The U.S. Department of Education's Office for Civil Rights ("OCR") has decided to wade into the controversy surrounding the use of standardized tests in primary, secondary, and higher education.

In the spring last year, OCR began circulating "draft guidelines" on "Nondiscrimination in High-Stakes Testing." The draft was leaked, apparently by disgruntled college admission officers, to The Chronicle of Higher Education, which put the guidelines on its website. Much controversy ensued, culminating in OCR head Norma Cantu being brought in to testify before the House Education and Workforce Committee in July. Cantu protested that the draft was just preliminary and promised that future drafts would be more widely shared.

In December, a revised version — still just a draft — was released, with a considerably longer text and title: "The Use of Tests When Making High-Stakes Decisions for Students: A Resource Guide for Educators and Policymakers." The guidelines, which challenge the use of standardized tests when they have a "disparate impact" on racial or ethnic groups, remain deeply and fundamentally flawed, however.

Before talking specifically about the guidelines, it is useful to describe the three kinds of discrimination that can be held illegal under our federal civil rights laws.

The first kind is holding people to different standards, depending on the color of their skin or where their ancestors came from. If you have a double standard based on race or ethnicity, everyone would agree that this is discrimination in any normal use of the term.

The relevant statute here, that OCR is enforcing, is Title VI of the Civil Rights Act of 1964. It reads: "No person in the United States shall, on the ground of race, color, or national origin, be excluded from participation in, be denied the benefits of, or be subjected to discrimination under any program or activity receiving Federal financial assistance."

It has become something of an open secret these days that most selective colleges and universities use precisely these double standards. They require Asians and whites to score at a particular level on the SAT or ACT in order to be admitted, but hold African Americans and sometimes Hispanics to a lower standard. Evidence that this is taking place has been collected by the Center for Equal Opportunity in a series of published studies of different state systems. Those studies are available on CEO's website, <www.ceousa.org>.

OCR has not complained during the Clinton administration about this practice — which, again, any reasonable person would conclude is discrimination in violation of Title VI, the statute that OCR is charged with enforcing.

A second kind of discrimination that violates federal civil rights laws is when someone chooses a selection criterion because of the racial or ethnic impact it will have. For instance, if a school was told to desegregate and then suddenly decided to change its admission criteria in order to keep out blacks, that would clearly violate the law, even if the new criteria were neutral on their face.

Here is a more recent example. The U.S. Court of Appeals for the Fifth Circuit in Hopwood v. Texas, 78 F.3d 932 (5th Cir. 1996), cert. denied, 116 S. Ct. 2581 (1997), held that the state could not use racial and ethnic admissions preferences. Texas decided, in the wake of the decision, that it would no longer consider SAT scores for the top ten percent of each high school class. It made clear that it was changing the standard in order to ensure that more blacks and Hispanics, and thus fewer whites and Asians, were admitted. In doing so, then, Texas was clearly violating the law.

OCR made no complaint about the new Texas law either. In fact, OCR head Norma Cantu sent a letter attempting to intimidate Texas schools into ignoring the Hopwood decision.

This leaves a third kind of discrimination, namely "disparate impact." Under this approach, a selection device that is neutral on its face, and that is applied neutrally, and that was chosen with no discriminatory animus, is nonetheless presumed to be illegal if it has a disproportionate effect on some racial or ethnic group.

Very few ordinary Americans would consider a test in such circumstances to be "discrimination" under any reasonable definition of the term. But the Office for Civil Rights has decided to apply the theory of disparate impact, which originated in employment law, to the context of higher education. As a policy matter, this is a mistake.

No federal statute requires OCR to challenge selection criteria that are nondiscriminatory on their face and have been adopted without discriminatory intent. To the contrary, the Supreme Court has made clear that Title VI itself bans only intentional discrimination — that is, only the first two kinds of discrimination discussed. See United States v. Fordice, 505 U.S. 717, 732 n.7 (1992), citing Regents of the University of California v. Bakke, 463 U.S. 265 (1978), and Guardians Association v. Civil Service Commission of City of New York, 463 U.S. 582 (1983). See also Washington v. Davis, 426 U.S. 229 (1976), and Village of Arlington Heights v. Metropolitan Housing Development Corp., 429 U.S. 252 (1977).

Nonetheless, OCR has decided to challenge standardized tests if they presumptively have a "disparate impact." The disparate impact approach is dubious enough in employment law, and should not be extended to other areas, particularly education.

Not only is OCR wrong to extend the disparate impact approach to the educational context, the manner in which it does it so also inconsistent with the law. In a variety of ways, the OCR guidelines overstate the extent to which the disparate-impact approach can legitimately be used to challenge the use of standardized tests.

OCR's draft guidelines are a transparent attempt to intimidate colleges — and primary and secondary schools, too — into manipulating selection criteria with an eye on the racial and ethnic bottom line, or overlay any criteria with bald preferences based on race or ethnicity. These are examples of the first two kinds of discrimination discussed above — real discrimination — and each is patently violative of Title VI.

Their illegality aside, OCR's proposed guidelines are bad policy. The Clinton administration purports to champion rigorous academic standards, but then tries to intimidate schools that would use tests to ensure they are being met. As Abigail Thernstrom wrote in a New York Times op-ed (June 10, 1999), "Removing the tests simply shoots the messenger and undermines the drives to raise academic standards."

There are racial and ethnic gaps in educational achievement, and those gaps won't be closed by pretending they don't exist or attempting to "litigate them away" (as a Washington Post editorial [December 25, 1999] put it). Instead, competition and accountability among schools should be encouraged through choice; illegitimacy rates lowered (they are around 70 percent for blacks — triple that for non-Hispanic whites); and an end put to the notion that studying hard is "acting white."

The use of standardized tests can certainly raise difficult issues, but they are issues for educators and parents, not civil-rights lawyers.

Roger Clegg is vice president and general counsel of the Center for Equal Opportunity, a Washington, D.C._based research and educational organization. He served as a deputy in the U.S. Justice Department's Civil Rights Division from 1987 to 1991.