1 comment - Last on 07/13/2009

The Score: Test Rigging before Ricci

July 13, 2009 By Glenn Ricketts

Nearly twenty years ago, two National Association of Scholars members nailed an attempt by race preference zealots in the federal bureaucracy to rig hiring and promotion standards.   But for their efforts, mandatory race quotas would have been spirited into the 1991 Civil Rights Acts.   The goal of the preference advocates was to redefine “fairness” to require that minority candidates get extra points on standardized tests.  The subterfuge was called “race norming.”  NAS members Jan Blits and Linda Gottfredson, both at the University of Delaware, blew the whistle with the result that Congress amended the bill.  Since then the devious technique of “race norming” (explained below) has been illegal.

I mention this not out of mere love of history.  Rather, we have a new reason to celebrate Blits and Gottfredson’s signal victory.   It provided the statutory authority for Justice Kennedy’s majority opinion in the Supreme Court’s decision last month in Ricci v. DeStefano.   NAS has been fighting race preferences since its founding.  In our roll call of major civil rights victories where we played a significant part—Proposition 209 in California, Initiative 200 in Washington, the Michigan Civil Rights Initiative, the Nebraska Civil Rights Initiative—the Blits and Gottfredson exposure of race norming has a proud place. 

NAS has fought with uneven results many other battles against race preferences.  We filed about 70 Freedom of Information requests aimed at exposing race preferences in undergraduate admissions at public colleges and universities.  Sometimes we exposed illegal practices and helped put a stop to them.   More often we encountered evasive bureaucrats and more varieties of stone wall than divide all the farms and fields of New England.  We filed amicus briefs in the Gratz and Grutter cases and published a detailed refutation of the Gurin report (the edifice of bogus statistics that the University of Michigan put forward in Gratz to show that racial preferences make students into better thinkers)—but, alas, Justice Sandra Day O’Connor didn’t see things our way.  We have devoted conferences and journal issues to the critique of racial double standards and the corruption they breed in higher education.  Generally this has earned us opprobrium among the bien pensant of the university.  Recently we have been mixing it up with Virginia Tech over that institution’s plan to require faculty members to “demonstrate their commitment” to racial preferences as a condition for professional advancement.  Virginia Tech has adopted the strategy of hiding in the tall weeds until, it hopes, we go away. 

That record makes all the more gratifying Justice Kennedy’s recognition of the essential point scored by Blits and Gottfredson nearly two decades ago.  Not that the Ricci decision is all that we could have wished for. 

As Peter Wood observed here recently, we’ll probably be wrestling with its implications for some time to come.   As Peter noted the case probably won’t demolish the quota-driven hiring policies that are standard issue at most colleges and universities.  We’re happy, of course, that simple justice was meted out to the deserving New Haven firefighters who will now receive the promotions they earned.  What we would also have liked was the enunciation of a more general set of principles applicable to situations involving race relations such as those in Ricci.  

Specifically, we wish that the justices had addressed the central constitutional issue which came knocking once more:  Is the “disparate impact” standard in Title VII of the 1964 Civil Rights act compatible with the equal protection clause of the 14^th Amendment?  The court, through Justice Kennedy, sidestepped that question again, ruling more narrowly that the firefighters could be awarded the relief sought without addressing their 14^th amendment arguments.  But the landscape of “disparate impact” was not totally unaffected, and the most potentially significant aspect of Kennedy’s opinion rests squarely on the dogged spade work of nearly 20 years ago by Blits and Gottfredson.

The court’s majority, as Peter Wood notes,  resoundingly rejected New Haven’s contention that it was required to toss its test results simply because they failed to achieve proportional statistical balances among the black, Hispanic and white applicants for promotion to the officer ranks.  Remember, under the “disparate impact” standard, employers until now could be held liable for such a random racial mix even if they’d bent over backwards to avoid discriminating as New Haven obviously did.   This means that it won’t be nearly so easy to dispense with the many, many Riccis of the world simply to avoid litigation prodded by “disparate impact” claims. 

In support of the court’s ruling, Justice Kennedy cited the provision in the 1991 Civil Rights Act that amended Title VII by prohibiting “race-norming.”  That was the practice of altering job-related test scores on the basis of race.  Under this practice, adopted in 1981 by the United States Employment Services within the US Department of Labor, the test scores of blacks and Hispanics would be rated within each group, rather than against the aggregate of test takers within a particular cohort with everyone held to a single standard of evaluation.  Since these two groups frequently registered lower test scores, “race-norming” effectively and deceptively awarded individuals assigned to these groups additional points.  The intention of Congress in outlawing this practice, Kennedy held, was directly relevant to the circumstances in Ricci:

If an employer cannot rescore a test based on the candidates’ race, then it follows a fortiori that it may not take the greater step of discarding the test altogether to achieve a more desirable racial distribution of promotion – eligible candidates – absent a strong basis in evidence that the test was deficient and that discarding the results is necessary to avoid violating the disparate-impact provision.  Restricting an employer’s ability to discard test results (and thereby discriminate against qualified candidates on the basis of their race) also is in keeping with Title VII’s express protection of bona fide promotional examinations.

