This is how to measure bias in the workplace (and what to do when you find it)
Someone once famously said, “You can’t manage what you can’t measure.” This certainly holds true for biases.
We all have biases. Our conscious biases—or those we’re aware of—are fairly easy to recognize. However, our unconscious biases—or those we don’t realize we have—can be more challenging to detect.
Although unconscious biases are unavoidable, they can negatively affect the workplace when we act on them to someone’s disadvantage or use them to make unfair, or unethical, decisions. This can lead to discrimination and negatively impact company morale, innovation, and an organization’s ability to attract and retain talented employees.
One way to combat unconscious bias is to measure your workplace’s bias quotient—your employees’ ability to recognize and disrupt bias.
Measuring bias can give organizations a more concrete sense of where they are more prone to exist, as well as the types of biases employees are most susceptible to.
IT STARTS WITH AWARENESS
Over the past several years, many companies have made an effort to facilitate a more equitable workplace through new or updated diversity, equity, and inclusion policies and initiatives. However, these DEI policies and mission statements are of little use if they don’t work in practice.
A company may say it’s committed to hiring more diverse candidates. However, a closer look might show that, internally, the same types of people are receiving promotions or that when the company makes a diverse hire, it struggles to retain that person.
What might be lacking is awareness—looking beyond your organization’s policies and looking at what’s happening in your workplace. It’s asking the hard questions and evaluating how things in your workplace actually are, not how you would like them to be.
USING DATA TO MOVE PAST AWARENESS (AND TOWARD ACTION)
Data from a variety of sources and channels can be collected to help organizations determine their bias quotient. This data includes:
Behavioral training data
Exit and stay interviews
Helpline reports
Demographics
Employee surveys
Hiring, promotion, and attrition data
Policy incidents and violations
Keep in mind that not all data is created equal. For example, employee surveys can help capture sentiment data at a particular point in time but shouldn’t be the sole predictor of bias quotient. These surveys tend to get polarized responses—there’s generally a group of people who are upset, another group who don’t trust that the survey is anonymous and will tell you what they think you want to hear, and a group of employees who are happy. But the responses miss the vast majority of people who are somewhere in the middle and can provide more balanced insight into what’s working and the areas that need improvement.
Similarly, reports from demographics and internal controls are valuable for their raw data but rarely dig into the “why” beyond the numbers. An organization may meet its goal of hiring more minority candidates but fail to follow up on tracking the experience of those new employees. Bias quotient can’t be all data or all sentiment—it needs to be a thorough mix of both.
The ideal way to measure bias quotient is to find a way to measure day-to-day interactions because, ultimately, that is when bias occurs. However, this can be difficult to achieve and can be invasive.
Behavior-based training offers a way to uncover how employees may act when confronted with certain situations (e.g., hiring, promotions, interacting with people who are different from them) and teach them how to navigate and combat their unconscious biases to affect real change. Instead of simply lecturing learners, the best training challenges employees to visualize everyday scenarios and consider their actions. The goal isn’t to instantly eliminate bias, but to discover how bias influences decisions. Besides making employees more self-aware, behavior-based training produces meaningful data that can be used to measure your organization’s bias quotient.
TAKING ACTION
Although measuring your organization’s bias quotient is an important first step in creating a more equitable workplace, effectively using that data allows you to be successful in achieving that equity. Here are three ways to translate your data into something actionable:
1. Monitor your data
Monitoring your data keeps you on top of trends and ensures you’re responding to your learners’ performances and your organization’s specific training needs.
By evaluating choices that employees felt were appropriate in simulation, your team can understand the baseline for bias and determine a path to remediate misconceptions or biased behavior. A best practice is to evaluate:
How did my population perform in the aggregate to spot and avoid bias?
Are there areas in the organization where microcultures may exist (e.g., per business segment, region, or function)?
What are the most common misconceptions?
Do managers have the same susceptibility?
With this data in mind, you can establish a clear path forward for future training and development. And if you’re doing this in a measured way, you can identify what your organizational trends and propensities are, and if they’ve changed over time. Ultimately, you’ll be able to also provide this visibility to leadership in a thoughtful way.
2. Make sure your program is responsive to employees’ and organizational needs.
Pushing strong, behavior-focused training into the employee population is necessary to ensure change across the organization. Good training challenges participants to consider bias from multiple angles. It also helps employees recognize and challenge—in a safe environment—their own biases.
Consider nontraditional types of behavior-focused training, such as visualization exercises. For example, imagine you’re waiting in line for coffee, and the couple in front of you is taking a long time to order. After employees visualize this scene, ask them to examine it. What race is the visualized couple? What gender are they? Are they old or young? Not all employees have the same unconscious biases; open-ended visualization exercises can yield more personalized learning experiences, which will help the training resonate better.
When thinking through must-haves for effective unconscious bias training, ensure it is focused on provable behavior change—versus a check-the-box approach—and produces behavioral data.
3. Go beyond one-and-done training.
Addressing unconscious bias is a continual effort, not a once-a-year training course or something you roll out when a crisis forces your hand. Consider creating an ecosystem with tailored reinforcement, regular touchpoints (e.g., short videos), and periodic refreshers.
This ecosystem can also be an opportunity to reinforce tone—and, more importantly, action—from the top. For example, when encouraging engaged leadership, insight from behavior-based training can empower leaders to employ consistent two-way dialogue around findings and misconceptions their employees may have regarding such things as bias and microaggressions. Bringing leaders into the dialogue with concrete actions and measures shows your employees that the things they say and do everyday matter. The ways leaders conduct themselves and respond to incidents and potential incidents set the tone and help drive their organization’s work culture.
Whether you’re training leaders or non-supervisory employees, the more exposure employees have to navigate and avoid unconscious biases, the more prepared they’ll be to put what they’ve learned into practice should a situation in the workplace arise.
The goal of measuring and acting on your organization’s bias quotient is to strive for employee growth and continual improvement. Ultimately, you want to train employees to be better, more self-aware, versions of themselves. This can help lead to a workplace culture in which everyone feels respected and included.