Throughout human history, right up until the 20th Century, people have been selected for. It was Thomas Edison in the 1920’s who first started using interviews as a technique for selecting new hires. Prior to that people were most commonly selected for jobs based on their family connections. But he was looking for inquisitive minds, not social status. So he developed a 150 question interview that allowed him to rate applicants. Whilst the world of work has moved on a pace, the interview has not changed much in the past hundred years. This begs the question, is there a better way to select the right person for a role?
“Most interviews are useless” exclaims Laszlo Bock, Google’s Senior Vice President of People Operations. They certainly have enough data to back the statement up. The internet behemoth deals with more than a million applications each year. “We looked at tens of thousands of interviews” says Laszlo, “and everyone who had done the interviews and what they scored the candidate, and how that person ultimately performed in their job. We found zero relationship.” (Laszlo Bock – Business Insider)
Bock is not alone in his view that the common job interview doesn’t work. Indeed behavioural scientists thoroughly dismissed the case for the interview over a decade ago. Summing up over 85 years of research into the accuracy of different selection methods, Schmidt and Hunter (1998) showed that interviews select the right person about 14% of the time. That is less than a one in seven chance – would you gamble the success of your organisation on those odds?
Recent research by Jason Dana at the Yale School of Management goes further. His team found that in some cases interviews can actually diminish the ability to select the right people. Our own research confirms this. In our studies of a client selection process where psychometrics and interviews are used, we found the combined results are less effective than the psychometric testing alone. Summing up their findings in the Journal of Judgement and Decision Making, Dana says “Our simple recommendation for those who make screening decisions is not to use them”.
Perhaps most disconcerting is that despite the continued barrage of scientific evidence, many big businesses continue to take senior hiring decisions without objective and scientific measurement. Instead, businesses are routinely paying expensive experts to provide an “in-depth” interview. More often than not this amounts to a selection process that is simply a barrage of unstructured questions that lead to mostly inaccurate inferences.
A recent example was a senior executive candidate who told us that he was asked the disconcerting question, “Describe your 12 year old self”. Even if you can remember your 12 year old self, what is the relevance to future performance? The continued use of the interview as a selection method should be cause for concern.
A wrong decision is just the start. As we know, selecting the wrong candidate is a costly business. It is not just the time and capital invested in another recruitment initiative, but because during that time you may not have the corporate capability to achieve the business objectives set out in your strategy. At best you stand still, perhaps you cover the responsibility by diffusing it across the rest of the board, compromising the energy and focus they have for their own remit, or you have a temporary promotion, both are still less than ideal.
The main problem is that interviews measure how someone behaves, which is situational. Our behaviour in a job interview is likely to be very different from our behaviour once we get the job. To effectively predict job performance we need to understand people at a deeper, more essential level (see The hfi Onion for more).
So what’s the solution?
Can it be a simple one? If we are to replace, or at least augment the interview, we will need a standardised and objective alternative that effectively predicts future performance by measuring someone’s core personality and capabilities. Thomas Edison provided standardisation by asking all of his candidates the same 150 questions. Each question had a single right answer and so his scoring was also objective. But his questions only scratched at the surface of a person’s knowledge. The predictive validity of asking engineers such questions as “Where is the Sargasso Sea?” and “Who was Leonidas?” is not obvious.
For modern organisations the inclusion of a well-designed and standardised interview can add value. However using the same questions again and again also has its drawbacks. There are innumerable websites devoted to answers for standard interview questions. As J. M. Henderson from Forbes magazine (the most useless job interview question) pointed out that if an internet search provides agreement on what the desired answer to any given interview question should be, then the answers will often be analogous to one another and, therefore, not very enlightening.
The real solution, as shown by countless research studies, and attested to by our clients over the past 30 years, is the inclusion of effective psychometrics. Carefully selected personality and cognitive ability assessments repeatedly deliver far better predictive performance than interviewing. The future is psychometrics, not interviewing.
To start using psychometrics to select the best candidates for your organisation, contact us today!