Why Hiring Needs Science
According to Glassdoor, the average US employer spends approximately $4,000 and 24 days to hire a new employee. However, despite this investment, many companies continue to miss the mark on their hires, with Career Builder pegging the average cost of one bad hire to be nearly $15,000. Top reasons for bad hires include hasty hiring, taking a chance on a nice person, candidates lying about their qualifications, and clients lacking requisite skills, but thinking they could learn quickly. To a behavioral scientist, these reasons are not surprising because they are a representative outcome of the myriad of biases that impact traditional hiring methods:
This list is not intended to be exhaustive, but provides a window into the many barriers to making an objective hire. Structured interviews and/or interview panels are step in the right direction, but many hiring managers often misunderstand the science of what they are actually correcting for – missing the benefits that they sought. For example:
The science of question design
Hiring managers often ask multiple behavioral interview questions to draw out experiences and examples of a candidate’s knowledge and skill - “give me an example of when…” Although we may think that we are measuring different factors each time, the research shows that a person’s performance on any behavioral interview question is dominated by the same three qualities: experience, job knowledge, and social skills. Consequently, there’s no need for the repetitiveness of these types of questions, which do not provide any additional information.
The science of measurement
Independent, specific, and objective tests are better predictors of performance than subjective, holistic impressions. Although we target the former, we often rely on the latter for the final decision because we often don’t have Specific, Measurable, Achievable, Relevant, and Transparent (SMART) goals. The fields making up the behavioral sciences (specifically psychology, behavioral economics, and organizational behavior) have developed a number of these tests to measure personality, cognitive, and behavioral factors. These fields have also developed rigorous criteria for development and validation of these tests, ensuring validity (the test measures what it claims to measure), reliability (scores are consistent and stable over different periods and environments), and fairness (the test is appropriate for all examinees irrespective of race, religion, gender or age).
The science of evaluation
Real-world studies on manager evaluations using 360 reviews by peers, bosses, and direct reports showed that more than 50% of the variance in the manager’s rating could be explained by the unique rating patterns of the rater and not the performance of the individual or the source group doing the rating - this is referred to as the idiosyncratic rater effect. A tip – we’re much better when we’re asked to comparatively rank people versus judge them individually.