Place your Selection Bets - Personality or Performance

Place your Selection Bets: Personality or Performance?

Personality 'testing' has a huge following amongst busy HR professionals. It appears common-sense that some personalities will be better suited than others to high job performance. Furthermore, administering a personality inventory is a relatively cheap and efficient way of sorting applicants. Understandably enough, businesses order personality inventories as a supposedly sound and scientific way of choosing the most fitted applicants for their vacancies. But is it all too good to be true?

The short answer is 'yes'. The common-sense behind the use of personality inventories melts away on scrutiny and the empirical evidence to support them is distinctly shaky. First, the common-sense. Businesses want performance. Most selection methods (good and bad) try to get a direct insight on performance. These methods range from references and word-of-mouth recommendation through to internships and assessment centres. On the other hand personality inventories are self-reports of qualities that are thought to lie behind performance. There are three hazards here. First the self-report might be inaccurate. Second, the qualities might not lie behind performance. Third, the job situation might overwhelm the influence of personality so that the nudge in one direction from personality is more than counteracted by the push in another direction by circumstances. All three problems are overcome if we observe people directly in the job situation or its simulation in an assessment centre.

Of course, these theoretical problems could be put to one side in favour of the empirical test - Never mind the doubts, do personality inventories work? Well, no they don't. There has been an interesting set of articles in the leading journal, Personnel Psychology, over the winter months. The first article was by six past editors of the Journal of Applied Psychology and of Personnel Psychology, under the title 'Reconsidering the use of personality tests in personnel selection contexts'. These luminaries look at the evidence over the past 25 years on the use of personality tests and state that the validities "are still close to zero", later quoting an uncorrected validity coefficient of approximately .10 between personality and proficiency. This is a trivial figure.

Not surprisingly, this onslaught provoked a reaction and this was provided by the academics Robert Tett and Neil Christiansen as well as an advocate of integrity tests, Deniz Ones, and colleagues. However, these authors try and repudiate the criticisms of personality testing by resorting to various corrections to the meagre 'headline' relationship between personality and performance, such as correcting for unreliability in the measurement of performance. These corrections are not wholly convincing. As the journal editors comment, "when organizations use personality tests, they do not correct the scores. They use observed scores" and the relationship between uncorrected scores and performance "is very low and often close to zero".

There is a good deal more in these articles to keep the connoisseur of such debates interested for several hours but the HR practitioner should take away the bald truth that personality inventories need to be scrutinised very carefully indeed before you entrust to them decisions about people's suitability for key positions, such as your future leadership.

How can Human Assets Help?

We are a team of psychologists who are entirely independent of any particular product. We will recommend to you the method of choosing people that is the most likely to work in your particular circumstances. Using our innovative indicators of excellence methodology, we examine the roles for which you are selecting people. We then take account of all the practical circumstances before recommending the most suitable method of selection. We see our recommendations through to the follow-up evaluation. We are particularly well-known for our work with assessment centres, but we are certainly not wedded to them. Each client's requirement is considered on its merits and often an alternative to an assessment centre is the best solution. For example, in recommending to the Law Society and Bar Council how to award advocates with the accolade of Queen's Counsel, we steered them away from assessment centres to a sophisticated method of gathering peer and judicial references. Nor are we anti-personality. We just know that personality inventories are a dubious way of choosing people.

For a confidential and without obligation discussion of your selection procedures, please contact our consultants on +44(0)20 7434 2122 or by email at charles.woodruffe@humanassets.co.uk

References

Morgeson FP, Campion MA, Dipboye RL, Hollenbeck JR, Murphy K, Schmitt N. Reconsidering use of personality tests in personnel selection contexts. Personnel Psychology , Autumn 2007, Vol 60, No 3, pps 683-729.

Tett RP, Christiansen ND. Personality tests at the crossroads: A response to Morgeson, Campion, Dipboye, Hollenbeck, Murphy, and Schmitt (2007). Personnel Psychology , Winter 2007, Vol 60, No 4, pps 967-993.

Ones DS, Dilchert S, Viswesvaran C, Judge TA. In support of personality assessment in organizational settings. Personnel Psychology , Winter 2007, Vol 60, No 4, pps 995-1027.

Morgeson FP, Campion MA, Dipboye RL, Hollenbeck JR, Murphy K, Schmitt N. Are we getting fooled again? Coming to terms with limitations in the use of personality tests for personnel selection. Personnel Psychology , Winter 2007, Vol 60, No 4, pps 1029-1049.

Newsletter: January 2008