Human Assets

 

Nowadays, personality inventories form part of the selection process for many organisations. This is particularly the case when selecting for leadership or leadership potential.  However, we need to be aware that when a candidate wants a job there is a strong incentive for them to present themselves favourably.  This is understandable and can occur across the selection process.  For instance, how many candidates have told interviewers that their biggest weakness is "working too hard"?  Similarly candidates can be tempted to respond to personality questionnaires in a way they assume the employer wants.  If we consider a sales job, how many candidates will disagree with the statement "I enjoy talking to people"?  

 

 

One way to reduce faking in personality questionnaires is to use forced-choice formats whereby candidates are presented with two or more statements and must indicate a preference between them.  For example, a question might ask a candidate to choose whether they are more "hard working" or "creative".  Because the candidate cannot make him or herself look good on both the statements it has been widely thought that faking is reduced.  Forced-choice questions (often known as ipsative questions) are widely used in personality questionnaires designed for selection.  However, a recent study suggests that such forced-choice formats might be susceptible to faking after all. 

 

 

A team of psychologists in the US led by Dr Eric Heggasted looked at faking at the individual level rather than the group level that had been examined in previous research.  They asked participants to complete three personality questionnaires.  The first was a standard personality questionnaire which asked respondents how strongly they agreed or disagreed with statements like the "I enjoy talking to people" statement above (often known as a "Likert scale").  And then two questionnaires designed for the study that each had the same content but the format of the questionnaire was either forced-choice or used a Likert scale.  The participants were asked to complete the first questionnaire honestly and were then asked to complete the other two questionnaires while "pretending that you are a job applicant for a job you really want ... with the goal of making the company believe that you will make a good employee".   If forced-choice questionnaires are resistant to faking then we would expect a strong link between the respondents' results on the 'honest' questionnaire and on the 'dishonest' forced-choice questionnaire.  But the researchers actually found that the forced-choice format was no better at indicating 'true' preferences under the faking conditions than the traditional Likert-type format.  

 

 

The researchers state that using forced-choice questionnaires for selection can be expected to result in decisions that are no better than the traditional Likert-type format.  Heggestad and his colleagues conclude that they cannot recommend forced-choice questionnaires for use in personnel selection because faking will distort the outputs of the questionnaire and make it less likely that the best person will be chosen. 

 

 

One response to this study might be to think that personality questionnaires should not be used for selection.  However we are not so pessimistic! 

 

We prefer to use techniques such as assessment centres to check a candidate's actual behaviour rather than relying entirely on their own description of their personality.  Nonetheless, we will still recommend personality questionnaires to our clients when they are carefully chosen and integrated into a selection process.   Personality questionnaires are not 'perfect'  but neither is any selection method and organisations can't wait until perfection is achieved.  To use personality questionnaires effectively we always recommend that they are not used in isolation and that ideally the outputs should be explored in other ways.  One suggestion is to arrange an interview after an assessment centre to discuss behaviour at the centre as well as personality information in order to gain a more holistic understanding of the candidate.  This has two benefits: (1) If candidates know that their personality questionnaire results will be discussed and 'checked' then they may be less tempted to fake in the first place; (2) A trained and skilled interviewer can explore discrepancies between behaviour and preferences to help the employer understand how likely someone is to consistently demonstrate a behaviour and also what hinders or facilitates behaviour. 

 

All Human Assets consultants are occupational psychologists and we keep up-to-date with research, such as this study, which enables us to make informed recommendations and advise clients on the best way to use personality information.  In addition, Human Assets is not a test publisher so we are free to recommend the personality questionnaire that best meets a particular client's needs and have no incentive to pressure our clients into a "one size fits all" approach. 

 

If you would like to find out more please contact our consultants on +44 (0)20 7434 2122 or at enquiries@humanassets.co.uk

 

 

Heggestad, E.D., Morrison, M. & Reeve, C.L. (2006).  Forced-choice Assessments of Personality for Selection: Evaluating Issues of Normative Assessment and Faking Resistance.  Journal of Applied Psychology, 91 (1), 9-24.

 

 

Rust, J. & Golombok, S. (1999). Modern Psychometrics: The Science of Psychological Assessment. (2nd ed.). London: Routledge.  

 

Newsletter: March 2006