 |
 |
 |
 |
 |
 |
 |
 |
« Back to the May/June 2006 Explorer newsletter
BEST PRACTICES
Are your job candidates telling the truth about themselves?
Richard Griffith, Ph.D., Florida Institute of Technology
In recent practice, personality measures have become increasingly popular as a hiring tool available to HR professionals. These tools tend to be relatively inexpensive and have demonstrated useful utility in improving selection decisions. In an increasingly service-oriented economy, personality measures are also useful indicators of interpersonal traits that lead to success in prevalent job categories such as customer service and sales. Moreover, personality measures have been shown to predict important out-of-role behaviors such as citizenship, integrity, and tardiness.
Despite considerable benefits, personality-based selection measures have been criticized for their transparency. It has been noted that applicants can simply provide answers to the questions that would make them appear to be top candidates, regardless of any shortcomings they possess in interpersonal skills. Unlike traditional ability tests, which have right and wrong answers, personality measures simply ask applicants to endorse items in a way that best describes them.
While in a technical sense these items are not keyed with correct and incorrect responses, often the “right” answer is apparent to the applicant. Consider the following question that appears in a published integrity test designed to assess the reliability of potential job applicants: The applicant is asked to indicate his or her agreement with the statement “I smoke marijuana on special occasions.” Regardless of the behavior of the applicant, there is definitely a “right” answer to that question, especially in an organization that promotes itself as a drug-free workplace.
While disputed by some researchers, there is now considerable evidence to suggest that a significant portion of applicants fake their responses to personality measures. In fact, between 30 and 50 percent of applicants elevate their scores (Donovan, Dwight, & Hurtz, 2003; Griffith, Chmielowski, & Yoshita, in press). This supporting research does little to ease the minds of practitioners who have had a long history of concern regarding the impact of faking on their selection decisions. Why are practitioners concerned? What is it about someone who fakes his or her score that HR professionals find undesirable? Much of this anxiety regarding applicant faking stems from assumptions that we have about deception.
While both unconscious and conscious forms of faking behavior have been identified (e.g., Self-Deceptive Enhancement and Impression Management), a great deal of the discomfort on the part of HR professionals can be attributed to faking behavior being seen as volitional. In general, we tend to target the locus of human behavior within the individual. Applicants who faked have lied to us. They deceived us. They tricked us into giving them a job. Our socialization regarding this kind of behavior is strong, and in general we associate deception with a lack of integrity. This categorization may cause those concerned with faking to extend their attitudes about deception to the employment setting. The secret fear of some employers is that fakers may not stop at the application stage with their dishonest behavior; they may steal, lie to customers, and conceal mistakes.
So are fakers going to hurt my organization?
The assumptions regarding fakers as poor performers remain largely untested, and there currently is little empirical evidence to support the notion that dishonest behavior generalizes to the workplace. Deception is a complex activity, and applicant faking behavior is likely to have similarly complex dynamics. The truth is that little is known about the characteristics of the faker and how those characteristics, as well as situational influences, combine to form the phenomenon of faking. Recent models recognize the dynamic nature of the phenomenon, and incorporate a variety of possible influences on applicant faking. Some of these factors, such as integrity, are internal to the applicant. Other factors, such as the presence of warnings or the desirability of the job, may be a function of the situation.
Deception is relatively common in our society, with some research suggesting that, on average, we tell 3 lies per 10 minutes of conversation (Smith, 2004). Often, deception is expected in certain social situations. We are socialized to put our “best foot forward” and to emphasize our strengths in an employment setting. Therefore, applicants may not be “faking” in the sense that they are intentionally attempting to deceive the employer, but may be answering in a fashion that they perceive to be consistent with the demands of the situation. Most likely, the deciding factor of whether applicants will elevate their scores in an applicant setting will be an interaction of the situation and applicant characteristics.
Do personality tests work when people fake?
Regardless of the faking issue, personality measures are still useful predictors of performance for most jobs, especially those positions that require a lot of interaction. Overall, organizations can see positive gains in their hiring efficiency when using these measures. However, the predictive ability of these tests would be greatly increased if the amount of faking were reduced, or if fakers could be identified and removed from the pool of applicants.
