share interactive transcript request transcript/captions live captions
download
|
MyPlaylist
YAEL LEVITTE: Good afternoon. Good afternoon. Welcome to the sixth Robert L. Harris Jr ADVANCEments in Science Lecture. I'm Yael Levitte, Associate Vice Provost for Faculty Development and Diversity, and Executive Director for the CU ADVANCE Program, the program that has sponsored this talk.
For the past six years, CU Advanced program has sponsored the ADVANCEments in Science lectures, which brought nationally recognized speakers to Cornell to address campus and community audiences on gender and broader diversity issues in higher education. The CU Advanced program was established in 2006 with a $3.3 million Institutional Transformation Grant from the National Science Foundation. The grant's goals were to increase the recruitment, retention, and promotion into leadership positions of women in engineering and the sciences, and to institutionalize best practices to diversify our campus and improve its climate.
I'd like to recognize and thank first the two engineering faculty who wrote this grant, Professor Sheila Hemami. Are you here, Sheila? Not yet. And Professor Marjolein van der Meuleun, as well as Professor Kim Weeden, the third PI on the ADVANCE grant. Professor Weeden? Here she is. Professor Robert Harris, also one of the original PI and the person after whom this lecture is named, is also in the audience. And I'd like to recognize him. Where are you? Robert?
[APPLAUSE]
Next October, the NSF funded [INAUDIBLE] with successful outcomes in improving the recruitment and climate for women faculty across the university. In 2006, the PI set a goal of hiring 75 women in the sciences, and to increase the proportion of women in each department.
By the end of the grant, we've hired over 80 women in the sciences. In addition, Dr. Weeden, together with Dr. Marin Clarkberg, also in the audience I see, Director of Institutional Research-- have compared the faculty work-life surveys of 2005, before the grant, and 2008, and found that in a number of climate indicators, women who were unsatisfied or were less satisfied in 2005 and who took advantage of the ADVANCE program scored higher on the climate indicators.
Realizing that our work is not done but that we're in the right direction, Provost Fuchs institutionalized the program's efforts last year and established our office, the Office of Faculty Development and Diversity-- and more broadly, which he'll talk about in a second.
I'd like to invite now Provost Kent Fucks, who's also the principal investigator on the NSF-funded ADVANCE to join me in welcoming you to offer a few words about his and President [INAUDIBLE] institutional approach to campus diversity, and to introduce Dr. Steele. Thank you.
[APPLAUSE]
KENT FUCHS: Thank you, Yael. And thank you for your leadership as Executive Director of the ADVANCE program, and also now as an Associate Vice Provost. I want to also just begin by welcoming everyone. We've got a great audience here for a very, very special talk. Welcome to the Sixth Annual Robert L. Harris ADVANCEments in Science Lecture.
As Yael said, the ADVANCE program, for which she's been Executive Director in the faculty that led that are here in the audience, that program has now become part of something broader. It is now called-- it's part of an initiative that President Skorton announced last year called Toward New Destinations.
This initiative is part of a very structured program in which we're holding the senior leadership of the university-- the vice presidents, the vice provosts, and every one of the academic deans of colleges and schools both accountable but also letting them be creative in the ways that they enhance diversity in the programs and units that they hold. It's a new structure of goals, and as I said, accountability. It's also placing new emphasis on measuring results and holding those leaders accountable.
And under that initiative, it has the oversight of what we call the University Diversity Council, which was also reorganized a year ago. That council now has five diversity professionals, Yael being one of those, with four others. It also has myself, the president, the dean and the provost of the medical college, as well as a number of vice provosts and also vice presidents of the university are members of that council.
And the ADVANCE program, as I said, the initiatives launched over five years. The past five years are now a part of the work that Yael leads, as well as has oversight by the University Diversity Council.
Toward New Destinations actually has four areas of focus. And we've asked the senior leaders to develop initiatives that are specific and appropriate for the programs that they lead. Number one, it's in the area of composition-- specifically, the demographic makeup of a unit or the institution, broadly speaking.
The second is engagement, which reflects personal and social and professional involvement in the life and work of the university. Number three, it's inclusion. And that refers to the welcoming environment of our university for all parts of the community, including multicultural and interpersonal relations.
And lastly, achievement-- we want to have a university, an environment, where everyone can succeed at the highest level-- those from underrepresented groups and every one that is apart, whether you're a student, faculty member, staff, or even our alumni.
So today's speaker is a part of this initiative. Dr. Claude Steele is an expert on stereotype threat and its affect on groups broadly speaking, and including achievement, engagement, and inclusion in higher education. Dr. Steele comes to us today from Stanford, where he's the James Quillen Dean of the Graduate School of Education. I know him because he was also provost at Columbia University, and we were part of what's called the Ivy Plus Provost Group.
Dr. Steele is a leader in the field of social psychology. And I'm not going to reach you the fellowships and other publications that he has, but I'll just mention to you the National Academies he's been elected into as sort of a summary of his achievements. He's been elected to the National Academy of Sciences, the National Academy of Education, the American Academy of Arts and Sciences, and the American Philosophical Society. He's a member of the Board of the Social Science Research Council, and he's also on the John D. and Catherine T. MacArthur Foundation Board of Directors.
He's been at Stanford now-- he's had faculty positions in a number of places, but he's also now currently, as you know, at Stanford, where he has held previously appointments as a Lucy Stern Professor in Social Sciences. Has also served previously as the Director of the Center for Comparative Studies in Race and Ethnicity, and he's also been the Director for the Center for Advanced Study in Behavioral Sciences.
We use a book of his that will be the focus of the talk, called Whistling Vivaldi and Other Clues-- How Stereotypes Affect Us, and we're using this in different faculty groups. We're using it with some of the vice presidents, with staff, and different parts of the university with students. What I'd like you to do now is welcome Dr. Steele.
[APPLAUSE]
CLAUDE STEELE: Thank you. I can't thank you for that warm introduction. It's a real pleasure to be here. I an honor to give the Robert L. Har-- Pardon?
AUDIENCE: Harris.
CLAUDE STEELE: Harris. That's what I-- excuse me-- Robert L. Harris Lecture, and to also be just a part of your initiative, the university initiative that you're starting here. Just hearing a little bit about it as I just did from what Kent said, it really sounds like it's getting off to a great start, and it sounds like it has all the components of a robust initiative. So it's a pleasure to help launch that, so I'm grateful to be here for all those reasons.
My mission will just be to give you kind of an overview of our research. And hopefully you'll see some of the significance for the enterprises that you are about to undertake and have been undertaking. So think of it that way. I'll try to go for a while, and then preserve some time so that you can have some questions. We can have a kind of robust give-and-take. That might be the most fruitful thing to do.
