Why your HR practices might not be as inclusive as you think with Sonia Kang

Why Female Job Candidates Tend to Make Their Applications Sound Less Feminine – and Why It Backfires

“What we found in some of that work is when women are applying for  jobs in these traditionally male dominated fields, they will actually downplay their femininity by reducing the amount of feminine language they’re using in things like their cover letter. So when they’re submitting their application packages, they’re using less feminine language. And of course, this makes sense. It makes sense why people do this. They’re looking at these jobs and saying, “Okay, well, these are jobs where men have been valued or masculinity has been valued… Unfortunately, what happens is, that tends to backfire. Because we live in a society where our norms around gender are so deeply ingrained, we have expectations about how we want women to act, how we want them to present themselves even when it comes to something like job applications., When a hiring manager sees a cover letter from a woman that has less of this feminine language, they don’t like it as much.”

Why Accurate Data Collection is a Critical Foundation for Effective Diversity and Inclusion Strategies

“Of course, you don’t necessarily want to be collecting data that you think might be sensitive for your employees to have to fill out. But at the same time, if you’re not collecting that data, then you can’t actually track how you’re doing. You can’t make goals because you don’t even know where you’re starting from. Then how are you going to make any changes?”

Addressing Equity, Diversity and Inclusion – Even When the Business Case Isn’t There

“But there are spaces where you can’t really make the business case. And that doesn’t necessarily mean that it’s not worth doing. We need to, because it’s the right thing to do. So one really good example is women who are working in low wage jobs which have typically been very precarious. Let’s say women who are cleaning hotel rooms, it’s a really common example that’s used. As you said, the business case doesn’t really apply there because having diversity, inclusion, all of that stuff, in those spaces, it depends on what your definition is of productivity. It depends on what your definition is of creativity. It’s not so clear-cut.”

Showing Ambition and Resilience in Diversity Recruitment Practices

“Oftentimes, organizations give up really easily when they’re like, “Oh, we can’t find candidates from whatever group.” You have to look harder if you can’t right away, and mostly it’s because your network is not set up to help you find people other than people who already work there or are like yourself.”

Onboarding and Getting to a Sense of Belonging

“There’s a lot of uncertainty there that I think needs to be managed. To the extent that a company can help individuals to manage that uncertainty, make things very fluid for them, or really just make that process as clear and as simple as possible, that’s really going to go a long way, I think, to making people feel like they’re welcome, making people feel like they belong in that space because you are ready for them. If you take… I don’t know, a long time to get their email address set up, or different things fall through the cracks, I think that makes people feel like, “Oh, I don’t belong in this space yet.”

Fostering a Culture of Learning and Experimentation

“It’s really about getting vulnerable, getting curious, admitting ‘I don’t have all the answers, I’m really curious about what you need, what you want, what you think might be helpful, what are you missing?’ Then getting really ready to experiment. And that means giving people the resources, whoever your research leads are, wherever you want people to be doing this experimentation, give them the resources that they need to be able to run really rigorous experiments.”

Brooke: Hello, everyone. Welcome to the podcast of the Decision Lab, a socially conscious applied research firm that uses behavioral science to improve outcomes for all of society. My name is Brooke Struck, research director at TDL, and I’ll be your host for the discussion. My guest today is Sonia Kang, Chief Scientist at BEAR, that’s the Behavioural Economics in Action Research Center at Rotman, Canada Research Chair in Identity, Diversity, and Inclusion, and host of the ‘For the Love of Work’ podcast. In today’s episode, we’ll be talking about inclusion in the workplace, the effects of biases, the approaches that work, and the surprises that can trip us up. Sonia, thanks for joining us.

Sonia: Hey, Brooke. Thanks so much. Thanks for having me.

Brooke: Before we launch, tell us a bit about your work. What problems are keeping you up at night that you’re working to solve?

Sonia: Yeah, for sure. I’m really interested in stigma, inequality, and how people manage those things in the workplace. And really for me, my focus shifts away from what research in this area has previously focused on, which has really been on organization. So what are organizations doing to either cause these problems of discrimination, of applying bias where it shouldn’t be applied, or what are they doing to solve the problem? And on the other hand, how individual job seekers were reacting in the system that was cast in more of a passive light. So their perspectives hadn’t really been examined or understood as much.

That’s where my research really focuses. It’s on what’s going on with the people who are actually in the system. They’re not just hanging out and letting this discrimination wash over them, they’re trying to navigate, trying to overcome the system. They know the barriers that are there rising up against them, and so it’s about working around that system. Then at the same time, really thinking about how we can redesign the system so that we’re not leaving it up to those people to fend for themselves and figure it out. So really focusing as well on these structural, procedural changes that we can make in order to create systems where bias is less likely to happen.

