Lisa Burrell for Harvard Business Review : We Just Can’t Handle Diversity : August 10, 2016 Artwork: Roger Clarke, The Deadliest Toxins (DMM), 2009, Polyester Resin, Fiberglass, Varnish
It’s hard to argue with the benefits of diversity, given the decades’ worth of studies showing that a diverse workforce measurably improves decision making, problem solving, creativity, innovation, and flexibility.
Most of us also believe that hiring, development, and compensation decisions should come down to who deserves what. Although the two ideas don’t seem contradictory, they’re tough to reconcile in practice. Cognitive roadblocks keep getting in the way.
The Trouble with Merit While merit sounds like an easy, obvious filter for talent decisions, it’s anything but. We believe we know good talent when we see it, yet we usually don’t—we’re terrible at evaluating people objectively. That’s why orchestras started holding blind auditions decades ago. It’s why today algorithms often make smarter hires than people do. It’s why so many companies are searching for alternatives to traditional performance reviews. Even (and especially) when leaders proclaim a commitment to fairness in their organizations, stereotypes cause them to evaluate and treat equal performers differently, as Emilio Castilla, of MIT, and Stephen Benard, of Indiana University, have demonstrated in their well-known research on the “paradox of meritocracy.”
What’s tripping us up? Robert H. Frank, a Cornell economist and the author of Success and Luck, provides one explanation: We just don’t see the large role that chance events play in people’s life trajectories. If someone lands a great job and makes lots of money, we interpret those outcomes as evidence of smarts and hard work. (We look at our own lives the same way.) As for those who don’t thrive? Well, we tell ourselves, maybe they’ve caught a bad break here and there, but they could turn things around if they tugged on their bootstraps a bit harder.
If those in power think this world is basically fair and just, they won’t even recognize—much less worry about—systemic unfairness. Frank talks about inequity mostly in socioeconomic terms, but the implications for underrepresented demographic groups are clear. He dips in and out of the abundant social science findings suggesting that good fortune accounts for a great deal of success, and that we’re hell-bent on believing otherwise. Framing effects, or our immediate points of reference (living in suburbia or attending a posh school, for instance), shape how we perceive haves and have-nots in the wider world. Hindsight bias causes us to believe that random events are predictable and to manufacture explanations for the inevitability of our achievements. And winner-take-all markets—where “rewards tend to be highly concentrated in the hands of a few top performers”—intensify the consequences of our cognitive shortcuts.
Of course, believing that merit will be justly rewarded can come in handy for individuals. As Frank notes, it’s easier to muster the energy to overcome obstacles if you feel you’re on a well-earned, reasonably certain path to high achievement and if you have an inflated sense of your own abilities. But, he says, this mindset also keeps people from investing in public solutions that expand the economic pie for everyone. Perhaps the biggest reason we cling to it is that when it’s challenged, we feel attacked—as if our talent and effort are being dismissed. Talent and effort matter quite a bit, Frank acknowledges. But very often they’re not enough to ensure success. Changes in public policy and a dose of gratitude can help rectify inequities, he says—but we’re a far cry from living and working in a meritocracy, because our view of merit is so flawed.
In her book Pedigree, Lauren Rivera, of Northwestern University, also examines how we understand and evaluate merit and finds it to be a moving target. But where Frank applies a “macro” lens to society, Rivera looks specifically at how students from elite schools and backgrounds get elite jobs—and at how employers judge the people applying for those positions.
She studied hiring committees at professional services firms that believed they were ensuring rigor and counteracting bias through group discussions of job candidates from the school-recruitment pipeline. But those conversations actually dampened diversity by giving negative racial, ethnic, and gender stereotypes greater sway over decisions—particularly “in ambiguous situations, where the quality of a candidate [was] not clear.” In those cases, Rivera points out, “stereotypes served as an unconscious navigational system, guiding interviewers’ attention to where they should focus and look for clues in order to figure out if the candidate did or did not have the right stuff.” This gave evaluators “a common lens and shared language” when they didn’t immediately agree on someone’s value to the organization.
One consulting firm invited Rivera to sit in at various points in the selection process—first during “calibrations,” or discussions between pairs of interviewers about first-round candidates, and then during the group discussions in later rounds. At each stage she consistently found that evaluators had little or nothing to say about the “rock stars” or the “rejects.” They deliberated mainly about candidates in the middle, which is where stereotypes about women and minorities came into play.
In the calibrations, the most common criteria for moving candidates from the middle to either the “yes” or the “no” pile were communication skills (referred to as “polish”), analysis of a sample business case, the math used to support that analysis, and cultural fit. But the interviewers weighed and judged those criteria differently depending on the race, ethnicity, or gender of the candidates. For example, black and Hispanic men were often seen as lacking polish and moved to the reject pile, even when they were strong in other areas, whereas white men who lacked polish were deemed coachable and kept in the running. A similar pattern emerged among men who appeared shy, nervous, or understated: Nonwhites were rejected for being unassertive, but in whites, modesty was seen as a virtue. Among candidates who made minor mistakes in math, women were rejected for not having the right skills, and men were given a pass—interviewers assumed they were having an “off” day. (See “Why Diversity Programs Fail” by Frank Dobbin and Alexandra Kalev.)
Somewhat predictably, when discussing final decisions, evaluators shifted their focus away from the candidates’ performance and toward personal chemistry and gut instinct—their “feel” for people took over. They barely discussed technical skill toward the end unless they were evaluating women, particularly those in banking.
