Au, Wayne. (2009) Unequal by Design: High-Stakes Testing and the Standardization of Inequality. Reviewed by Victor L. Willson, Texas A&M University

Au, Wayne. (2009) Unequal by Design: High-Stakes Testing and the Standardization of Inequality. NY: Routledge Pp. 216         ISBN 0-415-99070-8 Reviewed by Victor L. Willson Texas A&M University July 29, 2009 Wayne Au critically evaluates the high-stakes testing enterprise in the United States through the lens of Marxism. Since Marxism assumes the answer before the question is posed, the interest here is on the interpretations it can give to the testing enterprise, even though many can be predicted before reading the book. Nonetheless, the book is worth the read. Au assumes that testing, and particularly No Child Left Behind (NCLB) testing, was both intended to and indeed does stratify children into economic niches, guaranteeing failure for the lower classes while nominally espousing a bootstrap sort of model for educational improvement. He begins in Chapter 2 with the early twentieth century perspectives of schools as production facilities designed to facilitate the economic growth of the industrial America. I doubt there are serious critics of this interpretation, even among conservative educators. Au’s conclusion that school boards simply bought into the capitalist production model is a bit of a stretch, but he was trying to fit the local control paradigm into his historical context. I do not doubt that for many big city schools the focus was clearly on fitting most students for manufacturing jobs, while in the country it was simply to get students sufficient learning to work profitably on the farms and small businesses of Main Street. I thought Au fell short, however, in discussing the post-WWII change in thinking about the purpose of schools, the GI Bill, and the shift to college-focused public school education. Perhaps that still fits the Marxist perspective, but it is not given much thought. Since that played such a large role in the movement toward the minimum competence testing in the early 1970s that in turn led to current high-stakes testing, the omission seems significant to me. The concern with revisions of school curricula in the Sputnik era by the National Science Foundation and integration of schools led to a conservative backlash that included increased emphasis on statewide testing, none of which is covered by Au. The focus of the end of Chapter 2 is on IQ testing and standardized testing as part of the capitalist ideological control mechanism. The recapitulation of development of the IQ test was not particularly necessary here, and again, a better focus would have been more current uses of IQ tests, which have largely been abandoned in regular education, at least compared to the pre-1970 era when ability tests were routinely given to all students. Now, the emphasis is on establishing special education categories. While that might have been profitably explored, particularly in the context of high stakes testing, where excluding low-performing students improves school ratings, Au did not follow that thread. Au takes aim at the testing industry, quite appropriately. His analysis of test development leaves something to be desired, however, and suggests he does not really understand the process of criterion-referenced test development very well. The testing industry is now big business and has far too few technical experts available to monitor and control test development, with resulting tests in many states that are of poor quality. Au’s perspective is that the tests merely reinforce the same segregations that the economic structure has generated for a century. Indirectly he is correct, but not from arguments he presentss in the book. Au follows the testing process so far, then ignores the other half of the NCLB promise that was not fulfilled- program improvement. In most states the high stakes tests focus in the early grades on reading and mathematics, gradually adding other topics such as writing, social studies and science at various levels. I am quite familiar with the development of these tests in Texas over the last 20 years, and as a reading researcher focusing on children’s reading development I did not have significant concerns that the tests were designed to segregate children. Gradually, they incorporated elements on which there is now good consensus that indicate reading growth. The reading tests, however, are quite limited in length, with only 4 or 5 questions available for measuring important components of reading level. Nevertheless, they are typically reasonable indicators of the level of reading performance. That they indeed separate groups of students at the classroom, school, district, and ethnic levels is indisputable. Similarly, the mathematics tests measure reasonable and publically available topics at each grade level, although the increased requirement for reading with grade raises a significant question about what is really being measured- but that is a question for the whole field, regardless of political stance. So if the tests are not designed specifically to segregate, why do they? The answer is in the failure of legislatures to do more than measure. The equivalent would be to require hospitals to conduct patient examinations to diagnose, but then never treat beyond the drugs and procedures available 10, 20, or 30 years ago. More on that below. In Au’s third chapter, he reviews more modern history of educational policy, specifically the post Nation at Risk period. I believe this is too late a time period to start. If he had begun with the Elementary and Secondary Education Acts of the post-Sputnik era, he might profitably have connected the dots between the curriculum reforms, attempts to revise tests to reflect them (again, the college-focused curriculum in the post-industrial age that was developing), and the backlash against both by conservatives, who put successively in place minimum competency, competency, and mastery test programs in various states, culminating in the pre-NCLB movement in Texas. I was part of designing the accountability system that included the testing