Delve podcast: How Organizations Can Increase Gender Diversity by Rethinking Job Recruitment, with Brian Rubineau (Read Transcript)

Delve podcast, January 20, 2023: How Organizations Can Increase Gender Diversity by Rethinking Job Recruitment, with Brian Rubineau
Robyn Fadden – host: In the past few years of the Covid pandemic, many people have left or lost their jobs and sought out new ones. Who has succeeded and who hasn’t depends not only on merit and ability, but on who you know. And who you know typically reflects your gender, your race, and other influential differences that in policy terms have come to be known as markers of diversity. We also live in a time when organizations are more aware of this than ever before: they’re implementing diversity, equity and inclusion initiatives and they’re listening not only to current employees but to market and management research that shows a more diverse workforce benefits both the organization and society.
Robyn Fadden – host: Welcome to the first Delve podcast of 2023! The Delve podcast draws real-world insights from new management research at McGill’s Desautels Faculty of Management and beyond. If you’d like to listen to past episodes of this podcast and read feature articles, go to For this episode, I’m your host Robyn Fadden. In this episode, we’re asking a complicated question: what role does gender play in job recruitment and hiring? Research by Desautels Faculty of Management Professor Brian Rubineau shows that gender is a factor not only in word-of-mouth recruiting – one of the most common ways that people learn about and are encouraged to apply for jobs – but in who applies for the job in the first place and who reapplies after they’ve been rejected. Welcome to the Delve podcast, Professor Rubineau – thank you so much for being here.
Brian Rubineau: Thank you. It’s a pleasure to be here.
Robyn Fadden – host: If you could first set the tone for the current state of diversity, equity, and inclusion initiatives in organizations in North America especially – what initiatives are you seeing in your research and what is currently missing?
Brian Rubineau: I believe organizations are hungry for identifying effective interventions to try to promote diversity. Diversity is an important value and goal for many organizations. And what’s missing are a range of effective tools to try to achieve those diversity goals. One of the things we observe is that organizations tend to grasp on to whatever is being put out there as likely to be effective, even if we don’t necessarily know if it’s effective or not. One good example is diversity training, which tends to be very popular, but the effectiveness of those interventions of diversity training is far from established. My research really tries to understand what are the social processes, what are the dynamics that influence diversity outcomes and organizations to try to identify what are some policies, interventions practices that could be more effective and reliably effective in promoting diversity in organizations.
Robyn Fadden – host: So rather than answer the very broad and possibly unanswerable question “Are these initiatives working?” let’s break this down into sectors. Are these diversity, equity and inclusion initiatives present in all sectors – and are there some sectors that are taking up these initiatives more than others?
Brian Rubineau: One of the pervasive features of work in North America is that it tends to be segregated. There are very few jobs that are, for example, if we consider gender segregation, there are very few jobs that are just 50% male, 50% female, reflecting the general population and the available employees. Most are either male-dominated or female-dominated. A lot of my research focuses on these male-dominated contexts. There’s a whole range of industries that are male-dominated: finance, interesting industries, tech industries, and things like that. And that’s where a lot of my work focuses.
Robyn Fadden – host: Because your research talks about the employment pipeline, I’d like to ask whether the organizations that have diversity, equity, and inclusion initiatives are taking research like yours into account. Are they really looking at all levels of the organization, including at recruitment and hiring, to solve these problems to achieve equity?
Brian Rubineau: Organizations recognize that our human resources are a flow through the organization, that we have people applying to enter an organization to become employees, they try to succeed, become promoted, achieve success in the organizations, then will exit those organizations. There’s a flow of people through organizations and large organizations—organizations that have been around for a while—recognize these flows and try to manage these flows towards their own organizational success, including their diversity goals.
Robyn Fadden – host: In your research paper Network Recruitment and the Glass Ceiling, you and your co-author show that a big part of increasing diversity, equity, and inclusion at work is how people become aware of job positions in the first place. Word-of-mouth is one of the main ways that people learn about jobs. What does word-of-mouth look like in the real world of job recruitment?
