Kieran Snyder was chatting with a friend about his performance reviews when he mentioned one of his direct reports was talented — but "abrasive." She was being considered for a promotion but her personality seemed to be a stumbling block for her manager.
"This is a really good guy, concerned with fairness," said Snyder. "The way he described one of the women on his team… made me wonder whether he’d actually write that into a formal document."
That conversation was the catalyst for Snyder — a Seattle-based linguist, technology professional and CEO of her own startup, Kidgrid — to do a little research.
"I’ve worked in tech for many years and held leadership roles at major tech companies," she said. "I have generally received strong reviews. But while the reviews often call out my business and technical contribution, they have frequently included personality feedback — I’ve heard words like abrasive, emotional (and) aggressive a lot. For many years, I thought it was just me. I’m sure I have things to work on, but I got curious about how pervasive the pattern was."
Snyder collected 248 performance reviews from 180 different people — 105 men and 75 women. Her findings? Women were much more likely to receive negative wording or comments about their personality traits.
"Women received plenty of straight constructive criticism too. It wasn’t all personality. But where personality feedback occurred, it occurred almost exclusively in women’s reviews. Men only got the constructive stuff. Women got both," she said.
That type of negative personality feedback occurred only twice in the 83 "critical" reviews belonging to men; it occurred in 71 of the 94 critical reviews received by women, found Snyder.
Sadly, those findings didn’t come as a surprise.
"I’ve worked in tech a long time. I’ve seen some terrible examples of institutional bias," she said.
The results were just one more example of something Marina Adshade has experienced in the education sector for quite some time.
"While there may not be a lot out there about reviews done by firms for their employees, there’s an enormous amount of research being done about how students evaluate their teachers, based on their gender," said Adshade, an economist and lecturer at the University of British Columbia in Vancouver. "We hear this all the time that a lot of the behaviour that is rewarded in men is actually penalized in women."
Conversely, when students evaluate their professors, the biggest rewards are given to women who conform to gender stereotypes, said Adshade.
"Deviating from gender stereotypes actually really hurts you in the classroom. And I suspect this is true at work as well."
Adshade did her own experiment last year, using a course she’s taught several times and consistently received good reviews for.
"I thought I would try an experiment in light of this thinking about conforming to gender stereotypes. So what I did is I taught the entire term wearing a dress. Because, I figured, nothing says, ‘I conform to gender stereotypes’ quite like a dress and pantyhose, which professors don’t normally wear. And I saw my course evaluations jump up by half a point on a five-point scale," she said.
"It was really shocking — in fact, it’s the highest marks I’ve ever been given for a course. Now, it’s not a controlled experiment, but I thought it was a really interesting outcome... What’s even more interesting was the course that I was teaching was actually ‘Women in the Economy’ and 85 per cent of my students were women."
Evaluator’s gender makes no difference
That raises another interesting point — and a key one — in Snyder’s findings. One-quarter (25 per cent) of the performance reviews she collected were written by women. And female managers accounted for just over 23 per cent of the negative feedback.
"I actually consider this the breakout finding of the study," she said. "I went back and looked at reviews that I had written after completing the analysis. Guess what? I follow the pattern too. I was tougher on women in my critical language than I was on men, even though they statistically got the same review scores. I didn’t use the specific word ‘abrasive’ but my criticism of women on my team reads tougher to me."
Both women and men can have these unconscious biases because they’re so embedded in the way we think, said Carolyn Lawrence, president and CEO of Women of Influence in Toronto.
"The gender lens is really critical in both recruiting and also performance evaluations processes. What’s really interesting in the work that we do in evaluating corporate culture, you can see how if it’s male-dominated, it’s not just in the culture, these processes are actually (embedded) in how you hire and promote people, and in succession plans," she said.
"If I’m a man, and I’m evaluating a man, I’m naturally marking him on the same markers of success that I evaluate myself (with), and probably embedded into the job description. So if I’m a man and I have a female employee, I’m marking her on my perception of success.
"Even more than that, if she is then trained to performance evaluate her employees, she’s trained in the male way."
Even really good and meticulous HR departments are generally not resourced to monitor for bias at the level of review language, said Snyder.
"The best departments look at statistical variation in numeric scores. Lots of places do that now. But the language that we use to describe our colleagues is another layer," she said.
"If you’re constantly getting rewarded for a particular behaviour, you’ll show more of it. If you’re constantly being punished, you’ll do less of it. Using negative language to characterize a behaviour is a form of punishment. We talk about sticks and stones, but names do undermine."
And whether it’s words like "abrasive," "bossy," "strident" and "emotional" or more positive descriptors, relying on personality feedback is evidence an appraisal system is broken, she said.
"Any kind of personality feedback, whether you’re calling someone abrasive or emotional or brilliant or collaborative, is inherently subjective... It’s true that performance reviews are always going to be a little bit subjective but ‘delivered X’ is a lot less subjective than ‘You are brilliant but impatient.’ We should be aiming for objective analysis. Personality adjectives are never going to be objective."
To begin identifying and addressing this, HR should take an active, critical look at the language used in reviews, said Snyder.
"(One) guy I used to work with requested that his HR department publish guidelines on review language. He acknowledged that not every manager would use them actively, but the mere fact of their publication would give employees and managers a framework for discussion," she said.
And — though it may be a bit uncomfortable — it’s important to take a long, hard look at the performance reviews you’ve written, said Adshade.
"Unless people are encouraged to think about the way they themselves are behaving, it’s always ‘the other’ — ‘The other person is behaving this way, it’s not me behaving this way.’ ‘These other people are bad, but I would never do that.’ What people need to do is they need to be prodded into thinking about this themselves," she said.
"For example, what would be really useful... is if people went back and sat down and read through all their evaluations they’ve provided in the past, and see if they can observe this for themselves. I think that would be a good learning experience for people — self-evaluation is the only way that this is ever going to change. Just telling people that other people do this is never going to be enough, because they all assume it’s not them."