Podcast Episode 1: David Pedulla

Transcript

Upon Further Review: Frontline Conversations with Dean Bobo
Episode 1 with David Pedulla

Division of Social Science
Faculty of Arts & Sciences
Harvard University

(Recorded 06/23/2021)

[00:00:06.07] LARRY BOBO: Hello, everyone. My name is Larry Bobo. I am the dean of social science at Harvard University and it’s my pleasure to welcome you to Upon Further Review: Frontline Conversations with Dean Bobo, where we will be holding talks with members of the Harvard faculty in the Division of Social Sciences. Our first guest is David Pedulla, who is Professor of sociology and a relatively new arrival here to Cambridge, Massachusetts. So welcome, David, and tell us a bit about this transition to the Boston Cambridge area which had to happen sadly during the COVID 19 era.

[00:00:43.92] DAVID PEDULLA: Excellent. Well, thank you so much, Larry, for having me today. And yeah, it’s wonderful to be here in Cambridge. I actually grew up in the Boston area. And so it kind of feels like a homecoming. And before coming to Harvard, I was on the faculty at Stanford in the sociology department there. And then last July, made the transition here to Cambridge. And so it’s been wonderful to be here. My colleagues have been incredibly welcoming and it’s been great to get to know students over Zoom, but I’m very excited for what lies ahead and getting to kind of get fully integrated into the community here at Harvard and actually get to do some in-person communication, some in-person meeting with faculty and students here.

[00:01:24.03]

[00:01:24.51] LARRY BOBO: That is terrific to hear. We were all so excited when you decided to come and your work is really enlivening the discipline and reaching beyond the boundaries of sociology. As we all know, American society the world at large in many respects is being reshaped by worsening economic inequality. And your work really as a stratification sociologist bears very directly on those questions though you’ve got it quite distinctive slant on it. And so maybe I would make my first question to you before we kind of really leap into the details of your new book. For you to tell us how the world of work has changed, who is it that finds good jobs, who does not and kind of why?

[00:02:19.44] DAVID PEDULLA: Absolutely. So when we think about the changing nature of work in the United States, there’s a lot of different axes that sociologists economists and social scientists care about. One of the main areas which we’ll get into when we discuss the book is the emergence of various types of non-standard contingent and precarious employment in the United States. And while these types of positions have existed for decades or even centuries since around the 1970s or 80s, we have seen an increase in the use of temporary help agencies and temporary employment, which is one thing that I look at in the book.

[00:02:55.96] So in addition to the rise and increase of the use of temporary help agency workers, we also see other axes of kind of precariousness or inequality in the labor market. So one of the things that a colleague of ours Danny Schneider over at the Kennedy School studies is schedule precarity and schedule uncertainty and how that impacts workers. And so that’s an access that’s gotten a lot more attention recently and that’s incredibly important.

[00:03:21.79] And then we can also think about the rise of the gig economy and platform jobs. And so we know that workers are laboring through things like TaskRabbit and Upwork and Uber and Lyft. And well, these jobs make up a relatively small proportion of the overall economy. They have a really important role in policy conversations and really thinking about what does the future look like in terms of technological platforms, and how workers are engaging through those different types of platforms, and what are the consequences of that going to be for various types of inequality?

[00:03:52.90] LARRY BOBO: So would it be fair to say, and kind of what’s increasingly becoming a common sense term for it, that your work is focused on the gig economy? Or is it a slightly different way we should be understanding what piece of this changing world of work that you’re really wrestling with?

[00:04:09.91] DAVID PEDULLA: So I actually take up a slightly different piece of the economy than the gig economy. And so oftentimes when I’m talking to journalists or family and friends, when I talk about studying non-standard deployment, immediately say, so you study the gig economy. And so it’s an interesting kind of distinction. The gig economy generally thinks about as I mentioned, platform based work. Technological work through Uber, Lyft, TaskRabbit, Upwork, et cetera.

[00:04:34.35] And we know it’s hard to get great estimates of the proportion of the labor force that works through those platforms, but the best estimates are somewhere at least before the pandemic of about 1%. I’m focused on other types of non-standard mismatched and precarious work. So things like part time employment, which again before the pandemic was somewhere between about 16 And 18% of the workforce temporary agency employment, which is usually about somewhere about two and 1/2 of the labor force.