So that will be my strategy. I will cover stereotype threat, social identity threat-- which is sort of a bigger canopy concept, but very similar-- and I will be making the case that these phenomenon, these social psychological phenomenon, are important and have big effects in all of our lives-- not just those of certain groups, but all of our lives-- in the sense that they affect big decisions that we make, like where we live and who we know and what careers we choose for ourselves and how we perform in certain kinds of situations. All of these things can be influenced in big ways at certain parts of our lives by these pressures, even though it's difficult to know that sometimes.
So I'll be making a case that these are powerful pressures-- not passing pressures, but powerful pressures. Still, I don't want to be depressing. So I don't want to leave the impression that these pressures are all determinative, that there are no ways around them and that people have never thought about ways of getting around them. They have. People do ingenious things to get around them. The title of the book, Whistling Vivaldi, is one ingenious strategy for getting around stereotype threat. So I'll hope to give you some good sense of that.
And also remedies-- what we can do in particular, both as institutions and as individuals, to get around these kinds of threats. I almost think of it as tool kits-- a tool kit for the institution, some strategies institutions can use, and tool kits for individuals to reduce the unwanted effects of these threats in our lives.
I think I can walk over here and be heard, so I'll do that. So that will be my general mission. I'd also like to give you, in honor of the science mission of this talk, some sense of the science of doing this research.
One of the big lessons for me in my role as a scientist is how humbling it is. You start out with a set of ideas, and almost every time when you put them to careful test, they're either wrong or very different than you thought they were. And that's certainly the case for the work on stereotype threat and what you're about to hear.
None of which you're about to hear are ideas that we had when we began this work. All of these ideas came from wrestling with a certain problem and then trying to come up with some explanation for it, putting that explanation to test, usually finding out that it was an inadequate explanation, and then trying to come up with a better one and put that to test. And that's the sequence of science, and that's what it can do for you.
It's the fun of it, once you get your ego under control. It's the fun of it. But it's also the way ideas-- I think I sort of trust them more when they come from this kind of process, and I want to convey that process to you as I talk today.
I won't make this a big scientific talk. My aim here will be to create a closer encounter with stereotype threat, to give you a phenomenological experience of it so that it's easier to recognize the experience of in your own lives. So that will be my strategy, a close encounter.
I'll begin with a description of the concrete world problem that we tried to figure out and that led to this work. Then I'll talk about stereotype threat a little bit to flesh out the phenomenon for you, give you some sense of what it is and how it's mediated. And then we'll get to remedies. And then at that point, I'll give a few, kind of get us started, and then we can open up into a broad discussion of what remedies are.
I'm sure you guys-- what is always the case is that I hear remedies I've never thought of before from audience members. So I'm looking forward to that part.
OK, well what's the problem that got us started? It's a very simple one, and it's a kind of mystery. It's this-- that for groups whose abilities are negatively stereotyped in the broader society, their abilities in some particular area are negatively stereotyped in the broader society, when they're performing really challenging work in that area, seem to underperform in relation to other groups, even when-- and this is the mysterious part-- even when they're equally prepared, have the same talents, the same preparation, the same motivation, and the like. Even when you equate for those things, there still seems to be this phenomenon of underperformance, as it's called.
I first discovered this many years ago-- I talk about this in the book-- at the University of Michigan, looking at grade point averages of Michigan students as a function of the SAT score they had when they came in. And what you see in general is that as kids come in with higher SAT scores, they tend to get someone higher grades at Michigan-- not in a really big way, but they tend to get somewhat higher grades. That's what you'd expect.
But another part of the graph broke out a line just for African-American students at the University of Michigan. And the surprising thing to me and the evidence that started all this work was the fact that at every level of entering SAT-- you think of that as a measure of preparation for college work-- every level of preparation for college work, the African-American students were getting lower grades at Michigan than were other students.
And that was mysterious to me. Why should that be the case? If the two groups, if you've got these two sets of kids and they're equally prepared and equally motivated, they're admitted to a great university like that, why should there be a difference in performance? What could be causing it?
I thought if you got preparation and skills and knowledge, all those things pretty much equal, then the game was over. There would be no differences. That's what it's all about. I thought that would be the end of it. But these data were puzzling to me, because they didn't show that.
And then within a brief time, we found that exactly the same thing occurred for women in advanced math courses at the University of Michigan. If you lined them up as to their SAT level-- especially in advanced math courses; not so much in entry-level math courses, but high-level math courses-- same SAT scores, same prior grades, same motivation, women were getting lower grades in those courses than men were. So again, why would that be the case, since their preparation, their skills, everything was essentially the same?
Then we had the sort of scientifically interesting-- maybe even exciting, but as a human being kind of depressing realization-- that this phenomenon of underperformance happens everywhere, certainly not just at the University of Michigan. It happens in almost any integrated classroom that you can find, almost at any level of schooling and at any place-- Harvard Medical School, Stanford Law School, the third grade in the school down the corner. It happens everywhere, this phenomenon of underperformance.
So it seemed then like this is an interesting and important kind of problem to understand. And we took out after it. I'll spare you the years sort of wandering in the wilderness, trying to come up with some kind of an explanation and then we did some very silly things. We did some things that we just thought were going to work, and probably five or six years later we started to get a picture that looks something like stereotype threat.
So sparing you all those missteps, that's what I'll say. Our argument, our explanation for this underperformance, is that in big part-- in big part-- we think it's due to stereotype threat. Maybe not entirely-- there are circumstances where I do think other things can be involved. But I think a lot of times, especially in the circumstances we are interested in, academic life, a lot of it has to do with stereotype threat.
Stereotype threat is a very simple kind of phenomenon. As I say many, many times, it's something that happens to everybody. There's nobody on earth that hasn't experienced stereotype threat. I think probably maybe a couple times a week, I'd say, I'd be really confident saying that. Sometimes I think it happens multiple times a day.
But it's this-- it's simply being in a situation or doing something for which a negative stereotype about one of your identities is relevant. When that happens and the thing you're doing is important to you-- it's important to your future, you love it, or it's important you, you care about it-- then the prospect of being judged by that negative stereotype that is relevant to you just because of your identity-- your identity brings online the relevance of this stereotype-- that prospect can be upsetting and distracting and can shadow you in that walk of life if you stay there. You have a sense of maybe discomfort.
That's probably how we experience it most frequently down on the ground, as sort of a sense-- sometimes vague, sometimes quite strong-- sense of discomfort. And you have a sense that it's going to be part of me being in this particular situation. In this particular situation, who I am brings online the stereotype that's relevant, and it could at any time potentiate a judgment of me or a treatment of me that I don't like and would make this walk of life difficult to be in. That's the definition of stereotype threat.
