I recently heard a leftist radio commentator lament, "We need our own thinktanks!" I immediately shouted at the radio, "You have them already — they're called universities!"

Yet as valuable as Greene's analysis is, he might, in some cases, have looked at broader categories of data. This is especially true about the issues surrounding consumer choice in education (which encompasses charter schools, private schools, home schooling, and most importantly voucher programs). He rightly notes that parents are very happy with vouchers when they can get them, but I think more light could be shed on this preference if we were to examine, not just testing and attendance data, but also measures of social pathology, such as rates of adolescent drug use, pregnancy, assaults and other violent crime, and so on. That is, we should look at data from criminology and sociology as well as pedagogy to determine in what other ways private and voucher schools may be superior to public schools.

I believe you will see that parents are just as interested in schools that are safe and that uphold high standards of conduct as they are in schools that generate high test scores, as important as those may be. Indeed, recent work by the psychologist Judith Harris strongly supports the theory that peer groups are even more important in older children's lives than parental training, and that vouchers enable poor parents to put their kids in schools where the peer groups are competing to see who gets into the best colleges, rather than who is handiest with a switchblade.

Another area where Greene's data sets might be enhanced is the experience that other countries have had with vouchers. While comparing an American system with some other country's system is subject to the danger of false analogy, it can be enlightening as well as rhetorically effective. For years, this country has debated the idea of privatizing retirement systems, but the fact that Chile and other countries have actually done it adds immensely to the argument for doing it here. For years, we have argued about the theoretical benefits of a flat tax, but the fact that Russia and several other Eastern European countries have successfully implemented flat taxes is immensely instructive. Again, the case for "loser-pays" tort reform is greatly strengthened by the fact that Britain and most other Western nations use it.

The Swedes have successfully introduced a voucher system. While Sweden is more racially homogeneous than America, it is (as we are) a country with economic classes. And its experience with vouchers has shown that they decrease segregation by class. The reason should be obvious: there, as here, the rich and the upper middle class already have full school choice and mobility, and can therefore avoid underperforming schools. But when vouchers give everyone mobility, and schools spring up specializing in science, the arts, and other subjects, both wealthier and poorer families are attracted to them. Swedish opponents of vouchers argued that they would suck all the good students out of the public schools. In fact, the public schools lost only about 15% of their students, because they quickly improved their quality of service.

Greene's book would have been better if he had made it broader, but it's more than good enough as it is. It is an indispensable compendium of data and argument on fundamental issues of American education.

© Copyright 2008, Liberty Foundation

Send editorial comments to letters@libertyunbound.com.
All letters to the editor are assumed to be for publication unless otherwise indicated.

Send web site comments to webmaster@libertyunbound.com.