revisions (for good or ill) through a Texas Legislature-charged committee that included members who became part of President Bush’s education advisory group. The later efforts in testing clearly did react to Nation at Risk, however incoherently. What was lacking at all political levels was any sense that without curriculum reform and significant investment in pedagogical reform, the tests would simply reflect the social inequalities in the schools and school districts, which they did and do. Au spends no time on school district expenditure across any given state, which can explain the same variance that class and ethnicity do. Nor does he consider any fine-grain analyses that have been conducted that illustrate schools in the minority conditions that are phenomenal successes, and why they succeed in spite of the tests and inequalities. That said, however, schools without textbooks are good candidates for failure, and the fact that they are found in minority neighborhoods is not chance. Au might have focused on the assumptions about high-stakes tests that they somehow would by themselves improve education. To some extent, they increased test performance as teachers learned what they needed to focus on, as to some extent also did students. Another area Au neglected was the common requirements that at the school level, performance by ethnic groups and by gender must be publically presented. That information at least had, and still has, the potential to require changes at the actual instructional level. School boards now have to respond to such differences, as do administrators. How they do that is quite important to students. Au’s Chapter 4 is perhaps the least convincing in the book, attempting to link control of what goes on in the classroom to quite abstract mechanisms. He was correct in listing teaching to the test as a major driver in classroom instruction, but his arguments about curriculum, content, pedagogy, bureaucracy, and discursive control are much less persuasive. For example, in curriculum and content control, Au seems to either ignore or discount the roles that professional organizations and societies have played in developing the reading and mathematics curricula in today’s schools. Marilyn Adams’ Learning to Read and the National Reading Panel report were much more influential in framing what is now a consensus model of reading development than some abstract corporate vision of reading control. The NCTM mathematics curriculum revisions of the late 1990s are much more influential (and controversial) in constructing mathematics curricula at the state level than Au appears willing to acknowledge. The political stance of the conservative literalist right wings on reading and math continues to shape test development also in ways that are not entirely economic. Finally, in Chapter 5: Devising Inequality: High-Stakes Testing and the Regulation of Consciousness, Au focuses on the test as regulator. Unfortunately, this is a misplaced understanding of the process. The high-stakes test is required to reflect the curriculum at the state level. Now I will certainly agree that 40 or so items is hardly an adequate representation of the entire curriculum of a year’s work, or in the case of exit exams, 8-10 years work. The test does not regulate the consciousness of teachers but the required curriculum, in Texas embodied in the Texas Essential Knowledge and Skills (TEKS) does. So now we come to the crux of the issue: who develops the curriculum and who should? If it is inherently racist and class-stratifying, who gets to change it? As in most states, in Texas the TEKS is developed by school teachers and college professors. While political persons (voted into office or appointed) do intrude in those deliberations in subtle or even overt ways, the final results are not changed in backroom legislative deals. Now Au can argue that the developers are all co-opted and fellow travelers in the great conspiracy, but it is a weak straw to build this particular house on. The net result, however, is that the high-stakes test does indeed classify and regulate who has access to the future, whether it is college, higher paying jobs, or increasingly even to upward mobility. Au has focused on high-stakes tests as the culprit, rather than the messenger. The message has different meaning than perhaps intended. It indicates the failure of states and the nation to respond to the stratification and classification they have created. The message that many children, including disproportionate numbers of poor minority children, cannot read well or numerate even minimally, is also real. It is not just a political ploy intended to keep lower classes in bondage. In Chapter 6 Au summarizes his conclusion, that the tests are designed to perpetuate the social and demographic stratifications of our society. It may be his conclusion, and maybe inevitable from a Marxist perspective, but it is not necessarily either true or valid. This is 2009 and with a new national administration that appears much more pragmatic in some areas and more progressive in educational thought, and with extensive resistance by states that now recognize the silliness of AYP (annual yearly progress) and the serious consequences of not meeting it on federal allocations, it is quite probable that things will change. The Marxist conclusion is that it will be rearranging deck chairs on the Titanic, but perhaps there is potential for real discussion in most states about what high-stakes tests tell legislators and parents about their schools and what they do to children. Unfortunately, in my state (Texas) I see no light at the end of any such tunnel. While Wayne Au’s book missed some opportunities, and its stance will turn off some people and even prevent its being read by the political right, it is a book that needed writing; and even if some shots go astray, it makes plenty of on-target hits about what is a process seriously in need of review and revision, if not extinction. About the Reviewer Victor L. Willson is Professor and Head of the Department of Educational Psychology, Texas A&M University. http://coe.tamu.edu/~vwillson/