Diversity Hiring's "Reality Gap"

March 23rd, 2017

Jon-Mark Sabel

Diversity hiring continues to be an issue that everyone gets behind on paper, but few actually follow through with. This is highlighted in Deloitte’s 2017 Human Capital Trends report, where 69% of executives rate diversity and inclusion as an “important” issue, while only 12% of organizations have a mature level of inclusivity.

In other words, there’s a reality gap.

This is not to say there is a lack of support from the top: 38% of executives report the primary sponsor of their organization’s diversity efforts is the CEO. And with inclusive companies generating 30% more revenue per employee, you would think investment in diversity hiring initiatives would shoot to the top of the priority list.

Again, there’s a reality gap.

Today’s organization is an international network, digitally interacting with clients, freelancers, and operatives around the globe. Diversity of gender, race, religion, and culture is a business reality that cannot be avoided. Yet only 6% of these globally connected organizations tie compensation directly to diversity outcomes.

Sounds like a reality gap.

Bridging the Gap by Tackling Unconscious Bias

You'd be hard-pressed to find a talent acquisition or HR professional who shamelessly excludes certain groups from consideration, be it in hiring or promotion. I would even take it a step further and say that most try their upmost to be fair and objective when making life-changing decisions.

Yet, as mentioned several times above, there is a reality gap. In most instances, unconscious bias is to blame - bias that is not rooted in our conscious decision-making, but in less logical parts of the human psyche. The University of San Francisco provides a great definition of unconsicous bias:

Unconscious biases are social stereotypes about certain groups of people that individuals form outside their own conscious awareness. Everyone holds unconscious beliefs about various social and identity groups, and these biases stem from one’s tendency to organize social worlds by categorizing.

In this article, we'll first look at the different ways unconscious bias plays a role in the hiring process, then create a three-part plan to confront it at its source.

Like Hires Like

When choosing between two candidates with equal qualifications, recruiters and hiring managers will almost always unconsciously choose the one they most identify with.

It is human nature to fill gaps in objective data with our own personal experience. This is not behavior that should be chastised, but it should also not be ignored. Instead, the decision making process should be carefully evaluated.

But sometimes unconscious bias takes a more insidious form. When Frank Dobbin and Alexandra Kalev of HBR interviewed a new HR director at a large West Coast food company, they discovered that white managers made only strangers (mostly minorities) take their supervisor assessment test while hiring their (mostly white) friends without testing them.

Loaded Assessments

In an investigation into job-skills assessment tests, Northwestern University professor Lauren Rivera discovered that even when tests were administered universally, hiring managers played unconscious favorites.

For example, when white men performed poorly on the math portion of the assessment, the hiring team paid little attention. But when minorities did just as poorly, their deficits were elevated. This data cherry-picking not only ensured a less diverse hiring pool, it also confirmed the team’s preexisting bias (enabling future cherry-picking).

In another instance, Rivera found that white men lacking communication skills were deemed “coachable,” while their black and Hispanic counterparts were moved to the reject pile. And in applicants considered “shy, nervous, and understated,” white applicants were praised as “modest” while nonwhites were rejected for being “unassertive.”

It is notable that the bias Rivera observed occurred almost exclusively during the evaluation of “middle of the road” candidates. Candidates considered “rockstars” were put through the process quickly, regardless of race or gender. Likewise, those considered “rejects” were ejected from the running with equal abandon.

In the above situations, no member of any hiring team believed themselves performing maliciously. So how can we tackle unconscious bias at its root?

A Complex Problem Requires a Complex Solution

As we've seen, unconscious bias can take many different forms. As a multifaceted problem, it is best confronted from multiple angles. In this article, we’ll look at the three big ones: Data, Exposure, and Training.

Data

There are numerous solutions vendors that can provide insight into the TA professional’s decision-making process. Deloitte’s Human Capital Trends Report identified three tools that monitor different stages of the hiring process:

SuccessFactors’ Human Capital Management suite leverages machine learning to review job descriptions, performance reviews, and other people processes for potential sources of bias.
Entelo’s recruitment management software can be tailored to find candidates from underrepresented groups.
HireVue’s video interviewing platform gives direct insight into how recruiters and hiring managers screen candidates. By identifying patterns in candidate acceptance and rejection, recruiters can objectively evaluate their own unconscious bias.

Unlike vague “recruit more inclusively” mantras, data provides an objective foundation on which to evaluate unconscious bias at the source. Metrics gleaned from this data can then be used to create specific, results-oriented gameplans.

When using data to drive results, it is important that it is not used as a tool for confrontation, but for evaluation. We’ll look at why in the “Training” section of this article.

Exposure

Working side-by-side with underrepresented groups breaks down stereotypes, reducing unconscious bias as a result. A fairly obvious way to accomplish this is hiring more diverse employees - but if that were easy, we probably wouldn’t be having this conversation in the first place. Instead, specific strategies need to be adopted to increase contact within the existing workforce. HBR identifies two:

Self-managed teams. These teams encourage employees in different roles and functions to collaborate on projects by their own volition. Since underrepresented groups tend to be concentrated in certain roles, letting teams self-manage enables greater contact between them than if they were managed from above.
Rotational programs. With a focus on management, these programs rotate trainees through departments as part of the learning process. Again, as different functions have different concentrations of underrepresented groups, these sorts of programs increase contact and break down bias.

HBR estimates that organizations with self-managed teams increase underrepresented groups in management positions by 3-6% over five years. Cross-training programs result in a similar change, 3-7%. It would not be a stretch to hypothesize that if both were implemented they would work together synergistically to bump those numbers even higher.

Training

Training is a cornerstone of diversity enablement, and is often the first program to be implemented after a lawsuit. This is not necessarily a good thing: the wrong training can actually have a negative impact on the inclusion of underrepresented groups in management roles.

