Diversity, inclusion and unconscious bias are buzzwords in recruitment right now. The benefits of increasing the diversity of your company workforce are beyond question (more on that later), and the Black Lives Matter protests have forced CEOs to look at their own hiring practices more closely. This involves learning more about the cognitive biases that affect big decisions such as who to hire.
The problem of bias is difficult to overcome precisely because it’s mostly unconscious. There are some common practices for preventing hiring bias including unconscious bias training and Artificial Intelligence, both of which have some major flaws. In this post, we’ve lifted the lid on 18 types of bias in recruiting and 5 tips for improving fairness in recruiting.
The list: 16 common hiring biases
Jump straight into a specific bias, click on the links below.
- Affinity bias
- Expectation anchor
- Confirmation bias
- Affect heuristics
- Halo effect
- Horn effect
- Overconfidence bias
- Illusory correlation
- Negative emphasis
- Beauty bias
- Conformity bias
- Contrast bias
- First impression
- Central tendency
The benefits of a fairer hiring process
Striving for fairer hiring isn’t just a moral choice, it’s a business one.
One consequence of fairer hiring process is a more diverse workforce. This brings more innovation, breadth of opinions, and a stronger employer brand. A recent study suggests that two-thirds of people say working in a diverse environment is an important factor when evaluating job offers.
Moreover, inclusive companies are 1.7 times more likely to be innovative leaders in their market. McKinsey studied 180 companies in France, Germany, the UK and the US and found companies that had a more diverse top team performed better financially.
The impact of a biased recruiting process
Preventing hiring bias is a real problem. In 2016, research institute IZA exposed the impact of bias in CV screening with a simple experiment. They sent three fake CVs to nearly 1,500 companies across Germany who were hiring for office employees. The CVs were based on three different profiles, but with the exact same qualifications.
Everything else being equal a female with a Turkish name who wears a headscarf has to send 4.5 times as many applications as an applicant with a German name and no headscarf to receive the same number of callbacks for interview. Sadly these experiments have been replicated across different geographies and industries.
It’s also not just CV sifting that’s prone to bias, interviews are also hotspots for potential bias. Findings from the Harvard Business Review found that if there’s only one woman in your candidate pool, there’s statistically zero chance she’ll be hired.
Of course, it’s impossible to be conscious of all 175 cognitive biases that influence human behaviour when hiring, but we can acknowledge the major ones and try to limit their impact on decision-making.
Here we explore how to prevent hiring bias with 16 of the most common types of hiring bias and provide you with 5 tips on how to overcome them.
16 common types of hiring bias
Affinity bias, also known as similarity bias, is the unconscious tendency to get along with people who are like ourselves. We tend to rate people with similar backgrounds to ourselves higher during subjective assessments and interviews. People are programmed to gravitate towards the familiar as this seems safer, but this is a survival instinct that has no place in a 21st-century workplace. This is often perpetuated by asking candidates about personal lives, hobbies and other non-job-related questions.
As a recruiter, you’ve got a hundred things to think about when hiring staff. It takes time, energy and hard work. Therefore it’s easy to narrow your judgement to just a few areas to form your whole opinion. These are your “expectation anchors”, and they could be the reason why you’re making bad hires. It’s like buying a used car based only on the number of miles on the clock. Okay, it’s probably better than picking a car because of its colour, but it’s not a fair representation of all it’s merits and flaws.
It’s human nature to make snap judgements about people we meet based on our “gut feeling”. Confirmation bias is the disposition to subconsciously seek evidence to confirm our initial impression of someone was correct. This also involves discounting evidence contrary to our initial judgement of someone. Hiring by gut feeling is full of confirmation bias; anyone who does this runs the risk of missing great candidates because they made a poor first impression.
For more information on confirmation bias download our practical guide to fairer hiring.
Affect heuristics is rushing to conclusions about someone before you have all the information to make a fully informed decision. In recruiting we seek this mental shortcut most frequently when we’re tired and when we’re bored of sitting through interview after interview.
It’s more efficient and cost-effective to assess candidates quickly. Recruiters only take a few seconds to review a CV. The trouble is, you’re more likely to make the wrong decision if this is your process. What’s worse, you’ll likely make this decision based on discriminatory prejudices based on someone’s appearance, name, or background.
This is having a positive impression of a person in one area which then influences your opinion or feelings in other areas. Just because someone does one thing really well doesn’t mean they are strong in all other areas. Although this makes you more likely to hire someone, it doesn’t help prevent hiring bias: you’re still overlooking more qualified candidates.
The flipside of the halo effect. This is where one perceived bad quality distorts your whole impression of someone. For example, if a candidate has a poorly designed and laid out CV, we could think they’re a bad candidate, even if the role is unlikely to include any word-processing or design elements. For halo effect think positively influenced and horn effects think negatively influenced.
This is where a recruiter is overconfident about their own ability to pick good candidates because they believe they have “great instincts”. This form of illusory superiority is why 90% of the population think of themselves as better than average drivers.
This is where the interviewer observes one factor and assumes that it correlates to another, possibly unrelated factor and uses it to judge job performance. For example someone being extroverted doesn’t always mean they’ll be good at sales. Often asking unstructured interview questions and having chit chat about a candidate’s personal life usually results in illusory correlations.
