How unconscious bias warps our sense of merit

Last time we considered an intriguing study led by Dr. Arin Reeves, testing whether and how law firm partners were influenced by race when reviewing associates’ work. A diverse group of law firm partners reviewed an identical research memorandum, with half of them told that the “author,” 3rd year associate Thomas Meyer, was Black, and the others told he was White. The results were that Black Thomas Meyer’s memo was treated as clearly inferior to the exact same memo by White Thomas Meyer – in other words, Thomas Meyer had less “merit” than Thomas Meyer. While these results may not have been startling, this particular finding was, at least to me:

In other words, the race, ethnicity, and gender of the reviewing partners had no significant impact on the different evaluation results for Black Thomas Meyer’s and White Thomas Meyer’s identical memo. How could this be?

Befuddled by this, I reached out to Dr. Reeves, who, BTW, is as delightful as she is impressive. After years of practicing law, Dr Reeves earned her doctorate in sociology at Northwestern University. She is now a best-selling author and the founder and managing director of the expert research and advisory firm Nextions.

I knew going into our conversation that unconscious bias can have an undeniable impact on my own thinking. I’m reminded of that each time I set foot on an out-of-service escalator and briefly feel – not merely expect, but physically feel – the stationary escalator move. And I’ve always liked Jonathan Haidt’s metaphor of the human mind as a small rider (our logical and rational prefrontal cortex for executive function) perched atop a large elephant (our emotional and intuitive amygdala/limbic system). The rider tries to steer the elephant, but instead often ends up simply rationalizing why the elephant went where it will.

Here’s what I learned from Dr. Reeves:

• Misdirection is needed to effectively test for how our minds unconsciously work. A classic example is the selective attention test in which video viewers are instructed to count how many times a basketball is passed, and are then asked if they noticed the person in the gorilla suit walking through the scene (half of first-time viewers did not … and if you’ve already viewed it and are now gorilla-vigilant, try this follow-up test). In Dr. Reeves’ implicit bias study, the participating partners were all Baby Boomers, likely skeptical of young lawyers’ writing abilities, and likely not so focused on the passing mention of the memorandum writer’s race. The conscious focus was on generational differences, and the planted, unconscious seed was race.
• In studies involving misdirection, participants can later feel embarrassed, or even betrayed, especially if the study’s results challenge a participant’s sense of self. So Dr. Reeves reached out to each of the study participants to debrief them afterwards. And there was indeed dismay, particularly for participating Black and Brown partners, and also for White partners who were sure they were very objective.
• Quite a few of the Black and Brown participants commented along the lines of “maybe I was harder on somebody I thought was African American because I knew that they needed to be better.” That explanation resonates for me, because “we have to be twice as good” is a widely held perspective. Recent research conducted for the ABA Commission on Women in the Profession and the Minority Corporate Counsel Association by the University of California, Hastings College of Law found that, among lawyers, women and Black and Brown men reported they have to go above and beyond to get the same recognition and respect as their lawyer colleagues.
  • Women of color reported Prove It Again (“PIA”) bias at a higher level than any other group, 35 percentage points higher than White men.
  • White women and men of color also reported high levels of PIA bias, 25 percentage points higher than White men.
  • Women of color reported that they are held to higher standards than their colleagues at a level 32 percentage points higher than White men.
• This type of thinking can trigger stereotype threat, which occurs when we are aware of a negative stereotype about our own social identity, and our environment brings this awareness to the fore.
• Dr. Reeves agrees that this thinking may well have influenced some participants who were conscious of race when reviewing the memo by Black Thomas Meyer. But she also stresses the unconscious impact of confirmation bias, triggered by the race of the person reviewed rather than the reviewer’s race.
• According to Dr. Reeves, our brains constantly create heuristics, or rules, for interpreting the world around us. We create these rules based on what we see in the world. It’s purely a numbers game – the more of something we see, the more likely it will end up as a rule for us (such as the thousands of times I’ve stepped on a moving escalator). That’s what leads to confirmation bias. And ironically, though we create these rules or norms for what is external to us, we view ourselves as the exception, somehow immune to the rules we unconsciously apply to others.
• The reviewing partners in Dr. Reeves study had spent their careers surrounded by lawyers who are predominantly White. And so, regardless of the reviewer’s race or ethnicity, the “numbers game” supports a heuristic of what a competent third year associate looks like. White Thomas Meyer fits the heuristic, and so we are less likely to notice his writing errors. Black Thomas Meyer does not, and so our unconscious risk level (what if his writing is not up to par?) is slightly higher, and we become more focused on his writing’s shortcomings. Dr. Reeves stresses that this all operates unconsciously, such that the participating partners, regardless of their race or ethnicity, would be sincerely surprised by the disparate results.

A key point I learned from Dr. Reeves is that we need to understand the difference between observation (how many errors do I see in Thomas Meyer’s writing?) and evaluation (given those errors, how do I evaluate Thomas Meyer’s writing ability?). Unconscious bias can affect both – not just how I evaluate, but also what I observe in the first place. If my unconscious heuristics or rules warp what I observe, then even an “objective” evaluation can be tainted. So, if I’m truly interested in being merit-based, I’d better be intentional in what I’m looking for, and in what I observe.

And in how I evaluate, another danger lurks – my use of proxies for merit as a shortcut for my decision-making. More on that next time.

P.S.: I’m going rogue with this postscript – before publishing I always provide interview posts in draft to the person interviewed, for their review and approval, but I’ve had another thought after receiving Dr. Reeves’ green light for this post. I’ve now reflected further on what I learned from Dr. Reeves, and I realize that in approaching her study I was doing exactly what she described.

I viewed the Written in Black & White study from the perspective of wondering why the reviewers treated Black Thomas Meyer so negatively. And then a light bulb lit. I’m now thinking that the reviews of Black Thomas Meyer’s memo seem pretty much on target. Remember, the identical memo was seeded with numerous errors, including 18 spelling and grammar errors, substantive technical writing errors, and errors in reciting the facts underlying the memo (four additional errors in analysis were not quantified in the findings). The reviewers of Black Thomas Meyer’s memo on average found most of those errors (15 of 18). So, it seems fair and reasonable to me that his writing was rated overall as 3.2 of 5, and “average at best.”

What I missed was that instead it was the reviews and rating of White Thomas Meyer’s identical memo that were off the mark. The partners reviewing exactly the same memo by White Thomas Meyer only flagged on average 10 of the 18 errors, and they rated White Thomas Meyer’s writing overall as 4.1 of 5, by a “generally good writer….”

My perspective on the study warped my view of its meaning. The issue isn’t why Black Thomas Meyer was treated negatively. The issue instead is why White Thomas Meyer was basically given a free pass on an apparently mediocre memo, with 18 errors in spelling and grammar, in substantive technical writing, and in reciting the facts.

Turns out that I too was unconsciously steered by a heuristic that centers White Thomas Meyer as the norm. Whiteness then becomes invisible for me, and my conscious focus shifts to the reviewers’ treatment of Black Thomas Meyer, instead of what was right in front of me – the anomalous, positive treatment of White Thomas Meyer. Given all of the errors in the identical memo, why was White Thomas Meyer reviewed so favorably, treating his memo’s mediocracy as merit? That’s where my conscious focus should be, if I’m to get better at observing and acting upon actual merit.

In other words, I’ve got more work to do in steering my own elephant.