January 7, 2009

Off the deep end on Griggs v. Duke Power

Before I get to the main topic this morning, my thanks to everyone who has participated in the reader survey, which will stay available through the weekend. It takes just a few minutes (or more if you want to give me lots of ideas for topics). Tell me what to do by taking the survey!

On Sunday, George Will decided to use a think-tank paper last year by Bryan O'Keefe and Richard Vedder to argue that policies have unintended consequences. Thanks, George: we never knew that without your help. But because Will accepts O'Keefe and Vedder's argument at face value, I have to correct the record.

O'Keefe and Vedder make an argument that Vedder has made repeatedly over the years: the Griggs v. Duke Power (1971) decision discouraged employers from using intelligence tests and therefore falsely magnified the credential value of college degrees as the easier way for businesses to make distinctions among applicants. In the case, 13 African American employees of Duke Power complained that after the Civil Rights Act, Duke Power changed its promotion criteria to eliminate references to race and to add a high-school credential requirement as well as specific scores on two tests. Because the combination of these disproportionately affected African American workers, the plaintiffs argued, Duke Power was using race-neutral means to maintain discriminatory outcomes. The Supreme Court accepted the reasoning of the plaintiffs, and Griggs was a landmark in disparate-impact litigation. O'Keefe and Vedder argue that because the Court said that credentials and tests had to be tied to business necessity, businesses began to turn from general IQ tests to college diplomas as the main screening device used in personnel decisions. 

There are several reasons why this argument holds little water, and let's start with the case itself. O'Keefe and Vedder are correct only if the Court discouraged IQ tests and let educational credentials alone. Without that distinction, there's no argument that businesses used college diplomas as a substitute for IQ tests. So let's peek into the crucial passage:

On the record before us, neither the high school completion requirement nor the general intelligence test is shown to bear a demonstrable relationship to successful performance of the jobs for which it was used. Both were adopted, as the Court of Appeals noted, without meaningful study of their relationship to job performance ability.

Maybe I'm misreading the case, but it looks as if the Supreme Court said both credentials and IQ tests were indefensible unless tied to job performance. I don't understand why Vedder has made this argument over the years without addressing the obvious problem with his line of reasoning.

But even if the Supreme Court had written differently, or even if HR professionals developed the same misreading that Vedder did (in which case the fault lies with them, not with the Court), it's a stretch to tie credentialism to a specific case. To believe that, we would have to believe that in the entire history of industrialization no one thought about using educational credentials as a screening tool until the 1970s and then--pow!--employers discovered that some applicants and employees had college degrees and others didn't.

In the paper, O'Keefe and Vedder do not even attempt to collect or display evidence that any industry started using college degrees after 1971 when they had used IQ tests before. And the reference they use to imply a broad historical sweep?--

In fact, according to Staffing Industry Report, a human resources newsletter, 65 percent of companies reported using some type of pre-employment screen, up from 34 percent in prior years. (p. 12)

--is from a 2008 New York Times story titled Dilbert the Inquisitor. I have no clue what "up from... in prior years" means, but it's not pre-1971. I know what business history is. I've read business history. Bryan and Richard, you are not business historians.

Keep in mind the broader uses of this argument that Vedder's shown: because college expanded in significant measure due to businesses' inability to use IQ tests, we have credential inflation and a greater use of college that is warranted strictly by human-capital needs. Ergo, we should invest a lot less in college.

Well, Richard, we already have: starting almost with the time of Griggs, states have dramatically shrunk their subsidies of undergraduate education at public colleges and universities. Students and their families have continued to see college as a good thing, even though they are having to acquire more debt as a private investment instead of a substantially public investment. Part of that is credentialism, but if so, I don't think you can blame Griggs. There are arguments to make about the problems of student debt and college waste, but O'Keefe and Vedder's argument is bad history.

Listen to this article
Tags: George Will, higher education, Richard Vedder
Posted in Education policy on January 7, 2009 10:15 AM |