California Continuation Education Association Document

Title:  A Horse Called NCLB

Author:  Anne C. Lewis (Phi Delta Kappa International)

Date:  November 2002

A Horse Called NCLB

by Anne C. Lewis

From the Phi Delta Kappan, Volume 84, Number 3, November 2002, pp. 179-180. See
http://www.pdkintl.org/kappan/k0211lew.htm

Illustration © Mario Noche


WHEN the horse dies, many school reformers advise riders to dismount. Don't buy a new whip, don't form a committee to study the horse, and don't blame horses. In other words, when things are not working, change them.

The same advice should be given to those who are riding a crazy horse that is galloping at full speed toward a cliff. Get off and find a mount that will get you where you want to go -- presumably not the bottom of a precipice, with you in a mangled heap.

The trouble is that one particularly crazy horse -- test-based accountability under the No Child Left Behind (NCLB) Act -- is being ridden by advocates for poor children and guardians of "rigorous standards" for all as if it were a Kentucky Derby winner instead of a maniac steed bent on self-destruction. Let's be brutally honest. The early evidence of the impact of NCLB's test-based accountability on the states indicates that it is undermining many good policies, fostering some bad ones, and creating resentments that will not ease until better policies are developed and put in place. The growing criticism of the policies does not mean that people reject the goals of the legislation or the goals of accountability per se. Rather, the criticism stems from a realization that current standardized, high-stakes testing narrows the whole enterprise of education and could halt the development of truly significant improvements in teaching and learning.

Undoubtedly, Richard Elmore of the Harvard Graduate School of Education speaks for many when he contends that internal accountability must precede external accountability. That is, the people inside a low-performing school, for example, must know what to do to improve student achievement and have a coherent system of beliefs and practices before any significant change can take place. Without substantial investments in schools' capacity, test-based accountability will probably aggravate existing inequalities, Elmore wrote in the spring 2002 issue of Education Next.

High-performing schools essentially reflect the social capital students bring to school. Low-performing schools most often cannot rely on these supports but, instead, must turn to their own organizational capacity, Elmore notes. While NCLB provides more money for these low-performing schools, it cannot match the intensity of effort applied to the accountability side of the issue, so the low-income schools will slip even further behind.

Once, when fairness was considered important, the movement for higher standards and better assessments was coupled with opportunity-to-learn standards. This last item dropped off the agenda when state governors realized that it came with a hefty price tag, especially in the form of increased investment in teacher quality and improved working conditions that few state leaders were willing to support. Nor were local leaders altogether willing to work with teacher unions to find a way to distribute experienced and highly qualified teachers to the neediest schools. There is a reason why educators who don't know what to do are clustered in low-income schools. It is because policy makers have put them there -- and now are demanding that they be held more accountable for improvement than those who teach in more affluent schools.

At the same time, NCLB assumes -- falsely -- that states have the capacity to fulfill its testing requirements. Not even half the states came close to meeting the NCLB mandate of testing every year at grades 3 through 8 when the legislation was passed. As the Clinton Administration was leaving office, it had found fewer than half of the states in compliance with the 1994 changes in Title I, changes that nudged, rather than bludgeoned, the states toward performance standards and assessments. Some states were doing quite well, but NCLB created new demands. Maryland, for example, could not afford to keep making improvements in its performance-based system of testing and still comply with NCLB. So it chucked the former and adopted a multiple-choice system.

In the rush to meet the NCLB deadlines, states are grabbing standardized tests off the shelf -- no matter whether they meet their learning standards or not. But one chilling factor that is often overlooked is that the commercial testing industry itself lacks the capacity to respond to NCLB. As if the scoring problems of the past few years were not enough to make parents and politicians worry, the tests themselves may not be able to meet high-quality standards, according to Steve Dunbar, a professor of measurement at the University of Iowa and an architect of the Iowa Tests of Basic Skills.

Speaking to the annual conference of the Center for Research on Evaluation, Standards, and Student Testing, Dunbar said that the selection of material for a test is the critical point. It is important to review, revise, and replace items, then go through the "three-R" process again. "If I don't throw away half of the materials developed, I'm cutting corners," he said, predicting that this is exactly what will happen as states fail to budget enough for testing and as testing companies have to charge higher prices to produce new versions of the tests for every grade, every year.

Proof that the testing mandates of NCLB are far ahead of the state of knowledge and practice about testing is obvious in the choices that now confront states. In working out the details of "adequate yearly progress" and determining which schools students could transfer from this fall, the states have exposed their comparative strengths and weaknesses. A student in Arkansas didn't have to fret over making any choices of alternative schools: no schools were considered failing by the state. In Michigan, though, 1,513 schools were on the "flit" list.

In essence, according to New York Times columnist Richard Rothstein, the federal rules are punishing states for having higher standards. Ohio and Louisiana already have lowered their standards. If California does not lower its quality index standards, it is facing the possibility that virtually all of its schools will not meet the proficiency standards required by NCLB, largely because of the mandate to have all subgroups of students reach proficiency.

As a way of checking whether all states have brought themselves into line with high standards, all will be required to participate in the reading and math assessments given by the National Assessment of Educational Progress (NAEP). Edward Haertel of Stanford University estimates that, at the current rate of improvement, it would take 100 years for all students in the nation to attain the proficient level on NAEP.

Calling the NCLB accountability system unrealistic, Eva Baker and Robert Linn, co-directors of CRESST, and Damian Betebenner of the University of Colorado have suggested that the goal should be for all students to reach the "basic" level of NAEP. If the percentage of students who achieved at that level or higher were used as a benchmark for state standards, the result would be less disparity among states, they say in the August/September 2002 issue of Educational Researcher.

The federal government should back off immediately, according to Rothstein, and grant waivers to states that have good plans for improving schools the way they want to, instead of using the "one best system" approach approved by Washington.

Perhaps the paranoid sentiment of many critics of the federal policies may turn out to be correct. That is, perhaps this is all a plot to discredit public education to the point where privatization and choice are seen as the only answers. If so, that's a horse of a different color, and one that we are trying to change in the middle of the stream.

ANNE C. LEWIS is a national education policy writer living in the Washington, D.C., area (e-mail: aclewis@crosslink.net).