Unconscious Bias in Hiring:7 Ways to Remove It
Your hiring process has bias built into it. Not because you are biased. Because human brains take shortcuts. Here are seven methods that replace gut feelings with evidence, backed by research from NBER, SHRM, and real hiring data.
6 Biases That Sabotage Hiring Decisions
Affinity Bias
Favoring candidates who look, sound, or think like you
Halo Effect
One strong trait (e.g. elite school) colors the entire evaluation
Name Bias
Judging candidates based on how their name sounds
Confirmation Bias
Seeking evidence that supports your first impression
Anchoring Bias
Over-weighting the first piece of information you see
Horn Effect
One weak trait (e.g. job gap) tanks the whole assessment
In 2003, economists Marianne Bertrand and Sendhil Mullainathan ran one of the most cited hiring studies ever. They sent identical resumes to employers. The only difference: the name at the top. Resumes with white-sounding names got 50% more callbacks than resumes with Black-sounding names.
Two decades later, a 2024 study from the University of Chicago and UC Berkeley submitted 83,000 fictitious applications to Fortune 500 companies. The gap shrank to 9% on average. Progress, but far from solved. And some companies still showed callback gaps of 24%.
Bias does not just affect race. Age, gender, disability status, school prestige, and even a candidate's hobbies influence decisions. A 2024 study from the University of Washington found that large language models used for resume screening showed significant racial and gender bias when names were included.
The cost is real. You lose qualified candidates. Your team stays homogeneous. And homogeneous teams make worse decisions. Here is how to fix it.
Strategy 1
Use Blind Resume Screening
Name bias is the most documented form of hiring bias. The fix is straightforward: remove the name before a human sees it. Blind screening strips out names, photos, school names, and addresses from applications before review.
The UK's civil service ran one of the largest blind hiring experiments. They found that blind screening increased the share of ethnic minority candidates advancing to interviews by 26%. Similar results show up in the private sector. Applied, a hiring platform, reported that blind screening led to 60% more diverse interview shortlists.
This does not mean you never learn a candidate's name. It means the decision to interview someone happens based on what they can do, not who they appear to be. Once a candidate reaches the interview stage, normal introductions resume.
Prepzo's AI screening evaluates every applicant against the job requirements you define. It scores resumes on skills and experience without weighting names, schools, or demographic signals. The recruiter sees a ranked shortlist based on qualifications.
Strategy 2
Run Structured Interviews With Scorecards
Unstructured interviews are a bias amplifier. When every interviewer asks different questions, there is no way to compare candidates fairly. The interviewer who "clicked" with a candidate gives a thumbs up. The one who didn't "feel it" gives a thumbs down. Neither is evaluating job performance.
Research from the Society for Human Resource Management confirms that structured interviews predict job performance twice as well as unstructured ones. They also reduce the influence of interviewer bias because every candidate answers the same questions on the same rubric.
Build a scorecard with 4 to 6 criteria tied directly to the role. Rate each on a 1 to 5 scale with written definitions for each score. "Meets expectations" is not a definition. "Can independently debug a production React app and explain their process" is a definition.
Every interviewer fills out their scorecard before seeing anyone else's scores. This prevents anchoring bias, where the first opinion shared dominates the discussion. Read our full guide on structured interviews for question templates and rubric examples.
Biased vs. Evidence-Based Hiring
Strategy 3
Replace Credential Filters With Skills Tests
Filtering by school name or degree type is a proxy for ability, not a measure of it. Research from Harvard Business School shows that credential-based screening disproportionately filters out candidates from underrepresented backgrounds. It also misses skilled candidates who took nontraditional paths.
Skills-based hiring flips the model. Instead of asking "Where did you go to school?" you ask "Can you do the job?" This means work sample tests, coding challenges, case studies, or portfolio reviews that directly simulate the work.
Keep assessments under 90 minutes. Respect the candidate's time. Longer take-home projects disproportionately disadvantage candidates with caregiving responsibilities, which introduces a different form of bias.
For a deeper look at this approach, see our guide on skills-based hiring.
Strategy 4
Build Diverse Interview Panels
A panel of three people from the same background will share the same blind spots. Diverse panels catch biases that homogeneous panels miss. They also signal to candidates that the company values different perspectives.
This does not mean tokenizing. It means intentionally including interviewers who differ in seniority, function, background, and thinking style. A junior engineer, a product manager, and a senior leader will each notice different things about a candidate.
Assign each panelist a specific evaluation area. One assesses technical skills. Another evaluates communication. A third focuses on problem-solving approach. This prevents overlap and ensures every angle of the role gets covered.
Strategy 5
Write Inclusive Job Descriptions
Bias starts before anyone applies. Research from the Bureau of Labor Statistics shows women make up 47% of the workforce but hold fewer than 40% of director-level positions. Part of the gap starts at the job posting.
Gendered language in job descriptions reduces the applicant pool. Words like "dominant," "competitive," and "crush it" skew masculine and discourage women from applying. Words like "supportive" and "collaborative" skew feminine. Neutral language performs best because it attracts the widest pool.
Keep requirements lists short. Research shows women tend to apply when they meet 100% of listed requirements, while men apply at 60%. Every "nice-to-have" you add to the requirements section narrows your pool. Separate must-haves from nice-to-haves clearly.
