In 2007, Vicky Sandy applied for a job as a cashier, bagger and stocker at a Kroger supermarket in West Virginia. As part of the application process, Sandy was asked to take a 50-question "personality test" that would predict whether she would be friendly and communicate well with customers.
The test, called a "Customer Service Assessment" ("CSA"), was designed by Kronos, a workforce management solutions company, and reportedly evaluates characteristics that could factor into a person's job performance. For example: Is she patient, a team player? Does he listen attentively and respectively?
Those with higher CSA scores are supposedly more cheerful and friendly, and are better able to listen carefully and communicate well with customers than those with lower scores.
Sandy, who is hearing- and speech-impaired, scored a 40 percent. Her post-test results showed that she was less likely than other applicants to "listen carefully, understand and remember" and suggested the job interviewer listen for "correct language" and "clear enunciation," the Wall Street Journal reported.
She was not hired, and subsequently filed a discrimination complaint with the Equal Employment Opportunity Commission. In court documents Kroger stated that it had based its decision in part on her low Assessment score.
While the EEOC is still investigating the claim, the issue has brought up larger questions about personality tests, which have become common in the hiring process and are often used to screen applicants in the finance, technology, healthcare and the retail industries.
"If you use these tests with all the other components—a good resume, a very solid interview process with behavioral-type interview questions, the way the candidate presents him or herself, and a personality test—you should have a full picture of the person through all of these lenses," Michele St. Laurent, a recruiting practice manager at Insight Performance, a human resource consultancy in Dedham, Mass., told ABC News. "That should predict the success of the candidate in the role for a long time."
According to a 2011 poll from The Society for Human Resource Management, 18 percent of 495 randomly selected HR professionals use some kind of personality test in the hiring or employee promotion process. Of these, 56 percent use them for mid-level managers, followed closely by executives (45 percent) and entry-level exempt jobs (43 percent). Seventy-one percent of them said that personality tests can be useful in predicting job-related behavior or organizational fit.
But what the SHRM survey did not say is that they can potentially be used to discriminate against certain potential employees because of, say, their race or gender. From Oct. 1, 2011 to Sept. 30, 2012, the EEOC received 164 charges of discrimination challenging an array of employment tests—including, but not limited to, personality tests, Justine S. Lisser, a spokesperson for the EEOC told ABC News. The EEOC received 100,000 charges of discrimination during the same period, she said.
"Are you using a test to screen out applicants, or to provide insights on people who you are interested in?" added Daniel Schwartz, an employment lawyer at Pullman and Comley, in Hartford, Conn. "Those are two different reasons."
Personality tests could be discriminatory, said Lisser, if they are intentionally used to treat members of a particular ethnic or religious group worse than others. "There would be disparate treatment discrimination, for example, if an employer only gave personality tests to Hispanics because he thought them inherently untrustworthy and did not give the tests to anyone else," she said. There would also be discrimination "if an employer gave a test to everyone, but the test disproportionately screened out African-Americans and did not predict job success."
Under the Americans with Disabilities Act, if a personality test involves a medical examination—that is, if it measures the individual's physical or mental impairments or health—the test can only be administered until after a job offer is made. (The Minnesota Multiphasic Personality Inventory, for example, is considered a medical examination).
But "if an employer uses a personality test that is a medical examination to screen out or withdraw an offer from someone because of an impairment, then the employer would have to show that the particular job applicant is either unable to perform the essential functions or fundamental duties of the job, or would pose a direct threat," she said.
Also at issue is whether these types of tests actually tell employers anything useful about how future employees might perform on the job. "My research persuaded me that they really don't," said Annie Murphy Paul, author of the Cult of Personality.
"These tests may make Human Resources people feel that they're doing a good job sorting the application pool, but because personality is situational and because these tests are actually not very reliable in terms of their results, they're not a good way to evaluate perspective employees."
Susan J. Stabile, a professor of law at the University of St. Thomas School of Law, in Minneapolis, agreed. "A lot of these tests that measure aspects of personality don't measure things that are particularly job-related," she said. "Let's say someone uses Myers-Briggs. So you know someone is an INFJ—that doesn't necessarily tell you anything about how they're going to do a particular job. There are values to tests like that, but I don't see evidence that they help you get better employees in the end."
Although there have been few lawsuits—primarily because most failed job seekers don't ask for their test results-- there have been some notable cases. According to the Wall Street Journal , in July, Leprino Foods Inc., a Denver-based mozzarella cheese maker and a government contractor, agreed to refund $550,000 in back pay to 253 African-American, Asian and Hispanic job seekers who were denied laborer positions after failing a job skills assessment test called WorkKeys
The government's Office of Federal Contract Compliance Program, which prohibits federal contractors and subcontractors from discriminating on the bases of race, color, religion, sex and national origin in their employment practices, determined that the test's focus on math and observation skills had no relevance to the entry-level jobs that were available.
So are these kinds of tests going to go the way of the pterodactyl any time soon? Unlikely. But they will force employers to rethink the types of tests they use, and whether the tests are even necessary for the available position.
"The main lesson is: Don't solely rely on those tests to hire somebody," Francine W. Breckenridge, a labor and employment partner at Strasburger Attorneys at Law, in Austin, Texas, told ABC News. "If you are going to use it as a factor in employment consideration, you need to make sure the tests results are not disproportionately impacting a protected class."
As far as the Sandy case goes, Sabine, the law professor, is unsure whether the case counts as discrimination. "The important question seems to me to be whether the test in some way disadvantages hearing- and speech-impaired persons," she said. "If the test itself was discriminatory, there is a good argument for an ADA violation. But if the test does not disadvantage hearing- and speech-impaired persons, it seems to me the employer can legitimately refuse to hire a person on the basis of the low CSA score."
A spokesperson for Kroger did not respond to phone calls from ABC News.