I think just thinking more simply, just you can imagine your life some circumstance where you really wanted to do well on something but the people that were watching your performance or keeping track of it somehow, you had a sense that they didn't believe you could do it or that you could be very good at it, or that you could really stay and achieve in that domain. That's what stereotype threat is. It's coming from this stereotype in a society about your group, and you have a sense that that stereotype about your group could lead people to have these feelings about you or expectations about you.
So that's kind of what it is. I think it's a pretty simple thing. It's probably a cousin of any kind of judgmental threat. Sometimes I give the example of-- this was sort of a family story of Aunt Ruby, who thought that our family showed off their culinary skills at Thanksgiving by always bringing some big, too generous, and too fancy a dish for Thanksgiving dinner. That was the allegation. And you knew that Aunt Ruby thought that about you.
So now, think about what's happening on Thanksgiving morning as you're preparing this dish. Aunt Ruby's like in the room with you. Every decision you make is like negotiating what she's going to think, and why does she think that? You're resisting it. You're [HYPERVENTILATES]. But there it is. You're contending with it.
Well, stereotype threat is like that, except it has some really dastardly features to it. I don't know another word to put it. But if the judgment is coming from a stereotype about your group, you know that not just Ruby has this view of you, but everybody in your world, in your environment, has this view of you or could have this view of you, because you know everybody in this environment knows the stereotype.
And when you're in a situation where the stereotype is relevant, they could be seeing you in terms of the stereotype. So when the judgment gets tied to a stereotype, it becomes a much more expansive possibility. That's the downside and the difference of stereotype threat per se.
Well, in trying to think of examples of this, I gave this talk with some of my former students last year. And they encouraged me to update some of mine demonstrations and the like. And so sort of trying to figure out-- let's come up with an example of stereotype threat that's really unusual. Like for example, where would a white male feel a stereotype threat about an intellectual performance? Where would that be? Can we think of a real example?
Well, we toyed with this for awhile and we came up with the idea of rapping.
[LAUGHTER]
Rap-- it's an African-American form. And there is in the culture a sort of sense that, well, white guys just can't really do that that well. The stereotype is white with a mic. And this led us to a very interesting clip, which I'll show you now, of the experience of the rapper Eminem, who is a great-- I mean, a really truly, even I who-- well, I won't date myself.
[LAUGHTER]
I'll just say I'm impressed with some of his lyrics. You'll see this clip I'm going to show you. We had to blip every other word to get it presentable to an audience. But it's kind of amazing use of language and so on.
So anyway, what you're going to see are the first few minutes of Eight Mile, which is his biography. And he is the actor in this, and this is a true story. Here the rapping situation, and why it's so difficult and challenging. Two rappers come onstage. It's called a battle. They're going to have a battle.
And they flip a coin, and one rapper goes first. And that rapper, what he has to do is think up the most horrendous, horrifying insults of the other guy that you could imagine, and he has to think them up on the spot. And he has to do it in time, and he has to do it in rhyme. And there's an audience there, pulsing and evaluating this rap as it goes on. So there's a lot of pressure there.
And then the poor other guy who lost the flip of the coin, he has to defend himself against all the allegations made by the first rapper. And he has to, again, make it all up on the spot, do it in time and do it in rhyme. So that's the pressure of the situation. And what you see in this video is Eminem's backstage experience just before he comes on, and then you see him come on and you see what happens.
One thing I want you to just pay attention to are the things that happened to him as he tries to get on stage, the cues that he experiences which send the message of the stereotype.
[VIDEO PLAYBACK]
-[INAUDIBLE] [END PLAYBACK]
CLAUDE STEELE: It's a battle. Well, you kind of get a sense of what that is. I wouldn't argue that the only pressure in that situation is stereotype threat, but it's probably a part of the pressure in that situation, maybe a significant part of the pressure.
And in an interesting way, the rest of the movie-- I'm sure many of you have seen it-- is kind of the story about how he overcomes it, and the experiences that eventually enable him to overcome it. And I think they too illustrate some of the things we'll talk about later as to remedies and strategies for coping with this kind of pressure. They're not all defeating kinds of pressures, but they are formidable barriers.
The question given the problem that we started out with, though, is do these stereotypes go to school? Does this kind of stereotype threat constitute a significant pressure in the academic performance of groups whose abilities, academic abilities, are negatively stereotyped? Could it be that powerful?
Well, there's another clip that I'm sure many of you have seen that I think illustrates this again in a kind of interesting way. It comes from the blue-eyed, brown-eyed experiment. You may have seen this in years past. It's over 40 years old this videotape.
It was made by Jane Elliott, who was a schoolteacher in Riceville, Iowa. And the day after Martin Luther King got assassinated, she wanted some demonstration in her class that would give kids a sense of what his life was about, of what the struggle was about.
And so she has a flash of an idea that what she's going to do is divide her class into blue-eyed kids and brown-eyed kids, and on alternate days she's going to stigmatize them based on their eye color. And she stays up late into the night making and ironing out collars to match the student's eye color so there can be no doubt as to what their eye color is.
And she goes in the next day, and she singles out the brown-eyed students and she has them put the collar on. And then she stands in front of the classroom and says that brown-eyed people are not good people. They're not intelligent people. They don't smell that good.
And she goes on and on like this. And I think they should sit in the back of the classroom. She has them stand up and move to the back of the classroom. And what you see in the ABC documentary is the reenactment of this situation. You could never do an experiment like this today.
[LAUGHTER]
This is unconscionable to do. But back there in those days, you could do some unconscionable but interesting experiments.
[LAUGHTER]
So you see how crushed these kids are by this treatment. And the only fairness in the whole event is that the next day, she turns the tables and she treats the blue-eyed students with exactly the same thing. They have to wear a collar. She says they're not good people, not intelligent people, they don't smell so hot. It's the same thing and the same kind of devastating reaction happens. To me, especially the point where kids with the collars on are sort of huddled in the corners during recess. I mean, they couldn't sort of socially function in the situation.
Well, if you're interested in what happens to these kids, it's all on YouTube. There's so much documentation about what happens here, and it's all kind of interesting. But the thing that interested me the most is something that didn't get paid a lot of attention to, is at a certain point in the documentary the kids are on the side of the room with her and they're working through a problem deck.
And she's keeping track of how fast the group of kids huddled around her at a table are working through the problem deck. And you can see there what effect having the collar on has on their cognitive scholastic performance in that situation. So I'll show you that little quick take, just to get a sense of it.
[VIDEO PLAYBACK]
-[INAUDIBLE]
[END PLAYBACK]
So collars and the kind of explicit treatment that she sets up there to implement the stigma-- one could say that's pretty heavy-handed. And still, the question could be open. Do identities-- just having an identity, walking into a classroom and having an identity that that has a negative stereotype attached to it-- could that alone be enough to affect cognitive performance in a situation, like this? This could be coming from the heavy-handedness of the demonstration, of the manipulation here, interesting though it is.