Nodding to the sociologist Randall Collins’s argument that “emotion is a critical basis of social sorting, selection, and stratification,” Rivera found that candidates in the “maybe” pool ultimately needed a passionate champion on the hiring committee in order to receive an offer. And evaluators advocated most fervently for people who were most like them. Perhaps because women and minorities were more vulnerable in their status at the firm, they championed fewer people than white men did—they chose their battles, as one female evaluator put it. (There’s something to that reluctance. As Stefanie Johnson and David Hekman, of the University of Colorado, have found in their field and lab research, women and minorities who actively push for diversity are punished by their organizations—they get lower performance ratings than those who don’t. Men who promote diversity don’t suffer the same penalty.)
So, with white men doing most of the championing and having the greatest influence during deliberations, candidates’ similarities to interviewers tilted the playing field heavily in favor of “white, affluent, athletic graduates of super-elite institutions.” Similarity to evaluators who deviated from that norm sometimes helped women or minorities land a role—but those were isolated cases.
The Trouble with Diversity As Rivera suggests, the hiring conversations at the consulting firm were ultimately more about reaching consensus than about vetting people accurately. To fix that kind of conceptual problem, it’s necessary to sort out (at least somewhat objectively) what constitutes merit in a given context. Assuming that’s possible, and that we can send our biases packing (a gigantic if when we consider how stubborn they are), will diversity naturally follow?
That’s difficult to say, since we don’t agree on how to define it. According to one Deloitte study, Millennials think of diversity and inclusion as valuing open participation by employees with different perspectives and personalities. In contrast, older workers think of it as equitable representation and assimilation of people from different demographic groups.
Even if we stick with the second, more traditional definition, how can we set goals and track our progress? As Ashleigh Shelby Rosette, of Duke University, pointed out at Wharton’s 2016 People Analytics Conference, we tend to boil things down into tidy dichotomies—male/female, white/black, dominant/minority, and so on. But reality is a lot messier than that. No one is just female, or just black, or just Muslim. Each person is “a whole package of interlocking attributes,” Rosette said, and that’s a lot harder to analyze and balance in an organization.
Further complicating matters, rhetorical framing skews how people perceive power. Rosette and her colleague Leigh Plunkett Tost, of the University of Michigan, discussed this dynamic in their Psychological Science article “Perceiving Social Inequity.” In general, describing inequities as privileges for certain groups (rather than disadvantages for others) gets our defenses up. Much like the notion of dumb luck that Frank writes about, it damages our self-image—haven’t we earned what we’ve got?—and makes us not want to see or rectify the problem.
Plus, power is variable for members of any group. People can have high status on some social-hierarchy dimensions but low status on others. That mix, Rosette and Tost’s research shows, may help individuals recognize the privileges they enjoy as part of a dominant group, as long as they also believe they’ve experienced disadvantages as members of other, subordinate groups (and thus can identify with people who feel disadvantaged in comparison with them). White women overall, for instance, are more likely than white men to view themselves as beneficiaries of racial privilege; they get it because they, too, have had to deal with discrimination.
Senior leaders need to recognize their organizations’ inequities—probably more than anyone else, since they have the power to make changes. But once they’ve climbed to their positions, they usually lose sight of what they had to overcome to get there. As a result, Rosette and Tost find, “they lack the motivation and perspective to actively consider the advantages that dominant-group members experience.” This is especially true of successful white women, who “reported [even] lower perceptions of White privilege than did highly successful White men.” It’s fascinating that their encounters with sexism don’t help them identify racial advantage after they’ve gotten ahead. Perhaps, the authors suggest, their hard-earned status feels so tenuous that they reflexively tighten their grip.
Beyond murkily defined concepts and somewhat defensive motivations, we have an even-higher-level conceptual obstacle to overcome: our bias against diversity itself. Recent research by Ohio State University’s Robert Lount Jr. and colleagues (Oliver Sheldon, of Rutgers; Floor Rink, of Groningen; and Katherine Phillips, of Columbia) shows that we assume diversity will spark interpersonal conflict. Participants in a series of experiments all read, watched, or listened to the exact same conversations among various groups. They consistently perceived the all-black or all-white groups as more harmonious than those with a combination of blacks and whites.
If we expect people to behave less constructively when they’re in diverse organizations or teams, how do we interpret and reward their actual performance? Under the influence of those flawed expectations? Quite possibly.
So, Is It Hopeless? According to the renowned behavioral economist Daniel Kahneman, trying to outsmart bias at the individual level is a bit of a fool’s errand, even with training. We are fundamentally overconfident, he says, so we make quick interpretations and automatic judgments. But organizations think and move much more slowly. They actually stand a chance of improving decision making.
Research by John Beshears and Francesca Gino, of Harvard Business School, supports that line of thought. As they have written in HBR, “It’s extraordinarily difficult to rewire the human brain,” but we can “alter the environment in which decisions are made.” This approach—known as choice architecture—involves mitigating biases, not reversing them, and Beshears and Gino have found that it can lead to better outcomes in a wide range of situations. The idea is to deliberately structure how you present information and options: You don’t take away individuals’ right to decide or tell them what they should do. You just make it easier for them to reach more-rational decisions. (For more on this idea, also see “Designing a Bias-Free Organization,” an interview with Harvard behavioral economist Iris Bohnet.)
There’s still an element of manipulation here: The organization sets the stage for certain kinds of choices. But that brings us back to what most of us can agree on, at least in the abstract: Diversity improves performance, and people who apply themselves and do good work should be treated fairly.
If the members of an organization could get behind those broad ideas, would it bother them that they were being nudged to do what they wanted to do anyway? It might—and that would be another cognitive roadblock to clear.
A version of this article appeared in the July–August 2016 issue (pp.70–74) of Harvard Business Review.
Lisa Burrell is a senior editor at Harvard Business Review