How Evidence-Based Hiring Can Reduce Bias and Discrimination

Can you find the answer to this riddle?

A father and son are in a horrible car crash, which kills the father. The son is rushed to the hospital. Just as he’s about to go under the knife, the surgeon says, “I can’t operate—this boy is my son!”

How is this possible?

Take a minute and try to figure out the answer.

It’s surprisingly simple - the surgeon is the boy’s mother.

Research from Boston University showed that only 14% of students were able to figure out the answer, and this included a large number of students who consider themselves feminists. They came up with all kinds of wild answers, including that the surgeon was a ghost, or that the whole thing was a dream.

So how come so many people have trouble figuring out the answer? Well, it has to do with preconceptions of what a typical surgeon looks like. When we think of a surgeon, most people tend to imagine a male surgeon.

We can’t avoid our biases, it’s human nature. But it’s important to understand why we are biased – and it’s not because we are stupid or uneducated. We are biased because when we lack information, we resort to stereotypes to fill in the missing pieces.

Another crucial thing to understand is that our biases can’t be avoided or turned off just because we are aware of them. If you think you can, then try this next exercise:

Don’t think of a polar bear.

Did you imagine a polar bear? Of course you did. Attempting to suppress certain thoughts makes them more likely to occur. Psychologists call this phenomenon ironic rebound or the white bear problem.

This is a big problem in hiring, as hiring managers tend to learn a little bit about a lot of people in a short time, and crucially they don’t know a lot of information about those people, so they tend to fill in the gaps with biases and stereotypes, albeit unconsciously and unintentionally.

Trying to reverse this process, using what’s sometimes called “positive discrimination”, doesn’t work. If, for example, you decide to hire people of a specific ethnicity, then you are technically still discriminating against all other ethnicities. Giving a positive point to one group of people is equivalent to giving a negative point to all other groups of people. There is no such thing as “positive” discrimination, it’s still just discrimination.

Fortunately, there is a way to solve this problem almost completely, which results in a hiring process that is quicker, more accurate and less biased.

Skip resume screening

Resume screening has traditionally been the first step in the hiring process. It shouldn’t be. Not only do most biases trigger during resume screening, but resumes themselves are mostly useless. This is because almost all information on a resume has no validity in predicting job performance. Things like age, interests and even job experience and education level poorly correlate with job performance.

Resumes are essentially a marketing tool used by candidates to try to sell themselves to employers, and lying on resumes is fairly common. After all, why shouldn’t people lie on their resumes? They have nothing to lose — they already don’t have the job. Perhaps most candidates wouldn’t outright lie, but one of the most common resume writing tips is to tailor the resume to the job you’re applying for. This means that candidates will highlight and embellish anything they think would help their chances, and leave out anything that doesn’t.

But regardless of the truthfulness of someone’s resume, people are notoriously terrible at evaluating them, and that’s without even considering bias.

So let’s consider bias. Resumes are a minefield that can trigger all kinds of biases, prejudices and stereotypes, most of them unconscious and unintentional. Even something as simple as the candidate’s name can trigger bias, with researchers noting that resumes with “white male names” tend to be rated higher by employers.

Automated resume filtering software and AI screening tools have been used as an attempt to try to solve these issues, but research has shown that not only does this not help, it makes things worse, as even AI can be biased if it’s trained using biased data. Part of the problem is bad software, but the bigger issue is that resume screening is simply a bad hiring method.

Resume screening isn’t just useless, it’s detrimental to your hiring efforts.

The first step

What if, instead of screening resumes, there was a different first step in the hiring process? One that isn’t subject to bias, and doesn’t fall prey to prejudice and stereotyping. There is, and it’s already widely used in many industries, though not always as the first step in the hiring process: skill tests.

Many companies already use skill tests in their hiring process. They can also be used as the very first step in the process and replace resume screening. Though in that case, the test should be relatively short and easy. The point is not to test the skill of candidates in-depth, but only the core skills required for the job. This way you can filter out the weakest candidates right from the beginning, and only invite better-qualified candidates to the interview.

Of course, you can still have a longer skill test later on for a more in-depth test of skill so that you can compare candidates and their abilities. But a short test of basic skills that candidates need to have for the job goes a long way to making the hiring process fairer for everyone, as well as much more accurate.

Research has shown that work-sample tests, which are skill tests that focus on samples of actual work, are one of the most accurate and effective hiring methods. The idea is simple: to test if a candidate will be good at work, give them a sample of actual work to do.

Skill tests that focus on samples of real work are a core concept of evidence-based hiring.

How to avoid bias in interviews

Skill tests are a good way to avoid resume bias, but bias also occurs in interviews. Fortunately, there is also a way to minimize interview bias while also making your hiring more transparent and objective: Structured interviews.

Structured interviews are a type of interview where questions are prepared in advance, and all candidates are asked the same questions in the same order. Marks are given to each answer so that all candidates can be easily compared to each other.

Bias is minimized in this case since the focus of structured interviews isn’t on the candidate itself, but rather on their knowledge and skills, since answers are the main criteria for evaluating candidates.

And, as mentioned earlier, if you use a short skill test before the interview, you’ll ensure that the candidates you invite to the interview are better qualified from the start.

Availability

Another issue that’s rarely mentioned is that employers can only hire people that apply for the job. For example, programming is mostly dominated by men, and employers are often criticized for primarily hiring male programmers. But what other options do they have? 90% of all programmers are men (even though programming was invented by a woman).

The only thing employers can do is ensure that their hiring process is fair and unbiased, but they don’t have control over who applies for the job, all they can do is select the best candidate, regardless of which group they belong to. If most candidates belong to one specific category of people, then the problem isn’t for employers to solve but lies somewhere else.

But that doesn’t mean employers shouldn’t do what they can to ensure that their hiring process is fair to all applicants, regardless of who the applicants are.

Final thoughts

Bias exists in the hiring process because people are biased by nature. It’s not something we can ever completely avoid, but there are ways to minimize it and focus on the main reason for hiring a candidate in the first place: their knowledge and skills. To do this, make sure that the methods used in the hiring process are accurate and objective, as shown by research. We’ve used this kind of data-driven, evidence-based approach at TestDome since 2014, and it’s been incredibly effective.