So you could ask that question. I think that for us is the next question. I just heard a talk a couple weeks ago that made me think maybe what Jane Elliott did in the classroom isn't so heavy-handed compared to reality. This was an anthropologist who'd done a classroom ethnography of ethnicities in junior high school in North Oakland. And she's looking at how kids use stereotypes.
And the first thing-- and this is a sort of classic California classroom, where you've got four groups in almost equal number-- whites, Asian, blacks, and Latinos. And she's looking at how they relate to each other.
And one thing that's very dramatic is that they like stereotypes. They think they're cool. It's almost as if the stereotypes are giving these young kids, who were seventh, eighth graders, their first sociological theory about how the world is organized and how it works. And they like it. They like using it, and they don't think of it as particularly bad.
And so the stereotypes, just the stereotypes that we know-- we could all probably write them down. What's the stereotype? The white and Asian kids are smart. The black kids-- they're not so smart, but they're really good athletes. And Latino kids-- yeah, but they're kind of gang-oriented. Those are the stereotypes. But they're using them in the mix, just as kind of everyday explanation of ongoing behavior.
And so her strategy of studying it is that she tries to observe when somebody behaves in a way against the stereotype, how to the kids explain that. So one day, one of the black guys gets the highest grade in the math test. And everybody's just so impressed with him. They really are pulling for him, and they say things like, man, that's fantastic. You must have some Asian in you, man.
[LAUGHTER]
So stereotypes are out there. And they're not so much coming necessarily from teachers or the school. They're just coming from every which way, and they are offered to people as ways of interpreting their experience. But at any rate, the question still is, does this make a difference?
So the first experiment we ever did, I think, had that kind of motive behind it. Suppose we just do the least possible. Could you get an effect like this? So what we did in that very first experiment-- now, when I say "first," this was the first experiment we did that took the form of a typical stereotype threat study. We had done oceans of other experiments looking at other ideas and so on. But this is the first one that tested something like stereotype threat.
We got women and men math students at Michigan-- sophomores, juniors, who were really, really, really, really good at math, as indicated by their SAT scores, by their prior grades, by their stated motivations about how important math was to them personally and to their career aspirations and all that. So these were top of the hill with regard to math skills and motivation. We brought them into the laboratory one at a time and gave them a very difficult math test, a half-hour section of the graduate record exam you take if you're a math major-- not the general quantitative section, but the math major section of the exam.
And our idea was that if all this reasoning is true-- is correct, is the better scientific way to put that-- then this would be a very different experience for a woman than it would be for a man. For a man, this experience would be frustrating. We'd set it up to be frustrating. And they could find out or worry that they're not as good at math as they thought they were. And that could be upsetting and distracting, and so on.
But for a woman, there would be this extra pressure in there-- that in addition to thinking that from the frustration, she might also think that, ah, jeez, am I sort of meeting this limitation and ability that the stereotype alleges? Am I meeting my sort of gender-based Waterloo with regard to this? Is Larry Summers right?
[LAUGHTER]
No, that's not true. He's an idiot. I hate him. So a lot of sort of rumination and cognition starts to happen to resist-- so this would be a very different situation for women than it would be for men, because it's hard. It's frustrating. That makes the stereotype about the group relevant as an interpretation of personal experience. That's a critical thing here. And that's the fact that's going to make this a different situation for women than it is for men. That was the idea.
We did the experiment, and sure enough, women performed like a full standard deviation worse than men. So we had completely bottled in the laboratory, recapitulated in the laboratory, what we found in the advanced math courses at Michigan, this mysterious underperformance of women in a circumstance.
Now, somebody could come out, somebody-- and there were plenty of them-- and say, well, you just supported Larry Summers, because Larry Summers said something-- the reason I use his name is because his quote was just so perfect for this. His quote was, the reason we don't have another women in our science faculty is, among a lot of things-- career issues and preferences, and so on-- among a lot of things, it could be that women just lack ability at the high end.
Well, we gave them a test at the high end and women didn't do as well as men. So who did we support, all the stereotype threat stuff or Larry Summers? And I can just tell you, there was a lot of tension in our laboratory in this era.
We know that we had to do an experiment to separate these two. It just was really hard to come up with a clear way of separating it. It took us a year, and finally we came up with a simple manipulation that would take the stereotype threat out of the situation, and then we could see what happened.
The manipulation was this. We redid the experiment, this time telling them just before they take the exam, look, you may have heard that women are not as good at math, especially difficult standardized math tests, that they don't do as well on difficult standardized math tests as men. You might have heard that. But that's not true for this test, not for the test you're taking today. This is a test on which women always do as well as men.
So the subtext here is any frustration you experience on this particular test cannot be taken as a confirmation of some gender-based limitation in math. It could be that you aren't as good at math as you thought, but that would be because of something about you as an individual, not because of you as a woman.
So with that little sentence, we changed the situation in one precise way, which is to just make the stereotype in our society about women's math ability irrelevant for these women to use in interpreting their experience in this test. So with that simple experiment sentence-- it sounds like, how could that make a difference? But it's making the stereotype out there in our society about women's math ability irrelevant as an interpretation of their performance on this particular test.
And as soon as we did that, to our great glee, women performed the same as equally skilled men in that situation. The difference went completely away. It was a revelation to us. We were grateful for it. I wouldn't be here if it weren't for that--
[LAUGHTER]
--finding. It was remarkable. We replicated it and replicated it. We'd take it to conferences, and people wouldn't believe it. But the difference went completely away.
Race studies-- same thing. I'll give you some other-- Josh Aronson and I did a number of them, but then some other people used the procedures. I just think it's a lot cleaner and has more generalizability to real life.
They gave black and white college students an IQ test, and it was a nonverbal IQ tests. It's a gold standard of IQ tests, the Raven's Progressive Matrices. Each item is a big square, and there's a pattern on the square. And then there's five little squares, and you have to pick which of those little squares has the same pattern as the pattern in the big square. That's it. So you're just matching patterns.
And it starts out pretty easy. You just are cooking along. You think, boy, I'm really smart. And then it starts to get really frustrating, and then it starts to get almost impossible. So frustration comes in. And you'll note, I'm emphasizing frustration because it's frustration that makes the stereotype about your group relevant as an interpretation of your own experience.
Without the frustration, in the easy math courses in college, with less frustration, if you're performing well within your own skills, there's no need for that stereotype to come online as an interpretation of what's happening to you, because you're not having any frustration. You're just doing fine. So that stereotype that says your group isn't that good at this is just irrelevant to the situation. So it's the frustration that is the critical thing here.
And sure enough, when nothing was said about this and they were just allowed to assume it was a test of cognitive abilities, or when they explicitly told them it was an IQ test, in either case black students performed a full standard deviation worse than white students, which is the exact size of the IQ difference between blacks and whites in the general population-- the exact size.