Brian Rubineau: Word-of-mouth is the primary way, job seekers connect with employers in the labor market, and this is true in the United States and North America broadly. It’s both the most common recruiting approach that employers use, when there’s a vacancy, it’s very common for employers to ask their current employees to try to identify good people that would be able to fill the vacancies that an organization has. In addition, it’s one of the most common search practices of job seekers — in addition to going to online tools and things like that, job seekers will also talk to their contacts, their own social network to try to say, Who do you know that’s hiring? What jobs are available out there? And use their contacts to try to identify these opportunities. So it’s the most common form of job matching that exists in the labor market.
Brian Rubineau: Now, this common form of job matching has some important consequences. One of the most reliable findings in social science is that our social contacts tend to resemble ourselves along many different dimensions of similarity. So gender, level of education, race, religion, language preference, a whole range of dimensions. Our social contexts tend to be similar to ourselves. And so when organizations that use word-of-mouth practices to try to fill their vacancies, which again is one of the most common forms of recruiting, it has obvious implications for diversity.
Robyn Fadden – host: I can see how that could have either negative or positive implications, depending on the composition of a workplace. But what does research, including your own, show about the current impact of word-of-mouth recruitment on diversity?
Brian Rubineau: The assumed relationship does not actually hold. So for a long time, the perception was that this was the old-boys network, right? The idea that if you have, say, mostly men in an organization and you have word of mouth recruiting that they’re going to replace themselves with most mostly men. Now, that tendency is there, and actually in the 1970s, a scholar named Rosabeth Moss Kanter at Harvard Business School, coined the term homo-social reproduction, the idea that the current generation of organizational members are reproducing and creating the next generation of organizational members. And because they’re creating, the next generation is very similar to themselves. It’s this homo-social reproduction aspect. But there’s also this possibility of introducing difference. And so just to give you an example of that, if you have mostly men in upper levels of an organization, and the men generate the kind of word-of-mouth contacts that they generate, as new employees are 90% male like themselves, and only 10% of the people they generate are female. Well, those 10% female contacts that are generated, those 10% applicants that are generated, those that are hired will then become able to reproduce themselves. And women are also more likely to produce other women in this kind of homo-social reproduction dynamic. And so, the process can actually be a tool towards integration towards desegregation.
Brian Rubineau: One of the insights of my earlier research was that if you can get underrepresented groups to refer — we call that word-of-mouth recruiting, sometimes referrals and referring — if you can get underrepresented groups to refer more, then you can turn this reproductive process into an integrating process, so that an underrepresented group can actually become more integrated. And you can desegregate a workplace that way. That was one of the insights of my earlier work. And since then, a number of organizations throughout the US — actually Silicon Valley has adopted this to quite an extent, also in the banking industry and elsewhere — have tried to use referring policies to try to promote diversity in the organizations. And that was an innovation coming out of this research that started probably around 2010-2015. Many organizations have embraced using word-of-mouth recruitment as a tool towards achieving diversity goals.
Robyn Fadden – host: So it became a conscious choice of organizations to recruit in that way, and really build word-of-mouth into their recruitment and hiring policies as a tool to achieve diversity goals. Yet there are still limitations to this approach. In your newest research on women’s underrepresentation in talent pipelines, what methods did you use to identify the limitations and what did you discover about those limitations?
Brian Rubineau: The initial work that I was doing to try to understand how word-of-mouth recruitment worked, and its relation with the segregation of the labor market, and to try to identify these opportunities to integrate the labor market, was done via a combination of mathematical modeling and computational modeling. The idea that word-of-mouth recruitment was this kind of homo-social reproduction process, I took that seriously. And we can just study it as a population process. So we can look at it, like an ecologist might look at dynamics of, of species in an ecological niche, or something like that. And you could look at various reproduction rates. And so what I was doing, what my research did was using mathematical models and computational models to look at these population processes to try to identify: Where are the opportunities to intervene? What are the levers that you can identify that could help promote diversity through word-of-mouth recruitment?
