Implicit gender bias

An interview with Toni Schmader

Books and Ideas: Despite efforts at the organizational level, gender biases continue to persist. How prevalent are these biases and why are they difficult to change? Why are the main reasons for this gap between active efforts to increase gender equality and persistent disparities?

Toni Schmader: I do a lot of my research looking at the experience of women working in highly male-dominated fields like computer science, some of the physical sciences, and engineering. And in these fields, in particular, we often think about disparities or bias as coming from explicit forms of discrimination or sexual harassment.

But what often gets missed is the degree to which there are more subtle biases, implicit biases that also persist and in some ways are much more pernicious, because what I mean by implicit biases or stereotypes is an automatic tendency to associate fields like engineering, let’s say, more with men than with women, and simply having that cognitive association can mean that people find it more surprising to see women working in these fields.

They find it more difficult to imagine that they can rise to the same levels of professional success, and those types of implicit stereotypes can undermine women’s ability to be as successful as they might be. And it’s much harder to change those types of stereotypes. In one sense, these kinds of implicit associations form because we do live in a world where we see men and women working in different kinds of occupations and taking on different kinds of roles.

And so it’s easy for our brains to, almost in a machine learning way, just automatically pick up on these associations, and then to assume that because this is the way things are, that it’s also the way things should be. And so you actually need to take active efforts to try to circumvent the degree to which those implicit associations play a role in shaping how you’re evaluating your colleagues and even how you’re seeing your own role as a woman working in these fields.

So, for example, when I talk about these implicit associations in one of the recent studies that we published a couple of years ago, we were able to measure the degree to which men and women had this automatic tendency. When they think about science or engineering, what comes to mind automatically is the concept of men or the concept of women.

We did this in a sample of over 1250 people working in STEM fields. And one of the interesting findings from that is that the degree to which men in particular in these organizations had these automatic stereotypes, these implicit stereotypes, they also reported being less likely to include their female colleagues in their social networks. That meant women were being excluded from these more informal conversations that they might be able to have with their male colleagues.

And when women reported feeling excluded from these social networks, they also reported feeling less engaged at work, higher levels of burnout, and also just higher awareness of gender while at work as well. And so we can see that these kinds of implicit stereotypes might be associated with meaningful professional outcomes for women in these fields.

So, one of the other questions is: to what degree or why is it so difficult to change these kinds of implicit biases and reduce gender disparities? And one of the reasons is that the very people who believe that disparities that exist are due to just the way men and women are, or due to the choices that men and women are taking, are the ones who are most likely to kind of create biased opportunities for men and women and to use bias decision making in their selections.

So, for example, in a unique study, I had the opportunity to collaborate with colleagues in France to look at the evaluation process that happens at the CNRS program for elite positions in science. And in this study, we were able to survey evaluators who make these selections of men and women into elite research positions and look at the degree to which their implicit associations that tendency to when I think about science, do I think about men more easily than women and their explicit beliefs to what degree do these predict the kinds of decisions that these adjudication committees were making?

This research really asks the question: To what degree do these stereotypes predict more biased selections against women? And the answer is that it was complicated because it really depends. And it depends on the degree to which members of those committees thought that the existing gender disparities in their academic discipline were fair or based on systemic disadvantages that women are facing.

The ones who thought the existing disparities were fair were more likely to show that their cognitive stereotypes predicted making the selection decisions that disfavored women’s outcomes. The ones who thought that there were still barriers that women face were more likely to check their biases, were more likely to, it seemed like, break the link between simply having these stereotypes in their minds and having them be associated with their final selection decisions.

Books and Ideas: What socio-cultural factors might explain a weaker attraction to science, technology, engineering, and mathematics disciplines by women?

Toni Schmader: I’ve already talked about these implicit associations, the tendency to associate science and technology more with men than with women. And research shows that kids pick up on those associations from a pretty young age, for example, and a really compelling study that was done analyzing or meta-analyzing all of the studies that had been done, asking kids to draw a scientist.

