2015 NCWIT Summit – Plenary III, Hacking Bias by Joan C. Williams

May 20, 2015

[upbeat pop-rock music]

BOBBY SCHNABEL: So, welcome back. We'll give people a minute or so to find their seats. I hope you all had a great day of workshops. You may have even noticed that Lucy has powers that you don't even realize, so we arranged for there to be slightly dubious weather in the afternoon so that people would be less tempted to be on the beach and would actually attend the workshops. And you certainly don't wanna be on the beach now because we have a great session coming up. To set the context for that just a little bit, one of the themes we hear a lot about at NCWIT is that one wants to address diversity in organizations by changing the culture of the organization, not by changing the culture of the people who may be different from that. And our next speaker, Dr. Joan Williams, is a wonderful person to help further our understanding of that issue. Dr. Williams is a distinguished professor of law at the University of California Hastings School of Law and also the founding director of the Center for WorkLife Law there. She's a leader in understanding how gender bias functions in the workplace and developing innovative strategies to, one of our favorite words, disrupt it. Her recommendations help organizations identify, measure, and address diversity-related bias in their daily interactions to create a more inclusive culture. Dr. Williams is the author of more than a hundred academic articles and book chapters and is a frequent contributor to the Harvard Business Review, most recently publishing a paper whose title I love, Hacking Tech's Diversity Problem, last October. So please join me in welcoming Dr. Joan Williams. [audience applause] [upbeat pop-rock music]

DR. JOAN WILLIAMS: Darren, I notice I don't have a clicker. I don't have a clicker, folks. This is a gorgeous slide, but I don't think we wanna see it for a whole hour. Where's my clicker?

AUDIENCE MEMBERS: There it is.