Brooke: That’s fascinating. And there’s so much in your research where you see this ‘measure and counter-measure’ of people trying to adapt to the system and the system adapting to them. A lot of the stuff that comes out of that, to me anyways, is really surprising. So there are a few of those surprising findings that I wanted to walk through with you so that we could just get a bit of a flavor for what this looks like in practice. One of the things that you’ve worked on quite a bit is women downplaying feminine attributes in their job applications. Can you talk us through some of the work you’ve done there?

Sonia: Yeah. That work really looks at how women present themselves when they’re applying for jobs in traditionally male dominated fields. For example, STEM, that’s a big area of focus for me. Again, it’s this idea of, okay, women are in these fields, they’re applying for these jobs, and they know that traditionally their identities have been devalued in that space, that this is a space that largely has been created by and for men. So the research really looks at, okay, well, what are women doing to navigate through those fields and get a holding in that field and then hopefully try to get ahead.

What we found in some of that work is that when women are applying for these jobs in these traditionally male dominated fields, they will actually downplay their femininity by reducing the amount of feminine language they’re using in things like their cover letter. So when they’re submitting their application packages, they’re using less feminine language. And of course, this makes sense. It makes sense why people do this. These women, they’re looking at these jobs and saying, “Okay, well, these are jobs where men have been valued or masculinity has been valued.”

So they’re downplaying any feminine attributes. They’re using less feminine language specifically. And that makes sense, right? You’re like, “Okay, yes, this makes sense. Why wouldn’t you do that? It makes total sense.” Unfortunately, what happens is, that tends to backfire. Because we live in a society where our norms around gender are so deeply ingrained, we have expectations about how we want women to act, how we want them to present themselves even when it comes to something like job applications – when we see something, when a hiring manager sees a cover letter from a woman that has less of this feminine language, they don’t like it as much.

They have this idea of what women should be like, this goes against that idea, and so then those women are less likely to be hired. We actually see that that backfires. So this very sensical strategy that they’re taking on actually makes it worse. It’s a common scenario that happens for women. You’re damned if you do, you’re damned if you don’t. What do you do in that? So again, that’s why we’re really trying to focus on, okay, well, how can we change something about either the jobs themselves or about the way that people are hired into those jobs that prevents those kinds of things from happening?

Brooke: Yeah. That finding itself really jumped out at me, that even in a professional setting, adhering to the gender norm is actually more impactful than adhering to the job norm, even though it’s about jobs.

Sonia: Yeah, absolutely. And that’s really what we wanted to look at next. It’s like this realignment of those two identities. You have the professional identity on the one hand and then you have people’s personal identities on the other. Some other work that we’ve done in the same vein is looking at how we can bring those two identities together by changing how those jobs are presented. I can talk about that later, because we’re talking about counterintuitive things right now. But that’s one way to align those things.

Brooke: Cool. Let’s continue down that rabbit hole of counter-intuitive stuff.

Sonia: Sure.

Collecting Meaningful Diversity Data

Brooke: You read it on the label, it’s like a company wants to start gathering more race-based data about how the demographics of their organization represents maybe the demographics of the community or this kind of thing, that sounds like an initiative that’s really pushing in the right direction, you’re starting to orient yourself so that you can make better decisions, and then employees themselves have to arrange themselves under this set of labels, and then things start to come apart. Tell us a little bit about how things come apart and where we go from there.

Sonia: Yeah. This is really interesting work because of how new it is. So shockingly, many companies, even my own employer, the University of Toronto, we’ve only recently started collecting race-based data. I think there’s been this idea that race is a taboo, we don’t want to talk about race, and so we’re not going to collect the data on race and we’re not going to ask people to check a box and let us know about your race. I get that, right? Of course, you don’t necessarily want to be collecting data that you think might be sensitive for your employees to have to fill out. But at the same time, if you’re not collecting those data, then you can’t actually track how you’re doing. You can’t make goals because you don’t even know where you’re starting from. Then how are you going to make any changes? So there’s this real tension that exists with organizations. There are goals on the one hand, which is to understand what their workplace demographics are, to be able to set goals, to be able to track progress towards those goals, and then that’s up against this individual or group-based interest on the other hand, which is to be defined or described in a way that reflects how you define or describe yourself.

So one of the big challenges that has come up in this work on collecting this race-based data is that companies will create these diversity surveys, or even more formally, the government has a census, and they’re using these umbrella terms, these superordinate terms that put a whole bunch of different categories together. So they’ll put ‘Asian’, right? And Asian subsumes a whole bunch of different identities. They’ll put-

Brooke: Literally billions of people.

Sonia: Exactly. It subsumes a whole bunch of different identities. There’s lots of terms like that, like black, white. All of these things are superordinate categories, they are these umbrella terms, and organizations need to do that. They can’t say, “We have four people from this very small village in India.” They can’t do that. So they really need to be able to get that data in the aggregate. But what happens is that when employees are filling out those kinds of labels or filling out those kinds of forms, they feel like those labels don’t reflect the way that they actually see themselves.

