For some mysterious reason, Larry Stedman, like most people, sees complacency as the only current response to the constant, often erroneous criticisms of American schools. Yet there is nothing in my work (Bracey 1992a, 1993, 1994a, 1994b), nor in that of the other researchers Stedman cites, nor even in the work of others he doesn't (Berliner 1993, Jaeger 1992), that implies complacency.
In the Third Bracey Report (1993), I quoted Israel Scheffler's definition of education: [Education is ...] the formation of habits of judgment and the development of character, the elevation of standards, the facilitation of understanding, the development of taste and discrimination, the stimulation of curiosity and wondering, the fostering of style and a sense of beauty, the growth of a thirst for new ideas and visions of the yet unknown. If we accept this definition, we must also accept that we simply do not know how our educational system is functioning, save for anecdote and impression. This is troubling. Critics, however, have claimed that the data we do have show the system has failed.
Stedman allows that the system is not deteriorating, but that the schools were awful in the past and are awful now. My contention has been that the data we have show no such thing. Let's consider each of the myths Stedman claims I and others have fashioned.
Myth 1: There really wasn't an SAT decline.
When I conducted a study on SAT trends, taking into account demographic changes, I actually found a 5-point gain in math, but a 22-point decline in verbal skills (Bracey 1990). Although this decline is small compared to the 76-point overall drop, my statement of “no decline” was half in error.
No one knows how much of this decline comes from changes in test-takers. We do know, though, that the 10,654 students in the Northeast who set the SAT standards in 1941 were 98 percent white and 60 percent male, and that 40 percent of them had attended private high schools. Currently, 31 percent of the more than 1 million SAT examinees are minorities, 52 percent are women, and many come from low-income families. It would, therefore, be astonishing if the SAT averages had not declined.
Stedman claims that the “most careful” analyses find 50 to 75 percent of the variance to be accounted for by changes in composition. Even these figures require whopping correlations —.71 to .87—between the relevant demographic variables and SAT scores. But no such correlations or regression analyses have been calculated. Stedman places the College Board Advisory Panel's analysis among the “most careful” studies, but its figure of 67 to 75 percent of the decline up to 1970 is conjecture: the board itself admitted no precise data existed (Advisory Panel 1977).
Because compositional change slowed between 1970 and 1977, the last year of the panel's study, the panel looked to other variables to account for the decline. And did it find them! One background paper listed no less than 89 hypotheses to account for what was then a 91-point decline in the overall average (Wharton 1976). Of course, demographic changes have accelerated again in recent years.
The proportion of students who score high (650 or above) on the SAT offers a different perspective. Whereas the proportion of students who scored high in the verbal section was originally 7 percent, it is now 3 percent. With pervasive television and other media, this should not be surprising, but it is not necessarily a cause for alarm. I once studied a group of 8th graders who were engaged in a project about finding the Titanic (Bracey 1992b). They watched movies and videotapes, read books and articles, played computer simulations, and videotaped an interview with Robert Ballard, the man who led the successful Titanic search expedition.
In the end, they produced a multimedia program to show the rest of the school—a pretty rigorous course of study. For purposes of the largely textbound SAT verbal section, however, they might have been better off simply reading material and writing a report.
The picture is different in math. In 1993, the proportion of high scorers on the SAT-M reached an all-time high of 11 percent, 65 percent greater than the proportion of standard setters who scored that high. This gain cannot be accounted for by Asian-American students; while their scores were higher than those of other ethnic groups, they made up only 8 percent of the test-takers.
Myth 2: Test scores are at all-time highs.
To bolster his assertion that we myth makers claim test scores are at all-time highs (we simply say some scores are at all-time highs), Stedman cites one journalist and one attorney on a school board.
