The Subtle Difference Between “You Should Develop This Job Skill” and “Stop Being so Bossy”
This study dovetails closely enough with both Vikram’s post on “bossy women” and my post on Julia Pierson that it seemed worth throwing into the conversation down here in Off the Cuff.
This week Kieren Snyder, who holds a PhD in linguistics and is currently the CEO of the small (I think one-person?) tech consulting company Textio, published an independent study on the way male and female job performances are reviewed at U.S. tech companies.
Before I even get into what it says, I want to be clear that this study appears to have been done entirely by Snyder without any kind of partnership with an academic institution. Indeed, I can’t see any evidence that it was peer-reviewed in any meaningful way. (The study was published in Fortune magazine.) Therefore, I think it’s important to take the findings with a grain of salt.[1]
In her study, Snyder reviewed 248 written performance reviews of 180 men and women employed by tech companies. In all, 28 companies of various sizes were represented. The results are extremely interesting, and so skewed by gender that I suspect half of America will read them as proof positive that they’ve been right all along, while the other half will read them as proof positive that Snyder cooked her books.
Here are the most important findings:
58.9% of men received critical feedback; 87.9% of women received critical feedback. According to Snyder, this pattern was consistent regardless of the company size.
Men and women received different kinds of critical feedback for similar shortfalls. Overwhelmingly, according to Snyder, men were given suggestions of ways they could develop “job skills” in areas where their performance was weak. Women were given feedback to similar performance shortfalls — but rather than referring to the shortfalls as “job skill” issues, they were referred to as the result of negative personality traits held by the employee. In fact, negative personality criticisms show up in just under 2% of the employment reviews given to men; it shows up in just under 70% of those given to women.
The reviewer’s gender is not a factor. The results that Snyder found happened regardless of whether it was a male or female doing the performance review. This includes the identification of “job skill” issues vs. “negative personality” issues.
Assuming that Snyder’s data is indicative of the companies she took data from, all of this suggests a rather systemic problem. It also potentially gives some potent ammunition to those that claim that there really is something to those who cite the “bossy” meme as a sign of systemic sexism in the workplace.
[H/T to both the Dish and readers Rebecca and Karen, who all pointed me to the Fortune article.]
[1] It should also be noted, however, that Snyder’s results do seem to jive with actual academic papers looking at similar kinds of data. Full disclosure here: My spouse, knittingniki, has been a key researcher on many of these academic studies published over the years, especially those that examine this phenomena in the STEM fields.
Follow Tod on Twitter, view his archive, or email him. Visit him at TodKelly.com
I’m likely to hire some people soon, for the first time in my life. The great thing about this study is that it’s actionable-it provides me with a behavior (discussing personality traits) that is sexist, and an alternate (discussing job skills) that is less sexist. Frankly, I think you’re better off discussing job skills rather than personality traits for a whole bunch of reasons. Skills are something that can be learned, but not everyone thinks personalities can be changed.
Information like this is so valuable, because I feel that it’s very likely I would have fallen in to this trap without it. So thanks for sharing it.
I am also relieved by point 3 – “the gender of the reviewer didn’t matter”. A lot of the stuff I read about sexism sort assumes that it’s the fault of, and perpetrated by, men. But no, we’re all in this together. I’m reminded of a study recently where they gave out STEM resumes for evaluation, changing only the name from a male name to a female name. All reviewers, male and female, ranked the female-named resumes lower.
We’ve got a problem all right, and this stuff actually helps solve it.Report
Wow, that’s fantastic feedback, @doctor-jay . It’s really great to see someone take these posts and fold it onto their professional lives in practical ways.
If you don’t mind, I’ll probably share this response with my wife. I know she’ll be interested to see that this was the first comment here.Report
a question:
would one mitigation be to discuss what might ostensibly be a personality trait – let’s say someone’s bad at disagreeing with people – as a job skill rather than a personality trait? in other words, “you need to learn the skill of how to disagree with people without insulting them?”
?Report
@dhex Well, kind of.
In most job performance reviews, the personality of the person isn’t supposed to be an issue so much as the job performance. For example, you can be an arrogant person and still communicate effectively with others. Therefore it’s considered inappropriate to say, “don’t be arrogant” instead of “you need to develop the ability to communicate effectively with your team members.”
One of those statements communicates that there is a skill you can develop to be successful working where you do. The other tends to communicate that you aren’t liked or reelected, and probably won’t be successful. It’s like the difference between telling an employee they need to learn how to do X in order to be successful in your department, and telling them they are stupid because they don’t know how to do X.Report
On #3; it’s baked in, both men and women have the cultural habit, developed through all history, to defer to men. Recognizing this takes a tremendous effort, and is to some great degree the part of political correctness that matters.
Think of how often a teenage boy might get asked, “So, you got a girlfriend?” I watched my gay brother get asked this hundreds of times, and he’s told me he’d cringe each time. It’s so easy, instead, to ask if you’re dating someone, have someone you’re keen on, etc.
It takes effort to be respectful; and we’re often unable to see how we offend with a thousand tiny presumptions of the norm.Report
low level genetic/instinctual things can be routed around.
Getting some really interesting data out of the Arab Spring, in particular.Report
I have a big concern about the methodology, if this study was in fact done by a single person, and in fact by a single person primed to find bias.