It’s not actually jiving with the way that they would describe their own identity. Then we see a number of things happening. Either they don’t check that box because they don’t realize it’s supposed to describe them, or they might not check the box because they don’t like being described like that. So in any different scenario, the organization is missing a lot of the people that they’re trying to capture with those data. My PhD student, Grusha Agarwal did a study recently where we looked at basically how people identified with different terms.

So terms like racialized or BIPOC or visible minority, all of these different terms. We found for all of those different superordinate groups, more than half of the people who replied who you would ostensibly think fit in those categories either didn’t describe themselves using those terms, knew that they fit into that category, but didn’t like it, or just flat out said, “I don’t identify with this.” So in any of those different categories, you’re missing, like I said, more than half of the people that you want to capture into that category. That’s why that really needs to be reformed.

And I think that when we’re thinking about, in the Canadian context at least, right there is the Employment Equity Act, which is there to correct disadvantages in employment that have existed. All of those groups are groups that have been traditionally underrepresented, they’ve been marginalized. So if you look at something like, let’s say disability status or indigenous identity, and all of a sudden your employer is asking you to define yourself along those terms, why would you describe yourself like that?

Of course, you know that this is a context in which you might experience discrimination based on that identity and all of a sudden you’re supposed to trust your employer with that information. It just doesn’t make sense. So this is another area where with Grusha and my former PhD student, Joyce He, who’s now a professor at UCLA, we’re trying to figure that out. We’re trying to figure out what are the category labels that are useful both to organizations, but also respectful of how individuals and groups want to describe themselves. That’s ongoing and it’s really, really an interesting tension, I think, because it touches so many different aspects of society.

Why We Can’t Just Rely on the Business Case for Diversity

Brooke: Talking about those who are traditionally underrepresented, those who are traditionally marginalized, I think that,at least in the echo chamber that I exist in out there in the news world, it seems to me that the business case for diversity, equity, and inclusion is pretty well accepted that having a more diverse population within your employee group tends to lead to more innovation and better business outcomes, and this kind of thing. But some of the work that you’ve been doing shows that applies for one type of work or maybe one broad swath of work.But when it comes to low wage jobs in particular where traditionally marginalized people are actually the most in need of support and the most in need of more equity, that business case doesn’t hold up the same way and potentially there’s a different conversation that we need to be having there.

Sonia: Yeah. So there’s a lot of different ways that people come to the equity, diversity, inclusion, whatever you want to call it, how people come to that conversation. The business case, which you mentioned, has been pushed. It’s been pushed a lot and it’s one of the most common reasons why people are interested in EDI in organizations because they know that it’s associated with greater productivity, greater creativity.

It has all of these really beneficial downstream consequences to increase diversity and make sure people feel included. But there are other reasons why people might be interested in EDI. Some people come to this space because of the legal case, they want to avoid being sued. Some people will come to it from just strictly that. They just want to know, what are the rules? I don’t really care why we do it or why we don’t, this is just why I’m interested in it. Then we also have the moral case. This is where people are there because they feel like this is the right thing to do. Why wouldn’t we make sure that all kinds of different people would feel welcome in our spaces? Your question is really interesting because the business case, like you said, it gets pushed so much. And I think that pushing that business case so much really makes it top of mind for people. So they’re always looking to see where’s the evidence that these EDI initiatives are resulting in better business. How are they helping the bottom line? 

But there are spaces where you can’t really make the business case. And that doesn’t necessarily mean that it’s not worth doing. We need to, because it’s the right thing to do. So one really good example, I think, is with women who are working in low wage jobs which have typically been very precarious, something like, let’s say women who are cleaning hotel rooms, is a really common example that’s used. As you said, the business case doesn’t really apply there because having diversity, inclusion, all of that stuff in those spaces, it depends on what your definition is of productivity. It depends on what your definition is of creativity. It’s not so clear-cut.

Another example that you brought up, which I really thought was very poignant, is an example of the Amazon fulfillment center worker. Of course, it makes sense to just get those people to work as quickly as possible, and when they want to not do that job anymore, they’re easily replaceable, and so we don’t care that much about this EDI stuff. That becomes really problematic because then it’s only the people who are already in these protected workspaces who are getting those benefits of EDI and the people who are in jobs where they are more vulnerable, where they’re already subject to marginalization, oppression, their rights are not as well established or well accepted in society that they don’t get those kinds of benefits.

So I think that we need to have a broader conversation around the business case. This is why in my work I write about how we can’t just only rely on the business case. We need to look beyond the business case for diversity because there’s so many other reasons that we do this work and there are so many other reasons that we should see as being important. It’s not just important because it’s making us more money, because as you said, it’s often in those cases where it’s not so clear that it’s making us more money that we need these kinds of initiatives the most.

