Why is Bias so Hard to Eliminate in Hiring?

Hiring Aug 27, 2020

Bias in hiring is a big problem, but it may not be what you think it is. What people usually think of when they hear the word ‘bias’ is social judgment, prejudice, racial profiling, stereotyping, etc. These are huge challenges, however they are only a small subset of what psychologists call “cognitive bias”.

In this article we’ll dig a little deeper into the science behind cognitive bias. Understanding where bias really comes from is key in figuring out how to reduce biases. But, just as crucially, to understand where and when it may be impossible to remove them entirely. In this article, we’ll make the case that the root cause of bias is so integrated into our physiology that asking a person not to be biassed is like telling a person to never feel sad or upset.

Because of this, if we want to eliminate bias in recruitment, we cannot simply rely on awareness or training. This is why in many cases anti bias  training doesn’t work [1]. We have to build new processes that are fundamentally less sensitive to human bias. At Swyg we have some ideas about how to approach this - but more about that later.

To solve bias in hiring, we have to build new processes that are fundamentally less sensitive to bias.

Bias fills in the blanks

The psychologist’s definition of cognitive bias is “a systematic pattern of deviation from rationality in judgment.” [2] It is a natural result of our brain’s limited processing power as well as the need to make decisions even under conditions of limited information. Our brains are constantly trying to ‘fill in the gaps’ in the sensory information we receive.

You might think that cognitive biases don’t affect you, because you’re a smart cookie - but you’d be wrong. Let’s look at an example.

Optical illusions make clear how our brains fill in the blanks to piece together a cohesive understanding of the world around us despite what is often ambiguous, conflicting, or limited  information. Take the Kanizsa triangle in the figure below.

A Kanizsa triangle, an example of “filling in the blanks”. Our brain assumes the three black shapes are circles and that the multiple pieces of missing pieces add up to a non-existent (illusory) triangle. Figure credit: Fibonacci - Own work, CC BY-SA 3.0

Our brain subconsciously assumes that the black shapes are filled circles when in fact they are merely Pacman-shaped disc fragments. Our brains “see” the illusory white triangle that is not really there because our brains subconsciously make assumptions about what we’re seeing.

A key insight here is that the heuristics that our brain uses here make sense in most cases: circles are more common than Pacmen so it is more likely that I’m seeing a partially occluded circle than that I’m seeing a Pacman. If I see part of a tiger next to a tree it makes a lot more sense to assume the rest of the tiger is behind the tree than it is to assume that I’m seeing a half-tiger floating in space.

"Pacmen" has become the accepted term for the black shapes in these figures even though Kanizsa published his work about 5 years prior to the famous video game.

Social biases

Social interaction is far more complex than interpreting geometric shapes, but the same mechanisms are at play. When our brains deal with ‘higher order’ things like social interactions, or making hiring decisions, we still subconsciously apply heuristics and extrapolate from limited information.

Stereotyping is a good example of a cognitive bias in a social setting. The origins of stereotypes are complex [3] but at a basic level we assume that the traits that may apply to some individuals in a group apply to all individuals in the group. In other words, we’re extrapolating.

Psychologist Perry Hinton in a recent paper in Nature states “Learning an association of large animals with danger might be “biased” against harmless large animals (who we run away from needlessly) but that is a very small cost to pay compared to a life-saving rapid decision to get out of the way of a dangerous beast.” [3]

What’s surprising is that many of these heuristics are culturally dependent. For example, a landmark study found that in countries like Germany, Switzerland, China, and Malaysia, smiling faces were rated as significantly more intelligent than non-smiling people. But in Japan, India, Iran, South Korea, and Russia, the smiling faces were considered significantly less intelligent. [4]

Cultural differences in bias have some implications for the next time you travel but are truly profound when making hiring decisions about a global talent pool.

Bias in Hiring

One of the biggest culprits of bad hiring is affinity bias. It is an inherent bias to identity with and favor people who share our own characteristics. People with a similar background are more likely to experience easy relatability, share similar values, be like minded. But hiring isn't like dating, so we shouldn't treat it as such.

This becomes a problem when we need to judge someone based on very limited information like a CV/resume. There is a heck of a lot we do not know about this person, but if I can see that they worked at the same company, or went to the same school as me, I will subconsciously assume they are more qualified than someone I don’t relate to as much.

Similarly, If we see one great accomplishment from a candidate we’re likely to see all their other experience in a more positive light - the converse is true for great failures. For example, if we see that a person has won an olympic medal for swimming we implicitly assume they must be great at other things too.

This effect in itself can cause a recruiter or interviewer to get tunnel vision. But it also systematically disadvantages people from cultures or regions other than their own.

For example, have you ever heard of Tsinghua University? Know anything about the University of Singapore? How about KU Leuven? I know I haven’t. Yet, according to this list, they all rank higher than the Ivy League’s Brown and Dartmouth Universities. I would never have known this without looking it up. No time for that if you only spend 2 minutes per CV.

This shows that our biases are a product of not only our inherently limited neurology but also our limited experiences. This points to the necessity of a hiring team that’s as diverse as your candidate pool.

To eliminate bias in interviews, the interviewer panel should be as diverse as the candidate pool.

Can we eliminate bias in hiring?

Biases are natural. We all have them and apply them all day to all tasks. This is not to say that biases are never a problem. They can be benign, or harmful, but the constant is that they are, without exception, always present.

In this article, we established that bias cannot be entirely removed from individual interviewers. Furthermore, the impact of bias is greatest when we lack good information. And bias differs between individuals and between cultures.Because of that it is unrealistic to think bias can be eliminated by training or by hiring more fair-minded recruiters. We need a new way to hire.

Therefore we should design new selection processes with these criteria:

  1. Assess candidate abilities, not their background. I.e., collect better data.
  2. The interviewer panel should be as diverse as the candidate pool.
  3. Multiple independent interviewers should be used to mitigate the bias of any one individual.
  4. Incorporate interviewing best-practices by design, not by training.

We developed the Swyg interviewing platform to incorporate these principles by design. Follow us on LinkedIn or Twitter to stay updated with the latest research from Swyg.

References

[1] Dobbin, Frank, and Alexandra Kalev. "Why Doesn't Diversity Training Work? The Challenge for Industry and Academia." Anthropology Now 10.2 (2018): 48-55. (pdf)

[2] Haselton MG, Nettle D, Andrews PW (2005). "The evolution of cognitive bias.". In Buss DM (ed.). The Handbook of Evolutionary Psychology. Hoboken, NJ, US: John Wiley & Sons Inc. pp. 724–746.

[3] Hinton, P. Implicit stereotypes and the predictive brain: cognition and culture in “biased” person perception. Palgrave Commun 3, 17086 (2017). https://doi.org/10.1057/palcomms.2017.86 (html)

[4] Krys, Kuba, et al. "Be careful where you smile: Culture shapes judgments of intelligence and honesty of smiling individuals." Journal of nonverbal behavior 40.2 (2016): 101-116. (pdf)

[5] Hamamura, Takeshi, Steven J. Heine, and Delroy L. Paulhus. "Cultural differences in response styles: The role of dialectical thinking." Personality and Individual differences 44.4 (2008): 932-942. (pdf)

Authors

Tara Lonij

Psychology Expert @ Swyg