Page 110 - Harvard Business Review (November-December, 2017)

FEATURE “NUMBERS TAKE US ONLY SO FAR”

Terrible as it felt at the time, her directness was useful to size, the n, is too small. Basically they’re saying, “If only
me. It meant I didn’t have to scour the facts looking for some there were more of you, we could tell you why there are so
other, nonracist rationale for her sudden rejection. few of you.”
Many people have been denied housing, bank loans, Companies have access to more data than they realize,
jobs, promotions, and more because of their race. But however. To supplement a small n, they can venture out
they’re rarely told that’s the reason, as I was—particularly and look at the larger context in which they operate. But
in the workplace. For one thing, such discrimination is il- data volume alone won’t give leaders the insight they need
legal. For another, executives tend to think—and have a to increase diversity in their organizations. They must also
strong desire to believe—that they’re hiring and promoting take a closer look at the individuals from underrepresented
people fairly when they aren’t. (Research shows that in- groups who work for them—those who barely register on
dividuals who view themselves as objective are often the the analytics radar.
ones who apply the most unconscious bias.) Though man-
agers don’t cite or (usually) even perceive race as a factor in
their decisions, they use ambiguous assessment criteria to SUPPLEMENTING THE N
filter out people who aren’t like them, research by Kellogg Nonprofit research organizations are doing important work
professor Lauren Rivera shows. People in marginalized ra- that sheds light on how bias shapes hiring and advance-
cial and ethnic groups are deemed more often than whites ment in various industries and sectors. For example, a
to be “not the right cultural fit” or “not ready” for high- study by the Ascend Foundation showed that in 2013 white
level roles; they’re taken out of the running because their men and white women in five major Silicon Valley firms
“communication style” is somehow off the mark. They’re were 154% more likely to become executives than their
left only with lingering suspicions that their identity is the Asian counterparts were. And though both race and gen-
real issue, especially when decision makers’ bias is masked der were factors in the glass ceiling for Asians, race had 3.7
by good intentions. times the impact that gender did.
I work in the field of diversity. I’ve also been black my It took two more years of research and analysis—using
whole life. So I know that underrepresented people in the data on several hundred thousand employees, drawn from
workplace yearn for two things: the EEOC’s aggregation of all
The first is to hear that they’re Bay Area technology firms and
not crazy to suspect, at times, from the individual reports of
tween negative treatment and Executives tend to 13 U.S. tech companies—before
that there’s a connection be-
Ascend determined how bias
bias. The second is to be offered think—and have a affected the prospects of blacks
institutional support. and Hispanics. Among those
to fulfillment. When we encoun- desire to believe— groups it again found that, over-
The first need has a clear path
all, race had a greater negative
ter colleagues or friends who they’re hiring impact than gender on advance-
have been mistreated and who ment from the professional to
believe that their identity may and promoting the executive level. In the Bay
be the reason, we should ac- Area white women fared worse
knowledge that it’s fair to be sus- people fairly when than white men but much better
picious. There’s no leap of faith than all Asians, Hispanics, and
here—numerous studies show blacks. Minority women faced
how pervasive such bias still is. they’re not. the biggest obstacle to entering
But how can we address the the executive ranks. Black and
second need? In an effort to find Hispanic women were severely
valid, scalable ways to counter- challenged by both their low
act or reverse bias and promote diversity, organizations numbers at the professional level and their lower chances
are turning to people analytics—a relatively new field in of rising from professional to executive. Asian women, who
business operations and talent management that replaces had more representation at the professional level than other
gut decisions with data-driven practices. People analytics minorities, had the lowest chances of moving up from pro-
aspires to be “evidence based.” And for some HR issues— fessional to executive. An analysis of national data found
such as figuring out how many job interviews are needed similar results.
to assess a candidate, or determining how employees’ work By analyzing industry or sector data on underrepre-
commutes affect their job satisfaction—it is. Statistically sented groups—and examining patterns in hiring, promo-
significant findings have led to some big changes in orga- tions, and other decisions about talent—we can better man-
nizations. Unfortunately, companies that try to apply ana- age the problems and risks in our own organizations. Tech
lytics to the challenges of underrepresented groups at work companies may look at the Ascend reports and say, “Hey,
often complain that the relevant data sets don’t include let’s think about what’s happening with our competitors’
enough people to produce reliable insights—the sample talent. There’s a good chance it’s happening here, too.”

144 HARVARD BUSINESS REVIEW NOVEMBER–DECEMBER 2017