Addressing Biases Throughout the HR Cycle – Starting With Recruitment

Brooke: All right. Let’s pivot a little bit now from all of the stuff that’ll blow your hair back about the surprising outcomes that we don’t expect. Getting more into the meat and potatoes of what it is that we can do about this. In the world of human resources, the full life cycle of an employee from the moment that you post a job ad and start trying to build awareness around recruitment to interviewing people, choosing who you’re going to hire, onboarding them, supporting them through the progression of their career, and eventually them being offboarded and transitioning into something else in their career, what are the main decision points where we need to be paying attention to biases the most?

Sonia: Yeah. I think that there’s points throughout the life cycle that are really important all the way through, like you said. Even when we start thinking about things like recruitment, even before people are in your company, I think you really need to think about, who are you reaching with your recruitment efforts? Where are you looking? Where are you not looking? There’s so many different things that you can miss in your recruitment. 

On our podcast, for example, we had an indigenous woman who was the first person from her community to work in a bank. She talked about how in her life this had never really crossed her mind as a possibility for her. She knew that that job existed to some extent, she didn’t really know what it was and she didn’t know that this was a possibility that was open to her. So she ended up being the first person from her community to get a job in a bank. Then that really deemed mystified that whole world for her and she was able to pass that on through her community. From there, she was able to really inform the recruitment practices of the bank to make sure that they were reaching out to those communities that may not have heard about these opportunities before.

Again, I said demystify, I think that’s a good way to describe this because for people who have never thought of these jobs as a possibility. Everyone has these kinds of jobs they’ve not thought of as a possibility. It’s just like I don’t even know where to begin. And a lot of times it’s just about having someone open your eyes to see what that world is like in order for you to see that as a possibility for yourself. So I think on the recruitment side, organizations can do more to make sure that their recruitment efforts are going to be successful in recruiting diverse candidates by looking in the right places and sending the right people to look and being more resilient, I would say.

Because oftentimes, organizations give up really easily when they’re like, “Oh, we can’t find candidates from whatever group.” You have to look harder if you can’t right away, and mostly it’s because your network is not set up to help you find people other than people who already work there or are like yourself. Another interesting thing I think about recruitment is that you can also think about recruitment starting even earlier on, so bringing people into different pathways that will lead them to the jobs that you’re interested in having them in.

So there was some work, for example, on recruiting more women into econ majors where they were able to increase the number of women who chose to major in econ by just sending these email reminders as the deadline approached with information about what that was like. And it’s just like those kinds of little nudges that can push people over their initial hesitations where they might feel like, “Oh, this isn’t really for me,” and then they start thinking about it more and more and then that can often help them to get engaged. And once they’re in there, they often realize that they’re really interested in this.

Then again, this is about increasing those pathways, making those pathways more accessible. So recruitment, I think, is definitely important. Then we have selection. Selection is a big one. There’s a lot of focus around selection, around how do we interview, how do we test people? All of these different kinds of assessment, evaluation, processes, all need to be, I think, really carefully handled. There’s some really interesting work on selection, for example, showing that the way that you make decisions is really impactful in terms of the final diversity of the group that you hire will be.

This is some work by Edward Chang. He’s at HBS now. This is some work that he did when he was a graduate student, really amazing work showing that if you make selection decisions in sets, so you hire let’s say like eight people altogether, it’s going to be a much more diverse group than if you are hiring eight people in sequence. So you hire a person for this job, a person for that job. This makes sense. I really think about this in terms of how I make decisions about meal planning. If you make your decisions about meal planning for a week at a time, you can see, like zoom out, you see what the week looks like. So you can have a lot more diversity there because you see like, “Oh, I can’t have the same thing on this night, on a Monday, that I’m going to have on Wednesday,” and you build more diversity into the system. But when you’re making those decisions one at a time, they become disjointed from each other. So one day you’re like, “Oh, we’ll just have pizza,” and then two days later, you’re like, “Well, Let’s just have pizza again.”

So, making those decisions individually leads to a lot more homogeneity because you just go to whatever’s the easiest, most accessible thing for you. But when you’re forced to zoom out a bit, it’s a much more active, conscious process, you’re much more deliberative, and so you’re able to create much more mindfully this more diverse group. So I think that’s another really promising, promising finding there. I feel like I’m just going to go through every part of the life cycle. Okay.

Looking for Recruitment Blindspot and Hiring in Sets

Brooke: Let’s sum up a little bit, just to help the readers or the listeners at home keep tally of where we’re at. The first, in terms of recruiting and awareness building, attention to your blind spots. Think about how those paths of least resistance to someone finding their way to your company are set up. There’s going to be a natural tendency that people who are in the network of your existing employees are just going to know more about the job opportunities that are there in your company than the people who are not in that network, all other things being considered.

