How to reduce bias in your hiring process
Some of these tips may take a bit more time or effort, but they can help you have a more fair, less biased hiring process — and ultimately build a stronger team.
1. Scrub job descriptions of biased language and requirements.
Making your job descriptions more inclusive doesn’t mean lowering the bar. It means widening the net. Your goal at this stage should be to attract as many qualified applicants as possible. You’ll have plenty of chances to get picky during the screening, interview, and test-project stages.
In your job descriptions:
- Eliminate unnecessary requirements. Google, for instance, has expanded the list of college majors they deem acceptable for certain roles. Does your open position even require a degree (i.e., does data show that having one leads to strong performance)? Do you really need a support specialist with Fortune 500 experience? Of course, some roles are so specialized that certain criteria are mandatory. But to attract the most diverse candidate pool possible, keep mandatory requirements to a minimum.
- Run job description language through a bias filter. Some websites and apps will grade your job description and point out potentially unwelcoming language. For example, they might suggest you change “dominate” (which may turn off some women) with the more neutral “influence.”
- If possible, accommodate remote, part-time, and/or flexible hours. Work configurations like these will attract qualified candidates who might otherwise struggle to make it to your office every day, like parents (especially single parents), people with disabilities, or people who can’t afford to live in an expensive city (if your office is in one).
- Signal a culture of inclusion. Add a statement about your team and/or company’s diversity mission. And mention any inclusive perks your company offers, like flex-time and opportunities to join diversity initiatives.
2. Find more diverse sources for qualified candidates.
Additional sourcing can be costly and time-consuming, but it’s often worth the trouble — you’ll not only find a broader range of candidates, but you also may beat others to untapped talent pools.
Consider posting the job ad on social-networking sites like Facebook and Twitter (and asking colleagues to, as well). These sites tend to reach more diverse candidates than dedicated professional networking sites.
3. Remove names and other identifying information on job applications and sample projects.
Names matter. Few of the resume reviewers creating these troubling statistics are aware of what they’re doing. It’s common to speed past a resume simply because a name is unfamiliar, or one we haven’t been socially conditioned to associate with the job title we’re screening for.
Just slowing down and reviewing resumes more deliberately can help mitigate bias, if you can make the time to do so. It’s also pretty easy to “blind” yourself to names and other identifying information on resumes, job applications, and sample work. Ask HR to replace candidates’ names with initials or identification numbers, so non-essential information is hidden from yourself. You can also remove names for the benefit of others on your evaluation team.
4. Learn how the skills and experiences of lesser-tapped communities may translate to your industry.
Some managers toss aside resumes from veterans or people from the nonprofit sector or other industries, assuming the skills these people have just aren’t relevant for the open job. But is that really the case? What if the managers simply don’t know how the unfamiliar job titles and experiences translate? Fair-minded managers acknowledge this possibility and proactively fill in knowledge gaps.
5. Standardize interview questions.
Standardized interviews give you the same precious information from everyone — whether it’s someone you have a lot in common with, or someone who’s a different race, gender, sexual orientation, religion, or age from you.
Before your first interview, create a list of behavioral and situational questions to ask each candidate. Behavioral questions focus on what they have done: “Describe the most significant written report or presentation that you had to complete.” Situational questions present hypothetical situations: “How would you handle a customer complaining that he had received insufficient service?” Both are better predictors of job performance than asking candidates open-ended questions like “Tell me about yourself.”
Also, try to maintain fairness by sticking to the same follow-up questions when you probe for more: “How did you approach X?”, “Can you tell me more about that?”, “What were the results?”
6. Use group interviews and test projects to ensure each candidate receives the same overall experience.
- Group/panel interviews. This is when a large portion — if not all — of your interview team is present, each taking turns asking questions from a single, predetermined list. Yes, these interviews are tough to coordinate and can be stressful for interviewees. But they tend to be the least biased type of interview, partly because social pressure keeps interviewers attentive and accountable. They’re also the most predictive of performance, partly because they tend to ensure standardization of questions across candidates.
- Test projects that simulate a potential job task. When every candidate receives the same test, it’s far easier to compare candidates’ work. And research finds that test projects are better at predicting job performance than any traditional interview question you could ask.
7. Standardize interview rating and evaluation forms.
Imagine you ask two candidates the same question to gauge their ability to adapt to change. Both give decent responses. But Candidate A looks tough, like he might have played football in college. Candidate B, on the other hand, looks frail, like she was picked last in gym class. You decide to go with Candidate A. However, unbeknownst to you, Candidate B is a cancer survivor and actually far more resilient. Whoops.