[00:05:02.43] And I also look at things like skills under utilization, which it’s hard depending on how you define skills utilization. It can be somewhere around 20% of the workforce, 20% of the college educated workforce might be defined as being in a job below their skill level or a job for which they’re overqualified. And so I’m kind of looking at a different slice of nonstandard and contingent work than would be normally captured by the gig economy.

[00:05:26.67] LARRY BOBO: I see. Very good. Now, the major title of your important new book is Making the Cut. Now, why Making the Cut?

[00:05:37.38] DAVID PEDULLA: So it’s actually really interesting titling books. And so we had a bunch of back and forth kind of thinking about what the title would be. Initially, the title of the book was Contingent, kind of thinking about contingent workers. And then the book really focuses on what we refer to as the demand side of the job matching process. So what I’m really interested in is how do employers make decisions about who to hire? And so making the cut really captures two aspects of the world of work that I’m interested in from these kind of employer side perspective.

[00:06:11.26] The first is how do employers actually make the cut between the appeal and the people? Who do they decide gets in and who do they decide stays out? So making the cut that way. And then thinking about making the cut kind of from the perspectives of the worker and what does a worker need to do to make the cut. What do they need to do to get into an organization to kind of maintain a positive valence within the eyes of the employers who are making these decisions. And so it had kind of a dual meaning both from the perspective of the worker as well as the perspective of the employer.

[00:06:42.85] LARRY BOBO: I see. So and as you’ve already stressed, this kind of issue of hiring process, making decisions about who’s really in line for a position and who finally gets a position in a way it’s something that’s been a concern for social scientists for a very, very long time, but there’s still something hard about getting a grasp on that for social sciences. What are the challenges when it comes to trying to examine hiring decisions and decision making?

[00:07:15.98] DAVID PEDULLA: It’s a really important question because as we hiring decisions are often made behind closed doors. They’re made by hiring managers and their teams and various recruiters within an organization and it’s very rare occasionally ethnographers or others will get access to those decision making rooms. But in general as scholars, we have to study the hiring decision making process from outside of those actual rooms where the decisions get made.

[00:07:39.49] And so there’s two key techniques that I use in the book and that scholars have used to really try to understand in depth the hiring process. The first is what’s often referred to as audit studies or correspondence audit studies, where what the researcher does is send fictitious job applications to or in the case of some original work in this area, actually sending out actors or testers to apply for a job. So you can either do it with actors or now it’s more common to apply for jobs online. So do it with online applications.

[00:08:11.86] And what the researcher can do is randomly manipulate certain aspects of the job application. And so you keep everything consistent except the one thing that you want to see whether that has an effect on a hiring decision. So these have been great at studying things like racial discrimination, gender discrimination. And what I do in the book is manipulate different employment histories on people’s applications. So randomly assigning someone a year of part time work or a year of unemployment or a year in a temp agency.

[00:08:40.09] And then what you can do is you can track employers responses to each of those applications, which we call a callback so did the applicant get a callback for a job and by comparing the callback rates between the different conditions that you’ve manipulated as the researcher, you can get a sense of the causal effect of one given attribute on hiring decisions. And that provides really nice insight into how these decisions get made and what the kind of outcomes for workers are with these different backgrounds and histories. So that’s one branch is really kind of the field experiments are audit studies.

[00:09:09.91] The other is using in-depth interviews and actually talking to people. And so I wasn’t able to get access to actually being in the room when these decisions were made, but I did get access to have in-depth conversations with hiring managers staff and consultants recruiters to really talk with them about how they make decisions, the things that they care about, really asking them what’s important to you during the hiring process.

[00:09:32.92] LARRY BOBO: I see. And so tell us a little bit more then about your field experiment that you were actually in a position to get a pretty large sample of test cases in effect.

[00:09:47.95] DAVID PEDULLA: Absolutely. So in the book, I present evidence from over 4,000 job applications that I sent to four different occupational groups across five different geographic areas in the United States. And there were five key employment histories that I manipulated on the resumes. So I signed some workers a full times standard kind of seamless employment history. Some workers got one year of part time work leading up to applying for a new job, some got a year of temporary agency employment, some got a year working in a job well below their skill level and some got a year of unemployment.

