Find answers to common questions
Find answers to common questions
Of the various ways we select new talent, interviews can be particularly laden with unconscious bias and susceptible to snap-judgments that get in the way of selecting the best person for the job. In fact, researchers believe that interviews are a valid predictor of job performance only 35% of the time. Common kinds of unconscious bias likely to show up during an interview include “affective heuristic” and “confirmation bias”.
describes our tendency to evaluate someone based on initial cues like appearance or body language or attitude and use those superficial criteria to make a conclusion about their character or potential job fit. Psychologists and sociologists have proven over and over our tendency to select individuals who are like ourselves—we connect more easily with persons with whom we share an alma mater or neighborhood, and in turn, may misinterpret those clues for how someone will perform in a job. The malicious inverse is that when we don’t recognize someone as reflecting our personal framework, we default to stereotypes and unconscious biases to make unfounded conclusions about their character.
This is when we take those heuristic-based snap-judgments and then spend the interview seeking out information that supports that notion. Help yourself side-step this unconscious bias by asking all your interviewees the same questions in the same order. Don’t skip any questions, and never cut an interview short. You can also keep at the forefront of your mind the criteria specified as most important for job performance when the requisition was opened – this criterion is reinforced when you score the interviewee following the interview and summarize your feedback.
Interviews are incredibly easy to mess up—every interaction between you and a candidate subtly skews your sense of who they are. All too often when it comes time to make our decision of “Would I like to work with this person or not?” we are thinking more on what we have in common with the person than how their skills align with the needs for the job.
The good news is that researchers have compiled a workaround list of interviewing best practices. And even better news is that TalVista makes it easy for you to stick to those best practices:
Competency-based interviewing is best practice interviewing. A phone call or in real life conversation with a stranger is subject to all sorts of bias. When it comes time to size up a person’s fit for a job, we’re thinking more about how a person made us feel than how they might perform in the role.
Years of understanding human evolution and fight or flight responses tells us that it’s unlikely we’ll ever learn to curb these unconscious gut-level leanings. Instead, relying on competency-based interviews is the best antidote to the implicit bias entrenched in every decision we make.
Practicing competency-based interviewing is simple, and TalVista does it for you:
HR manuals and popular blog posts are great references for interview questions you should never ask—are illegal, all are bad at predicting job performance. These include:
Simply put, making a hiring decision based on marital or parenthood status (including future plans!), religion, race, ethnicity or disability is illegal. So just don’t go there.
But you probably already knew that. In addition to those, below are five less commonly deplored, but equally bad, interview questions that you should avoid.
Whether it’s leadership or technical skills, research shows that people from underrepresented backgrounds in a particular field tend to underestimate their true ability. Shelley Correll at Stanford University, for example, found that men assess their math ability higher than women who perform at the same ability level according to grades and test scores. So avoid asking candidates to rate their own abilities; chances are, they’re not very good at it.
While many believe that people whose interests match the content of their jobs have higher job performance, research suggests this is true only to a very limited extent (a rigorous meta-analysis found that the correlation between interests and job performance is only 0.1). Here’s why: Interests influence which kind of careers people try to enter. But once individuals pick a job or career, the quality of their job performance is determined mostly by their mental ability and certain personality traits, like integrity, not by their interests. So stop wasting time trying to suss out whether a candidate is so passionate about accounting because he blogs about it on weekends. The truth is her ability to learn new skills will be more predictive of future job success.
Questions about personal preferences that have nothing to do with core job competencies should be avoided like the plague for several reasons. First, there is a mountain of research showing that interviewers already commonly suffer from “similar-to-me” bias. That is, interviewers are more likely to select a candidate for a job if they have similar demographic characteristics, regardless of actual qualifications. Like many biases, “similar-to-me” bias is unconscious and therefore difficult to avoid. Don’t make selecting the right candidate for the job even more difficult because you’re unduly swayed by a mutual love for ultimate Frisbee.
