Race-norming’s likeliest successor is something called “banding.” If you see references to a “diversity-based sliding band,” do not expect to encounter something as agreeable as a Dixieland ensemble. No, the term is only a euphemism for the latest subterfuge to scuttle rank-order selection of top scorers on tests for hiring and promotion. It’s better, you see, to advance lower-scoring members (^f politically favored minority groups.
Under banding, employers tap employees from within a narrow band of mediocre scorers—say the 70-to-80 percent correct range—on objective tests of job skills. The band is designated for its demographic correctness. That is, job seekers are clustered within the band in racial/ethnic proportion to the local labor pool. Band selection thus effects the racial quotas mandated in the 1991 Civil Rights Act that President Bush so foolishly signed. As with race-norming (which evaluated blacks and Hispanics against lower standards than whites and Asians), the employer pays a considerable price in decreased worker productivity but supposedly comes off a little better than by simply drawing names out of a hat.
Washington’s bigwigs naturally are not cranking out press releases to announce that race-norming’s evil heart continues to beat inside banding’s bosom. To wit: former New York Mayor Ed Koch’s interest was piqued by my article last February in Chronicles entitled “Dirty Secrets: Race-Norming Lives On.” Mr. Koch (a Democrat who, to his credit, has taken a principled stand against racial quotas) wrote to Secretary of Labor Lynn Martin asking, in essence, “What gives?” She wrote back in the blandest of bureaucratese to say that the Department of Labor had terminated race-norming in compliance with Section 106 of the Civil Rights Act. She even sent him a copy of a Federal Register notice from last December 18 as proof.
What the Secretary said was true enough. (The Federal Register notice was released after the deadline for my article in Chronicles.) What she did not explain was why the DOL had dragged its feet for more than a year in halting the race-norming of the General Aptitude Test Battery, administered annually to millions of job seekers. Her predecessor, Elizabeth Dole, had published a notice of suspension of the GATB in July 1990 following disclosure by the Richmond Times-Dispatch of the secret race-norming tables. Nor did Martin explain win the DOL dillydallied another two months after Section 106 went into effect before putting out the word to job agencies and employers to stop rigging test scores. (Congressional quotacrats wrote Section 106 into the law after concluding it was pointless to try to defend race-norming. But the rest of the bill essentially requires what was outlawed: that test results be fiddled with or set aside if they produce an “adverse impact” on designated minority groups.)
Nor, finally, did Secretary Martin tell former Masor Koch about what heats the fevered brows of Washington’s civil-rights elite these days: banding, a benign-sounding subterfuge they hope will pass legal muster. No, banding has not yet become official practice (at least as far as is known) at job agencies or at the corporate personnel offices they hector. But if a closely watched California court decision is sustained on appeal, racial banders will be beating the drums big time.
In a 39-page (!) decision, U.S. District Judge Robert Peckham ruled that San Francisco could depart from a merit- based, top-down system of promoting cops to sergeant in favor of hiring instead from within a narrow band of lower scorers. The judge noted enthusiastically that banding “decreases adverse impact by creating the possibility that minorities within the band who otherwise would not have been selected, will be promoted.” And he breathlessly observed that “the percentage of qualified minority candidates available for promotion will increase the farther down the list the City moves to make its appointments”! Which raises the limbostick question: how looooow will they go? Exactly how far down are egalitarians willing to drive considerations of individual merit—let alone considerations of public health and safety—in pursuit of statistical equality?
Banding’s proponents make the audacious claim that the science of probability legitimizes their techniques. Basically, they argue that because tests are not perfect and because individual scores may fluctuate each time a test is administered, there is a range of “statistical comparability” in which lower scorers may be leapfrogged over higher scorers. If scores run from 80 to 90 within a band, a score of 80 is deemed just as good as a score of 90. (That may be true in very specific instances—a test taker had a bad day, say. But averaged over thousands of job applications, higher scores have to be better than lower scores, if the tests are valid. And the nondiscriminatory way to go is with the high scorers.)
I’ll confess that much of this arcane psychometrics passes over my head, but it is good enough for me that the University of Iowa’s Frank Schmidt, one of the nation’s foremost testing gurus, has pronounced banding to be just as much a crock as race-norming. It utterly lacks scientific justification, he has written.
Well, you ask, didn’t Congress intend Section 106 to bury such pernicious subterfuges as banding along with race-norming? Ha! Who knows what Congress intended, least of all that decrepit institution itself. The San Francisco case is on appeal to the 9th U.S. Circuit Court of Appeals, where judges are again asked to decide what they think Congress thought.
The city’s police association argues that banding is “nothing but a thinly disguised (illegal) quota system.” However, the ubiquitous Lawyers Committee for Human Rights shoots back that Section 106 only forbade employers from “adjusting” or “altering” the results of job tests. “Congress in no wav intended to bar employers from banding test scores for affirmative action purposes,” say these rightsters/shysters. Indeed, they add, scores are not tinkered with at all, and “the proposed banding method is completely blind to race. It docs no more than to determine which actual scores are statistically equivalent, thus eliminating the potential for statistical error of measurement.” This is an outrageous stretch, given that the bands arc selected precisely because of their racial composition.
Finally, then, to the question Ed Koch asked me after receiving his reply from Secretary Martin: “Do you believe that the circle is now closed and that race-norming no longer exists?” Answer: no. Race-norming by any other name is still race-norming. And one of its most perverse results is that strivers and achievers of all races may be bypassed in favor of the mediocre. With diversity-based bands sliding ever southward, no longer will it pay to shoot for excellence.
Leave a Reply