The “Wrong” Training

As mentioned above, diversity training implemented as the result of a lawsuit is unfortunately common. With this sort of framing, most diversity training carries a negative “or else” undertone - leading to further discrimination.

HBR estimates that 75% of companies using diversity training use negative messaging in their attempts to inspire change. In this training framework, legal disputes and settlements take center stage, sending a profound message: diversify - or else. Needless to say, these “negative incentives” don’t drive a whole lot of change, and in most cases make the situation worse.

Mandating diversity training goes arm-in-arm with negative messaging. Companies that made their diversity training mandatory saw no improvement in the makeup of management over the course of five years, while the number of black women and Asian-Americans in management roles actually decreased (by 9% and 5%, respectively).

In other words, the “wrong training” consists of efforts to control manager’s behavior, rather than enable it to change.

The “Right” Training

The “right training” puts managers in the position to be “diversity champions”, leading the charge for change themselves.

Voluntary training focused on positive messaging is critical to creating real change with diversity programs. When managers elect to take voluntary training, they already are beginning to see themselves as the type of person who cares about diversity - and will hold themselves to a higher standard as a result.

To illustrate, a University of Toronto study had white subjects read a brochure analyzing prejudice toward blacks. The subjects were then split into two groups: the first group was pressured to agree with the brochure, while the second group was given the freedom to form their own opinion. Oddly enough, those who were pressured to agree actually had their preexisting biases toward blacks strengthened, while those given a choice had their bias reduced.

So how do organizations put their managers and recruiters in a position to become “diversity champions”? The Harvard Business Review provides two frameworks:

Diversity recruiting programs. When diversity is expected as a result (rather than enforced as an “or else”), TA professionals rise to the occasion. Programs that focus on diversity recruitment consistently increase the proportion of underrepresented groups by around 10%. Compare this to the negative effect of mandatory training.
Mentoring programs. When managers are given proteges to tutor and train, they go to bat for them regardless of their diverse status. This creates a virtuous cycle, where underrepresented groups find representation in upper management, resulting in promotions that enable them to mentor diverse proteges of their own. HBR estimates that mentoring programs increase the representation of diverse groups by 9-24%.

Up until now we’ve thought of “diversity” as traditionally underrepresented groups. But Deloitte has a slightly different conception, including age and thought in its diversity model.

A More Inclusive Model of Diversity

Technology has done a marvelous job of pulling us together… except when it pushes us apart. Every year sees Millennials take up a larger portion of the working population. Members of the Baby Boomer generation who continue to work often feel marginalized at the expense of those considered young and innovative.

Deloitte also identifies “diversity of thought” as a future sticking point. This encompasses not only different viewpoints, but also the differences in cognition that are found in conditions like autism.

Too Much Inclusion?

Some would argue that expanding the diversity model to this extent devalues existing notions of diversity. If the elderly are given the same “weight” as traditionally underrepresented groups, and contribute to the same "diversity number”, it is not hard to imagine the creation of a workforce made up of elderly white men. Which kind of defeats the purpose.

But remember the source of the problem in diversity hiring: unconscious bias. If we move away from the concept of diversity as a “reporting goal” and replace it with the notion of “business necessity,” the above objection becomes moot. If diversity is leveraged as way to increase innovation, productivity, and reputation, there is no benefit to gaming the system.

By removing bias present in the hiring process, the door is opened for all members of diverse groups, both traditional and non-traditional.

Where to Begin?

In this piece we’ve covered several methods designed to tackle unconscious bias. Given the range of tech solutions designed to improve diversity hiring, it might be tempting to simply throw money at the problem rather than address its core causes.

But without uncovering the causes and sources of bias first, no solution can be used to its full effect. If you intend to adopt a “tech first, evaluate later” mentality, start with a solution that gives insight into the activities and tendencies of decision makers throughout the hiring process.

For organizations more hesitant to adopt one of the range of hiring solutions available, start with the simple semantic distinction made in the Training section of this article: consider diversity an “expectation” rather than a “metric”. Put frontline recruiters and hiring managers in the position to be diversity champions, and trust them to go above and beyond.

Stats to Reflect On

We've referenced a lot of statistics through the course of this article, courtesy of Deloitte's 2017 Human Capital Trends Report and Harvard Business Review's Latest Research: Diversity. For quick reference and reflection, here are some of the most relevant:

Trends within Organizations:

38% of executives report the primary sponsor of their organization's diversity efforts is the CEO. (Deloitte)
69% of executives rate diversity and inclusion as an "important" issue. (Deloitte)
12% of organizations have a "mature" level of inclusivity. (Deloitte)
6% of organizations tie compensation directly to diversity outcomes. (Deloitte)
75% of companies using diversity training use negative messaging. (HBR)

Diversity Outcomes:

3-6% increase in underrepresented groups in management with self-managed teams. (HBR)
3-7% increase in underrepresented groups in management with cross-training programs. (HBR)
9% decrease in the number of black women in management roles with mandatory diversity training. (HBR)
5% decrease in the number of Asian Americans in management roles with mandatory diversity training. (HBR)
10% increase in underrepresented groups in management with diversity recruitment programs. (HBR)
9-24% increase in underrepresented groups in management with mentoring programs. (HBR)
30% more revenue per employee is generated by inclusive organizations. (Deloitte)

The Diversity Rules Have Changed

Deloitte accurately points out the reality gap between what organizations say and who they hire. But it’s 2017: Glassdoor is closing in on 35 million users, and an increasingly politicized populace leans heavily on the reviews of their peers in the decision-making process.

Companies that work to bridge the gap are not only on the right side of history, they put themselves in the position to out-invent their peers in an era where innovation is king. Just ask Salesforce.