Another reason to avoid these types of questions is that they may signal certain social class norms that can alienate some qualified candidates. One job description asked applicants to be prepared to pick and defend their favorite coffee brand. I’m sure the question was meant to be a silly ice-breaker. But it also sent a sign that employees at this start-up can afford, and value, a cup of $6 single-origin Kirinyaga Peaberry coffee from Kenya. Not all candidates will share that value. And, most importantly, it’s probably not relevant for job success.
While this question is timely and would probably make for an interesting discussion among friends, it’s not appropriate for a professional job interview. Discriminating against a job candidate based on political affiliation is forbidden, so avoid questions that can be interpreted as trying to ascertain political beliefs.
In some states, like California, employers are only allowed to ask applicants whether they’ve ever been convicted of a crime, not whether they’ve ever been arrested. Since people who have been arrested are often cleared of any wrongdoing, it’s unfair for employers to discriminate against individuals who have been arrested. (And, going quite a bit further, a recent studyfrom Harvard and the University of Massachusetts – Amherst questions the common belief that convicted felons are bad employees. Based on a sample of thousands of military employees, the study found that those with felony criminal backgrounds were actually promoted faster through the ranks and more often made it to the level of sergeant than recruits who weren’t offenders.) If you are concerned about an applicant’s’ integrity—a characteristic that has been shown to be a reasonably good predictor of job performance—ask specific questions that directly evaluate that quality instead.
An interview is an information-gathering session made up of verbal and nonverbal exchanges between you and the interviewee. Because it’s hard to make a complete character and skill assessment of a stranger in just one hour, our brains rely more on those non-verbal, gut-level cues and feelings to determine whether we want to work with this person or not. Because of this, interviews risk becoming more a reflection of who you are than who the interviewee is.
If we’re really connecting with someone, we may ask soft toss interview questions the candidate can easily skate through; and when we’re not getting a good connection with someone, we may ask curveball interview questions to subconsciously prove they wouldn’t do well in the job.
Because of this, structure is critical to an effective, unbiased interview and TalVista helps you follow a single interview template for everyone you interview.
Research has demonstrated the effectiveness of a structured interview template in sidestepping bias by studying interviews with pregnant women and individuals with physical disability. Structured interviews and pre-assigned templates help us get less distracted by extraneous cues and instead surface the most important information needed to synthesize about a candidate.
Here’s the three key strategies of structured interviews:
Recording your interview notes has two key benefits:
Each job description should be written to encourage the broadest and most diverse talent pool to apply for the open position. The TalVista Job Description Optimizer will identify problematic terms and offer up suggestions to replace problematic terms without changing the context of the job description.
The main thing to keep in mind is: Make your content meaningful—it is critical that your job description includes only criteria that is truly important to job performance—every line you include wields a huge impact on the diversity of the individuals choosing to apply for that job.
In an internal study at Hewlett Packard,women applied only for jobs for which they believed they met 100% of the criteria listed; while men in the study applied for the job if they believed they met 60% of the required criteria—that means while men are likely to apply when they meet most of the requirements you’ve listed, women are likely to review the required and preferred list and then only apply if they meet all of them. Listing less than ten total qualifications and three or fewer requirements is a critical job description best practice to follow.
As described above, women tend to apply for jobs only when they feel they meet 100% of the criteria. Imagine Mary in this real-life example: She reads your job description’s requirement for “3-5 years experience working in a service provider environment.” Only having 2.5 years experience and interpreting your requirement very literally, she hesitates to apply. Now Paul, also with just 2.5 years experience, applies for that job.
Consider also the possibility that Mary who didn’t apply had a better overall resume than Paul who you are now interviewing—all because of a slightly narrowing choice made at the job description stage. Avoiding listing years or only listing the minimum number rather than a range is a critically important job description best practice.