Stedman attributes the record rise in scores on the Iowa Test of Basic Skills (ITBS) among 3rd through 8th graders to teaching to the test. But trends nationwide closely match those in Iowa. This is important because the Iowa State Testing Program has been around since the 1930s. It is not a recently imposed, high-stakes program where someone's salary or job hangs in the balance. Moreover, unlike other commercial achievement tests, new forms of the ITBS are carefully equated to old forms, making the tracking of trend data much more reliable.
In addition, it is often difficult to interpret state-level test scores over time because so many variables change. For instance, recent immigration flows into Florida and California render these state's test score trends impossible to decipher. But Iowa is, in some ways, frozen in time. It has no large cities and their attendant problems. It remains 98 percent white. And Iowa has the highest SAT scores and NAEP mathematics scores, and the second highest ACT scores and NAEP reading scores. All the evidence from assessments suggests that ITBS and ITED performance in Iowa is matched by national trends.
Stedman is deceptively selective in reporting NAEP scores. He plays down reading and math, emphasizing instead the science scores of 17-year-olds. (A Nation At Risk did the same thing.) Could this be because the science scores of 9- and 13-year-olds are at all-time highs, as are math scores for all ages and reading scores for ages 13 and 17?
Myth 3: Our top half is internationally competitive.
Stedman cites studies by Ian Westbury (1992, 1993) and claims that the top half of American classes were comparable to the top half of Japanese classes “only in arithmetic.” In fact, Westbury argued that he compared whole grades only for arithmetic because only in that area are the two curriculums comparable. In algebra, however, he found the top 20 percent of American students comparable to the top 20 percent of Japanese students (Westbury 1992).
Stedman fails to report some important recent analyses that combine data from the Second International Assessment of Educational Progress (IAEP-2) with data from NAEP reporting categories (Educational Testing Service 1992, National Center for Education Statistics 1993). These data also bear on Stedman's fourth myth. As Figure 1 shows, Asian-American students outscore everyone, while white students in American schools tie Hungarian students for third place. In the United States, Asian and white students together make up over 70 percent of the K–12 student body. Thus, even in mathematics, where we are such reputed dolts, most American students are world-class. On the other hand, some American students are not even Third World (see fig. 2).
Figure 1. Highest Scorers in 8th Grade Mathematics (IAEP-2 and NAEP Tests)
Response / Stedman's Myths Miss the Mark - table
| Asian Students, U. S. Schools | 287 |
| Taiwan | 285 |
| Korea | 283 |
| Advantage urban setting students, U.S. schools | 283 |
| White students, U.S. schools | 277 |
| Hungary | 277 |
Figure 2. Lowest Scorers in 8th Grade Mathematics (IAEP-2 and NAEP Tests)
Response / Stedman's Myths Miss the Mark - table 2
| Jordan | 246 |
| Mississippi | 246 |
| Hispanic students, U.S. schools | 245 |
| Disadvantaged urban setting students, U.S. schools | 239 |
| Black students, U.S. schools | 236 |
Based on the statistics, it makes little sense to speak of nations' relative standings based on average test scores because the variability within any nation is huge. Jordan and Taiwan, for example, are separated by 39 NAEP scale points. But Taiwan's worst students and best students are separated by 150 points (National Center for Education Statistics 1993).
Stedman also employs the usual scare tactic on global competition, alluding to the “connection between academic achievement and economic competitiveness.” My four reports have demonstrated that this connection is itself a myth. To begin with, the economy seems to have rebounded. Naturally, the schools, which had been blamed for the economic slump of the '80s, received no credit for the recovery. The link between economic competitiveness and education is “the great school scam,” as Larry Cuban has called it (1994). The eminent education historian Lawrence Cremin put it this way: American economic competitiveness with Japan and other nations is to a considerable degree a function of monetary policy, and decisions made by the President and Congress, the Federal Reserve Board, and the federal departments of the Treasury and Commerce and Labor. Therefore, to contend that [these] problems can be solved by educational reform, especially educational reform defined solely as school reform, is not merely Utopian and milennialist, it is at best foolish and at worst a crass effort to direct attention away from those truly responsible for doing something about competitiveness.... It is a device that has been used repeatedly in the history of American education (1989). Whether the proponents or opponents are right about the recently signed General Agreement on Tariffs and Trade (GATT), it will have a far greater impact on our international competitiveness than all of the National Education Goals combined.