In document analysis like this, the proper technique is to have a panel of reviewers review the language, without knowledge of whether the language refers to a male or a female. That way you 1) avoid relying merely on the interpretation of a single person, and can examine the results for intercoder reliability, and 2) you avoid the priming effect of the coder being inclined to see what they expect to see.
Maybe the researcher did in fact do this, and just did not report it in the popular press article (or had an editor excise all that boring methodology stuff), but that’s all pretty crucial if we’re to have any confidence in her findings.Report
I second this critique, and I could have noted the limitations in my own post, but in my defense, I was juxtaposing an actual attempt to analyze the question, however clumsily, with the memory of a single cartoonist.Report
Well you guys work at Uni’s, check the library, see if you have access to the paper?Report
I had a long talk about this stuff with my wife after she had a series of disheartening interactions with her team at her current company and it prompted some serious soul searching and self examination.
As a manager, I’ve always had to work very hard to make sure that I’m being truly fair to people because I recognize the bias in myself. When I need to decide who gets the high visibility important task that’s the interesting career maker, my gut says, “Stick somebody who reminds you of yourself in there! You’d be great at this!” The “reminds me of me” thing could be obvious things like “thinks the same way I do” to things we never want to acknowledge like “looks like me” or “has things culturally in common with me.” I spend a lot of time stepping back and asking myself whether my tendency to trust this person with the keys to the rocket ship is because they have objectively the best track record and deserve a crack at it, or if it’s just because I’m more comfortable with them on a gut level.
Even though my wife is an engineer and my faculty advisor and respected mentor in engineering school was a woman, I notice that I have a slightly more intense visceral reaction to criticism from women at work than I do from men, all else held equal. It’s a disturbing thing to realize, but I’m pretty sure it’s there, and it makes it even more important to tamp down the instinctive reaction to criticism (“Having this flaw pointed out will reduce my status among the monkey tribe!”) and let intellect triumph over emotion. Nobody likes being criticized, but the best of us use it as a tool instead of dealing with it as an attack.
So much of what we do is instinctive and based on heuristics that are really subtle and built deeply into our brains that I don’t think we realize how often it skews our behavior and even our perception of reality. I’m trying to find the study, but it was done in software development teams and it found that both men and women tended to overestimate the number of bugs caused by women while underestimating the number of bugs caused by men. It was an interesting metric because bug tracking is usually done in database with good record keeping but there’s a lot going on, so our perceptions of how many problems exist and where they came from aren’t often perfectly in line with reality, so it’s a space that’s ripe for bias to creep in and create a new reality that can easily be checked against concrete data. It’s also scary to think that at the end of a rocky project when the manager is doing reviews, they’re very likely to let their own biases color the data about what really went wrong, who shined, and who dragged the team down.Report
@troublesome-frog
That’s a very good comment. I also worry about how my reactions to my coworkers and supervisors are in part conditioned by my own gendered (and racial) assumptions. I mentioned this before, but I work in a female majority workplace in a field in which women, while they may not necessarily be a majority, are probably less disadvantaged than in a field like engineering. (That’s my sense, at least.)
But still, I know that in practice I give my male colleagues a little more deference than I do my female ones. And although most (maybe all) of my superiors are women, I do think if any were men, I’d probably show a little more respect to them. It’s not that I disrespect my women bosses, it’s just that there’s some subtle thing where I act probably differently from how I would otherwise.
I try to account for that in my actions. Fortunately, I supervise very few people and probably have no direct reports. (The “probably” comes from the fact that lines of authority and responsibility at my work place are less than clear, even when it comes to something one would think should be clear, such as who is a direct report to whom. The formal answer is that I have no direct reports. The informal answer is that I probably have one, at least in practice.)Report
A colleague of mine is doing a similar review of our report cards write-ups, looking for gendered language. She hasn’t completed it yet, but early indications are, “Holy shit!”
The major divide she finds has to do with effort vs skills. Girls who struggle are assumed to lack the fundamental skills; boys who struggle are assumed to be putting forth insufficient effort (with the underlying assumption being that they have the skills but simply aren’t applying them).Report
Kazzy,
I’d be interested in hearing more about this when the study is done, if it’s something you can share. My guess is that “holy shit!” will be pretty dang appropriate.Report
Is it assumption, or are the teachers actually observing them screwing around when they’re supposed to be working?Report
With the caveat that science reporting is usually pretty lousy and not necessarily a reliable account of primary research, which itself is often not replicable, here’s an article in the Atlantic that describes research suggesting that a conscientiousness gap does exist between school-age girls and boys.
That doesn’t mean that you can simply assume that boys’ failures are due to insufficient conscientiousness and girls’ failures are due to insufficient cognitive skills, since of course there’s going to be a great deal of intrasex variation, but neither does a gender gap in teacher assessments necessarily imply bias.Report
You have to focus on both halves of the coin here; the boys are screwing around, and could do better if they focused. The girls can’t do better, they don’t have the ability.Report
@brandon-berg
Hard to say. But the trend seems very real.
@stillwater
I may not be able to share the results in any formal manner, but can probably do a brief write up summarizing what she is finding.Report
Interesting. I hope she goes back some years; change in that pattern might surface from the shift in praising the child to praising the effort best practices, too.
I do have this weird thought that if the pattern hasn’t changed, how does this reflect on girls greater academic achievement in college now? That’s counterintuitive to what I want to read out of this.Report