DR. JOAN WILLIAMS: This is it? Okay. I'm gonna talk to you about what works for women at work and to talk to you about what individual strategies women can use to navigate successfully through subtle gender bias. And what you can do walking out of here tomorrow to help interrupt bias. And then at the end of the talk, how organizations need to restructure themselves in order to eliminate this kind of bias. So this is all from a book that I wrote with my daughter Rachel Dempsey, who just graduated yesterday from Yale Law School. Yay Rachel. [audience applause] The book called What Works for Women at Work is all about individual strategies, because about five years ago I thought, you know what, I've been workin' on gender bias for 20 years and very little has changed, and I am over it. I'm not gonna work on organizational change anymore. I'm just gonna talk to women and find out what worked for them. And that's what this book is about. Then, after I had done that, the editors of the Harvard Business Review kept badgering me and saying, yeah, but how about organizational change. So what I did is I forged the approach that I developed in the book into an organizational change strategy, and I'll be talking about that and institutional solutions at the end of the talk. For all of this, what I did is I took hundreds of experimental social psychology studies, and those are what I call the 40 white college student studies. They are done by social psychologists who, as part of their classes, tell their college students, you have to come in and take a psychology experiment. And so, there was a huge question about whether these studies actually described what went on at work. So I did something very simple. I took several hundred of these, organized them into a very simple protocol where I simply described the findings of the experimental studies and I asked a very simple question. Any of that sound familiar? 96% of the women I interviewed said yes. So 96% of the women I interviewed reported experiencing one or more of these patterns of bias. But the other thing I did was organize this very complicated nuanced literature into four distinct patterns. And you see from the circles at the bottom, women experience these four patterns quite differently and with different incidents. I'm first gonna talk to you about how bias plays out and individual strategies that you can use to interrupt bias both for yourself and on behalf of others. And then I'll talk about the model I call metrics-based bias interrupters, which is what was the article in last October's Harvard Business Review. The first of these patterns of bias I call prove-it-again. That's simply because women have to prove themselves more so than men in order to be seen as equally competent. And you can see here that two-thirds of the women we interviewed reported this. We subsequently did a study of 557 women, a survey, of 557 women in STEM. Two-thirds of them reported it too. It stems from lack of fit. This is just an example for lawyers, it could equally be for computer scientists. Only 17%, it's actually bumped up a little bit, 17% of equity partners are women. When people think of the computer scientist, they tend to think of kind of a slightly tall balding white guy. That's actually a quote from one of the interviews. And so, women don't seem to fit quite as well, not to pick on anyone in particular, right? [audience laughter] And so they have to provide more evidence of competence than men in order to be seen as equally competent. This is a race-gender pattern. It's triggered by race as well as gender. Blacks also have to provide more evidence of competence than men in order to be seen as equally competent with Whites. One of the ways this plays out is that women tend to be judged on their performance, men on their potential, and here's an example. "I was constantly told, you have the network, "but you don't have the work. "Whereas a man could take out clients for "dinner or golfing, he was given credit for that." You notice how the man was being judged on his potential to bring in clients, the women had to have the clients in hand. There's another pattern that's called he's skilled, she's lucky. Men's successes tend to be attributed to skill, women's to luck. On the other hand, women's mistakes tend to be noticed more and remembered longer. This is somebody who had a, was doing great professionally, and had a stumble, and wow, it had profound consequences. She said, "How quickly do you take her hero robe off?" The studies suggest far more quickly for women than for men. This is what Rachel, my daughter, calls the stolen idea. It says, "That's an excellent suggestion, Miss Trigg. "Perhaps one of the men here would like to make it." [audience laughter] Raise your hand if you have been in a room where you saw the stolen idea. It's very, very, very common. That also is attribution bias. You assumed the brilliant idea's gonna come from someone who kinda fits your automatic image of a computer scientist. This is called leniency bias, when objective rules yield different results. Objective rules tend to be applied rigorously to women but leniently to men. So when a woman loses a big client, for example, if she's in sales, her competence to plummet immediately. Men are given a lot more runway, objective rules applied quite differently. And then there's double jeopardy. Double jeopardy is quite simply that the experience of gender bias differs by race. For one thing, Black women are more likely to report gender bias than are the other three groups of women, Whites, Latinas, and Asian-Americans. 