Include salary ranges. The EEOC has noted that pay transparency reduces negotiation gaps that disproportionately affect women and minorities. More states now require it by law. Get ahead of the mandate. See our guide on writing job descriptions for templates.
Bias-Free Hiring Pipeline
Application received
Name, photo, school hiddenAI screens skills
Scores based on requirementsStructured interview
Same questions, scored rubricPanel decision
Scores compared, bias flaggedBest candidate hired
Based on evidence, not gutStrategy 6
Use AI Screening to Remove Human Shortcuts
A recruiter reviewing 200 resumes in two hours spends about 36 seconds per resume. At that speed, decisions rely on pattern matching: recognizable school names, familiar company logos, gaps that trigger suspicion. This is where bias thrives.
AI screening can evaluate every resume against the same criteria with the same depth. No fatigue effect. No name bias. No anchoring to the first resume in the stack. The key is how the AI is built. Systems trained on historical hiring data can replicate past biases. Systems that score against predefined job requirements avoid this trap.
Prepzo's AI screening evaluates candidates against the skills and experience requirements you set for each role. It does not learn from your past hiring patterns. It scores what the resume says against what the job needs. Every candidate gets the same evaluation.
AI is not a silver bullet. It is one layer in a multi-layer system. Pair it with structured interviews, diverse panels, and skills assessments for the strongest results.
Strategy 7
Track and Audit Your Hiring Data
You cannot fix what you do not measure. Most companies track time-to-hire and cost-per-hire but ignore the metrics that reveal bias: pass-through rates by demographic group, offer acceptance rates by gender, and interviewer scoring patterns.
Run a quarterly hiring audit. Look at each stage of your funnel. If 40% of applicants are women but only 15% of hires are women, the bias is in your process, not in the talent pool. Pinpoint the stage where the drop-off happens. That is where you intervene.
Check individual interviewer patterns. If one interviewer consistently rates candidates from certain backgrounds lower, that is a coaching opportunity. Without data, you would never know.
Prepzo's analytics dashboard tracks candidate flow through every pipeline stage. You can see where candidates drop off and identify patterns that might indicate bias in your process. For more on which metrics matter, see our guide to recruitment metrics and KPIs.
Start This Week: A 5-Step Bias Reduction Checklist
You do not need to overhaul your entire process at once. Start with the highest-impact changes and build from there.
Audit one recent hire
Review the last 5 hires. Were structured scorecards used? Were all candidates asked the same questions? Document what you find.
Strip names from the screening stage
Use your ATS to hide candidate names during initial review. If your ATS cannot do this, switch to one that can.
Write scorecards for your top 3 open roles
Define 4 to 6 criteria per role. Write specific descriptions for each score level. Share with all interviewers before interviews start.
Review your job descriptions for gendered language
Run your top 5 job posts through a gender decoder tool. Replace loaded language with neutral terms. Cut your requirements list to true must-haves.
Set up a quarterly hiring audit
Track pass-through rates by stage. Compare demographic breakdowns at application vs. hire. Look for drop-off points.
The Business Case for Removing Bias
This is not just about doing the right thing. Bias directly hurts hiring outcomes. Teams that use structured, bias-aware processes hire faster, retain longer, and build more effective teams.
McKinsey's "Diversity Wins" report found that companies in the top quartile for ethnic diversity were 36% more likely to outperform on profitability. Companies in the top quartile for gender diversity were 25% more likely. The correlation holds across industries and geographies.
There is also a legal dimension. The EEOC requires that employment tests and selection procedures do not have a disparate impact on protected groups. If your screening process disproportionately filters out candidates of a certain race, gender, or age, you are exposed to legal risk even without intent.
Reducing bias is not a separate initiative from improving your hiring process. It is the same initiative. Every method in this guide also makes your process more predictable, more evidence-based, and more efficient.
Common Questions
FAQ
What is unconscious bias in hiring?
Unconscious bias in hiring refers to automatic mental shortcuts that cause recruiters and hiring managers to favor or reject candidates based on factors unrelated to job performance. These include name bias, affinity bias, the halo effect, and confirmation bias. They operate below conscious awareness and affect every stage from resume screening to final offers.
How does unconscious bias affect the hiring process?
Bias narrows your candidate pool, reduces diversity, and leads to worse hires. Research from the National Bureau of Economic Research found that candidates with white-sounding names received 50% more callbacks than identical resumes with Black-sounding names. Bias also shows up in interview scoring, salary offers, and promotion decisions after hire.
Can AI eliminate hiring bias?
AI can significantly reduce bias when designed correctly. AI screening tools evaluate candidates against predefined skill criteria without seeing names, photos, or demographic information. However, AI trained on biased historical data can replicate those biases. The key is using AI that scores against job requirements rather than pattern-matching against past hires.
What is blind resume screening?
Blind resume screening removes identifying information like names, photos, schools, and addresses from applications before a human reviewer sees them. This forces evaluators to focus on skills, experience, and qualifications. Studies show blind screening increases diversity in interview pools by 25-40%.
How do structured interviews reduce bias?
Structured interviews use identical questions for every candidate, evaluated against a standardized rubric. Research published in the Journal of Applied Psychology found structured interviews are 2x more predictive of job performance than unstructured conversations. They reduce bias because every candidate is measured on the same criteria by the same standard.
Hire based on skills, not assumptions
Prepzo's AI screening evaluates every candidate against your job requirements. No name bias. No gut feelings. Just evidence.
Start hiringAbout the Author