So if there are specific groups that you want to go and engage with, a helpful and very practical question you can ask yourself is, “Where do people like that hang out already that I can just go and say, “Hey, by the way, there’s a job opportunity and we would like to make you aware that this is here and we’d be potentially interested in offering an info session so that people in this group can learn more about what it would be to work in this job, to work in this company, and to consider this as a place of business.”” That’s one.

The second in recruitment, I really like this, this practical tip that you gave about hiring in sets. If we consider all hiring decisions in series, just like one thing after another, so to speak, then we’re likely to reach for that thing that’s easiest to hand, which because of the way that our networks work and our biases work, you’re just more likely to choose somebody like you, and like the people who already work there, than someone who is unlike you or unlike the people that you have traditionally hired.

So hiring in sets or shifting to a set-based approach to hiring can help to overcome that because you’ll start to see that pattern emerge and you can make more intentional decisions about the kind of mix you want to build in the group of employees.

Sonia: I think the set thing is also really interesting because you can start looking at multiple different people’s decisions together. You can look at five different hiring managers and make sure that across the set of people they’re hiring, you have good diversity there. So in that way, you’re able to cover each other’s blind spots. So when you’re zooming out even further, you can be even more intentional and make sure that you’re really, really, across a large number of hires, making sure that you’re being true to the group or the representation that you’re trying to mirror in your company.

Brooke: Yeah. That actually reminds me of something that we’ve done internally here at The Decision Lab. In the last year, maybe year and a half, we’ve been thinking a lot about our hiring processes. One of the things that we’ve reworked is the way that we assess candidates. When we were just getting that new system off the ground, the first thing we wanted to figure out is, well, does the system actually work? How reliable is it at identifying good candidates? Initially, the closest proxy we had to that is, how consistently do different staff members at The Decision Lab assess different incoming candidates? That started us on this path of building up a data set of previous candidates to assess. But that initial review where it was a whole bunch of people actually going to look at a whole bunch of candidates, and trying to triangulate how convergent or how divergent the different evaluations of individual candidates, that was a really, really interesting exercise and it actually led us a bit organically to the kind of process that you’re talking about, where in fact we are hiring by sets.

One of the things that we have taken from that is, if we need to do an individual hire, we will always go back to that old data and say, “Okay, well, if we were thinking about this person as a member of a previous set, would that change our perspective?” Because it’s a really different question to ask yourself, should I hire this person or not hire this person? As opposed to, “Among this group of people, which one or which ones should I be hiring?” They’re really, really different conversations and we found that it leads to very different decision dynamics at the end of the road.

Sonia: I mean, you can let me know from your own experience, I think as well it doesn’t really take that much more time because you’re doing this altogether as well. You’re able to see those patterns and you get better at it. Of course, when you first start doing this, it’s like, ‘Okay, we have to be a little bit more careful and slower’, but then when you get a little bit more comfortable in that system, it becomes really natural. I think a lot of companies, they hesitate to make changes like this because they’re like, “Oh, our process is so streamlined. It’s so efficient now. If we make a change, it’s going to add all this time.” Yeah, it might at the beginning, but it’s not… people adapt really quickly, I think, to those kinds of new processes when they see that they make sense.

Brooke: Yeah. The curve is definitely much less painful in the going, than it was intimidating in the prospect of it. We anticipated it was going to be really hard to adapt, and then when we actually did it, it didn’t take nearly as much time or as much coordination as we thought it would. And we learned some really valuable things quite early on that allowed us to streamline the process. Okay, you want everybody to be completely evaluated twice, that can be very time consuming. Surely, there’s some threshold below which you don’t want to invest in that full-fledged assessment. And you can learn pretty quickly, if you have one person do an assessment first, where is the lower threshold below which you say, “We don’t even need to go to a second assessment. We’ve already learned what it is that we need to learn.”? So if you’re scoring out of five, for instance, and you say, “Okay, well, roughly what we’ve seen is when candidates perform at a below three and a half out of five, they don’t gel with the team, they don’t fit into the workflow, this kind of thing.”

If your first reviewer scores a two and a half out of five for that candidate, you don’t need to send them necessarily to a second reviewer if you know that you have a decent level of convergence between the different assessments where it’s like, okay, a full one point out of five difference. Sometimes that happens, but you’d need a two point difference to just meet that minimum threshold of three and a half. And it’s just so unlikely from the data that we’ve been collecting, we say, “Okay, well, we know that this is not an airtight or hermetic system.”

Clearly, there’s always going to be inefficiency because there’s always so much you don’t know. But we’re pretty confident that we are not going to be making grossly negligent decisions on a regular basis if we move forward with something like this.