[00:10:20.48] And then what I did is as a sociologist, I really care a lot about key axes of social stratification. So issues of race, gender and how social categories map onto and shape evaluation processes. So I overlaid those five different employment histories with race and gender, which I signaled on the job applications using racialized and gendered names.

[00:10:41.20] And so what I was able to do is look at what is the effect of having a part time employment history for say, a man versus a woman? Or what’s the consequence of having a spell of unemployment for a white individual or an individual that’s likely to be perceived as white, versus an individual that’s likely to be perceived as African-American? How might there kind of spell of unemployment be read differently by employers. And so the book enabled me to kind of get really nice causal traction on how those different categories affect callbacks and then how they vary with the race and gender of the job applicant.

[00:11:13.19] LARRY BOBO: I see. So let me wrestle with a piece of this, then, of what’s going on in the framing of both pieces of your work. And that’s kind of an idea– not so much a theme, but an idea that’s core to your work. And that is that virtually all of us are working with some notion of what the good job is, or what the good job used to be, as a more common experience in the American economy. So when you’re writing, and then when you end up talking with employers and what we think are in potential employees and employers’ minds, what is the good job? What is that normative vision?

[00:11:59.83] DAVID PEDULLA: So generally when employers think about a good job, they think about a job that is full-time, that has kind of in addition to being full-time is kind of has an expectation of continued employment. So it’s not temporary. And that it’s well-matched to the worker’s skill level. So those are a few of the key aspects that often employers see as a good job but also something and importantly, that many workers see as a good job.

[00:12:23.42] Many workers want to be in a position where they expect continued employment, where they’re working the number of hours that they want, which for many workers or most workers is a full-time job, and that really is utilizing their skills, providing them with opportunities for advancement, et cetera. And so what I find is that when workers deviate from this kind, of normative conception of a good job that many workers and many employers both hold, oftentimes, although not universally, they face various types of stigma or various penalties when employers are evaluating them.

[00:12:57.89] LARRY BOBO: And we’ve seen a pretty big transformation I take it, in the extent to which the labor force writ large is characterized by that type of employment, the employment that is full-time, that is expected to continue, and that say, provides benefits and real stability for a person in the labor force?

[00:13:19.60] DAVID PEDULLA: So we have seen a lot of changes in the labor market. We know very clearly that there’s been a significant increase in temporary employment in the United States. And again, it’s a little tricky to figure out skills underutilization and how it’s transformed over time. But there are some nice estimates that indicate that increasingly particularly amongst college-educated workers, an increasing fraction of college-educated workers are in jobs below their skill level. And we see workers experience unemployment, and that fluctuates significantly with the business cycle. So that’s going to move in and out over time depending on kind of where we’re at with the broader economy.

[00:13:54.99] And so these things really have come together where even though it can be really challenging for workers to find these quote-unquote good jobs or the jobs that align with a normative conception of a good job, many of them end up in positions that deviate from that good job, outside of their own desires or even their own control. They may lose a job and then end up having to take a position below their skill level or only be able to find part-time work or only be able to find employment through a temp agency. And so yeah, we see workers that are kind of ending up in these positions through no fault of their own and then can really struggle to try to find employment in the future. As the evidence in the book demonstrates.

[00:14:35.11] LARRY BOBO: I see. So if you were to just do a slice kind of across the top, for the moment setting aside how it may differ between men and women or how it may differ between Blacks and whites, which one of these different employment trajectories kind of has the biggest stigma attached to it? Would it be some nonstandard employment, the mismatch condition, or someone who’s really had a very precarious employment history, what puts you furthest behind?

[00:15:06.61] DAVID PEDULLA: It’s a great question. And so one of the things that I argue in the book is that it’s actually quite difficult to fully disentangle these types of employment histories from the social categories of the workers who embody them because things like part-time work are so heavily gendered and unemployment is so heavily racialized. That being said, I do find consistent evidence that taking a job below your skill level is as penalizing for workers in general as a year of unemployment. And so we do see–

[00:15:35.69] LARRY BOBO: Wow. Really?

[00:15:36.38] DAVID PEDULLA: Yes.

[00:15:36.65] LARRY BOBO: That’s pretty strong.

