Psychology professor Richard Griffith is on a mission to stop "fakers."
To Dr. Griffith, of the Florida Institute of Technology, fakers are people who misrepresent themselves on personality tests increasingly used to screen applicants for entry-level jobs at call centers, retail stores and other customer-service positions. The tests typically ask candidates to agree or disagree with statements about their character and personality traits, hoping to shed light on how the job applicants approach tasks and problems.
Researchers such as Dr. Griffith say it's too easy to lie on some of these tests, because applicants try to predict the "right" answer, which can vary depending on the job. Applicants can also research employment tests on the Web and ask friends who have taken them previously.
In his research, Dr. Griffith collects personality-test responses from community-college students twice, about six weeks apart. The first time, the students' instructor administers the test and tells students to answer honestly. The second time, Dr. Griffith or his graduate assistants pose as recruiters, and distribute tests containing many of the same questions.
About 30% of respondents change their answers to achieve significantly higher scores. Faking "substantially harms the ability of the test to predict" people's performance, Dr. Griffith says. (The students are later told it's a research project.)
Despite the tests' flaws, Dr. Griffith says pre-employment tests are generally more reliable predictors of performance than an interview alone. But he says they could be more accurate.
Test makers analyze a job for the personality traits that would make workers successful, and then design questions to detect those traits, Dr. Griffith says. The right answer depends on what the employer is looking for; what's good for one job could be bad for another. The tests are designed to reveal an applicant's behavior patterns, not necessarily what someone would do in any particular situation, he adds. "For any given instance, I wouldn't bet a dime" that the assessment would necessarily predict a person's actions, he says.
Facing the perennial challenge of hiring the right applicant, more employers have adopted prehire tests in recent years, thanks in part to the Internet making distribution cheaper and easier. About 70% of entry- and midlevel jobs at big companies now include testing, says Scott Erker, a senior vice president at Development Dimensions International, a Bridgeville, Pa., human-resources consultancy.
Sherri Merbach, senior director for organizational development for Orange Lake Resorts, which sells time shares in Florida, says prehire personality tests are especially helpful in hiring salespeople. Sales applicants are usually able to make a good impression -- and thus interview well -- but sometimes offer answers that don't display great urgency, another quality that Orange Lake is seeking in its sales force.
Ms. Merbach has used bad test scores to dissuade colleagues from offering jobs to some suave candidates. "You're less likely to make a bad hiring decision," she says.
Employers and testing companies are aware that some applicants give misleading answers. So they include questions designed to weed out fakers. But Dr. Griffith says some of these tactics can backfire.
One common question designed to trap fakers asks applicants whether they have or haven't done something common but frowned upon, such as "I have never looked at a dirty book or magazine." In theory, liars are more apt to say they never have, while honest people admit the behavior.
But savvy applicants know to confess. "They fake those in the reverse direction," says Dr. Griffith. Another problem: Applicants who answer truthfully that they've never looked a naughty magazine can be labeled liars.
Ms. Merbach, from the time-share company, says the test she uses, by CraftSystems Inc., is sophisticated enough to reduce the likelihood of faking and is a good predictor of job performance.
CraftSystems, Bradenton, Fla., works with Dr. Griffith and has conducted studies to weed out questions that lend themselves to fake answers. Statements such as "I am highly motivated to achieve outstanding results" and "I am more outgoing than shy" were deemed too transparent. Respondents could tell what the question was looking for and answered accordingly.
Some employers say tests are useful, but need to be used together with other ways of judging applicants. Steve Suggs, president of sales at Sales Manage Solutions LLC, a Knoxville, Tenn., consulting firm, advises clients to use assessments to help make hiring decisions. But he cautions clients not to base hiring decisions solely on the test, because the answers don't always reflect all of an applicant's characteristics.
For example, he says test results may suggest that an applicant is cold and self-centered. In person, though, the applicant may demonstrate that he's learned to control these tendencies and could be effective. In a hiring decision, "there are other factors that come into play," Mr. Suggs says.
Email your comments to firstname.lastname@example.org.