Sonia: I think that’s really great from the employee or the applicant ,side as well because you’re also not wasting their time, which I see a lot. Applicants, are dragged through these multi-step processes…this happens too in academic publishing! You’re dragged along for so long and it’s just a waste of time. And you shouldn’t. If your process is good and you’ve worked out the kinks, you’ve made sure that you are dealing with bias very early on, then you shouldn’t be wasting people’s time like that. If you’re applying for a company and it seems like the process is inefficient and it’s annoying, frankly, I think that’s a really good indication of what your life will be like working there because that’s one of the most important processes that a company has, how they hire. 

Brooke: Yeah. Well, I will toot our own horn very briefly here and say I was really surprised when the first cohort of applicants who went through this new process, lots of people during the application process and especially during the interview phase, unprompted, made remarks like “Wow! This application process is really smooth.” And we’d never heard that before. So all of a sudden, we start to get positive feedback from candidates about the process itself. Even people who didn’t get hired told us that they really appreciated the hiring process. To me, that really stood out. It’s like we must be doing something here that’s working if we’re getting feedback from people who are saying they enjoy something that traditionally people hate.

Sonia: Yeah. I find that fairness, procedural fairness, and that feeling of fluidity, are very highly correlated. If something seems like it’s inefficient and sketchy, it’s probably not very fair.

Onboarding & Getting to A Sense of Belonging

Brooke: Yeah. All right. Let’s move on from the actual hiring decision. Now we’ve chosen who it is we’re going to hire and the next thing we need to do is onboard.

Sonia: The next step with onboarding, I think this is one of the most overlooked steps. Organizations, I’ll say this even in academia, put so much effort and time, and there’s so much emphasis put on selection. Then often, when people actually start working at the company, it’s like you forget about them. This happens all the time in academic hiring. We go through this whole huge process, there’s so much, often, controversy involved, there’s a lot of heat in that process, and people are fighting for these different candidates. Then the person starts, it’s like you forget that they are even there.

So that process of onboarding, I think, is the first real feeling that people are going to get of feeling included in your space. And we all know what it’s like starting a new job. Even the most awesome first day, there’s some awkwardness. There’s a lot of uncertainty there that I think needs to be managed. To the extent that a company can help individuals to manage that uncertainty, make things very fluid for them, or really just make that process as clear and as simple as possible, that’s really going to go a long way, I think, to making people feel like they’re welcome, making people feel like they belong in that space because you are ready for them.

If you take… I don’t know, a long time to get their email address set up, or different things fall through the cracks, I think that makes people feel like, “Oh, I don’t belong in this space yet.” So just little things like that in terms of onboarding really, it’s all about uncertainty reduction. I would say that step is really about helping people to know the logistics, just where things are, what I’m supposed to do at different times, and then really helping them settle into their work.

That’s all about setting people up for success to make sure that you’ve actually started them off on a good footing where they’re not just thinking about “How do I do this?” Where do we do this? What’s this? What’s that?” They’re really thinking about, “Okay, all of these things are really well spelled out and now I can just focus on my work.”

Brooke: In terms of diversity and inclusion, where’s the angle there to ensure that it’s not just the archetypal, the white male walking into a white-collar job who feels included and feels warmly welcomed and received into their new company, but people who are coming from all different walks of life, all feel very welcomed?

Sonia: Yeah. When thinking about this kind of stuff, I draw on the principle of universal design, really trying to think about how you can design that onboarding process so that it’s good for as wide of a group of people as possible. So you want to make sure the same things that you’re doing in terms of recruitment and selection, really doing that co-creation of those processes with the different people that you’re trying to bring in, you want to do that with your onboarding as well and make sure that employees that have been through that process really have an opportunity to give feedback and shape that process, really make it iterative so that you’re trying out different things as well.

I think it can also go a long way to ask people what they might need in advance. So saying, “These are the things that we have set up for you, is there anything else I can do?” Just to make it seem that there is this element of personalization and individualization, because a lot of times these things can seem very generic. If you can actually communicate that you care and do those small steps, nobody’s going to ask for something like, “I need a hot tub in my office.” It’s going to be very small things, and so they’re often very easy to accommodate. So it’s worth asking what people need to feel comfortable in the space and feel like they belong.

Brooke: Yeah. You mentioned being highly iterative and that really struck a chord with me. And the question I was asking myself is, what’s the outcome indicator that we’re going to innovate towards? I think when it comes to recruitment, it’s a bit more clear. If we’re collecting race-based data, for instance, it’s like, “Okay, well, we’ve set up this set of filters for applicants. A whole bunch of applicants come in the top of the funnel and out of the bottom of the funnel come those that we choose.” So we have some race-based data to assess how fairly represented are the people coming in the top of the funnel once we get to the bottom of the funnel. When it comes to onboarding, I think it’s a really different story. I’m not entirely sure what we would be measuring against.

Sonia: Yeah. I think for onboarding, the most important metric there is going to be belonging. Because if you think about when someone’s outside of your organization, their sense of belonging should be low. They shouldn’t necessarily feel like they belong. So you want to see how quickly you can get them from a sense of not belonging to a sense of belonging and feeling comfortable and valued in that space? I think doing things like individualization and making sure that their needs are taken care of is a huge step in terms of feeling valued.

