Unconscious Bias in Hiring: How to Reduce It

Workplace diversity. Many companies are using this term as their mission and goal. Apart from a better company reputation, higher innovation, and faster problem-solving, diversity improves hiring results and reduces employee turnover. A report from McKinsey & Company shows that diverse organizations simply perform better. But also, a mission like this is threatened by other factors, such as biases. So, how can we overcome our biases and succeed with unbiased recruitment? Blinding assessment tools could be the first step, but does it really make our hiring process unbiased?

 

 

 

 

The way we think

 

 

Let’s start where it all begins – in our minds. Daniel Kahneman and Amos Tversky conducted a groundbreaking work, where they explain how we make decisions. Their research shows, that our brain has two operating systems, which are referred to as System 1 and System 2. (Kahneman & Tversky, 1974).

 

 

“It is a commonplace observation that there seem to be two kinds of thinking. One fast and intuitive and the other slow and deliberative.” (Evans, 2011)

 

 

 

 

A: System 1

 

 

 

 

 

 

 

 

“System 1 is essentially what automatically comes up in your memory. For instance, when I say 2+2, something will come to your mind.“ Says Daniel Kahneman in his video interview, where he explains his findings.

 

 

This type of thinking has a large capacity, which allows us to process large amounts of information in a short time. (Evans, 2011) On average we all have about 35,000 decisions to make each day. It could be a decision about whether you will take the stairs or elevator, use the blue or black pen, etc. Some of the decisions are more important, and some of them are more difficult.

 

 

The task of our System 1 is to take care of decisions that are more familiar, by turning them into routines and habits. In fact, our System 1 will go through the information, choices, and ideas. Sometimes without us even noticing and therefore, sometimes we make decisions without controlling them.

 

 

As the article by Scientific American stated, System 1 is superior in its ability to automatically and effortlessly direct everyday life. It is important to state that neither of the Systems is all bad or all good. System 2 helps us do more reasonable and logical decisions based on facts. And System 1 helps us to save our cognitive capacity by creating decision rules.

 

 

 

 

B: System 2

 

 

The reference page, Suebehavioural design based on the theory of Kahneman & Tversky, stated that System 2 makes 2% of all our thinking (Kahneman & Tversky, 1974). They described this type of thinking as slow, rational, conscious, effortful, and risk assessing. As a result, our judgments and decisions are based on examination which is influenced by facts, logic, and evidence.

 

 

“Sometimes System 2 enables us to see things that System 1 doesn’t.“- Daniel Kahneman.

 

 

 

 

 

 

 

 

As a result, the most important ones are then sent and processed by System 2. What are those heuristics which process the majority of our decisions? Let’s find out.

 

 

 

 

Heuristics – our automated brain work

 

 

 

 

 

 

 

 

So, as mentioned before, System 1 helps to prevent our brain from cognitive overload. By using shortcuts that take care of habitual and less difficult decisions.

 

 

Suebehavioral design provided a concrete definition of heuristics:” Any approach to problem-solving that employs a practical method, not guaranteed to be perfect or rational. But instead sufficient for reaching an immediate goal.”

 

 

Let’s look at their work in practice.

 

 

 

 

Anchor heuristics

 

 

One of the examples could be the anchor heuristics. People tend to perceive the first available piece of information they have received as a reference point and unconsciously use it to “anchor“ or build their decision-making process on that piece of information. And we use it even if the initial information is incorrect or incomplete (Tversky & Kahneman, 1974) To explain how this works in practice, we will take a look at an example.

 

 

You want to buy a car and therefore checks the average price online, which is 100.000,- kr. You go to a dealership where you are offered a price of 90.000,- kr., you quickly accept the price because it is below your expectation, i.e. your anchor. The problem is, however, that your swift decision to purchase ruled out the possibility to check out a different dealership, which offers the car for 85.000,- kr. Instead of searching for every available information, we tend to favor the initial information provided to us.

 

 

 

 

Biases – the wrong heuristics

 

 

The issue with heuristics is that they might be potentially wrong. Heuristics are decisions made without complete information. Therefore, we have to have in mind that they are just mental shortcuts of which purpose is to find a solution as fast as possible while ignoring other aspects of the problem. As a result, heuristics affect our decision making and they could lead to a systematic deviation from rational behavior – called biases.

 

 

 

 

Cognitive biases

 

 

Do we really use biases while we think? The answer is yes. First, we need to explain what exactly are those biases. According to the article by Psychology Today: “Cognitive biases are repeated patterns of thinking that can lead to inaccurate or unreasonable conclusions.“ A bias can be described as a tendency or prejudice against someone or something. But the thing is that biases are often based on stereotypes, rather than actual knowledge.

 

 

 

 

Bias – the challenge of a transparent recruitment

 

 

Think about recruitment. When a recruiter is having an interview with an applicant, the decision of whether the person will be hired can be made in the first 10 minutes. And why is that? Because of our heuristics and biases. As soon as the person enters the room, our system 1 makes a first mostly unconscious judgment based on our heuristics. This decision then leads to a certain bias, for instance, it could be a stereotyping bias (person wears glasses, so we think he/she is smart) or liking bias.

 

 

Now that we understand how we think and how we make decisions, we can move forward and talk about recruitment biases in depth. How do biases affect the impartial recruitment process and what actions could be taken in order to prevent their use?