There’s still a notable tendency for kids to be more likely to draw a scientist as a man rather than a woman. That does seem to be changing a bit over time, but it also increases as kids get older, so teenagers are more likely to associate science or scientists as being male rather than female. So the degree that kids develop these kinds of associations, that also has the ability to shape their own self concepts and their own views of what is an appropriate field for them to go into. But the other thing, too, is that these associations can also shape the kinds of opportunities that parents and teachers and other types of mentors provide to people as ways to get training.

So, for example, in some really recent work that’s not even published from our lab yet, we’re finding that it’s not just believing that there’s a difference between men and women in their interests, but believing what explains that difference, that really predicts the kinds of affordances or opportunities that people provide.

So, for example, imagine you’re a manager in a company and you’re providing an internship to potential undergraduates, students who might be very different in their actual abilities. What we’re finding is that when people believe that gender differences in interests are due to biological factors, that they’re more likely to provide women with opportunities that are about working with people, and men with opportunities that are more about working with things or systems. And interestingly, when you have people who are in the same role, but who believe that gender differences are socioculturally based, they show no gender disparities in the opportunities they provide.

So these stereotypes that can be entrenched, that can be implicit to the degree we justify the differences that men and women have, can be providing dramatically different opportunities for learning and training. That, of course, over the course of someone’s lifetime really accumulate over time.

There’s this question about just what do we mean by inclusion related to this idea of what are the sociocultural variables that can erode people’s feeling of inclusion, or women’s feelings of inclusion in male dominated spaces? And we’ve made these conceptual distinctions between different ways in which environments can signal a sense of fit to the self.

Because I think for all of us, we can sort of admit that we thrive in those places where we feel like this kind of fits who I am. It allows me to do the things that I want to do, and the people here respect and validate the way I like to see myself. So we talk about these as three distinct kinds of fit, a sense of self-concept, fit, accusing the environment, just sort of.

So I feel like home to me. You know, I’m in the library right now and if you like books, like just being in that space and having those cues sort of activates the way you like to see yourself. Goal fit is the sense to which you’re working within, let’s say, an institutional structure or learning in a classroom where the rules for how things work kind of fit with the way I like to learn or the way I like to work and then social fit.

The people here see me as I see myself. And when environments are made for and by one group of people, we can sort of see them as being tailored to a cultural default in male environments. I’m not sure and talk about these as being masculine defaults, but for those who don’t fit that default, they can feel like they have to contort themselves into something that they’re not.

So, for example, in not a study that we did, but a study that others have done showed that women are less likely to dress in a feminine way when they’re attending their science and engineering classes, trying to conform themselves to a more masculine default. So it’s easy to take for granted as a member of an advantage group that the places where you work or where you learn are going to signal these three types of fit.

Because these institutions are probably created for and by people like you. But when you’re a member of a marginalized group, when you’re a woman working or learning in a male dominated field, you’re more likely to encounter cues that you don’t quite fit or what other people have called belonging that you don’t belong. We just prefer the term “fit” because we feel like we can distinguish between these three different types of fit.

You know, walking into an academic department to give a talk and seeing the row of photos of former department chairs, example who are all men, right? Is, is sort of a subtle signal of who fits here, who’s a leader here. Right. And these signals of fit are one of those subtle cues to not just the implicit stereotypes in the minds of people, but the way they get encoded into the physical structures, the institutional structures, and just the social environment at large.

Depending on what the problem is, you might have a different solution. Say you have a problem of women being more likely to drop out or do poorly in a computer science program. If you found out that it’s because those women are experiencing that they’re actively sort of disrespected or even just ignored, like I talked about the kind of being shut out of social networks, if the women are feeling that’s happening from their male colleagues, or so forth, that’s a lack of social fit, right? If it’s that the course assignments are being designed in a way that they focus on programing, things that are more sort of male focused, male valued, maybe competitive games as opposed to cooperative games, something like this, that just women are less likely to be socialized to care about then. That’s a lack of goal fit, not a lack of social fit. Right. So it calls for a different remedy. And that’s why I think there’s value in distinguishing between these different kinds of fit so that we can better understand exactly the nature of the problem.

