Can Job Applicants Outsmart Personality Tests?

By Kathleen Groll Connolly

Reprinted by Permission.

"Faking" is probably the biggest apprehension employers have about using personality tests during the hiring process. These concerns are well founded. Job applicants are confronted by more tests all the time. They are motivated to "pass the test" and come closer to a job offer. This should be no surprise, especially in tight economic times. To get the most from your investment in tests, you need to become skilled in test administration and interpretation.

Lynn McFarland, Ph.D., an assistant professor at George Mason University's industrial/organizational psychology program, studies the "faking" phenomenon. She works primarily with self-reported measures, such as personality tests and biodata. "Not just my research but the literature in general indicates that applicant groups score higher on 'social desirability' on tests than groups we can be fairly certain are responding honestly," she says. "Job incumbents or college students, for instance, have no consequence tied to the test. Therefore, it seems that at least some applicants do try to "fake" the test and are successful at doing so."

How Big a Problem?

Employers have forever advertised some clear preferences in personality traits of candidates: Total honesty is a no-brainer, closely followed by the ability to control one's own anger. Beyond this, conscientiousness, ability to work in teams and persistence are three top examples of universally desirable traits. Extraversion is also more attractive than introversion for many jobs, particularly when direct client contact is involved. If the "right" answers are an open secret, is testing a valid tool in selection? To answer that, it's important to look at some of the subtleties that experienced testers come to know.

Case 1: Some faking is devious and intentional. Let's say an applicant for a retail position intends to shoplift. He or she does not want (lack of) integrity to come out during the hiring process. The applicant might try to "beat the test" by providing what he or she believes to be the "correct" answers.

Case 2: Now consider the example of an applicant who seeks to please the employer with the "correct" answers so they can get past the test phase and get to second or third interviews. They may feel that testing doesn't show their real strengths, that it may exclude them from consideration for the wrong reasons.

Case 3: Finally, consider applicants whose self-awareness is very optimistic. He or she believes they have many socially desirable characteristics. Whether they really do or not is a question for the skilled interviewer to ascertain. Careful review of educational and work history might affirm the possibility that the optimism is well founded-or not.

For testing to be useful as an adjunct to interviewing and other hiring tools, each of the above situations need to be detectable by the employer.

How Tests Detect Faking

Most test publishers would not dispute that test profiles can be elevated or falsified. So they strive to overcome the problem.

" 'What's the point? Anybody can fake them!' We hear people ask that question," says test designer Robert Hogan, Ph.D., president of Hogan Assessment Systems. He claims this can be overcome with the "faking profile" that his company's tests generate. "We're doing a whole personality profile," he says of his company's tests. "They may be able to fake a whole scale (a single aspect of the measurement), but they won't fake a whole profile."

George Mason University's Dr. McFarland affirms that faking often is reported through special test scales: "I think the most common strategy to detect faking is to use what's called a social desirability scale, also called a lie scale. These scales ask the test-taker to indicate the extent to which a number of statements are true of them. The trick is that these statements refer to behaviors that are very common, but undesirable."

For instance, an example of such an item is "I have never been untruthful, even to save someone's feelings."

"A test-taker who denies many of these undesirable behaviors that are extremely common will receive a high socially desirable score," says McFarland. "The employer may use this to identify fakers." She has some doubts about the effectiveness of this strategy, however. For instance, some people may have unusually high scores on "social desirability" scales, but may simply be trying to please, may have poor reading skills, or may have overly optimistic beliefs about themselves.

Questionnaire wording is one of the primary tools in the battle to thwart would-be deceivers. People who give extreme answers to questions are often - but not always -overcompensating to hide their true intent than those who do not answer with the extreme choices.

Steps for Employers

What separates the experienced, effective user of tests from the ineffective? The dividing line is often defined by the test-givers ability to read the subtleties. It is also determined by following some preventive disciplines.

First, employers must cover the basics. They must choose a test that incorporates a "lie scale." They must choose a test with proven reliability and validity. They must become knowledgeable about the test or tests, usually through the publisher's certification programs. If this is not possible for a company staff member, then an outside consultant who is certified and experienced should be considered. Some other key measures include:

Consider the requirements of the job. Jobs have "personality requirements." For instance, the ability to work long hours without social interaction is far more "natural" to an introvert than an extrovert Likewise, the elevated confidence of an effective salesperson is often exactly what's called for in that job. The personality requirements of the job description can be examined through job analysis. This can help you assess the real importance of elevated scores in terms of your requirements.

Monitor the process. Though it may be convenient to have applicants take on-line tests at home, this can invite outright cheating. For instance, the applicant could have someone else take the test. (They could also be recording the test questions for distribution to other applicants.) All tests should probably be taken with adequate monitoring.

Use two tests. Let's say you are hiring for a job where anger-proneness would immediately disqualify a candidate. Two tests may be advisable. The chances of someone successfully deceiving two tests are considerably slimmer than one. Hogan Assessments, for instance, offers companion tests-one that deals with the applicant's visible, positive personality characteristics -and a second test to look at the darker sides of workplace behavior, such as tendencies to lose one's temper, to become disruptive, or to be disrespectful of authority or coworkers.

Tell participants they will be discussing their answers at a later date. Dr. McFarland says this strategy shows some promise. It holds an applicant responsible for consistent answers at a later date. "Applicants can be told their answers will be discussed in an interview if they are selected to go on," says McFarland.

For those who pass the testing screen, conduct interviews that are at least partially based on the test results. Some test publishers provide interview recommendations with the test reports, which can save you time. If someone has intentionally provided false or inflated answers, you may spot discomfort or hesitation during the interview.

Most applicants are not informed consumers of tests. If tests become more and more widely used, it is possible that some applicants will become experienced at beating the tests. For the time being, though, faking reports and other current strategies offer some assurance to employers. They must be aware that not all results that look like faking are actually untrue. Armed with these skills, you are ready to gain the advantages of pre-employment personality testing. Dr. McFarland summarizes, "I would use personality tests to select applicants. They have been shown to be valid selection measures. However, I would not base selection decisions solely on such tests."

Kathleen Groll Connolly writes on a variety of human resources topics and is a partner in Performance Programs, Inc., a firm specializing in human resources surveys and measurement. She can be reached at www.performanceprograms.com or call 1-800-565-4223. Copyright Performance Programs, Inc. 2003-2004.