[00:15:36.97] DAVID PEDULLA: So it’s quite strong. And so it might be easiest to think about the white male condition, we’re looking at kind of white men who move into a job below their skill level, the callback rate for workers who– for white male workers who move into jobs below their skill level is about half of that– the callback rate of white men who remain in full-time jobs at their skill level. And the callback rate is very similar between the skills underutilization condition and the unemployment condition for those white men. And so we’re seeing really strong negative penalties particularly for white men in that skills underutilization condition, but really across the board, we see we see penalties for skills underutilization

[00:16:20.03] LARRY BOBO: I see, wow. So if we were to now fold in more substantially a concern with how the experience of men and women differ, it seems– I don’t want to say it’s day for night, but part-time unemployment or part-time work doesn’t seem to carry near the cost for women in terms of future job prospects or movement as it does for men.

[00:16:43.09] DAVID PEDULLA: And that’s really what I find in the book. I find that particularly when you look at white individuals or individuals whose names are likely to signal whiteness or not be particularly racialized for employers, I see that men who move into part-time jobs face penalties similar to unemployment again. So their callback rates are cut roughly in half from the full-time condition. Whereas for women, I really don’t see strong penalties of part-time work.

[00:17:10.28] And I was really interested in this finding. So I got this finding from the field experiment, and part of what I wanted to do in the book is help to unpack that and really understand what’s going on in employers’ minds, how are employers thinking about these different types of positions? And so the in-depth interviews were incredibly useful in really trying to unpack and understand the mechanisms that might be driving those effects in the field experiment. And particularly, this gender in part-time work finding was one that I really wanted to explore with the hiring managers and recruiters.

[00:17:40.52] And what I found is that for women who move into a part-time job, employers have a really quick kind of narrative that they can use to explain that movement, which is around parenthood and caretaking responsibilities for women. And we know that parenting and caretaking responsibilities for women have huge negative consequences in other domains of the labor market in terms of wage setting, and promotions, and all sorts of employers’ conceptions about women and motherhood. And so I want to– I think it’s important to acknowledge that this kind of narrative around women and motherhood and caretaking certainly has costs.

[00:18:14.06] In this particular case, what I find is that it provides a quick narrative and script for employers to dismiss womens’ part-time work as not being particularly problematic. Whereas for men what I found is a man who took a part-time job led to questions about their competence and ability. Real concerns about why were they in this part-time job? What’s quote-unquote what’s wrong with them or what’s up with this guy that he couldn’t get a full-time job? And really leaves these kind of looming questions that employers fill with kind of negative ideas about the male worker who was in a part-time job.

[00:18:49.19] LARRY BOBO: I see. So there are kind of two threads here I want to pick up on, one is that different types of employment histories send a signal or at least are read as sending a signal. And at the same time, employers are in their own minds kind of crafting a story, I think you call them stratification narratives or stratification stories, about what that signal means for them. Can you elaborate on those points?

[00:19:19.68] DAVID PEDULLA: Absolutely. And so yeah, I use the term stratified stories in the book to really talk about the divergent narratives that employers weave about employment histories depending on the sociodemographic characteristics of the worker. And so in this case with part-time work, part-time work sends a signal, and employers make meaning from a part-time employment history but the meaning that gets constructed is shaped by the social position of the worker. And so essentially, for women, it’s a story about caretaking and parenthood.

[00:19:49.04] For men, it’s a story about competence, ability, or being a dud or having something wrong with them. And so the same employment history gets read and kind of ascribed meaning and given a narrative in a very different way for a man and a woman. And that’s the kind of sort of data that I’m able to draw on in the book to help understand why we see different patterns in the field experiment for say men and women or for white applicants and African-American job applicants.

[00:20:15.77] [MUSIC PLAYING]

[00:20:29.37] LARRY BOBO: Now, we’ve done a little bit of a dive more fully into the gender differences that operate there. Now, how about with respect to race and perhaps in particular, with the need to have this more intersectional eye, the intersection of race and gender here. So how do the experiences from African-American men say differ from those of African-American women and not just from their white gender counterparts?