Books and Ideas: So your research highlights that assuming awareness of implicit stereotypes and attitudes will eliminate them as a central pitfall of anti-bias interventions. Why isn’t awareness enough? Can early interventions in schools or high schools help prevent biased expression and foster inclusion?

Toni Schmader: There’s been a lot of public debate and attention focused recently on diversity training, the value of it, the efficacy of it, backlash against it. And a lot of the training that’s being done these days does focus on educating people about implicit bias. Some of the things that I’ve talked about, under an assumption that making people aware of what implicit biases, may be making them aware of their own biases would do enough to change those biased beliefs or perceptions.
And the research shows that that it’s just not enough. And it’s not enough, in part because it really takes understanding that bias is something that results from a process of cognitions and reactions, and awareness is only one part of that process. So if you really want to change the process, you have to be understanding or cognizant of all the components.

So one component is what are people’s motivations, even if stereotypes that earlier I talked about this tendency to think science, think male. So if you think about that automatic tendency to make that association and that it can come to mind for someone in order to put it aside in your judgment, decision making, like when you’re evaluating candidates for a job, let’s say, first of all, you have to be motivated to do that, right?

So you have to have an egalitarian motivation to be fair minded. Not everybody does, or even not everybody does in any given moment. For example, you might feel justified with whatever beliefs. So motivation is critical when it does come to mind. You do need to be aware that it’s happening, but be aware that you have biases in general doesn’t necessarily mean you’re aware in the moment when they come to mind. So you need to be cognizant and really trained on what are the types of situations when this is likely to be an issue. And then you have to have the ability, both in terms of not being too tired, not being to fatigued, to really focus on paying attention to details about rather than falling back on a stereotype as a quick and dirty heuristic.

And you have to have good strategies in place for what to do. So I think for training to be efficacious, it’s going to take into mind all of these different components. And ideally the training that an organization might be trying to do might need to be tailored to the type of problem that you’re encountering in that particular organization.

So you might be encountering a problem that many of your employees just don’t see the point of gender equality or creating gender inclusion. That’s a lack of motivation that takes a different kind of intervention than if everyone’s on board, but they don’t quite know how to put their beliefs into action. So we’ve made an argument for tailored interventions that try to isolate the type of issue and are then geared toward targeting that type of issue.

We did a lot of our research looking at the experience that men and women have in organizations. But of course, there’s a real opportunity in having interventions targeted with kids earlier on when it can make a real difference in changing sort of the habits that people develop.

Tricia Devine of the University of Wisconsin really talks about implicit bias as a habit to be broken. And so, as with any kind of behavior change, it’s good to start young. So I do think that there’s a real opportunity to think about what kind of training can be provided to young kids to, again, instill the kind of motivation, to be inclusive and egalitarian, to be aware of how stereotypes can bias judgment in decision making, but also provide people with those kinds of skills that they need to seek out diverse others in order to expand one’s mindset and embracing the difference and diversity that exists in the world.

I think another mistake that’s sometimes made in interventions or diversity training programs is to focus on how to react to the biases that we have, how to react to the biases we see in others. And that certainly has a place. It’s not that we don’t want to try to make changes when we see problems, we definitely do. But it is only part of of what we can be doing. So we’ve also been doing a lot of work and writing a lot in our lab about the benefit of more proactive efforts toward creating inclusion. And we can do this at three different levels.

We can do it for ourselves when we choose to read a diversity of books or watch a diversity of movies or have a diversity of friends, that is a way in which we try to expand our own viewpoints to include those of others.

But we can also be proactive in our social relationships in the kinds of connections like earlier I had mentioned that in working in STEM environments, women sometimes feel socially excluded from the social networks that their male colleagues have established.

And so seeking out those diverse connections to foster inclusion interpersonally is another thing that we can do.

And the third thing we can do is to try to create inclusive organizational policies and structures. So, for example, in one of our recent studies, we found that when women tell us that they work in an organization that has more gender inclusive organizational policies and practices in place, they are actually more likely to have those kind of respectful and inclusive interpersonal conversations with their male colleagues and maybe as a result, experience more positive workplace outcomes so their relationships between gender inclusion, that’s experienced at the organizational level, what happens interpersonally and how that impacts people as individuals as well.