Brian Rubineau: And so a lot of this was kind of theoretical, initially identifying that word of mouth recruitment could promote diversity, by encouraging it or representative groups to refer more. As organizations started practicing these policies, then we went out — my coauthor, Roberto Fernandez, who’s at MIT Sloan, and I collected data from several organizations that were using word-of-mouth recruitment for their recruitment policies, whether they were trying to use it to promote diversity or not was largely irrelevant. But we were trying to identify what the effects of the word-of-mouth recruitment were for the demographic composition of the organization. We could follow some organizations over time and see what it was we see what was happening.
Robyn Fadden – host: You were looking at quite a bit of data to get worthwhile insights. But did your findings align with your expectations?
Brian Rubineau: What we found was a bit surprising. We kind of had these slightly contradictory findings. On the one hand, more word-of-mouth recruitment was generating more integration, was desegregating the workforce. But we were finding the strongest effects at the lowest levels of an organization. And thus the least strong effects at the highest levels of the organization. What the problem was when we actually looked for the data supporting the research that we had come up with, in our findings from our mathematical and computational models, we were finding that to some extent the processes we were observing were still recreating glass-ceiling phenomenon where women appear to have more barriers as you went to higher levels in the organization.
Brian Rubineau: One of the things that we had missed was that in our models – there’s this principle in modeling that you keep things as simple as possible, but no simpler — and so we had in our models we were looking at organizations with only one job. We kind of had this idealized organization with just one job. And we were looking at the gender composition in those single job organizations. Real organizations have multiple jobs and multiple levels. And one of the processes that highly influences the gender integrating effects of word-of-mouth recruitment is the fact that not only do people tend to reproduce job candidates who are like themselves, but there’s a level of the organization dynamics that comes into play. And that is, when we refer others for job opportunities, we tend to refer others for jobs that are at our level or below. And very rarely do we refer others for job opportunities that are at a higher level than our own job. And because we have this level specific referring tendency, then the upper levels of the organization, which tend to be more male dominated than the lower levels of the organization, preserve that male domination, that that male over representation, that’s, it’s more enduring than it is for lower levels of the organization. So the integrating effects of word of mouth recruitment are strongest at the lowest level and weakest at the upper level. And so that revealed a limitation and the opportunities of using word-of-mouth recruitment to promote diversity, because it’s going to work best for the lowest levels. And the lowest levels are already the ones that tend to be most gender integrated. And it’s less effective to promote integration at the highest levels of the organization.
Robyn Fadden – host: And the higher levels of organizations already have a problem with gender equality already, this is what other research has already shown, right?
Brian Rubineau: The research that has documentation about who serve as CEOs, for example of fortune 500, companies have shown very few women, I believe the proportion is something around 14%, that might be that might be too high, it might be 7%, or something like that – it is extremely low, and that’s serving the CEO. And then you can look at the C suite, and you can look at the top management teams, and you can look at upper level of the organization. And it’s still women are far underrepresented relative to their capability, their present the workforce, their presence in those industries, the training and capabilities that they have. So yes, the upper levels of organizations tend to be more male-dominated.
Robyn Fadden – host: Is word-of-mouth understood differently when recruiting for these higher-level positions? Is the recruitment process different at the executive level versus entry level, even thought people are still basically talking with each other about these positions. Since management studies is so interdisciplinary, it’s using data to show the facts, but it’s also shining a light on sociology and psychology around biases in the workplace. Why is it that word-of-mouth functions differently in the recruitment process for different levels of work positions?