[00:20:56.43] DAVID PEDULLA: Absolutely. So what I find in the book is that racial discrimination against African-American job applicants is severe and very strong and persistent. And this is backed up by a huge number of studies across the social sciences by different research teams. And there’s some great meta-analytic work by Lincoln Quillian and colleagues that came out in PNAS a few years ago, where they actually pooled all of the audit studies of racial discrimination in hiring that they could find in the United States. And then they looked at the trend over time in those discrimination estimates. And what they found is that discrimination against African-Americans has not declined since 1989 in the United States across these different audit studies.

[00:21:37.16] And so it wasn’t particularly surprising, although, it’s quite troubling that I saw this kind of persistent discrimination against African-Americans. When I kind of merged that in with gender and these employment histories we get some really interesting patterns that emerge. So what I find for African-American men is that discrimination is so severe at the hiring interface that in most conditions you see very little variation in the callback rates for African-American men because there’s so much discrimination kind of baked-in to the kind of callback rates that they receive already.

[00:22:16.46] So I find essentially no difference in the callback rate between African-American men who were in kind of seamless full-time continuous employment, and African-American men who had a year of unemployment, a spell of skills underutilization, or a part-time job. For African-American men, the one condition that kind of jumped out is that when they had moved through a temporary help agency, they actually received higher callback rates than they did in the kind of full-time condition.

[00:22:44.08] LARRY BOBO: Yeah, that was fascinating as if the idea that a temporary agency had taken a gamble on you reduced the likelihood that you fit the negative stereotype in the eyes of another potential employer?

[00:22:58.33] DAVID PEDULLA: Absolutely. And that’s really what came out of the interviews again, is talking to a bunch of the hiring managers and recruiters because I did the interviews after the field experiment, one of the nice things was at the end of the interview right after I had let them kind of talk about all of these things in their own terms, and give me their own narratives and the way that they think about these things. I actually was able to present some of the findings from my field experiment to the hiring managers and have them talk to me about how they would interpret them, how they would think about them. Which methodologically for me was fascinating as a scholar to have the actual people I was studying in the field experiment, look at the data with me and then try to kind make sense of it.

[00:23:34.92] And one of the things that really emerged out of that in this case around temporary help agency employment and African-American men, was this idea of, yeah, a temp agency took a risk on this person. And also that temp agencies often have very high levels of kind of screening, criminal background checks, drug testing. And so there’s a set of overlapping stereotypes that come out in the literature on employers have stereotypes of African-American men that may overlap with having a criminal history, et cetera. And that there may be something offsetting about a temp agency having prescreened this person or provide some sort of sense that they’ve already been through a vetting process that may deactivate some of those stereotypes.

[00:24:13.55] LARRY BOBO: I see. That is really interesting. Now, if I recall correctly, and you get to tell me whether or not this was really a statistically significant difference, for the experimental condition in which you had an African-American male who was employed full-time in a credential appropriate position, had a 50% lower likelihood of getting a callback than a comparable white male was that correct?

[00:24:37.65] DAVID PEDULLA: That is– so yes, in the– between the African-American men and the white man in the full-time seamless continuous employment condition, it was about half the callback rate. That’s exactly right.

[00:24:47.89] LARRY BOBO: Wow.

[00:24:48.08] DAVID PEDULLA: Yeah.

[00:24:48.32] LARRY BOBO: Wow. That is a pretty steep penalty.

[00:24:49.97] DAVID PEDULLA: It’s a steep penalty. And again, that effect size is similar to what we’ve seen in previous studies as well.

[00:24:57.35] LARRY BOBO: Yeah. And now what about the story for Black women, what’s happening there?

[00:25:03.40] DAVID PEDULLA: So for Black women, what I find is that there’s discrimination and penalties that they face compared to white women. What’s interesting in the case of Black women is that the penalty is relatively consistent across all the different employment categories. And so I find racial discrimination against African-American compared to white women that’s diffuse as you might want to think about it, that kind of isn’t as contingent on the employment history. That kind of the racial discrimination has picked up across the board.

[00:25:32.00] LARRY BOBO: So if you were reflecting back on the data you collected, is there anything that really kind of hit you as a surprise or a departure from what you expected upon entering this, collecting these data, and pulling together what your results were showing?