77% of Black women report prove-it-again bias. Only 66% of the other three groups of women. That's 'cause Black women are triggering two sets of negative competence assumptions. Here's some individual strategies and interrupters. What if you're sitting in a room and you see men being judged on their potential, women on their performance. Or it could be Blacks and Whites. You can just say very moderately, you know, I think we finally recognized what we're actually looking for in this candidate. It's A, B, and C, so let's go back to the top of the pile and make sure we've picked up everybody who has those qualities. A very gentle, what I call an individual bias interrupter that you can use tomorrow. What if you're sitting in a room and you see the stolen idea. Again, you don't have to be confrontative, you don't have to embarrass anyone, I've been thinking about that ever since Pam first said it. [audience laughter] You've added something important. Here's the next step. An important thing for women is not to make themselves prove it again. For example, studies show that if there are nine requirements for promotion, women tend to wait until they have 14. You know, men go up when they have six. They have to show that they are men to be reckoned with. They're under their own pressures. And so, women have to learn to operate a little bit more outside of your comfort zone. Also, if you think about it, if people are gonna tend to notice your mistakes and overlook your successes, you gotta remind them. So it's important to keep careful, real-time records of all objective metrics met and to make sure your network knows about those. Sponsorship is super important. When you're in an environment where women have a fragile hold and many CS environments are like that, having a sponsor is one of the main ways that you can overcome that disadvantage. A sponsor is simply a mentor who's willing to spend his political points for you. That's what sponsor is. That's the prove-it-again bias. Here we come to what we call the tightrope. And as you can see, it was the most common form of bias amongst our informants. About three-fourths of women reported it. Remember that brilliant computer scientist? Not only is that automatic image of a man, it's of someone who has qualities that are coded by the culture as masculine. So he's very technical, he may be slightly Asperger's, he doesn't have a lot of social skills, he has leadership qualities, though, because he's so brilliant at what he does. So in order to be taken seriously, women have to behave in masculine ways. The only problem is that women are expected to be feminine. And so, women find themselves walking a tightrope between being seen as too masculine to be likable, that's the too masculine problem, or too feminine to be seen as competent, that's the too feminine problem. Part of this stems from inside women. A lot of it stems from outside women. Because women are under a lot of pressures to behave in what are seen as suitably feminine ways. One important study found that women are expected to do more of what is called as a technical matter, organizational citizenship behavior. I just call it the office housework. Women are expected to do more of it, and they get less pay-off for doing it than men. What is the office housework? Some of it's literal housework, or administrative work. Taking notes in a meeting, organizing parties, cleaning up the debris. Some of it is real work, it's just undervalued work. For example, one of the things I know, I live in San Francisco, when I go around to women in tech and say what's the office housework in tech? They go like, organizing off-sites. It's important, it's important for team building, but it's not considered a career enhancing move. A perfect example of office housework. In some tech companies actually, diversity work is office housework because it's important work, but it's kinda not counted when your performance evaluation comes around. And it should not be structured that way, but if it is, you gotta ask yourself some tough questions about how much time you're gonna spend on it. On the other hand, if you're direct, outspoken, assertive, and competitive, what's your name begin with?