So do how you get stereotype threat out of this situation? Well, this is why I like this experiment. It has the nicety that these items look like you're playing a puzzle. It doesn't have to be read as an IQ test, as a test of cognitive abilities.
So in the other condition, the experimenters told the subjects, look, we're just working with a puzzle today. It's just a puzzle. It doesn't have anything to do with your abilities or how smart you or anything like this. It's just a puzzle. We just want you to have fun with it, do the best you can with it. Just play with it. Have fun with it. Boom, they go off doing that.
Well, , under that instruction, black students perform exactly the same as white students on that test-- exactly the same. These are the-- after so many years of trying to figure something out and then you start getting this really strong pattern of data, your conviction deepens that, oh, I think there is something here.
And I also think that the stereotype threat is a big enough and powerful enough a force that it can affect things, like really important cognitive performances. I might have, in my other frame of mind, said, look, when you've got a test that's as important as an IQ test or a test in school, people are just going to power through that. This kind of ephemeral thing of this relevance to the stereotype is just not going to be enough to repress people, because it's so important to them they're just going to motivate themselves through it.
But as it turned out, that explanation is both right and wrong. The underperformance is caused by people pushing themselves through it, trying to beat the stereotype. I have a chapter of the book that's sort of my favorite chapter. It's called "The Efforting Life." Being under stereotype threat in a domain where you really want to achieve in it and you're invested in it, you're always trying to beat that stereotype on top of doing the thing that's in the domain.
And that's the problem here in these examples I give you. I think the simplest metaphor to use in trying to understand the experience of stereotype threat and how it interferes with performance is multitasking. You're doing two things at once. You're trying to take the test and figure that out, and at the same time another part of your brain is trying to resist the stereotype and disprove the stereotype and counter-argue the stereotype and say it's not true and prove that it's not true. But then there's another frustration, and that raises it again.
So you've got a lot of stuff going on at the same time that the clock is ticking and you just don't have the same attentional resources to pay to the manifest task at hand of doing the test. That's how I think it works.
We know an awful lot now about how this phenomenon is mediated. It's mediated pretty much as I've said. There's a kind of division of attention between the task and the business of refuting and disproving the stereotype. That's what the person is trying very hard to do, is refute this stereotype.
This pressured multitasking is reflected in all kinds of physiological symptomatology-- accelerated heart rate, blood pressure, suppression of the prefrontal cortex, excitement of the amygdala, which is a part of the brain much more vigilant to threat. So you get a very vivid pattern of physiological reaction among people who are experiencing stereotype threat.
But interestingly-- interestingly-- they don't self-report that. They don't say it. They don't seem to realize it. I remember the first studies we did, we would see that this kid had just screwed up the test and that they were really trying hard.
And we'd go in and we'd talk to them, and they say, no, no. I know how to deal with these situations. I just redouble my efforts, and I just put my energies in there. They didn't have a subjective sense of it, even though you could tell from the physiological arousal that they were very aroused in this situation.
So there's that kind of a feature to the threat. We don't enjoy a great deal of access, I guess, or ability to cognitively recognize that we're under this threat when we are in fact under it. It's undermining our performance. It's causing a big physiological reaction. But subjectively, we don't have a sense that it's there.
Well, another feature of it that is important and that I took another long time to bring into view-- but I try to focus the book more on this part of it-- is that stereotype threat is a contingency of identity. It's something that goes with your identity in specific situations, and you can't just simply disprove it once and get rid of it. It's a feature tied to your identity in certain situations.
I'll talk in a minute about thinking about how history visits our present lives through the forms of stereotype threat it imposes on all of us. But one critical aspect of that is to recognize that it's tied to our identity. It goes with our identity. I'll play you a little tape here.
This is an interview that Bill Maher is doing of Charles Blow. Charles Blow is an African-American editorialist for The New York Times, does their graphic editorial on Saturdays. And he's talking about his reaction to the Trayvon Martin circumstance. And you get some sense of how this kind of threat is a chronic feature of life.
[VIDEO PLAYBACK]
-[INAUDIBLE] and it is a very heavy weight on their shoulders all the time, because you cannot behave like all of the rest of your friends. It's almost exhausting--
-You can't wear a hoodie.
-Well, not only can you not wear a hoodie, but you literally are constantly paying attention to the way you move your body. It's a physically exhausting exercise to know, I can't run or make a sudden movement, because now I see some cop standing in the corner and he's looking this way. Or I can't put my hands in my waistband, because George Zimmerman says that Trayvon Martin had his hands in his waistband, and that may signal that I have some sort of weapon.
The kind of constant having to think about how you are positioning your body so that no one takes advantage of that and views it as suspicious and maybe takes your life or takes you to jail.
[END PLAYBACK]
CLAUDE STEELE: Well, keep that in mind and then think about this experiment. We bring into the laboratory white guys one at a time, and we let them understand that they're going to be a conversation with two other students, and the experiment is about a conversation. And then they see the two pictures of the two people they're going to be in conversation with.
And in one condition, one group, the two pictures are of two black guys. And in the other group, the two pictures are of two white guys. And then they find out that they're going to either talk about love and relationships, which people can talk pretty easily about with almost anybody apparently, or they're going to talk about racial profiling.
So then the experimenter says you're going to talk to two white guys, two black eyes, either about racial profiling or about love and relationships. The experimenter then says, look, I'm going to go down in the hall and I'm going to get your two conversation partners and bring them back. And by the way, would you just hang around for a minute and arrange the three chairs in the room for that conversation? Would you just do that while I'm gone?
And as you might imagine, when they do that, the experiment's over. You're interested to see, how do they arrange the chairs as a function of what conversation they expect to have? And you can probably predict the results.
When they're going to talk to two white guys about anything, the three chairs are very close together. And when they're going to talk to the two black guys about love and relationships, the three chairs are very close together. But when they're going to talk to two black guys about racial profiling-- these are two strangers, black guys, about racial profiling-- they put the two black guys over here and they put themselves over here. They put a distance between themselves here.
And the interesting thing about it-- we measure what's on the top of their brains, so to speak, by a series of instruments that are a lot like Rorschach tests. And what's on the top of their brains when they're in that condition-- going to talk to two black guys about racial profiling-- is, I don't want to be seen as a racist. I don't want to be seen as a racist. I don't know these guys.
That's the stereotype threat that the white participants experience in this interracial interaction, the threat of being seen in terms of the overarching stereotypes about whites, that maybe they could see me as racist or racially insensitive or unknowing in some important ways, and that would be just terrible.
And also interesting-- we also had their levels of prejudice mentioned in several ways. This is not a highly-- these are Stanford students. They don't have a lot of highly prejudiced people, but they do have a range of prejudice there. And the interesting thing is, who do you think in that critical condition, where a white guy's going to talk to two black guys about racial profiling, who kept their chairs farthest away? Was it the least prejudice or the most prejudice? It's the least prejudiced.