As I have pointed out in some detail elsewhere (Bracey 1992a, 1994a, 1994c), from a strictly economic standpoint, our level of education already exceeds our professional opportunities. We graduate a higher proportion of students from college than any other nation, and 26 percent of these graduates take jobs that require no college education.
Myth 4: This crisis in education is not general; it concerns inner-city schools and poverty.
Stedman reports many data to allege that overall performance is low, including NAEP data showing that only 37 percent of our seniors demonstrated reading proficiency. But he also notes that among students from 27 nations, American 9-year-olds were second in reading; while among 31 nations, American 14-year-olds were tied for eighth place. The 14-year-olds also had reading scores that were as close to first place as the 9-year-olds—the high scoring countries were tightly bunched. Thus, if the United States has a reading problem, so does the world, although no one, to my knowledge, has ever invoked the “international literacy crisis.”
It is worth noting, as an indication of cultural differences in attitudes toward schools, that in this international reading test, German students ages 9 and 14 both finished in the middle of the pack. Astonished Germans noted that “German standards were exceeded even in America.” German authorities placed the blame squarely on the family (German Research Service 1992). As for the attitude toward schools in the United States, it was significant that the media gave much ink and air time to IAEP-2, where our rankings were low, but none to the reading study at the time of its release.
America as a nation scores well in reading. The data in Figures 1 and 2 show that in mathematics, low scores are largely confined to areas of poverty, although in most of the categories ethnicity serves as a proxy variable for social class. There is no NAEP category “disadvantaged rural”; if there were, there would be another entry in Figure 2. As I showed in The Second Bracey Report (1992a), rural poverty is even more dire than its urban counterpart.
Do We Ask Too Much?
- People who dwell in universities have unrealistic notions of intellectual performance.
- People who do not actually take certain tests consistently overestimate how easy they are.
- Adults often impose unreasonable standards on the young, standards they themselves cannot meet.
While I was director of testing for Virginia, the legislature approved a minimum competency test for high school graduation. Before it was used, I invited all media representatives in Virginia and the District of Columbia to take the test. Only two did. Neither got a perfect score, and both wrote that the tests were reasonable measures of what they attempted to assess. All other articles emphasized how easy the tests were. Later, these same reporters were stunned to find that 11 percent of white students and 39 percent of black students had failed to attain the 60 percent passing mark.
A dramatic example of my third point occurred at a press conference called by the president of the National Geographic Society, Gilbert Grosvenor. There he announced the results of a survey concerning how much high school students know about geography. Grosvenor jokingly declared that he had found a new “lost” generation: “They don't know where they are.” A reporter then asked him to identify the states that adjoin Texas. He could not do so (Cremin 1989).
The Media's Dark Lens
Americans think that local schools are okay, their own kids' schools are fine, but that there is a crisis in American education. While some have characterized these opinions as schizophrenic, they actually make perfect sense. Americans get their news about the state of national schools almost exclusively from newspapers and television (Hunter 1994). Given the chronic negativity of these two sources, a pessimistic opinion is inevitable. When it comes to local schools, though, people consult friends, neighbors, their children, school newsletters, and local school officials (Hunter 1994).
Hudson Institute Fellow Denis Doyle once characterized Americans' favorable opinions of their local schools as “scientific evidence that ignorance is bliss” (Doyle 1993). The data, however, seem to corroborate another aphorism: Seeing is believing.
In my recent book, which largely discusses ways in which schools need to be improved, I wrote, “Let me be clear about my position: One need not assume school failure to propose school reform.” (Bracey 1994b). One need not assume low quality either.