In the absence of objective criteria to measure against, your biases can go wild, tricking you into over-relying on visual cues and assumptions. That’s why you and the rest of your interview team need to work from a mutual understanding of what you’re assessing. Write a definition for each competency you plan to assess. Then, put the competencies into a scoring sheet. For instance, you might use a scale of 1 to 6, and give an example of what each rating along the spectrum looks like.
8. Diversify and prep your team of interviewers.
Even if you’re making the final decision, a biased interview team means biased opinions that might sway you (or fail to provide healthy challenges to your thinking). Consider the following to create a team prepared to minimize bias:
- Diversify. Do a quick gauge of the gender, age, and ethnic balance of your interview team. Is it lopsided? And don’t overlook remote team members just because they’re remote; they’ll likely have unique insights and will also need to work effectively with your new hire.
- Sync with interviewers on the competencies you’re looking for and the intent of the interviews. Three different interviewers will have three different ideas of what “good communication skills” entail, so go over competency definitions and ratings scales ahead of time. Also be clear that you’re looking for interviewers to gather actual evidence for how a candidate rates in each competency, not just form surface-level impressions.
- Insist interviewers take notes. Their memories are no less susceptible to bias than your own.
- Prep interviewers on how bias affects interviews. Share this article, plus any additional knowledge you have. Also mention that they shouldn’t discuss candidates (and risk biasing each other) until after all evaluation forms have been submitted.
9. Schedule time right after interviews to fill out evaluation forms.
Even if you’re taking good notes, your memory of details will fade quickly. And when it does, your biases will be waiting to suggest what a candidate — probably, maybe — said.
Let’s say you estimate that it should take 30 minutes to complete the evaluation forms — schedule that into the block of interview time. (You can always go back and tweak the evaluation later if, say, on reflection you realize your original impressions were clouded by a candidate’s charm.)
If this scheduling feat isn’t possible, at least ask the interview team to prioritize evaluations — preferably completing them in the same day — without discussing candidates in the meantime. And if you do group interviews, schedule the debrief only after everyone has independently finished their evaluations.
10. Follow a process, not your intuition, when making your decision.
“I can’t decide between these two candidates, so I’m going with my gut.” Sound familiar?
Unfortunately, this approach makes you highly vulnerable to bias. In close-call situations, it’s generally best to rely on the deliberate process you’ve put in place. As much as we love to believe that our intuitions about people are uncannily perceptive, they usually aren’t as helpful as the evidence we’ve collected.
To make better, less-biased decisions:
- Use the evaluation forms as intended. Many managers start ignoring evaluation scores that don’t line up with their inclinations (e.g., “He doesn’t have the highest scores, but I really like him”) or cherry-pick data to validate a hunch. How were you planning to use the evaluations? To average the scores and deliberate the two candidates with the highest scores? Great! That’s one of many reasonable plans. So stick with it.
- Be certain any nagging concerns are backed up with evidence. For example, if one interviewer says, “I don’t feel like he’s a good culture fit,” ask, “What leads you to believe that?” If the response is along the lines of, “It’s a gut feeling,” then follow up and ask for specific examples. That’s just not enough of a reason.
- Don’t speculate. Which of the two candidates is more likely to stick around for three years? If you realize you still have unanswered questions, go back to the candidate and ask. Or ask their references.
- Make your decision when you’re focused and mentally sharp. This is not an activity you want to “power through” when you’re tired, hungry, or distracted. Biases are more likely to guide you when your defenses are down.
11. Don’t make assumptions about candidates based on whether (or how) they negotiate compensation.
When you make an offer, a candidate might be thinking anything from “It’s my duty to myself to ask for more” to “I’m just happy to have a job — I don’t want to make my new manager angry.” There would be no issue if everyone was equally likely to have one or the other mindset. But they’re not. Those from privileged groups are more likely to start negotiations and be successful in them.
The only way to truly level the field is through a no-negotiation policy, which is likely out of your control. Regardless, you can proactively avoid a common negotiation bias: viewing women who negotiate firmly as “aggressive” rather than shrewd, even when they use the same tactics as men.
And gender aside, don’t let any annoyance you feel toward tough negotiators lead to hasty assumptions about their character. Research finds that many people change their expectations and attitudes during negotiations. These people likely aren’t jerks, but rather are adapting to the behavior they think the situation calls for.