Who shows the biggest math impairment under stereotype threat, the women who are the strongest and most committed to math or the women who are less committed to math? It's the ones who are the strongest and the most committed to math that show the biggest effect. Which African-Americans show-- same.
In all of these stereotype threat paradigms, the members of the group that show the biggest impact of possibly being stereotyped in a way that you don't want to be based on one of your identities are the people who care the most about performing well in the domain where the stereotype applies. That's the tragedy of, say, Eminem, is that he was foolhardy enough to fully identify with a domain of performance in which his particular group is negatively stereotyped. That's what makes him vulnerable to the pressure of stereotype threat. It's not that he has low self-esteem or that he doesn't have the skills.
As a psychologist, I can tell you we looked at all those possibilities first. I thought those were just things you take off the shelf and look at. But none of those things work. Every time we looked, it was the members of the group who were the strongest in these ways that were showing the effects.
And I think this experiment I just described about white stereotype really illustrates the same principle, that it's the group that cares the most about not being seen as racist that is distancing themselves the most under the possibility to deflect the threat the most. Stay away from it.
So you've got Charles Blow, and you've got this experiment. If that is not American history visiting us in contemporary life, I don't know what is. This is our history, and this is how these kinds of mechanisms, these stereotypes, bring that history from the past right into our relationships with each other. And they do it through this particular mechanism.
And we're oftentimes not looking at this mechanism. We are using a completely different paradigm for understanding it. And that paradigm is that the big problem between groups in the United States is prejudice-- prejudice, direct disliking. What you can see in that experiment I just described, it wasn't prejudice that put the distance between the participant and the two conversation partners. It wasn't prejudice. It was the fear, the worry, about being seen in terms of their historical stereotype, that did that.
Now, I would argue-- I was just saying this in a session earlier today-- just for the sake of being clear and just to push the point really far-- farther than I actually believe, but just for the sake of clarity, I'll say this phenomenon is more important in our daily lives than this phenomenon. This apprehension about doing something that's going to cause us to be seen in a particular way in terms of some stereotype about our identity is the more powerful struggle in American life today than it is-- that that's really a strong statement; I recognize that-- but that this particular pressure is stronger than this worry about prejudice and direct behaving on the basis of prejudice.
What makes this pressure strong? What makes it weak? If we can answer that question, we can begin to get to remedies. Quick story-- Sandra Day O'Connor, interviewed by Nina Totenberg on the radio about her autobiography. Totenberg says, well, what was it like to be the first woman on the Supreme Court?
O'Connor says, it was asphyxiating. It was just asphyxiating. Everywhere I went, the press followed me. They talked about my intelligence. They talked about my wisdom. Was I feminist? Was I not feminist enough? They followed me in restaurants, up the steps of my house. It was asphyxiating. It was horrible. Every time a decision happened, it was reviewed ruthlessly in the papers. And I just felt suffocated.
Well, Totenberg says, what happened when Ruth Bader Ginsburg got there? She says, well, everything changed. Everything changed. I just stopped getting all that kind of attention. There were two women on the Supreme Court now. It just didn't make sense for the press to ask all those same kinds of questions about two women, especially women who were ideologically different. Didn't seem to make sense. And so pretty soon, that attention went away and we were just like two normal women on the Supreme Court.
So as I say in the book, I heard about this about a month before the decision on affirmative action in 2003 was going to be announced. And I thought I knew how that decision was going to go, because Sandra Day O'Connor was the deciding justice in that. All the other justices were equally divided. She was the swing vote.
And I thought as soon as I heard that interview that I knew how it was going to go, because Michigan's defense of affirmative action, the University of Michigan's defense, was that it was important to have critical mass, that you couldn't expect minorities and women in certain areas to be just a tiny minority, because that would be so much identity pressure that it would distract from their functioning in school. And that's why we needed a policy of affirmative action.
Now, to a lot of people, as we all know, that rationale doesn't have a lot of traction. But I knew it would have some traction for Sandra Day O'Connor, because she had been on the Supreme Court without critical mass, and she'd been on the Supreme Court with critical mass. As soon as Ruth Bader Ginsberg comes on the Supreme Court, she has critical mass. Her identity as a woman is no longer the sole focus of attention. It's such a big, reified thing in the environment and in her life.
So she could understand this. She had an experience in life which would enable her to understand that. Well, that's the kind of thing that makes these pressures of stereotype threat stronger or weaker, are the cues in the situation. Think of that as a cue in a situation that signals to you how significant this identity is going to be in the situation. If you're the only identity in this situation-- the only old guy, in some situations I've been in recently-- you kind of start to count.
And you say, well, why am I counting? You're trying to figure out whether this identity makes a difference or not. So the circumstances around which the cues in the situation are the source of this threat-- that's what we sort of shifted to in trying to understand the kind of factors that make this thing weak or not.
Think about how diversity is presented. If you present diversity as something that is essentially in competition with excellence, How are you going to feel if you're a minority in a situation? Were you going to feel like my identity, who I am, is not really bringing anything to the table. In fact, I could be detracting from it-- or at least, I could be seen as detracting from it-- because people think excellence is something over here and diversity is something over there. And in fact, maybe there's a zero sum game. Maybe the more we do for diversity, the less we're going to take away from excellence, and we're going to have that kind of competition.
Well, if that becomes the kind of cultural framework that people use to think about the issues and talk about the issues, then you could see that that's going to increase the sense of identity threat in a situation. The intentions may be quite the contrary, but the fact of it is going to be to increase the sense of threat. So that's another feature of a situation, cue in a situation, like critical mass, that can have an effect of amplifying a threat.
A much better approach is to say that diversity is essential to excellence-- that you really cannot have even, in the most rarefied of disciplines-- excellence if you have just too much homogeneity, that you need diversity of perspectives to have an excellent enterprise. I believe that.
That's where we as a society have always had a kind of secret weapon, is that we're such a diverse society and we've had so many perspectives to come to bear on so many of the problems, from the mundane and concrete to the political and philosophical that society faces. We've had so many perspectives to draw from that we benefit from that diversity.
But we haven't acknowledged it as such. That framework about diversity, that it really is-- this is the Scott Page argument, that adding new perspectives to an area is as important to the excellence of an area as adding increments in skill.
I see this in psychology. When I came into social psychology, it was a male-- there was hardly any women at all, and we were going to mathematicize all of psychology. We all remembered our high school algebra, and we were going to describe all these processes in terms of mathematical models. And if you had a process that couldn't be captured in an algebraic model, you just thought it wasn't scientific. You just ruled it out of order.