Some progress has been made in reducing faking. Warning applicants that if they fake they will be identified - and that there will be consequences - tends to reduce the effects of faking somewhat (Dwight & Donovan, 2003). However, some applicants will choose to ignore those warnings, and are actually more likely to be hired under those conditions. When the majority of applicants attend to the warnings, they will generally have lower scores than those fakers who ignore the warnings, and the fakers rise to the top of the scoring range. If you are only hiring a small number of employees, you now are likely to end up with the applicants who intentionally lied to you, and are more likely to present later employment problems. Some personality measures employ forced choice scales which make it more difficult to fake, and these scales show some promise in the overall reduction of applicant faking behavior.
Don’t psychologists have questions designed to catch fakers?
The use of lie detection scales to catch possible applicant fakers is one of the most ironic practices in personnel selection. Over the years many scales have been developed that are designed to identify fakers. These scales are typically referred to as social desirability scales and generally ask questions about behaviors that we all engage in but that we don’t like to admit to. The assumption is that a faker would deny these types of activities at a higher rate than the general population. The problem is that these scales tend to be more transparent, and thus more fakable, than the selection based personality measure we intend to use (Viswesvaran & Ones, 1999). To date, not a single study has demonstrated the ability of social desirability scales to effectively identify fakers. In fact, our current research suggests that they are completely unrelated to applicant faking behavior. So when we are telling applicants that we can catch them if they fake, we are actually lying to them!
So how do HR practitioners address the faking issue?
Ideally, when choosing a personality measure it is best to use inventories that are job related and have profiles specifically tailored to the particular job in question. Nonetheless, it is likely that the best defense against fakers is the faker themselves. While applicants may be able to fake one scale, they are less likely to effectively fake an entire profile unless they have extensive knowledge of the job. In many cases the faker will guess wrong on the items and actually reduce their score, essentially faking their way out of the job.
When shopping for a personality measure, ask the distributor about faking behavior and how it impacts the effectiveness of the instrument. If they tell you that applicants can’t fake, or that the test is designed to catch fakers, use a little healthy skepticism in your review. A reputable test publisher will tell you that while some applicants fake, overall the measures are effective when they are job related and used in combination with other selection tools such as ability tests, interviews, and background checks. Be careful, or test publishers may do a little faking on their own.
References
Donovan, J.J., Dwight, S.A., & Hurtz, G.M. (2003). An assessment of the prevalence, severity, and verifiability of entry-level applicant faking using the randomized response technique. Human Performance, 16, 81-106.
Dwight, S. A. & Donovan, J. J. (2003). Do warnings not to fake reduce faking? Human Performance, 16, 1-23.
Griffith, R. L., Chmielowski, T. S., & Yoshita, Y. (in press) Do Applicants Fake? An Examination of the Frequency of Applicant Faking Behavior. Personnel Review.
Smith, D.L. (2004). Why we lie. The Evolutionary roots of deception and the unconscious mind. New York: St. Martin’s Press.
Viswesvaran, C. & Ones, D. S. (1999). Meta-analysis of fakeability estimates: Implications for personality measurement. Educational and Psychological Measurement, 59, 197-210.
About the Author
Richard Griffith, Ph.D. is currently the Program Chair of the Industrial Organizational Psychology program at Florida Tech. Richard is an expert in the areas of testing and assessment of individual differences. He is the author of over a dozen publications and a recent edited book on the topic of applicant faking behavior. He has designed and provided training in the area of developmental feedback, employee retention, leadership, and team building for Fortune 500 companies. He also has extensive experience in conducting needs analysis, developing training curriculum, and delivering training in non-profit organizations. Dr. Griffith has designed and conducted team-building and leadership programs in the military, public, and private sectors. In addition, he has developed selection systems for Florida Law Enforcement, and served as a training and retention advisor to the Florida Police Chiefs Association. He received his degree in Industrial/Organizational Psychology from The University of Akron in 1997. Richard can be contacted at .
* * *