Bridging the knowing-doing gap in hiring
Most hiring managers are truly clueless
Most hiring processes are shockingly dysfunctional
If you ask successful business leaders what they think are the most important causes for their success, they almost always give credit to the people they have surrounded themselves with. Successful managers are good at getting “the right people on the bus and the right people on the key seats”, as Jim Collins puts it in his book Good to Great. It should come as no surprise that teams that consist of the right mix of people perform better (a concept I explained in more detail in this article). Indeed, the scientific management literature recognizes human capital as a key driver of competitive advantage. The company with the best-performing employees often beats other companies in the quest for competitive advantage: after all, it is great people who are behind great products and services. The business relevance of making good hiring decisions is clear. It is an important skill that all managers should possess.
How is it possible, then, that most managers make woefully wrong hiring decisions? It has been well-documented over decades of scientific research that, when left to their own devices, most managers make hiring decisions that are so bad that they might as well select candidates randomly and hope for the best. It is true that there are many fields in management where science and practice diverge, but the gap is seldom as large as in hiring. In fact, recruitment and selection has recently been called “the supreme problem of applied psychology”. Not a title to be jealous of.
This problem can be referred to as the knowing-doing gap of hiring. In a way, it is strange that this problem even exists. After all, there is a very broad consensus amongst scientists and psychologists on how we can make sure we select the best candidates — those candidates who are most likely to perform well. However, in practice, we see that hiring managers simply don’t adopt those best practices. Why is that? What can we do to solve this knowing-doing gap? These are the questions that I will address in this article, based on decades of scientific research and insights from the world’s best organizational psychologists.
What we know is best
First, let us establish the goal of hiring, which is to select those candidates that are most likely to perform the best. In other words: in hiring, we want to screen for things that predict good job performance. Of course, job performance itself is somewhat of an ambiguous concept. It can be defined both objectively and subjectively. However, for the purpose of this article, that does not matter — no matter how you define job performance, its predictors remain mostly the same.
What are the most important predictors of job performance? First of all, we must make a distinction between jobs that are low in complexity and jobs that are high in complexity. In this article, I will only examine high-complexity jobs, because that umbrella term includes most jobs that are relevant in a managerial sense (for example accountants, consultants, salespeople, marketers, and so on). Job performance in high-complexity jobs is best predicted by a combination of four factors: general cognitive ability, personality traits, job satisfaction, and mindset. I will explain each of these four factors shortly.
The most important predictor is general cognitive ability or intelligence. This cannot be debated. Smarter people are able to process more complex information faster. Intelligence, while definitely not the only important factor, plays a much larger role than most people want to admit. Over decades of psychological research, it has been found time and time again that more intelligent people do better in school and perform better in their job, especially if their job is complex. Intelligence can best be measured by a general cognitive ability test, otherwise known as an IQ test.
The second most important predictor is personality. Several personality traits predict either a higher or lower job performance. All else being equal, candidates who score high on conscientiousness (which basically expresses how hard and how structured a person likes to work) tend to perform better. Furthermore, candidates who score low on neuroticism — or on the contrary, high on emotional stability — are more resistant to the stress that often accompanies highly complex work and perform better as a result. Now, not all personality tests are equally suited to measure these traits. Tests that are based on the five-factor model of personality (the ‘big 5’) have been found to be most valid.
Another important predictor is job satisfaction. This should come as no surprise: people who enjoy their job are more likely to want to work hard and work long hours, and therefore they outperform those that don’t like their job as much. However, it can be hard to screen for expected job satisfaction before someone is actually hired. It is probably best to use energy and enthusiasm as a proxy — if someone seems to be very excited and passionate about a potential job opportunity, it is no stretch to assume that that person will enjoy the job more than someone who seems indifferent about it. Enthusiasm can be screened in an interview, but preferably a structured interview instead of an unstructured one. I will explain the massive drawbacks of unstructured interviews later.
The final important predictor is mindset. This predictor is not as important as the previous three, but some recent research by well-known psychologists such as Caroline Dweck and Angela Duckworth suggests that people who possess a ‘growth mindset’ tend to outperform people who have a ‘fixed mindset’. The difference is that people with a growth mindset believe that they can improve their skills through conscious practice, while those with a fixed mindset believe that their skills and traits are fixed — either they are good at something, or they are not. Mindset can probably best be screened in a structured interview as well.