People really want to feel like having gone through this whole recruitment process and doing all your selection things, that you chose them for who they are. I also think cohorting is really important. If you are hiring more than one person, making sure that those people all know each other, they’re well connected, setting them up with a mentor, and really making that meaningful mentorship time. Not just asking “What’s the address for the Slack team?” or something like that. Make sure it’s a meaningful connection. That can be done, again, by making sure that you’re co-creating that mentorship experience for the person by asking them, “What are you looking for in a mentor? What are the kinds of things that you are hoping to develop?” 

Training & Development

Sonia: That segues a little bit into the next phase, which is training and development. The modern thinking around training and development is that it’s not just this thing where you take a day to do your training and development. Now it’s more that almost every interaction, if not every interaction, is this opportunity for training. So I think you really need to start thinking about who gets the opportunity to have those interactions, making sure that you’re facilitating those interactions in a way that everyone has access to those opportunities and the right opportunities for them. That’s why I keep on saying co-creation, but what I mean by that is really that, you’re not just looking at your employee and saying, “Okay, this is how I want to shape you,” you’re including them in that decision and saying, “What are the things that are important for you to develop?

Where do you see yourself going in this company? What are your goals, aspirations, personal, professional?” And helping them create this mentorship or sponsorship experience that’s going to help them get there. I think that also starts right in the onboarding phase, to make sure they’re matched with someone who is going to be responsible for them.

The Power of Defaults

Brooke: Drawing on some other work of yours that I’ve read, I think defaults probably have a really valuable role to play there as well. But these kinds of things are not opt-in because there are going to be certain types of people who quite predictably will opt-in at much lower rates than others.

Sonia: Yeah, absolutely.

Brooke: All right. Speaking of opting in, you’ve done a lot of work around opting in for considerations for promotions and other areas of professional development. Let’s dig into that because I think that that’s a topic that a lot of people are really thinking a lot about these days as well. How can we make sure that there is not only room to develop, but room to succeed within our organizations and to grow and advance within an organization?

Sonia: Yeah, absolutely. This actually works really well because I feel like the next step after this, the performance, development training step that we just talked about, is promotion. How do people advance in the company? This work is with my former PhD student who I already mentioned, Joyce Lee. A lot of this work is from her dissertation. Basically, what we started thinking about was how there is this trope that women are less competitive than men. There’s this idea that, okay, yeah, so women aren’t represented in these higher levels of organizations because they’re just not taking those risks. They’re just not as competitive as men, they’re not putting themselves out there, and that’s why this exists. So we started thinking about that, and well, okay, maybe it’s not the case that women are more or less competitive or that men are too competitive, those kinds of things. Maybe it’s something more about the way the decision is set up that’s activating these norms around how women should act and suppressing their entrance into those competitions.

So when you think about your traditional promotion in a workplace setting, that is usually almost exclusively an opt-in decision. What that means is that the default is if you’re just sitting at your desk, the default is that you just continue doing your work, you just go on with your life. You have to actively do something to put yourself up for that promotion. You’ll have to apply, perhaps you have to tell someone that you’re interested. There’s something that you have to do, so that’s an opt-in. There’s something active that needs to happen.

When you look at promotions or when you look at competitions like that where there is this opt-in process, there is this gender gap that we observed, which is quite large, where men are much more likely to enter into those kinds of competitions than women. So we really started looking at the research on defaults and how we can change defaults in order to change those kinds of outcomes in terms of a lot of different decisions. A really common one is retirement savings programs at workplaces.

Workplaces that have an opt-in retirement savings program where people have to actively sign up have much lower rates of enrollment in those retirement savings programs than in companies where it’s an opt-out. Basically, everyone across the board is a member of that retirement savings program. If they want to, they can opt out, but otherwise they’re in it. Typically, when people are in, like an opt-out, they just stick with the default. There are really classic examples of this. Like I have an iPhone. If you have an iPhone, you and I probably have the same ringtone because people just stick with defaults. 

Making Promotion Competition the Default

So we thought, okay, can we apply this default principle to this competitive promotion space? We did a series of lab experiments, we’ve done a field experiment now, where we looked at this question of whether we can change that entrance into that competition from this opt-in to an opt-out and maybe see some changes in the proportion of men and women who are actually enrolled in the competition. What we found is that when we have our traditional opt-in where you have to actively do something to enter into the competition, we see the traditional gender gap where women compete at about 25%., 25% of women compete, and 75% of men are competing. That’s a huge difference. You can think about that and how that’s given rise to all of the inequity that we see, especially as we go higher and higher in organizations. In our opt-out condition, where basically we had so that everyone who passed a certain threshold of performance was automatically enrolled in the competition, they always maintain the freedom to opt out if they want. 

