Employment testing on trial

Dear HR Executive,

If you pay attention to the news, you know one of the big stories of the day is President Obama’s nomination of Sonia Sotomayor to the Supreme Court.

And you probably also know that a lot of attention is being paid to a decision regarding employment testing made by the federal appeals court that she sits on – Ricci v. DeStefano. That decision has been bumped up to the current Supreme Court, where a group of white firefighters argued they weren’t promoted by the city of New Haven, CT, but should have been.

The headlines in the general press are about reverse racial discrimination. But for HR folks, the focus should perhaps be elsewhere – on the employment testing that caused the trouble in the first place. The white firefighters passed the test, but when no black candidates did, the city threw it out.

What to watch out for with employment testing
The city was properly concerned about employment testing that appeared to have an adverse impact on minorities. That’s one caution that the EEOC and employment lawyers have frequently issued to employers about employment tests.

Pending the Ricci decision, though, the way that point is going to apply in the future isn’t yet clear.

Even so, there are four important points about employment testing and aptitude assessment that will help you stay on the right side of the law if you bear them in mind.

1. Match the employment testing to meet your needs. The test should measure the quality you want measured, and not something else. Too many organizations use blunt instruments when a scalpel is needed.  For example, a knowledge test probably won’t tell you a thing about a candidate’s ability to learn, lead or solve problems. An integrity test will speak to an applicant’s attendance and avoidance of drugs, but it won’t give you much of an idea how well she’ll work in a team. Also, make sure the test doesn’t demand more skill or effort than the job. The courts have labeled as discriminatory tests that call for applicants to show more physical strength than a job requires, for example.
2. Make sure the test is valid. Whether you buy a test from a commercial vendor or create one yourself, you must ensure that it’s valid. That’s to say, it must actually predict job performance. In many recent testing-discrimination lawsuits, a test’s validity has been successfully challenged. A test’s validity coefficient, which your vendor should be able to provide, can theoretically range from 0 to 1.0. Specialists caution against placing confidence in a coefficient of over .70. And if the coefficient is below .10, the test is next to useless. Generally speaking, anything above .20 is useful, and above .30 is very useful. 
3. Use a variety of assessments. It’s a mistake to rely too much on a test in isolation. A good rule of thumb is that pre-employment testing should make up no more than one third of your selection process. Structured interviews, job simulations, and careful reference checks will help you balance the results of a candidate’s test(s). And it’s a good idea to do your testing early in the evaluation process, so you have ample opportunity to follow up if the applicant’s score raises questions.
4. Consider job simulations as part of employment testing. Of all the employment assessment tools available, job simulations may be the most versatile. They can assess learning abilities, customer service skills, teamwork potential, leadership ability, motivation and other qualities. The simulation can either be an actual performance of specific job-related tasks, or a multiple-choice online test featuring video simulations of various situations.

We’ll keep you posted about the Ricci decision and its implications for employment testing and adverse impact as soon as the Supreme Court announces its ruling.

Dave Clemens
Editor-in-Chief
The HR Café Newsletter

UPCOMING AUDIO CONFERENCE: Don't miss B21’s upcoming audio conference , "No More Performance Reviews! A Revolutionary Approach to Performance Feedback" Most employees hate performance evaluations. Most managers dread giving them. Yet more than 70 million U.S. workers receive them each year. Why? Advocates of performance reviews say they raise productivity, differentiate salaries and protect against lawsuits. But it’s not true. No wonder just 5% of HR professionals are “very satisfied” with their performance management systems. On Wednesday, June 24, our speaker, Gary Markle, one of the nation’s most outspoken experts on performance management systems, will explain how you cope with this challenge.