Personality: measurement vs common sense
I am often challenged by people inside and outside of this industry as to whether personality does actually predict performance at work. Whenever this happens, I find it helpful to consider the converse: I think it would come as a great surprise to most of us if anyone asserted that someone's personality had nothing to do with behaviour and performance in work.
The notion that people behave rather consistently across different environments and over their lifetime (i.e. that they have an underlying personality) is something that we just intuitively 'know'.
However, there are many things that we intuitively 'know' that aren't necessarily based on evidence: goldfish actually do have a memory longer than five seconds; reading in dim light doesn't actually ruin your eyesight; allowing children to try alcohol under parental supervision may or may not prevent them from become teenage binge drinkers.
In the case of personality, our intuitive grasp is often borne out by the science - but the science can also iron out the errors introduced by gut instinct.
Of course there's going to be a difference in personality between, say, your average computer scientist and a typical salesperson. When thinking about career possibilities, it is completely natural to have a look at someone's individual likes, dislikes and preferences in trying to figure out what kind of career they might be happy in.
Yet if you mention psychometrics to 'the person on the street', s/he will often be extremely sceptical. This is understandable in somebody who isn't familiar with the scientific discipline of psychology, but puzzling when you contrast psychometrics with a method that is generally accepted: namely, the interview.
Psychometrics provide a baseline of measurable and independent criteria about someone. This is in contrast to the kind of personal judgements put to use during an interview; these tend to be informal instincts about how people differ, but suffer from a number of predictable pitfalls. For example:
- The 'Horns and Halo' effect: if someone does one thing badly (or well), we then assume that this is representative of them as a whole, rather than seeing that one incident within a wider context; this effect is even more powerful if the notable event happens early in an interview
- 'Similarity Bias': we tend to prefer people who are 'like us', rather than necessarily looking at how well they will do the job
- 'Non-threat Bias': there tends to be a reluctance to recruit someone who will have a competitive set of skills to our own (but might nonetheless round out a team's overall skill set).
So the beauty of psychometric measurement of personality, when applied appropriately, is that it the results are repeatable and that it does not suffer from the kinds of preconceptions that people's day-to-day judgements often show.
A recent and prominent example of this is Alan Sugar, who 'fired' his last two contestants in the Apprentice because a) he is unimpressed with qualifications, and b) doesn't like engineers. If Lord Sugar were to use psychometrics for recruitment and retention, it could help him to back up his intuition where it is correct, but also to challenge the inevitable and counterproductive biases that intuition alone can produce.
However, it might not make quite such good TV!