AUDIENCE MEMBERS: B

DR. JOAN WILLIAMS: You got it, right? That's the too masculine problem. Respected, but not liked. So if you're stern or you say no, the immediate reaction is to call that woman a bitch, right? If you're a man, it's just a no. This is that women, forceful people are commanding, but women are expected to do what a colleague of mine calls the nice work. Women are expected to be nice, modest, and self-effacing. And that's that tightrope that they have to walk. Anger is also a key danger point for women. Of the 127 women we interviewed, only three had ever shown open anger on the job. And that's 'cause they were very highly successful women and showing anger is very dangerous for women. Showing anger tends to increase the perceived status of a man but decrease the perceived status of a woman. Handle with care. So is self-promotion. When people who engage in self-promotion get further than people who don't, but women who engage in self-promotion often encounter push back. This is, we run a leadership academy in San Francisco. One woman found out that a man, she thought he was very similar, was earning $200,000 a year more than she was. So she went to a member of the compensation committee and said, I wanna familiarize you with what I did last year and I grew this account and I did this, and the comp committee member said, "Well, you think highly of yourself, don't you?" [audience laughter] Those programs where you teach women to negotiate harder and to promote themselves are goose-stepping women into this kind of bias. They are not, without more, they are not helpful. Like prove-it-again, the tightrope differs by race in super interesting ways. First of all, African-American women are allowed to behave in more dominant ways than White women are, although at a certain point, you go too far, and then what are you called?

[Audience Members] [shouting answers]. [laughter]

DR. JOAN WILLIAMS: A BB, you got it. [audience laughter] And that's not a good place to be, or an angry Black woman. Latinas who behave in very assertive, dominant ways are often called crazy, too emotional. You're triggering that racial stereotype of the hot-blooded Latina. Similarly, with Asian-Americans, Asian-Americans report far more pressure to behave in feminine ways than any of the other groups of women, and far more push back if they don't. And if they do behave in ways that are considered to be too dominant, unseemly in a woman, sometimes they also trigger racial stereotypes of being seen as untrustworthy and sly. Here's some strategies and interrupters. What if you are in a room where somebody was going like, oh my God, she's so outspoken. She really has no sense of limits. You can just say, again, moderately, "I wonder if we would be saying this if she were a man?" Sometimes that will work, sometimes it won't. How about if you have to, how about changing your own behavior in order to dodge this kind of bias. The first point here is, if it ain't broke, don't fix it. I mean, because what you're proposing to do here is to change deeply identified parts of yourself in order to fit in. Should we have to be doing this, ladies and gentlemen? No. But that's a different book. [audience laughter] That book is What Should Work for Women at Work. What I'm talking today is about what does work for women at work, and the fact that women have to be doing all this stuff shows you why women often retire in their early 50s. They go like, I have had it. But if you find people aren't taking you seriously, if you're getting stuck with large loads of office housework or if you're getting push back because you're outspoken, here's some a things to consider. First of all, when it comes to office housework, what if your, the advice for women is don't volunteer, but even more important, the advice for managers is don't staff this stuff by asking for volunteers. Because if you do, the women will be under subtle pressures to do it and the men will be under subtle pressures not to do it. Instead, establish a rotation. Whether it's taking notes, cleaning up after the meeting, all of those things, you should be establishing a rotation. And if you're the woman, you do it once, you do it graciously, and then you work behind the scenes to establish a rotation. How about if it's work that's actual work? The off-site, for example. But you know that running, I was working with one woman who was with a major tech company and she was supposed to be diversity on top of her regular job, and when it came time for a performance evaluation, nobody mentioned the diversity work but said, well, you're not really putting in as long hours as we'd like on your real job, you're outta there. You are outta there. How do you get outta there? You use the strategy we call the strategic no. You go out and get some of these devalued work assignments that are important, and that can help you build your network, for example, within the company. So you go out and you get some that, a few that are gonna work for you, and then when the next opportunity comes around to serve for the sixth year in the row on the paperclips committee, you go, I have loved paperclips. It has been a stretch for me. However, now I'm working with Tom on this major strategic initiative. You know who would be perfect for paperclips? Ted, down the hall. [audience laughter] That's the way you do it. One of the too feminine problems has to do with speech patterns. If you put men and women in a room together, the women tend to interrupt less, they tend to make fewer task suggestions, we should do this, and they tend to look people in the eye less. Women don't do this when they're with other women. But these are feminine conversation patterns. They are also the conversation patterns of every subordinate group. And so, it's very important not to be sending those signals. On the other hand, one of the reasons women get in that situation is that if a woman interrupts, she might be running into that tightrope problem of like, who does she think she is. Here's an example from someone who was an investment banker whose mentor said, she said, I can't get a word in edgewise. He said, here's how you do it. When someone is powering down towards the end of a sentence, and you think they're about to finish, you say, you start to talk over them. And then, if they're not finished, you say, oh, I'm sorry, I thought you were done. That's actually what I call gender judo. It's doing a masculine thing, interrupting, but in a feminine way in order to control the backlash. Here's another one of these feminine traditions. Is this a masculine stance or a feminine stance? Is this a masculine stance or a [laughter] I call this doing the Don Draper, right? [laughter] This stance, off-balance, and making yourself small, is actually the stance that every primate uses to signal submission. This is the stance that every primate uses to signal dominance. Sometimes if you feel you're extremely threatening, you're gonna wanna do this, but not most of the time. Here's more gender judo. This is actually from a tech executive. She said, "What I'd do is I'd be warm Ms. Mother "95% of the time, so that the 5% of the time "when you need to be tough, you can be." You sometimes hear that women leaders are more communal. That's not because women are more communal. That's because if a woman leader doesn't behave in a communal way, she will never get to be a leader because her name will begin with

