Asset Managers are increasingly viewing recruitment as a cost to be cut, either by bringing headhunting in-house or advertising on platforms like LinkedIn. This approach may save money, but it often leads to an over-reliance on interviews as the primary method of assessing candidates.
But are interviews alone effective? Research suggests not. The key research on this is still Schmidt and Hunter’s 1984 meta-analysis, which examined the validity of various selection methods, including job interviews. They repeated their research in 1998, finding similar results.
They found that the typical validity coefficient for unstructured interviews in predicting job performance was approximately 0.14. For non-statisticians: the coefficient of 0.14 corresponds to the proportion of variance in job performance that can be explained by interview performance. To put this in percentage terms, interview performance explains about 1.96% of the variance in job performance. In other words:
Interviews alone only increase the chances of making a successful hire by 1.96% over a random appointment.
The odds can be slightly increased by using Structured Interviews, which are interviews where questions are standardised and asked in a consistent manner across candidates. My guess is that most DIY recruitment processes do not use structured interviews but, if they do, the validity coefficient increases to 0.34 – or, in percentage terms, the chances of predicting success increase to 11.6%. However, interviews alone remain a weak predictor.
Schmidt and Hunter found that combining interviews with other selection methods, such as aptitude testing, behavioural profiling and reference checking, significantly improves predictive odds. They recommended using structured interviews and combining them with other valid predictors to enhance accuracy. These include:
Cognitive or Aptitude testing: the highest validity coefficient, of about 0.51;
Personality and Behavioural testing: validity coefficient from 0.22 to 0.33 depending on the trait being measured;
Reference checks: validity coefficient from 0.26 to 0.34 (though we would argue that confidential references have higher validity).
A 1991 study by Barrick and Mount also recommended using work samples and job simulations, where possible.
The key takeaway is the need for systematic, evidence-based approaches to hiring decisions rather than relying solely on subjective assessments like unstructured interviews. It’s critical to use a combination of tools to help increase the predictive odds of a successful hire.
That’s why, at Godliman, we use all the recommended tools to assess candidates. As a result, 79% of all candidates placed since 2013 are still with the same company. And 51% of hires since 2003 are still with the same company.
Since the costs of a bad hire can far outweigh those of an effective recruitment process, the recent trend for DIY recruitment looks like a risky decision.
When recruitment is seen as a cost to be eliminated, the focus on efficiency often results in a haphazard approach to selection: you might as well be throwing darts at a list of names on a pinboard.
For more information on how Godliman can supplement your own Recruitment Processes, please contact us at hello@godliman.com