You Validate Parking, Why Not Tests? Part 1

Lately, I have been so wordy that I have had to separate articles into two parts. This is yet another example of having so much to say about hiring and promotion practices that it will take more than one article to do it. In this first article, Part 1, I discuss a major lawsuit involving a large corporation and use it as a learning example. In part two, I’ll discuss how common HR practices will inevitably lead to similar challenges for other large companies. Atlanta Journal Constitution, Friday, May 3, 2002. Headline: BellSouth Faces Bias Suit. “Five African-Americans accuse BellSouth of using unvalidated tests and using those tests in a discriminatory manner to deny African-American employees opportunity for advancement.” I’m not an attorney and I don’t know all the details of this case, but this article reported that 1) a federal judge was considering whether to extend the bias suit to cover 15,000 (!) black employees, 2) the company was mounting a defense and, 3) four law firms were forming a money-line queue. Expensive news! Now, realistically, winning or losing this suit is not important. It will not keep BellSouth’s name out of the front-page business news, it will not save legal fees, and there will probably be a multimillion-dollar settlement before it is all over. Nevertheless, we’ll use its allegations to make an important point about hiring. The plaintiffs claim that:

Blacks always had to take a specific test, but whites were occasionally exempted.

Blacks tended to fail parts of the test (compared with whites) in a disproportionate number.
The company dropped a specific section of the test that blacks tended to pass.
The test did not reflect skills needed for the job.

Does any of this sound familiar? Be truthful. Are some people in your organization being given tests while others are exempted? Do you know if minorities fail certain tests disproportionately? Are you doing anything about it? Are people making promotion decisions based on clear job requirements? Do hiring and promotion tests reflect skills needed for the job? Can you prove that a certain test score predicts job performance? Do you realize that a “test” is any decision-making method (including interview questions)? In short, is your organization about to share the headlines with BellSouth because someone in HR is clueless about hiring and promotion regulations? I’ve consistently maintained that every HR professional ought to be thoroughly familiar with the Uniform Guidelines on Employee Selection Procedures. More often than not, I usually get an, “Oh yeah?” “Who says?” “Who cares?” or “You’re just trying to scare people.” Tsk. Tsk. Hiding one’s head in the sand or stoning the messenger is counterproductive. I’ll bet the HR people at BellSouth now have a greater appreciation for the Guidelines. Also, the people at Coca Cola, Lockheed Martin, The Home Depot, Waffle House and a host of other organizations you can find listed on the EEOC’s PR-release website (www.eeoc.gov/pr.html). Yep. I’ll bet they all care. But most of these errors never would have happened if HR departments just followed a few good employment and promotion practices that have been around for almost 25 years. Good employment practices start by knowing what the EEOC usually does when investigating these kinds of charges:

They look for systematic job analyses that defined job requirements (no, compensation banding and job descriptions won’t work).
They look for tests that directly relate to job requirements.
They examine each test for adverse impact and ask to see studies showing that test scores predict job performance.
They check to see if test results are monitored and if other methods with less adverse impact are investigated.
They look to see if everyone is treated the same.

You’ll notice that the government never says organizations should hire unqualified people. They just want organizations to ignore factors that have nothing to do with job performance?? i.e., age, gender, religion, and color?? you know, so you can keep your applicant pool as broad as possible and only offer jobs to fully qualified people. It’s simple, really. No great government conspiracy. Just do your homework and make sure tests relate to job performance while you continually search for better test methods. Isn’t that just good business? What? You say you don’t use tests? Just job descriptions and interviews? Sorry, you just don’t get it. A job description is only a definition of what a job is supposed to produce, the scope and depth of authority, and so forth. And as far as interviews are concerned, what is an interview other than a verbal test? Bottom line: unless you hire (or promote) everyone, you are, by definition, using some kind of test. Willing or not, you are “in the game” and your organization’s legal exposure and overall performance will be directly proportional to HR’s ability to understand and apply the Guidelines. In Part 2, I’ll discuss how common HR practices can lead to similar challenges for other large companies.