The Twitter-sphere generated a host of anecdotal evidence of the effect of job description language on female applicants when a Tweeter posted the question “If you’re a woman or minority in your field, what language turns you away from job descriptions?” From “rockstar” to “work hard play hard,” hundreds of replies voiced examples of off-putting gendered language that has become increasingly common in job descriptions.
In 2011, Aaron Kay from Duke University and Danielle Gaucher and Justin Friesen from University of Waterloo released a seminal study demonstrating that subtle wording differences in job descriptions can affect who applies for a given job. The researchers found that including more feminine-themed language in a job description, or minimizing masculine-themed language in a job description results in greater interest among potential female applicants.
Consider the differences between these two variations on a sentence in the same registered nurse job description:
“We are determined to deliver superior medical treatment tailored to each individual patient.”
“We are committed to providing top quality health care that is sympathetic to the needs of our patients.”
The results showed that descriptions with more feminine wording were more appealing and would get more female applicants than descriptions with more masculine wording.
We know that individuals considering applying for a job are looking as much in a job description for clues about a company’s environment as they are at the role itself. For example, a woman reading a job description boasting a company’s “beer Thursdays” or “foosball competitions” may assume a company is most friendly to people from a certain type of background. Use TalVista’s getting started to create a job description that strikes a gender-neutral tone and that better conveys to your potential applicants: “We are a company that values all kinds of people”.
To expand on the list of terms that Dr. Kay published as “masculine” and “feminine”, TalVista has surveyed thousands of participants on the gender-leanings of the most commonly used terms in job descriptions on the web today. In addition, we’ve conducted surveys that have found that including growth mindset language and talking about a company’s social contribution can attract other under-represented groups including people of color.
We help you look out for these words by labeling terms “inclusive” and “problematic.” Keep in mind that when Dr. Kay’s team looked at the effect of job descriptions with an inordinate number of feminine words, men’s interest in applying for those jobs went unchanged; in other words, it is likely that you can’t add too many inclusive terms. Also, you don’t need to get rid of all uses of the problematic terms, just add inclusive terms to counterbalance them so that you achieve an overall balance of inclusive and problematic language.
Research economists Andreas Leibbrandt and John List showed that a major cause of the pay gap between men and women is that women are less likely to negotiate. They uncovered in the study that when a job description mentioned “salary negotiable”, the pay gap between men and women hires closed by 45% because women became more likely to negotiate their starting salary.
The algorithm calculating your job’s effectiveness helps you assess how well you’ve implemented job description best practices. The score adds points for content that will attract a broader pool of applicants and subtracts points for content likely to alienate some applicants.
Positive points are earned by having inclusive language and other job description best practices for attracting more female and minority applicants; negative points are earned from problematic language, like citing “years” required too often and listing too many “requirements”.
Every new hire either adds to or detracts from your organization’s values. One of the easiest ways to impact the strategic success of your organization is to hire people who match your organization’s values.
To ensure organizational values don’t get diluted, companies must create a culture of values-based hiring excellence that:
Each employee brings with them values and culture from their previous roles. Only by ensuring a level of integrity between a new employee and your organization’s values can your company grow without diluting its culture.
How does values-based hiring impact an organization?
Read more about Talent Sonar’s values-based hiring services offering here.
A fascinating study by researchers Emilio J. Castilla (MIT) and Stephen Bernard (Indiana University) showed that the more an organization explicitly presents itself as a meritocracy, the more gender imbalances exist in that company, especially when it comes to pay gaps. They coined the term “paradox of meritocracy” to describe the phenomenon in which managers in organizations that promote themselves as meritocratic show greater bias in favoring men over equally performing women for raises and bonuses.
Further, emphasizing meritocracy can turn off minority candidates, especially when a company has low levels of visible minority representation. For example, a Yale-led study found that advocating a “colorblind” policy (as opposed to explicitly valuing diversity) when minority representation in recruiting materials was low led African-American managers to experience heightened distrust and discomfort with an organization. (There was no effect of advocating a colorblind policy vs. diversity when minority representation was high).