Brian Rubineau: One of the questions is: Is there more or less word of mouth recruitment at higher levels in the organization versus lower levels in the organization? And the best evidence is that it’s present and important in all levels, but it might be present and important in slightly different ways. So it might be the case that word of mouth recruitments are actually more important at the upper levels, because in these positions where there are fewer positions, fewer opportunities that open up, and they tend to be higher paid positions, positions that have more authority and trust, then there’s the desire to try to check references more and make sure that if you don’t know the person you’re hiring, at least you know the people who know them. And that can be even more important. At the same time at these lower levels of lower level jobs, entry level jobs and things like that. A lot of times a referral provides a useful signal of reliability that can really distinguish between two otherwise very similar candidates. When you’re looking at hundreds of applicants, a referral can definitely make a difference. And that tends to make sure people who get referred are far more likely to be interviewed for a position — and it tends to be a fast track to an interview. So it tends to be important across all levels of the organization, but how it’s important can be different.
Robyn Fadden – host: So there’s still a glass ceiling when it comes to executive positions and gender. How could the limitations of word-of-mouth recruitment be countered at an administrative level? Whether through HR data or certain administrative processes? What does this show you about how the gender imbalance could be solved?
Brian Rubineau: One of the things it shows, and this is something to accent we knew before this research also, which is the processes that create gender segregation and other kinds of inequality and lack of diversity in the workplace come from a wide variety of processes. And so there’s not going to be just one tool that’s going to fix it. So we need multiple tools. So I think word of mouth recruitment and the idea of trying to get underrepresented groups to refer more is one such tool, but it’s incomplete. And we’re going to need other tools. So that so integration and promotion of diversity at upper levels, the organization can also be robust. And that word-of-mouth recruitment is not going to be the only tool to achieve that.
Robyn Fadden – host: Word-of-mouth helps job seekers get the interview and get in the door to an organization. But before reaching that point, there are a few more barriers to overcome before even applying for the job. There are gender differences between who applies for positions in the first place, who gets rejected, and who resubmits their applications. You and your co-authors dove into these questions in a recent research paper “Reject and resubmit: A formal analysis of gender differences in reapplication and their contribution to women’s presence in talent pipelines.” What did you find?
Brian Rubineau: Beyond the labor market itself, across a right wide range of realms, there’s this finding that when men and women seek a particular opportunity, and they get rejected, men are more likely to try again for that same opportunity. And women are less likely relative to men to seek that same opportunity. And we’ve seen this in patenting applications. You see this in Kickstarter projects. I see this in grant funding applications. And we see this in the labor market as well. There was a paper even saying that if college students fail a course, who tries to take the course, again, right, we see this gender difference. And so it has an obvious potential implication for women’s under representation in these talent pipelines.
Brian Rubineau: Some colleagues and I were looking at this, again, we were starting out with a mathematical model to try to understand the dynamics of these processes to try to look for opportunities to intervene and what might be the levers be. And so one of the things we find is that this can be a contributor to gender inequality in talent pipelines. But one of the important enabling factors is because this is a gender difference following your rejection. The segregating impact of this process really only kicks into gear in contexts where there’s actually a very high rejection rate. Because the differences between men’s and women’s reapplication rates might be significant, but they’re not huge, let’s say, you know, 13% of men reapply, and only 10% of women reapply. So there might be a significant gender difference, but it’s not necessarily huge. And so only in context, we have very high rejection rates. Will this come into play? Now one of those contexts where there’s very high rejection rates is the labor market — for any position rejection rates commonly exceed 80% to 90%. So this actually can play an important role in contributing to gender inequalities and women’s under representation in work contexts, particularly male-dominated work contexts.
Robyn Fadden – host: Could you give an example of those contexts, such as whether this unwillingness to reapply happens across levels of an organization or whether this happens in different fields?
Brian Rubineau: The data we were using to validate our mathematical model, our labor market example comes from an executive search firm. So this was actually a firm that does basically headhunting for high level management positions across organizations. The people who are who are applying through this executive search firm are accomplished managers, across the board, and they’re men and women. And the research found that once rejected for an opportunity and asked again whether they would be willing to apply for another opportunity, men were more likely to say yes, and women were less likely to say yes.
Robyn Fadden – host: These are people who already have many accomplishments under their belts, and who have access to a network where word-of-mouth could benefit them and encourage them to apply. Yet despite that, there are still many factors at play that deter them from reapplying, from the confidence of the job seeker to even a perceived bias of the interviewer.