We were able to completely eliminate that gender gap in enrollment. We had 75% of men and women enrolled in the competition now. So when we think about that whole story, you can see it’s not that crazy what we did. We just simply changed the way that that decision is framed. So typically, it’s an opt-in, you have to do something. In our intervention, it’s just an opt-out. The competition itself is the same, it’s just what you have to do to be in. I think that taking advantage of little tweaks like that is really important for organizations to play around with, because it’s often these kinds of small changes where each one of them seems very small, but when you start putting these things together across the life cycle of the employee, all of a sudden you probably start seeing really big results.

Another thing that I want to mention about that work that’s really important is that when we made this change and when we saw more women enrolling in competition, we also looked at whether there were negative consequences of this, whether maybe women didn’t feel as good about the decision or whether their performance suffered or they were less likely to be hired, and we didn’t find any of those things. So basically, you can do this without making women more anxious, without decreasing their performance. Their likelihood of getting hired is still very good. It’s not just that you put them in there and then they were going to hire a man anyway. We’re able to do this without those negative consequences that we’ve thought of. I think that’s really promising in terms of adopting that into a wider practice, across different positions, especially when you’re looking at these higher level positions.

What Leaders Can Do to Encourage Experimentation

Brooke: Speaking of higher level positions, I want to change course again slightly now and just ask. For senior leaders or senior executives in a company who are listening to this and are just like, “Oh my gosh! All of what you’re talking about now is so desperately what my organization needs” What is something really powerful that they can start doing tomorrow morning to get the ball rolling on this?

Sonia: I think at the senior leadership level, it’s really about… And I mentioned this earlier, gathering data from the people who are on the ground. I think it’s about that. It’s really about getting vulnerable, getting curious, admitting “I don’t have all the answers, I’m really curious about what you need, what you want, what you think might be helpful, what are you missing?” Then getting really ready to experiment. And that means giving people the resources, whoever your research leads are, wherever you want people to be doing this experimentation, give them the resources that they need to be able to run really rigorous experiments. This doesn’t have to take forever either.

You can run a series of well powered experiments to figure out the best ways of doing things that answer those needs and requirements that come up when you’re asking people, “What are we missing here?” I think that’s it. It’s really going into this with genuine openness and curiosity about what people need to happen, and then having the courage to let people have the freedom to experiment around what’s going on. That goes back actually to what we were talking about almost at the beginning of this conversation, which is about collecting data, about making sure that you’re collecting that data around the things that you want to actually change and really understanding those.

Brooke: Yeah. Have you seen any particular barriers that tend to stand in the way of this action or this curiosity and experimentation?

Sonia: I’ve done a lot of work in hiring. Honestly, one of the main barriers in hiring, and this sounds so silly, is the people management softwares that companies use. At U of T, I think we use something called PeopleSoft. First of all, it’s a third-party software, so you don’t really have a lot of control over what people see. For example, if we come in and we make a suggestion that at this stage the resumes need to be anonymized, often companies can’t do that. It’s the same for me at U of T when I’m hiring for, let’s say, a work study position. In order to open the person’s application, I have to double-click on their name. That’s it, that’s where it is. I hope that the developers for those softwares are going to start thinking about building in these EDI functionalities more so. 

In the meantime, there are other really great companies that do this. One of my favorites is this company called BE Applied. Karen Glazebrook, she’s the founder of it and she’s one of my favorite people in this space. They do this for you. So they will help you to anonymize your resumes, they will present you the resumes horizontally rather than vertically, which means that… Normally, when you look at a resume, you look at it vertically. So you read the name on the top, you do the education, you read their experiences, and you move you down. That’s really opening you up to the halo effect, where if you see something that you like, then for the rest of it you’re like, “This is great.” And if you see something that you don’t like, you’re going to discard the rest of it. But instead, they set it up so that you’re looking at everything horizontally.

So you have a predetermined criteria for, let’s say, prior skills, training or education, and then you’re rating everyone’s education all at once horizontally, disconnected from the rest of the information, and then you do experience disconnected from the rest of it. So it’s just these kinds of small things that they can help you do to overcome biases. I think that’s a really important part of this. Bias is always going to be present in any system that is interacting with humans, even if there’s AI there, because it’s programmed by humans. So it’s not necessarily about getting rid of that bias, it’s about creating a system in which that bias is less likely to affect your outcomes.

Brooke: I think that that is an excellent note on which to end, Sonia.

Sonia: Great.

Brooke: Thank you so much for your time and your insights today. I’m sure that our readers and listeners will get a lot out of this conversation.

Sonia: Great. Thanks again for having me. It was fun!.

Brooke: Take care.

We want to hear from you! If you are enjoying these podcasts, please let us know. Email our editor with your comments, suggestions, recommendations, and thoughts about the discussion.