AUDIENCE MEMBERS: B.

DR. JOAN WILLIAMS: Right. [audience laughter] You also hear sometimes that women don't ask. But the reason women don't ask is, one of the studies finds, that women who do negotiate hard for starting salary tend to be disliked and hired less. It's another tightrope problem. What do you do? The solution is not to not negotiate. The solution is to negotiate and if you feel that there's push back, to use some gender judo. Make the request communal. It's really important for my team that they see I got this promotion. Express concern for relationships. Blame it on someone else. A really good example of this is the way Sheryl Sandberg framed her negotiation with Mark Zuckerburg, it's described in Lean In. Great example of gender judo. How about the promotion problem? The best solution here is what we call the posse. And that's to find a group of men, as well as women, about at your level or a little bit above, and you celebrate each other's successes. There's two advantages. One, you get your successes out there and you avoid push back. And the second is, you become a decent human being. [audience laughter] So the posse is a very important strategy both for women and for people of color, and make sure you include both majority and minority members in your posse. If you're a manager, one of the ways that you can help women around you is to become a member of their posse. It's a very easy thing to do. You can just say, you know, I think you're fabulous, I would love to help your career. When you have a success, when you've met an objective metric, I would love to hear about it. That's all you have to say. And then, you can share it, and that's a very powerful way that you can help women. If you're a senior woman, you can also praise your team. Now, make it clear you led the team, but you can praise your team. If you're a woman and you're getting angry on the job, what you're doing is avoiding triggering this women as id stereotype. There's literally a formulation from one of the psychological studies, says, "If I look angry, "it's because I am angry. "And the reason I'm angry is that you have "jeopardized [insert shared business goal here]." And you notice how deep I pitched my voice and how slowly I talked. Talking slowly communicates, I'm entitled to take up your time. So it's a very important power move. [audience laughter] I talk fast, that's just 'cause I grew up around New York. A different issue. If you do have a sponsor, someone who is willing to spend his political points for you, your sponsor is part of your posse. Is this weird? No, it's not weird. You know what he's called? He's called a work colleague. It is not weird. It's not weird for him, it is not weird for you. That's the way it is. Here's the third pattern, it's called the maternal wall. The maternal wall is less common. You can see, 59% of the mothers reported it. On the other hand, once it's triggered, it's an order of magnitude larger than the glass ceiling patterns I just described. The famous study is by Shelley Correll. Identical resumes, one but not the other a mother. Identical, right, identical resumes. The only difference is that the mothers mention membership in the PTA. That single difference, four words, was enough to mean that the mother was 79% less likely to be hired, half as likely to be promoted, offered an average of $11,000 less in salary, these were management consultants, and held to higher performance and punctuality standards. Motherhood triggers very, very strong negative competence and commitment assumptions. On the other hand, Shelley and Steve Benard did another study where they had a little vignette and it was of a woman who was indisputably competent, indisputably committed, but also a mother. And they found that women, not so much men, but women tended to see the highly-accomplished mother as less likable and tended to hold her to higher performance standards. This is another form of tightrope bias. The good woman, the good mother is someone who's always available to her children. The good worker is someone who's always available to her employer, and never the twain shall meet. Here's another form of maternal wall bias. This one is benevolently meant. It's, you know, I didn't consider you for this assignment, even though it's a stretch assignment and would be perfect for you, because I know it's not a good time for you with the two small kids. Benevolently meant but the message still is, good mothers wouldn't want this assignment. In that sense, it's still prescriptive bias. And then, there's the flexibility stigma for people who take leave or adopt a flexible schedule. For mothers, the flexibility stigma is simply a trigger for maternal wall bias. For men, the flexibility stigma, there's three different studies that show that it's actually even stronger than for mothers. The flexibility stigma is a femininity stigma. Men who take time off are seen as unsuitably feminine. Okay, strategies. Here's some individual strategies and interrupters. What if you hear somebody saying, "Remember that time she wasn't here?" You can say, very moderately, "Yeah, I do. "And I also remember the 17 times afterwards when she was." Or, "She's only part-time." Yes, but she's doing the following important assignment and she's doing it very well. Her husband supports her. This is irrelevant. [audience laughter] And saying that could get us sued, I tell you that as an employment lawyer. [audience laughter] I think it's very important for us to give managers a script when it comes to motherhood. Because I think that many supervisors, male as well as female, are apprehensive. They wanna do, they don't wanna overstress the mother, but on the other hand, they don't wanna take away the opportunity. So it's very simple. If there is an assignment that is very suitable for a mother, you say this. "Here's an assignment that is a good stretch "assignment for you, you would be perfect for this. "But you know, if this isn't a good time for you, "not a problem, these things come around from time to time." And then shut up. That's all you need to say. And then, you put the ball in her court and so she can make the decision. For mothers themselves, when you come back from maternity leave you counter bias with information. When you return from maternity leave, if you're still willing to travel, say so. And I highly recommend it, because then you can eat your own french fries, right? [audience laughter] If you're the primary earner, say so. If your partner is willing to follow you, this is really important in some corporate contexts, say so. Because if you don't say so, people will assume all the opposite, sometimes to your detriment. Also, if you want to cut back on your hours, the way you do it is you don't say, oh my God, I'm so overwhelmed, I'm just gonna die, please give me part-time. No. This is a business proposition. If you can show them how your department can function with you working fewer hours, they will give it to you. If you can't show them, it doesn't matter how much you need it. So the fact that you need it is not part of this discussion. The discussion goes as follows. I really need to cut back to 80% or whatever it is. This