The statutory provision on which Kennedy bases this finding, 2000e -2(l), hasn’t been much mentioned since passage of the 1991 Civil Rights Act, but was apparently seized on by counsel for the firefighters,  and included in their brief submitted to the Supreme Court.  Its original enactment into law - and subsequent inclusion in the Ricci ruling – rests squarely on the rigorous research and unflinching tenacity of Jan Blits and Linda Gottfredson. 

If  you’ve never heard of “race-norming,” don’t be surprised:  neither had just about anyone outside of the Department of Labor either, until it was exposed by Blits and Gottfredson in a pair of articles (1, 2) published in 1990.  Here they described the development and metastasis of a devious and dubiously legal bureaucratic contrivance which, with no public debate or Congressional consultation, quickly came to wield ever-increasing clout in the regulation of employment and hiring policies in the United States.   Employers in both the public and private sectors were often unaware that “race-norming” was at work, and believed that the test scores of all job applicants referred to them from the Labor Department had been ranked according to a uniform standard.  This was especially significant, as Blits and Gottfredson noted, in light of the reliance employers often placed on test outcomes to predict the likely performance of prospective employees.   

In 1986, “race-norming” was challenged by the US Justice Department, which alleged that the policy unconstitutionally discriminated against job applicants who were not black or Hispanic.  The Labor Department, in response, agreed to delay further extension of the policy and commissioned a review of the policy by a panel from the National Academy of Sciences – NAS, one could say – which would then present policy recommendations back to the department.  The panel, citing its purportedly “scientific” findings, subsequently recommended that “race-norming continue in use. 

Blits and Gottfredson, however, closely scrutinized the panel’s “impartial” conclusions and found essentially subjective political preferences masquerading as disinterested scientific analysis.  The panel’s report, they observed, abounded with the confused, equivocal use of familiar terms associated with “fairness” in the hiring process. “Equal opportunity,” in its usage, for example, had evidently been redefined to encompass racially proportionate group outcomes in place of the traditional view that all job applicants should be treated according to the same standard.   Same label on the bottle, with very different medicine inside.  Blits and Gottfredson also found numerous implausible or highly debatable statistical interpretations of test score data, all of which just happened to support the practice of “race-norming” with the commanding authority of a “scientific” evaluation.

Not surprisingly, the panel’s recommendations strengthened the hand of employers and personnel directors already inclined to adjust test scores for achieving racial parity.  Just as quickly, it also became an unofficial but potent enforcement instrument for eager EEOC junior staffers, who could credibly threaten employers with “disparate impact” litigation if their workforces did not reflect the requisite racial proportionalism. 

In December 1990, Gottfredson brought the issue to the wider public, describing the “race-norming” and its supporting pseudo-science in the pages of the Wall Street Journal, just as Congress was deliberating over the content of a proposed new Civil Rights Bill.  Especially relevant was the probable construction of the bill’s “fairness” language (“Explanation of the Legislation,” as termed by the Senate Labor Committee), in view of the sleight-of-hand alterations which had occurred within the Labor Department and the EEOC.  The Committee’s interpretive language, as Gottfredson demonstrated, clearly incorporated the EEOC’s reading of “fairness,” which placed a very heavy burden on employers to avoid “disparate impact” outcomes in the hiring process.  If the bill were enacted as written, “race-norming” and hiring quotas would be imposed by statutory obligation in place of bureaucratic stealth. 

An appropriate public uproar arose as a result, to the acute embarrassment of the bill’s sponsors, who had steadfastly dismissed critics’ claims that it would mandate hiring quotas.  Now, however, they were obliged to re-write the statute with new language specifically prohibiting the use of “race-norming.”  Thereafter, the issue subsided fairly quickly.  The modified statutory provision, however, was permanently in place, and was at hand in 2009 to provide helpful legislative authority for Justice Kennedy’s majority opinion.   It’s no exaggeration to say that our colleagues in Delaware made it possible.

Blits and Gottfredson have remained vigilant in the fight against bureaucratic subterfuge. 

 You may recall that they more recently performed a useful service by outing the University of Delaware’s Orwellian Residence Life Program, with its mandatory multicultural indoctrination sessions and attempts to turn student resident assistants into thought police. They’ve been fighting the good fight for a long time though, and the Ricci decision actually owes a great deal to their efforts of two decades ago.