And our field was really very narrow for that reason. And women begin to come into the field, and they had a kind of impatience with this. They felt that the field could offer a lot, but that it was just too narrow. And they brought in other topics, and other things started to happen. And topic-wise, methodology-wise-- in a whole variety of ways-- the field got broader and broader and broader.
So with that conception of diversity-- I can go on in this way. I want to sort of bring this to a close here. But presented that way, one realizes one's identity contributes to a situation, as opposed to being a detriment to a situation. And that, again, reduces the threats.
Well, I could go in this vein. The important thing to stress is that I don't think stereotype threat reflects an internalized vulnerability. For awhile I did think that, but the data just didn't support it in just the ways I've just described. I think it does come from the features of a situation, the degree to which they raise the possibility that based on my identity, I could be seen stereotypically in a situation. How significant is this identity in a situation, and what kind of image does it have in a situation? The cues tell me this in a situation.
And so when we think about remedying a situation and you think institutionally, I think you want to think about cues and how an institution presents itself and uses many degrees of freedom, as many opportunities as you can think of in a situation, to change those cues around so that they don't send that kind of a signal to people.
I just talked about how diversity is represented. That's a big deal. That's something that institutions have some control over. And the whole rationale for diversity, I think, can be very different than the most typical one that institutions use. That's an example.
I do think role models, existence proofs, are critical. Being in a country as an African-American where Obama is a president, where somebody of your identity can be the president, is very different than being in the country where somebody of your identity cannot be the president. It just makes a huge difference.
It doesn't turn over the whole racial organization, historically-produced racial organization of society. It doesn't do that, but it does represent something significant. I think that's another kind of thing that people can do.
I'm a little worried about going over my time here. How much time do I have? Am I over my time? I have zero.
[LAUGHTER]
-I'd like to leave some time for questions.
CLAUDE STEELE: Yeah. Well, let me just say, if you ask me questions about how individuals can cope with stereotype threat, I might be able to give an answer. But I will stop at this point. So you guys can--
[APPLAUSE]
YAEL LEVITTE: We'd like to open it to questions. And I will repeat the question in case the audience can't hear. Kate?
-Could you speak directly to the experience of teachers, [INAUDIBLE]?
YAEL LEVITTE: So the question was experience of the teachers, and how can they mitigate the stereotype threat?
CLAUDE STEELE: Well, one thing I can say is that my wife has just finished a book on that. It's called Identity Safety, and it is a kind of compendium of strategies. She's focusing especially on K through 12 teachers. And so this will be coming out in September, I think-- Dorothy Steele.
And in there, there are a variety of things. We were talking about this earlier today, and there are a variety of particular strategies. I'll try to give a big summary construct, which is that trust-- and I think a lot of teachers do this really well anyway. One of the remarkable things in that videotape about Jane Elliott is how much the students trusted her, even though they'd just gone through this kind of experience. They really liked that woman.
And students have that capacity to really like and love their teachers. And I think many teachers are very good at understanding that that is a critical part of teaching, building a trust between they and their students-- and to recognize that these identity differences, especially as kids get older, have to be dealt with, that they have the potential of interfering trust, and that ways of assuring people that you understand their circumstances, that you empathize with their circumstances, and that you think a good deal of them and expect a good deal of them. It's sort of a combination of high demands, this is serious, this work is valuable.
I think that has to be projected. But I think you can achieve this. That combination tells a student that you're not diminishing because of the stereotype. You're expecting a lot out of them, and you think that they can meet those standards. That's been something that in our research is a very amazingly powerful strategy in these circumstances.
I don't think that it requires a huge amount of-- this can sound maybe, I don't know what, bad, to say this. But I don't think it requires a lot of sophistication about race and so on. I remember my own advisor-- I used the example of this in the book-- who was very effective in my case. I think I was suffering an extreme case of stereotype threat.
But he just related to me very directly, but he related to me in that I knew he believed in me. He took me seriously. He demanded a lot. He cared about what I thought. He'd walk down the hall, stick his head in the office, and say, what do you think about this? And that just told me everything. He thinks that what I think about this could be-- that's true for any student. But given the kind of extra stresses I felt in that situation, it was a particularly powerful thing to do.
And so I think teachers have a good, powerful repertoire already in hand that can be very effective in these situations. And there are probably many teachers that know that, that have had some success there using these kinds of strategies.
YAEL LEVITTE: Any other questions? Yes, the second row there?
AUDIENCE: So you spoke about stereotype threat in terms of race and then in terms of gender. I'd like to know your thoughts on the intersections of race and gender, and how stereotype threat affects women of color specifically.
YAEL LEVITTE: The question was about the intersection of race and gender stereotype.
CLAUDE STEELE: Yeah. There has been interest in the research literature. Is there a double whammy there so that go this far down with one, and then that far? It's hard to answer that in a definitive way.
It doesn't seem to be quite a double whammy, but there does seem to be a slightly additive effect. So yes, when you intersect identities like that, there does seem to be that kind of an impact. But the good news is-- or at least, good news in certain circumstances-- is that oftentimes, making oneself aware of the positive side of the identity can reduce that effect.
For example, we just stumbled on this finding years ago replicating those experiments on women in math. And when we would remind women that they were Stanford students, the effect got a lot smaller. And then when we reminded women of other women that were really successful, the effect got a lot smaller. So a lot of it depends on kind of reminding oneself of the positive side of the identity. That can be very effective in reducing that.
YAEL LEVITTE: Back there?
AUDIENCE: You mentioned how stereotype threat can affect students at college level. Do you think that stereotype threat affects [INAUDIBLE]?
YAEL LEVITTE: The question was whether the same stereotype threats are applicable in the high school level, and not only in the college level?
CLAUDE STEELE: Yes, I do. The evidence seems to support that. The younger you go, I think the evidence gets more ambiguous. It takes a more powerful circumstance to bring this online for younger kids. Although, I do think it happens in very, very young kids sometimes, because even you get these doll study findings.
And that happens in kids who are as young as four years of age, who somehow know-- I have two three-year-old, almost four-year-old grandsons. And they already have a sense of how-- they just pick up these little jokes about chocolate and so on. So they have a sense very early that their group is seen a certain way, and what's cool and what's not cool.
So the doll study findings, I think, do tell us-- I don't think they tell us what we originally thought they told us, that there's some early internalization of these negative stereotypes. I just think it's a picking up of the social environment. And so I think that's where stereotype threat lives, in that social environment. And it can happen very early.
YAEL LEVITTE: Back there, and then I'll go to this. And I'll take three more questions. We have a reception afterwards. You can engage with Dr. Steele over hor d'oeuvres. So back there?
-I was making a connection between what you said about the physiological effects during the tests that you did [INAUDIBLE]?