is how the department can function with me working in that way. Or I need to work from home on Fridays. I will be there if there's an important meeting, otherwise I will be home on Fridays, and this is exactly how it would work. It's very important to present a flexible work arrangement in that way. What if you are in a situation, and this is actually a direct quote, somebody said, "Oh yeah, my partner said this," and this is her business partner, "I don't know how you can work such long hours. "My wife could never leave her kids like that." The way you handle that is, if anybody's read Anna Karenina, I call this the Tolstoy was wrong strategy. Happy families are not all alike. You know, I'm sure that's right for you and your family, but I'm doing what's right for me and my family. And as always, watch the tone. [audience laughter] Practice it, you know, vent to your posse before you actually say it. Here's the last form of gender bias. I call it the tug of war. It's when gender bias against women fuels conflicts among women, you can see it was reported by 55% of our informants. First of all, it doesn't always happen. In our survey of women scientists, 75% said, women usually support each other in my environment. Now Black women were sharply less likely to say this, only 56% of Black women did. But the other three groups of women thought that the tug of war was relatively uncommon. But here's what it looks like. "Opportunities for women here are very zero-sum," said one scientist. "If one woman gets a prized position, another woman won't. "And so it breeds a sense of competition." This is often called the queen bee syndrome, as if this were the problem of some woman who's name began with B and it's not bee. [audience laughter] It's not a personality problem of an individual woman. This is a woman who is being politically savvy in the context of gender bias when she knows that other women are competing with her for the one woman spot. The studies show that women who have experienced discrimination early in their careers, and I guarantee you, a lot of women in computer science have probably experienced discrimination early in their careers, those women tend to distance themselves from other women. Again, why? That's only politically savvy. Marissa Mayer, when she was Google said, "I'm not a girl at Google, I'm a geek at Google." And you notice how she identifies herself with the in group, geeks, and distances herself from the out group, girls. This is an artifact of gender bias in the environment. Sometimes prove-it-again bias is passed through. "Females are harder on their female assistants, "more detail oriented, and they have to prove "themselves harder so they put that on you." This was a secretary talking. And that produces conflict between admins and professionals. Sometimes tightrope bias is passed through. "I've seen lots of senior women behave that way, not only working long hours, but adopting male mannerisms and being very aggressive." This is a woman saying another woman is too masculine. I'm gonna skip that one. And then finally, there are mommy wars where, typically this is women my age who look at women who are younger who says, I don't know why you need that long family leave, or flexible work arrangement, I worked full-time my whole career, and my kids are fine. So this is identity threat. All of these tug of war dynamics have to do with identity threat of like, I worked really long hours, but that's the only way to be a good computer scientist. So if you're not, that means I'm not a good mother. No, this isn't about you. This is about me, I just wanna take parental leave. But that's the way these dynamics sometimes arise. So if there are conflicts among women, ask yourself, are they a symptom of gender bias? It's also important to recognize the limits of sisterhood, after all, do men always support men? No, they're human, we don't call them Bs, we just call them people. [audience laughter] On the other hand, if someone is undercutting you, it's very important to put a stop to it, and there's a literal passage in the book that shows how you ask for a private meeting and you say, "My sense, and I may not be right, "is that you don't always see me in the best light. "It's my intention to support you and I "want to do that going forward, "so let's try to turn over a new leaf." Women my age need to remember that younger women's experience is simply quite different from ours. That's success. But on the other hand, younger women need to remember that women my age sometimes don't help you as much as you think we should. We don't have the power that you think we do. It's important to get women working together with other women, and not just on women's issues, because then you'll lose the Marissa Mayers of the world. With your admin, it's very important to realize that there are both gender and racial dynamics. Sometimes admins have a lot higher expectations of women professionals than they do of male professionals, and this produces some racial context, as well, where there's a tradition among White women of sharing personal troubles, and not among Black women. So that is another tug of war dynamic. So all of that are things that you can do as an individual woman, or things that you can do as a manager. Those are the individual interrupters. Here comes organizational change. This is from the hacking tech's diversity article. What you do is, and I'll just do this very quickly so we can have just a few minutes for questions, you basically take these four patterns and you do an assessment, asking the women, here are these patterns, any of that sound familiar? First step. Second step is they say, yeah, we have a huge office housework problem. Well, you may, you may not. Develop an objective metric and measure to see whether you do. Third, you introduce a bias interrupter. I'll talk more about that in a minute. And then, fourth, you go back to your objective metric and see if the office housework problem has improved and if it hasn't, you ratchet up to a stronger bias interrupter. What is a bias interrupter? A very gentle interrupter for an office housework problem, first of all, you count who's doing these assignments and secondly, you have a training where you say, women are doing them at this rate and men are doing them at this rate and that's kind of not a thing. And so we're gonna have a rotation from now on of these assignments, and for these assignments here's gonna be the process. That's a good example of a bias interrupter. You can ratchet up if the training doesn't work to a stronger interrupter that may include more formal assignment systems. That's actually what organizations should be doing in order to interrupt these patterns of bias. They need to be redesigning their basic business systems to interrupt the bias in real time. There's been a lot of excitement about that HBR article, but I can't tell you that since October the climate has changed, ladies. You can forget everything else I said. No, no, unfortunately, unfortunately these organizations change very, very, very slowly. That's why I think it's very important to have this advice, and I hope you find it useful. Thank you. [audience applause]