We’re not calling you biased, but today’s leading researchers are. To be human is to be subject to unconscious bias in every decision and assumption we make. As Sheryl Sandberg aptly summarizes,
Most people would agree that gender bias exists… in others. We, however, would never be swayed by such superficial and unenlightened opinions. Except we are. Our preconceived notions about masculinity and femininity influence how we interact with and evaluate colleagues in the workplace. … gender bias influences how we view performance and typically raises our assessment of men while lowering our assessment of women.
In a cross-disciplinary study out of Yale, science faculty compared applicants for a lab manager position. The applicants were identical except for being named John or Jennifer. Even across male and female reviewers, John consistently received higher rating and a higher offer for starting salary than did Jennifer. Consider the similar study in which African-American sounding names got 50% fewer callbacks than white-sounding names for interviews. If you’re interested in understanding more about your unconscious bias, check out Harvard’s Implicit Association Test.
A landmark study by Yale University psychologists unraveled the subtleties of job discrimination that happens as a result of stereotyping. The study showed that even when we think we’re reviewing a candidate for totally objective screening criteria like education or experience, a gut-level idea about who would traditionally fill a job often have actually influenced who we select.
In the experiment, participants reviewed resumes that presented a male and female candidate for police chief. The resumes were actually identical with two alternating variables: whether the resume belonged to Michael or Michelle, and whether the candidate had more formal education or more experience. Michael’s ratings consistently came out on top. When his was the resume without college education but more experience participants said they picked him for “having more street smarts”; and they still picked Michael when Michelle had more experience and Michael had gone to college, saying that Michelle wasn’t a good candidate due to lack of education.
The good news is that the study also showed how to mitigate our predisposition for prejudice. Asking reviewers to establish screening criteria importance beforehand eliminated this shifting merits effect. With Talent Sonar, resume reviewers are asked to “pre-commit” to which qualities are most important for job performance, helping eliminate the possibility that you will be swayed by other content in a particular candidate’s resume, such as their race, gender, or where they went to school.
Howard versus Heidi, Greg versus Jamal—researchers have replicated countless times that the inferred race or gender (or parenthood or religious status) of an applicant affects the way we rate their qualifications; this is especially true when it comes to resume screening. When study participants were shown completely identical resumes save one fact—one has the name Heidi, and one Has Howard—Howard gets higher ratings. We see the same effect when comparing a white-sounding name with a black-sounding name: Greg gets more call backs than Jamal. By scoring a resume before getting identity information, Talent Sonar helps users ensure an added measure of objectivity.
If you watched season three of “Mozart and the Jungle” you’ll recall New York Symphony candidates auditioning behind a white screen for the open oboe chair. In the 1970s, U.S. orchestras were only 5% female. As orchestras became aware of the concept of gender bias in their auditions, most started using a screen for auditions by the end of the 1980s. Today they are well above 30%, and researchers credit “the screen” with up to 55% of the increase in female new hires.
A similar rationale is at play in the popular reality TV show “The Voice”. Judges have to vote for candidates with their backs turned to them, judging candidates purely on their vocal talent—a self-proclaimed rejection of the superficiality that too often defines a singer’s success.
And when it comes to resume screenings, when researchers manipulate “perceived race” by changing resumes from white-sounding names like “Emily” and “Greg” to non-white sounding names like “Lakisha” and “Jamal”, white names are called back 50% more than non-white ones.
Evidence continues to pile up about the benefits of diversity. We know that today’s leading teams need balanced representation from all groups. To name a recent few:
Our industry has a lot of work to do when it comes to hiring for diversity. Understand how your company compares with others and spread the word when you find strategies that work! Here is a great one-stop resource that aggregates the diversity numbers from tech leaders; also see this valuable summary of diversity in tech companies. This list, maintained by engineer Tracy Chou, provides data for over 200 companies. You can add your company’s data and track industry progress!