Brian Rubineau: So it’s both. So women might be less likely to be hired. And in many, in a lot of male dominated contexts, women are less likely to get hired. But one of the important thing about this particular process is it doesn’t actually matter if women are more or less likely to get hired, is that most of the people who are applying get rejected, and then more women than men are removing themselves from continuing in that particular talent pipeline.
Robyn Fadden – host: It’s partly the odds, and if you reapply your odds are higher for eventually getting the job. So knowing this, should organizations encourage women to reapply after rejection? If organizations have robust diversity, equity and inclusion initiatives, should this be included as well?
Brian Rubineau: I’m so glad you asked that question, because that’s one of the obvious first responses to this finding that women are less likely to reapply than men. So a common intuitive response is, how can we get women to reapply more at the levels of men, so that can erase that gap? One of the things that we find in our research, which was surprising to us, and a nice opportunity was that that may not be the best intervention. It’s unclear whether and how an organization could influence who reapplies and at what rate. We were able to find one paper, a working paper, it might be published at this time, that was looking at the rejection letters, and it varied some of the text in the rejection letters, and it actually did find that that influenced who reapplied and it could change the reapplication behavior. So potentially, that opportunity might be there.
Brian Rubineau: But one of the things that we find is that just by trying to do something like lowering the rejection rate — we had said earlier that this dynamic was really kicked off by high rejection rate contexts — and to the extent that hiring or other kinds of selection contexts can lower their rejection rate, then it can really reduce the gender inequality that’s coming from this process. In a way that’s an intervention that’s really gender neutral. When we’re saying lowering the rejection rate. We’re saying lowering it for everybody. And so there’s a question of like, well, how do you do this some of the contexts I was describing, so things like patent applications, Kickstarter, and grants and things like that, those selection contexts are already trying to lower their rejection rate across the board for everybody. Right? The the patent office is trying to give applicants information to make it the most likely as possible that their application will succeed. Kickstarter similarly tries to provide information and opportunities to try to make it the most likely that everybody who’s posting their projects are going to succeed to try to reduce the rejection rate across the board. In hiring, it’s kind of interesting, because generally, you have you know, a single opportunity, you have a vacancy and you’re trying to fill it and you want to have a lot of people applying. So, what are you what are ways that you can reduce reach lectionaries In that context, well, there’s increasing. One of the problems with the selection process in the labor market is it can be very hard to identify good workers just from like a paper application, right, just the information on paper. And this is actually part of why there’s an AI revolution and train in selecting people.
Brian Rubineau: But what the research shows is that more and more if you can have job samples, work samples, realistic job previews, or if you can have job tryouts, then you can increase the likelihood of hiring somebody good. This idea of trying to broaden job tryouts. These alternate selection processes for jobs generally chances are you’ll have fewer people applying. Because right now, when you have a vacancy, many organizations have hundreds of applications per vacancy, because there’s very low cost to applying, you can just post your materials on a job site, and it’ll get sent to all the different job vacancies, if you’re actually needing to do some kind of job tryout, that requires you to actually do something and show up, it’s going to be a smaller number of people. And that’s an example of how you could actually both reduce the rejection rate overall, likely increase the quality of the decision making in the selection, and so that can have double benefits for promoting diversity.
Robyn Fadden – host: That’s an incredibly useful research finding. I’m wondering if organizations are taking new research like this into account in their Diversity Equity and Inclusion policies. Should their policies be updated more regularly to reflect not only market changes but research findings?
Brian Rubineau: Diversity in organizations is a goal like any other organizational goal in terms of productivity, efficiency, and those kinds of things. And so an on any other organizational front, organizations, ideally, set goals, ask, are we achieving our goals? And when we are we say yay, and say, How can we keep this up? And when we’re not we say, Okay, well, let’s examine why we’re not and try to improve things. And so like any other organizational objective, the promotion of diversity, promotion, a culture of inclusion, and these kinds of things, need to be constantly evaluated, are we achieving our goals? How can we improve? So research like mine and my colleagues’ are trying to give organizations tools and ideas to try to do that, it’s really unlikely that there’s going to be a single policy that’s going to be inviolate and unchanging, that’s going to help all organizations. What needs to happen is we need to constantly check, evaluate, and try to improve as we go, like an organization does with any other aspect of its functioning.