YAEL LEVITTE: So the question was about the health effect of the physical pressures of stereotype threat on individuals who are experiencing-- so the heightened level of blood pressure, the stress, and how it affects the general health, right?
CLAUDE STEELE: Yeah. I'd say there's only conjecture at this point. I don't know of a direct-- there's the literature that shows the effects of stereotype threat on physiological reactivity, and it also shows the fact that people don't enjoy much access to that. So it has the same kind of subliminal features that hypertension has. But I haven't seen a real empirical connection between the two.
I can conjecture that one might be there. I do in the book review the evidence of race and social class on hypertension. And so I raise that possibility, but I don't know if it's been formally tested yet. But it's a good guess. Right there?
AUDIENCE: So a couple years ago at an academic conference, a social psychologist, Jon Haidt, famously argued that some of these same processes happen for political conservatives in our field and in academic instutitions of higher learning more generally, and sort of said that we needed systemic solutions to this problem. And as sort of a liberal problem child that he's talking about, I struggled with this because I understand that a diversity of perspectives is important. But what about when that perspective is that a diversity of perspectives isn't important? So I'm wondering what your take on that is.
YAEL LEVITTE: Do I need to repeat the question? OK. So the gentleman here quoted Jon Hadit, who said that the experience of stereotype threat also occurs in academia for conservatives. And I think you identify yourself as a liberal. He's trying to think how can you systemically approach that, especially when diversity is-- can you help me repeat this?
[LAUGHTER]
CLAUDE STEELE: Well, that part I lost a little bit. I got the at thrust of the question. First, I agree probably in most social sciences, if you're conservative among social scientists, you're going to feel a lot of stereotype threat, because most social scientists are not conservative. And so if you express that kind of a view, you would be very right to assume that you could be judged in terms of a negative stereotype about conservatives.
And I'm surprised Jon Haidt gave us that much credit, or at least characterized the problem that way, because I do think that's a good way to characterize the problem, is that is a form a stereotype threat that when you want a social science to have a full reach of perspectives, this form of stereotype threat is going to make it harder to happen.
But then I missed--
AUDIENCE: Well, do you think the same sort of solutions would apply to fixing this?
CLAUDE STEELE: Oh, yeah.
YAEL LEVITTE: Do you think the same kind of solutions will apply to fixing--?
CLAUDE STEELE: I would hope so. I would hope so. I would in that instance-- one of the things that is a kind of a chicken soup for stereotype threat as an individual strategy for reducing stereotype threat is to think of whatever it is that is alleged in the stereotype as an expandable thing. This is sort of a Carol Dweck import of that idea into this literature as a remedy for stereotype threat.
But if you think of intelligence, for example, or math ability, let's say, as expandable-- it's not something that is limited by a kind of genetic capacity that you're sort of given. That's an American ideology. I'll use that word. That's an American ideology. It's not all over the world, but in America and in some other Western societies we tend to think of ability that way.
But if you think availability as something-- a lot of Asian societies think of math ability as hard work, man. They're not focused so much on a limited capacity. How much did you work? How organized are you? So that's a very different underlying assumption to have. And if you have that underlying assumption, being in a group whose ability isn't seen as good just doesn't have the same meaning. It doesn't have the same consequence. I could get good. Maybe I haven't decided to get good. But I could get good. So it reduces the impact a little.
So I don't know where I'm going with regard to conservatives here.
[LAUGHTER]
But maybe they could think of themselves as open to influence. I don't know what the heck they can do. I feel for them.
YAEL LEVITTE: I'll take the last question there, in far right. And then we'll conclude. Yeah?
AUDIENCE: [INAUDIBLE]
YAEL LEVITTE: So the question was about computer science here at Cornell. Retention of women students in computer science is a challenge, and whether there's a way to test for that earlier on.
CLAUDE STEELE: Yeah, there is a good deal of stereotype threat research on women in computer science in particular. Sapna Cheryan is a social psychologist at the University of Washington in Seattle. And she has these geek studies where she gives them coding tasks to do, and the room is full of sort of geek signs or it's not full of geek signs. And she shows performance differences and motivation to go with this work as a career, differences as a function of that. So the experimental evidence is very suggestive of something like this.
One fact I always remember about that is that the verbal score on the SAT is the best predictor of computer science, better predictor than the math score. And women have better verbal scores than do men.
But I do think, even in the 12-year-olds-- I wish this weren't the case-- but I think even among 12-year-olds, they begin to sense that this is more of a boy's domain. And there have been differences in play habits and things that people-- there's a lot of socialization that's going on to kind of genderize these domains. I don't know if at 12 years old it's stereotype threat that's driving that, or maybe just straight ahead gender preferences. That could be a compelling factor there.
But when you get to college, and if they've been interested that long and you start to see a turning away from it, stereotype threat starts to be a candidate. Stereotype threat is a candidate for white males in computer science at Stanford, because most of the students-- I mean, I've interviewed this and looked into this pretty closely. Most of the students are Asian, Southeast Asian.
I can remember a sort of amazingly poignant story of interviewing white males who left, dropped out of the course. I had their academic records. I knew that they really good. And one typical thing they would say is that, well, I went to the course, and I'm in a big university now. I was good in this is high school and so forth, but I just don't know if I want to work that hard. So that's how they experienced the fact of being an identity minority in this situation, given the reputational differences.
We've done stereotype threat experiments where you give-- this is maybe more of a West Coast kind of thing than it is here. I just don't know. But we've done things at Stanford where we've given graduate students in engineering a really difficult math test. And to create stereotype threat for white engineering students, we would say just before the test, look, this is a good test. We want you to do the best you can on this. This is a test on which Asians tend to do better than whites. Here, do the best you can.
[LAUGHTER]
So there is out there this stereotype, that this is an Asian domain. It's increasingly seen that way in the high schools in San Francisco and a little bit in Stanford and Berkeley. And so then you get another group that you'd never think of it kind of withdrawing interest in that. So I think these kind of processes do-- sorting out all the causes. It's not the only cause, but it could be in some situations a significant one.
YAEL LEVITTE: Well, please join me in thanking Dr. Steele for a fascinating talk.
CLAUDE STEELE: Thank you.
[APPLAUSE]
Claude M. Steele, internationally renowned social scientist and Dean of the Graduate School of Education at Stanford, delivered the Robert L. Harris, Jr. ADVANCEments in Science Lecture on April 11, 2013. He discusses his theory of stereotype threat, which has been the focus of much of his research and writing throughout his academic career.
The theory examines how people from different groups, being threatened by different stereotypes, can have quite different experiences in the same situation. It has also been used to understand group differences in performance ranging from the intellectual to the athletic. Steele's recent book, Whistling Vivaldi: And Other Clues to How Stereotypes Affect Us and what we Can Do, published in 2010, was based on this research and lays out a plan to mitigate the negative effects of "stereotype threat."