Negative emphasis bias
This is judging someone negatively based on personal, irrelevant preferences, such as height, weight or hairstyle. For example, studies have shown that taller men earn more money. According to University of Florida research, they earn $789 per year extra for every inch of height. Similarly, obese candidates are rated significantly less competent than non-obese candidates. All of this is based on cultural ideals of beauty and holds no indication of an applicant’s ability to perform job roles.
Research has shown that beautiful people have certain advantages. They’re more likely to get a well-paid job, more likely to be popular at school and, if arrested, more likely to get a shorter prison sentence. Although beauty is subjective, it’s hard not to unconsciously correlate someone’s appearance to their predicted job performance.
Evidence that suggests beautiful people are on average, more successful in the world of work is most likely the result of this bias in action. Of course, there’s no real link between beauty and capability, unless it’s something that can be statistically verified by the company – but even then you’re on tricky ground like Abercrombie and Fitch discovered when their hiring policies came under scrutiny for hiring primarily on physical appearance.
Conformity bis is caused by peer pressure affecting hiring decisions. This can happen in assessment centres and group interviews when assessors let their decisions be influenced by other views and opinions on a candidate. It is really beneficial to not talk about or reveal sentiment towards candidates until all exercises or interviews have been completed and scoring has taken place.
This is caused by judging performance against the person(s) that came before them and not the hiring criteria. There are some pretty significant statistics to support the significant contribution of contrast bias to the recruitment process:
– We’re more likely to look favourably on someone’s CV if it’s reviewed directly after a poor one.
– We’re more likely to score a candidate higher at interview if the preceding candidate scored poorly.
This happens when sifting CVs one after another and when interviewing in close succession. The danger is that, with a constantly shifting benchmark, unsuitable candidates will get through, while good candidates will be missed.
Misreading body language can lead us to the wrong conclusions. According to Dr Albert Mehrabia our opinions are based on:
– 55% body language e.g. how some looks and acts
– 38% the sound and tone of someone’s voice
– 7% on what they are actually saying
There’s a saying in recruitment which is the best interviewer gets the job, not necessarily the best candidate. Non-verbal bias, or body language bias is where we’re overly affected by someone’s body language and charisma. It can be reduced by introducing online assessments and blind auditions into your hiring process.
First impression bias
It only takes seconds to make an impression. This is hardly enough time for a candidate to have shown any indication of whether or not they’ll be a great employee. First impression bias is making up your mind in the first few moments of meeting someone. It’s particularly misleading as nerves often give interviewers a false first impression of a candidate, especially in a stressful interview situation. First impressions can also be established before the interview, for example via CVs or social media, which are usually not good indicators of job performance.
Central tendency bias
Imagine you have a scorecard that rates candidate’s competencies on a scale of 1-5. Central tendency is the tendency for assessors to place most items in the middle of a rating scale and neglect the extremes. This effectively limits your range of judgement and will make it harder to distinguish between your candidates. Don’t just score candidates 3 out of 5 when you aren’t sure, use the full range of the scoring and take time to evaluate evidence fairly.
This is the “gut feeling” you get when meeting people. It’s common for recruiters to use this intuition as evidence of their hiring ability (see overconfidence bias). However, this intuition is more than likely formed by some of the biases we’ve already discussed, such as confirmation, beauty, affinity bias and expectation anchors. Drop the intuition and get to the facts.
How to prevent hiring bias – 5 tips
1. Blind Resumes
Remove information that leads to bias including names, pictures, hobbies and interests. This kind of information sections off applications while offering little value in terms of predicting job performance. When implemented, this approach also prevents problematic pre-interview candidate research.
2. Pre-hire assessments for candidate screening
Not only are manual CV sifts and interviews prone to bias, as an assessment method they’re also one of the lowest predictors of job performance. Realistic, job-relevant pre-hire assessments are proven to help companies screen candidates more effectively and fairly than any CV or resume based approach. That’s why companies using technology-based pre-hire assessments experienced a 39% lower employee turnover.
3. Avoid AI trained on existing hiring data
Artificial Intelligence can be useful but it typically relies on past data. In other words, you always run the risk of repeating old hiring patterns. If you want to encourage a more diverse and inclusive workforce, measuring against previous records can be dangerous. We know from decades of audit studies that employers tend to discriminate against women and ethnic minorities, and a recent meta-analysis suggests that little has improved over the past 25 years. AI will only perpetuate this bias as Amazon found out in 2015.
4. Diverse hiring panels
Diversity in your recruitment team is a surefire way to get diversity in your workforce. Each interviewer will have their own mental model – our deeply held beliefs about how the world works – so will approach the interview from a different perspective. This will help prevent bias blind spots.
5. Avoid unstructured interview questions
Questions such as, ”what’s your biggest weakness” or “tell me about yourself” offer little value in terms of predicting job performance. The answers will be qualitative and only lead you down the path of drawing conclusions based on your own biases.
It’s a lot to take in, but these five tips are at least a good start in making your hiring process fairer and more inclusive. Just remember that unconscious biases are exactly that, unconscious. Therefore they can’t be washed away with training, but they can be mitigated by creating systems that limit their impact.
Some of these are pulled from our e-book, Beyond gut feeling: a practical guide to fairer hiring. You can download that by clicking the image below.