Robyn Fadden – host: We need to say here that these DEI policies are tied to organizational outcomes, that DEI strategy doesn’t stand alone and isn’t or shouldn’t be siloed. So in many cases organizations typically tie these policies to issues of talent, such as what talent are we missing out on? They don’t want to miss out on people who could further the organization’s goals, who could help the organization grow or adapt, and depending on the organization, actually benefit society in one way or another.
Brian Rubineau: Absolutely. The idea of underutilizing the human capital of large portions of the working population is neglected, inefficient, and you know, impoverishes us all. And I said earlier on that, you know, workers are flows, you can consider kind of employee human resources as flows to organizations. But also that flow is constantly changing, we’re having demographic shifts in terms of composition of workforce is immigration and the whole range of things. And so we want to try to make sure that, you know, all organizations are using the human capital as fully as possible, and that all people who want to contribute are able to contribute as fully as their capabilities. And so the idea of trying to make sure that that’s can be realized across organizations. I mean, it just makes sense. It’s part of the UN sustainability goals. It’s just good organizational sense. It doesn’t make sense to leave any kind of capital on the table. Unused or underused, including human capital.
Robyn Fadden – host: We’re talking capital here in a less cynical sense of using humans to create capital, but recognizing the human worth of people as people – and what they can contribute if provided with opportunities and environments that support them. So at this point we’ve touched on only two segments of your wider research that addresses equity, diversity, and inclusion – it’s such a large and complex social issue, but we have seen movement on it. We’ve seen organizations embrace both the research and what their employees are telling them. The problem is in the process of being solved piece by piece and across fields, including in management studies, but there’s obviously much work to still be done.
Brian Rubineau: Presumably, and one of the things that we found, is that our understandings of diversity constantly evolve. One of the revolutions that we’ve seen, just in the last decade or so are revolutions and understandings of gender and how the workplace has adapted to that. And as we grow as a society and as people, those are constantly changing. I think the effort to try to be inclusive is also going to be a constantly changing process.
Robyn Fadden – host: Absolutely. And that includes gender identity inclusion as well, such as human resources intake forms that include options for genders other than male and female. But that data and the amount of it, is, in relative terms, still fairly new.
Brian Rubineau: And my research, which does mathematical modeling, computational modeling, we’ve had, historically, and what I’ve published, has adopted kind of the binary assumption of gender. And one of the things that we recognize is will not that’s problematic, and so we need to revisit how we treat that currently, right, the numbers of people who identify outside the binary are small. And so a lot of the findings that are made under the binary assumption are can still be useful. But it’s important to have the research itself be inclusive and, and practical and applicable to real world settings to make sure that their research is also inclusive of our evolving understandings of gender.
Robyn Fadden – host: And that even though we’re talking about changing the binary, your research is shedding light on the fact that we still live in that binary and in a male-dominated arena when it comes to hiring and promotions, even though much is being done and has been done by organizations to address those biases. Thank you very much for speaking with me on the Delve podcast about these issues and how management research contributes to solving them.
Brian Rubineau: Thanks so much, Robyn. It’s really been a pleasure.
Robyn Fadden – host: Our guest today on the Delve podcast was Desautels Faculty of Management professor Brian Rubineau, talking about his recent research on gender inequalities in job recruiting and hiring. You can find out more about this research in an article at Thank you for listening to the Delve podcast, produced by Delve, the thought leadership platform of the Desautels Faculty of Management at McGill University. You can follow DelveMcGill on Facebook, LinkedIn, Twitter and Instagram. And subscribe to the DelveMcGill podcast on your favourite podcasting app.