AUDIENCE MEMBERS: Thank you, hello? That was very interesting. So as a university administrator, one of the things I've noticed in some of my faculty is that men sort of do a lousy job at some of the housekeeping, and I think they do it intentionally, that way they don't have to do it. Whereas the women do a good job. So how do you deal with that? It's not like service is really counted for much in the university setting and so you don't have a lot of leverage.

DR. JOAN WILLIAMS: I think that the men, and I don't know whether this is conscious or unconscious, and I actually don't care, I think they do do a very poor job because they know that's a way to stop those office housework assignments coming around. I think you can do a couple things. Number one, you can establish a rotation so that everybody does it, and number two, you can say, this is important. If you are in charge of rank and tenure this year and the files are of very poor quality, there's gonna be consequences. I know that as a university administrator, it's not easy to give consequences to other professors. It's not easy. But it is possible. And professors do get consequences. Shaming is one. Having to do it again next year is another. [audience laughter] Talking to the university general council who this person has put at risk of defending a discrimination law suit is a third. There are consequences. But simply saying, having the work gravitate to the women because they're under subtle pressures to do it well, whereas the men are not, because they know they'll be given a pass, that is just, in the gentlest, kindest way, perpetuating gender bias.

AUDIENCE MEMBER: [faint voice, no microphone].

DR. JOAN WILLIAMS: I'll repeat it.

AUDIENCE MEMBER: I have the mic [laughs]. That's okay.

AUDIENCE MEMBER: [faint voice, no microphone].

DR. JOAN WILLIAMS: Oh, sorry. [audience laughter] Please excuse me.

AUDIENCE MEMBER: I’m sorry [laughs].

DR. JOAN WILLIAMS: You're gonna be bumped by the system. [audience laughter]

AUDIENCE MEMBER: Apparently I've been favored here.

DR JOAN WILLIAMS: [LAUGHS] I can hardly see you 'cause of the lights.

AUDIENCE MEMBER: I hope you can hear me. Thank you, this is very interesting. I had a couple questions. The most prominent one that came to my mind is, and I recognize that in many of these settings women are the minority, and so it makes sense to focus on how they experience their employment, but I wonder if you think it would be enlightening, and maybe there are other studies that do this, that look at how men experience the employment situations. Because, my suspicion is that there are plenty of men who would point to other comparable kinds of ways in which they get pigeon-holed and are held back relative to the ones who succeed. It might be an interesting counterpoint to consider that, because I think all of these are issues of poor management, more generally, in organizations, and maybe there are studies that focus on that. So that was one question, another which is related, I think, is that in some professions like law and medicine women have made much more progress than in tech, and I'm curious whether you see these bias issues playing out differently in those different sectors.

DR. JOAN WILLIAMS: Okay, on the first issue, you would use a different methodology and in fact I'm exploring a partnership with the Society for Women Engineers to do the kind of study that would answer the question you've posed. There is a study in law that, for example, asked women and men, people of color and Whites, do you think that you have experienced this kind of bias because of race, because of gender. The White men came out with like 2% and the women came out with like 30%. Not to say that if you're a White man it's instantly heaven. I mean, White men are disadvantaged by class, they're disadvantaged in some contexts by being introverts, society is very, very complicated. But there is a different study, it would be a different methodology. In tech, first of all, things are not that great in law. Only 17% of law firm partners are women, and we have been 50% of the incoming women lawyers for decades. That said, tech is worse, tech is worse. And it's worse in a lot of different ways. First of all, the kinds of sexual harassment, and you saw this in Gamer Gate, it's really bad, I admit, in gaming. The level of sexual harassment and the kinds of sexual harassment, the pervasiveness, is very, very different in tech than it is law, for example. Because there are so few women in tech, a lot of these problems become much more salient. Now you can talk.

AUDIENCE MEMBER: My early career as an HR director was in Europe, and there there's a lot better employment law framework to really protect everyone's employment rights, men and women. Do you see any correlation between our lack of some of those employment protection rights here in the U.S. around this equality piece, or maybe the way that some of this bias is showing up here in the U.S. Is there anything that.

DR. JOAN WILLIAMS: I actually disagree with your premise. I think that the work-family issues are much, much better in Europe. I think that the bias issues are much better in the United States. The United States is definitely the best place to be if you are a woman with a disposition that is masculine who wants to live the life pattern of a man. Best place to be is the U.S. [audience laughter] Some women are like that, right? Nothin' wrong with bein' like that. If you're a mother, the United States is the worst place to be, hands down, absolutely. So I think it's a little bit more complicated picture. In the U.S. we have much stronger employment law, anti-discrimination law. We have much weaker workers' rights, including work-family reconciliation. So that's the way the pattern is. I once wrote an article and said, "Want equality? "Die childless at 30." [audience laughter]

AUDIENCE MEMBER: Hi, so the thing that I hear most often from people that I work with, what they wanna know is, whether we can bring in some type of workshop on negotiation that is specifically for women. You gave some great examples about how to reframe how you negotiate. I was wondering if there are any good books or workshops or resources that touch on this.