What we actually do
All of that stands in stark contrast to what we actually tend to do in hiring. First and foremost, a typical hiring manager does not even clearly define what qualities he or she is looking for. Managers often hold a fuzzy image of an ‘ideal candidate’ in their head, or they look for some sort of ‘X factor’. Others insist that they ‘know a good candidate when they see one’ or that they have ‘an eye for talent’. Once again, psychology shows us that we are mistaken: approaching a hiring process in such a haphazard way inevitably leads to bias. Therefore, step one should be to clearly define what you are actually screening for. It makes the most sense to screen for the aforementioned predictors of job performance.
Furthermore, the methods that hiring managers use to actually screen candidates are abysmal. The most common selection method is the unstructured interview. It is the typical job interview, in which the interviewer comes up with a more or less random set of questions that is supposed to reveal some important insights about the candidate. The candidate, of course, answers these questions to the best of his ability — but science tells us that it might not even matter what the candidate answers exactly. That is because unstructured interviews are subject to massive bias. Truly irrelevant factors, such as the weather, the color of the shirt the candidate wears, the time of day, and the mood of the interviewer have a massive impact on the outcome of the interview. Does any of those things predict job performance to a meaningful degree? Does it make sense to accept or reject candidates based on how sunny it is on the day of their interview? Of course not. Luckily, there is a great alternative: structured interviews. In a structured interview, the interviewer uses the same set of questions for each candidate and assigns a score to each possible answer. In the end, the interviewer counts each candidate’s score and selects the candidate with the highest score. Perhaps surprisingly, it doesn’t even matter too much how those answers are graded exactly — it is the simple act of scoring and grading that removes the unconscious biases that impact unstructured interviews.
Another mistake many hiring managers make is to overemphasize the importance of experience. Admittedly, experience predicts job performance to a minor degree. However, the higher the complexity of the job, the less experience has been found to make a difference. Some experience is nice to have, but to focus on experience to the extent that most hiring managers do is a mistake. A final grave mistake is to use unvalidated or even nonsensical tests, such as the Myers-Briggs Type Indicator. At best, these tests don’t give the hiring manager any meaningful information about the future job performance of the candidate. At worst, they actively mislead the hiring manager by introducing yet another dose of bias into the process.
Why don’t we change our behavior?
So far, I have explained what the proven predictors of job performance are and how one can best screen for them. I have contrasted those best practices with the reality of hiring, which is plagued by bias. Why is there such a large gap? For several reasons, really. Of course, one major reason could be that many hiring managers simply are unaware of the evidence. They don’t know what they should do — and they surely are not taught these best practices in school or on the work floor.
Another major reason is hiring managers’ need for autonomy. It is human nature to strive for autonomy and hiring is no exception. Especially if a hiring manager needs to select a person to work on his own team, the manager is likely to feel some resistance to the idea of purely using tests and structured interviews. What if those tools lead to a result that the manager doesn’t like? Research tells us that we should trust these tools anyway. It is too easy for our brains to fool us, and our unconscious biases and prejudices will surface quickly unless they are controlled by the objectivity of validated tests and structured interviews.
A third reason is some managers’ mistaken belief that they have ‘an eye for talent’. Now, this is not such a strange thought. People who hire lots of employees must become pretty good at selecting the best candidates, right? Wrong. Once again, study after study tells us that there is almost no difference between experienced hiring managers using unstructured interviews and random selection. Our biases are simply that strong.
A final reason I want to highlight is closely related to the ‘eye for talent’ belief. Some hiring managers think that their peers would perceive them to be either incompetent or lazy if they would rely on intelligence and personality tests and structured interviews. After all, needing such tools implies that you don’t really have an eye for talent, right? By now, you should know how this is a mistaken belief.
This is how you should hire your next employee
To conclude this article, I will present some evidence-based guidelines on how to hire your next employee.
- Recognize the massive role that bias can play in the selection process
Drop the belief that you have a special eye for talent or that you know a good candidate when you see one. Bias severely distorts your judgment, so you should never be too quick to trust yourself.
- Immediately drop unstructured interviews and unvalidated tests (e.g. MBTI)
Unstructured interviews and unvalidated tests, such as the Myers-Briggs Type Indicator, are major sources of bias. At best, they provide you no valuable information at all. At worst, they actively mislead you. Drop them immediately.
- Screen for intelligence by using GCA tests
Intelligence is the most important predictor of complex job performance. It can easily be screened with a validated general cognitive ability test.
- Screen for certain personality traits by using FFM personality tests
People who score high on conscientiousness and low on neuroticism perform better. To test for these traits, use a validated personality test, such as one based on the five-factor model of personality.
- Screen for enthusiasm and mindset in a structured interview
Job performance and mindset are additional important predictors of job performance. They can probably best be screened in an interview — but make sure to use a structured approach instead of a haphazard unstructured approach.