 

 

At the beginning of this blog post, we talked about how important workplace diversity is for every company. The article by Crowdstaffing made it clear: “The hidden danger in the form of recruiting bias means that you are limiting your candidate selection during the hiring process.“ To say it differently, you are limiting your workplace diversity.

 

 

So, what kind of biases do we use when we assess applicants?

 

 

 

 

Similarity bias

 

 

Similarity bias also referred to as Ingroup bias, means you will hire those applicants who are more similar to you. For instance, you have similar hobbies, interests or you attended the same high school, etc. Making friends is one thing, but making sure that you will create a successful environment by diversifying the workplace is something else. You need to have in mind that all positions have different competencies and hobbies are not the right factor to predict how good will the person perform in their future position.

 

 

 

 

Halo effect

 

 

The halo effect can be observed when a positive first impression of an applicant influences the overall perception of them. Basically, if you know the person is good at A. You automatically assume she or he is good at B, C, and D. For example, a good-looking person will be automatically considered to be an overall good human being who is intelligent and funny.

 

 

 

 

Attribution error

 

 

According to Simply Psychology: “Attribution error is a tendency for people to over-emphasize dispositional or personality-based explanations for behaviors observed in others while under-emphasizing situational explanations.“

 

 

In other words, people have a cognitive bias to attribute other people’s actions to their personality. Rather than taking into consideration environmental and social forces that might be influencing the person’s behavior.

 

 

For instance, a person could be late for a job interview due to a car accident or an unexpected family emergency. The recruiter will automatically think the person is irresponsible and not capable of time management.

 

 

 

 

Intuition – the tool of recruiters

 

 

Highouse (2008) wrote an interesting article about how recruiters believe that the prediction of human behavior can be improved through experience and that recruiters to some degree relied on intuition (System 1) in decision-making.

 

 

“Perhaps the greatest technological achievement in organizational psychology over the past year is the development of decision aids (tests, cases, structured interviews) that substantially reduce error in the prediction of employee performance (Schmidt & Hunter, 1998). Arguably, the greatest failure of organizational psychology has been the inability to convince employers to use them.“ (Highhouse, 2008).

 

 

 

 

 

 

 

 

Blinded resumé, will the bias disappear?

 

 

Behaghel (2014) analyzed the experiment on the effects of anonymous applications. Recruiters couldn’t read names, addresses, nationality, or other identifying indicators which are usually included in the resumés. The results showed that some of the applicants were actually harmed by using anonymous applications.

 

 

 

 

 

 

 

 

In the previous article, we talked about assessment tools and their ability to predict future job performance. When it comes to resumés, work experience and educational background are the most observed and assessed sections.

 

 

But is the resumé the best way how to predict whether is the candidate a good fit? Can we limit biases while using this assessment tool?

 

 

 

 

The starting point is awareness

 

 

We can see clearly that the combination of biases and resumés is quite a common thing. Whether is it a good resumé or the hobbies which are included match one of yours. That is the thing about biases. Even though we feel confident about not having them, it is a commonplace observation that everybody has biases, whether they are conscious or unconscious. The good thing is that we can limit our biases if we are aware of having them. But blinding the assessment tools might not be the most effective solution. Therefore the best solution would be to use completely unbiased assessment tools such as case-based screening.

 

 

 

 

 

 

 

 

This created a common belief that a matter of experience and intuition (System 1) is a way how to do a good hire. But is it really so? Will we rely on predicting a candidate’s likelihood of success based on the ability “to read between the lines”? The research on predicting human behavior shows that experience does not improve predictions. (Highhouse, 2008)

 

 

 

 

The bias impact on recruitment

 

 

First of all, we need to say that Heuristics aren’t necessarily bad – we can make good decisions using heuristics. But biases are always bad and they lead to systematic errors. Let’s take a look at the possible consequences of making biased decisions.

 

 

Bertrand & Mullainathan performed a field experiment to measure racial discrimination in 2003. They responded to help-wanted ads in newspapers in Boston and Chicago with fake resumés. Some of the resumés had better quality (more experience, etc.), and some of them disposed of with lower quality (less experience, no higher education, etc). They randomly assigned either a very African American sounded name or a very White sounded name to each resumé. The results show that African-American names received 50% fewer callbacks than the resumés with other names. (Betrand & Mullainathan, 2003) We can clearly see how important it is to make biases disappear from the hiring process.

 

 

The blog post from Forbes also talks about a study, conducted by Yale University, that found that female scientists, both trained to be objective, also used biases while recruiting. Even though they were both trained to be objective, they were more likely to hire men and consider them more competent than women.

 

 

 

 

Be unbiased

 

 

As we can see biases are part of us even though we don’t know about them. But being aware that we have them can lead to their successful limitation. Blinding of the resumés and other assessment tools might be the option. But is blindness a 100% reliable factor to overcome our biases? As we mentioned before. Anonymously made resumés could actually harm some of the applicants. There is a lot of assessment tools to choose from, and some of them can provide you with a totally unbiased hiring process. Think about cases, case-based screening will include nothing but the solution which is based on the questions built on the job or task in question.

 

 

As we mentioned at the beginning, diversity is a driving factor to achieve positive future prospects and results. Don’t let biased recruitment influence your firm’s outlook. And give people the opportunity to showcase their talent with a fair and transparent recruitment process.

 

 

 

 

*The article was based on a conducted meta-analysis by Schmidt & Hunter (1998), where they looked through 85 years of research findings.