DR. JOAN WILLIAMS: I have really mixed feelings about that way of framing the issue, because that implies that women aren't getting ahead 'cause they don't negotiate well. I think women aren't getting ahead because there's gender bias in hiring, in performance evaluations, in assignments, in compensation, that's why women aren't getting ahead. I really think it's time to stop bringing in these, let's fix the women workshops. [audience applause] That said, those workshops are, can be useful, for example, if you have a woman who's going like this and she's talking like this, you know, she's not gonna be very effective no matter what she's saying. It's just a fact. And so, if you're having a workshop to familiarize women with types of feminine traditions that tend to undercut women, I think that can be a very helpful workshop. It's probably best framed as, if you're shy and hate public speaking, this is a good workshop for you. The reason that workshops teach women how to negotiate, I really have a problem with, is because they tend to assume that women's problem is that they should negotiate hard, and in a direct, hard driving way, the way men do, and when women do that, they tend to trigger tightrope bias. The workshops for women, I think there's a place for them. The problem is that bias is being constantly transmitted in these basic business systems. If we don't want that to happen, we have to change the basic business systems. What I am doing now, and a number of other people are doing, are developing bias interrupters that, for example, you can use if you have an office housework problem. Or you can use, I mean, one of the recent studies of women in tech found that, looked at, compared performance evaluations in San Francisco, men and women, and found that women were wildly more likely to get negative comments and overwhelmingly more likely to get comments that attributed, that faulted them, for personality issues. Men almost never got those kinds of, that's a tightrope problem, right? You're not going to solve that problem by teaching the women to negotiate more. You're gonna solve those problems by having somebody read the performance evaluations who knows how to pick up that particular pattern of gender bias and others, and interrupts that pattern. And that's what we're trying to do with companies now.

AUDIENCE MEMBER: Over here to your right, thank you. I'd like to kinda follow up on that with something that's related. Some academic scientists recently came out, from Cornell psychologists, who are claiming from their study in some very prominent op eds, one, the New York Times, late last year that there is no sexism in academic science. And then just recently, a CCN op ed saying that there's a myth about women in science. So their study supposedly was specific to academic science, but I don't think it's that far removed from them saying that there's no sexism in, perhaps, industry. I'd like to know if you can comment on this, because I and some other people on the grassroots movements are trying to combat this and bring attention to this, because I feel that the science communications they've done is pretty egregious by overextending their analyses, but don't have the position that you do or the platform to perhaps give an academic response to the work of Ceci and Williams at Cornell.

DR. JOAN WILLIAMS: Ceci and Williams, I promise you I will write this. This is Ceci and Williams cottage industry showing that there's no bias against women in science. First of all, I just talked to 60 women of color in science, and 100% of them described to me patterns of gender bias that track exactly what the studies call gender bias. There are methodological problems with the Ceci and Williams study. Number one is, if you look at the prompt, they gave a woman who worked long hours and was extraordinarily qualified. Well, first of all, one of the things I edited out due to time is that women superstars actually tend to get a little higher evaluations than men who knew a woman could do it. The problem comes with the merely excellent, right? And that prompt was measuring the superstar. So all it proved is that Sheryl Sandberg can be a success. But Sheryl Sandberg could be God, I mean. [audience laughter] That doesn't prove anything. So that's one methodological problem. The other methodological problem is that was done by a natural scientist and a developmental psychologist, not someone who is not trained in studying gender bias or bias at all. What they did is, another methodological flaw, and there's a lot of them, I will write it, I promise you, is that it was very clear from the wording that this was a study of gender bias. So what it was an elegant study of is, if you go to scientists and tell them you're studying gender bias, whether they will tell you that they're gender biased. That is what the study said. They won't, they won't. [audience laughter] But that does not prove that in a context where what the people they're doing is actually choosing a candidate that these biases will not play out. There's a lot more, it gets very wonky. The fact that that is such a wild, gets that kind of wild coverage, the easiest way, and I spend half my life talking to reporters and then I will shut up, the easiest way to get covered in my business is to purvey stereotypes. And they do. Thank you. [audience applause] Thank you, oh, okay. [audience applause]

BOBBY SCHNABEL: I just wanted to thank you for what was both an interesting and very useful and helpful talk, and we have a very small gift for you.

DR. JOAN WILLIAMS: Thank you.

BOBBY SCHNABEL: Thank you very much. And I should also thank you, Joan [audience applause] for perhaps making me useful for the first time at an NCWIT meeting, because when you made the comment about old, balding, and White men, there was a whole segment of the room over there that immediately pointed to me. [audience laughter]