[00:25:50.96] DAVID PEDULLA: So I have to say I was a little bit surprised at the magnitude of the effect that white men faced for part-time employment, and the limit, the really no negative effect for women. I thought there might be something in the middle. And we really see these kind of strong divergence by gender. I also have to say I was quite surprised that the callback rates for African-Americans, when you pool men and women for African-Americans in full-time conditions, was virtually identical to the callback rate for African-Americans who have been unemployed for a year. And that those were actually almost identical in terms of their callback rates was a little bit of a surprise to me. I thought that we might see again, something a little bit different there.

[00:26:35.92] And then in terms of temporary help agency employment, I was really struck by this finding for African-American men, where we see this positive effect. And it’ll be interesting to see if that shows up in other research or if that emerges in other studies as well. And then I think I have to say, I was a little surprised that I don’t find strong negative effects of temp agencies for any workers. I don’t find for white men, white women, Black women, Black men, there’s no group for whom temporary help agency employment is statistically significantly a negative effect. And that was a little bit surprising to me, that employers are not seeing this as negatively as some of these other employment histories.

[00:27:17.64] LARRY BOBO: Yeah. Wow, that’s really interesting. So some of your interpretation of the results here suggests– or let me put the question more provocatively and do it this way, in your, especially your in-depth interviews with employers or those making hiring decisions, especially if you present them with the– after you’ve presented them with the data you’ve collected and what you found, do any of them develop kind of a sociological insight, do they get to a different appreciation?

[00:27:49.74] Because it seems to me that part of what’s going on here is sort of the classic actor-observer bias, right? When we look at someone else and an action they’ve taken, or in this instance a description of their employment history on the page, we’re assuming that that history is driven by their characteristics and choices, not by the larger context and circumstances in which they’re embedded, right? And so is there any sign that those making hiring decisions learn to think a little more sociologically, and maybe invest a little more time and energy in reaching judgments about who might be able to perform a particular type of job?

[00:28:31.90] DAVID PEDULLA: So it’s really interesting, I did not find in general, a lot of like serious moves towards sociological thinking about these things in the employers that I talked to. But I will say that there were certainly a few employers I spoke with who did kind of from the start have more of a kind of macro-level understanding of these things. And it was– doing interviews with employers was really interesting because I did find that they were very hesitant to talk about certain things. So they were very hesitant to talk about race.

[00:29:04.74] In terms of anything in kind of the early parts of the interview before I had presented the findings from the field experiment, just asking them general questions, thinking about things they looked for, asking very broad questions about might that be different for you depending on race or how might race play into this for some other people in your field? Trying to get at it in a bunch of different ways. And they were very hesitant to kind of talk about race. They were a little bit more willing to talk about gender.

[00:29:31.47] And then one thing that was fascinating that I wasn’t expecting and I don’t have age variation in the book, and so I wasn’t really keened in on age variation. But age came up in the interviews a bunch as employers kind of being willing to talk about not wanting to hire older workers or things like that. And they were much less guarded about the age axis of variation than they were about race and gender, which I thought was quite interesting.

[00:29:55.74] LARRY BOBO: Yeah. That is fascinating. So out of all your findings now and your reflection on this, if you were giving advice to job seekers, people who are putting their resumes into these various online systems for job applications, what should they be thinking, what should they be doing? Should they add a little paragraph on an explanation of my job history for example?

[00:30:24.78] DAVID PEDULLA: Absolutely. And this is something I really struggle with that people often ask, friends and family who have read the book are like, oh, I have a son who lost his job or a daughter who’s doing this, what should I tell them? And it’s so complicated to offer individual-level advice given that what I’m able to look at in the book are really these aggregate patterns and aggregate trends. I do think there may be some utility in offering explanations for say, an employment gap, or why someone was working part-time in a cover letter for instance, and that sort of thing.

[00:30:54.15] Although, I’m a little hesitant to offer that advice given there’s some great work by Kate Weisshaar at UNC, who’s found that when people have taken time out of the labor force and signal in their cover letter that it was for caretaking responsibilities, they actually face stronger penalties than workers who were just unemployed without giving a reason. And that’s what gives me a little pause about encouraging people to tell these narratives because you never know exactly how it’s going to be received or what inferences an employer is going to make about the type of narrative that you apply. And so to pivot a little bit, I think there’s some great room in the social sciences to do work in this area really testing out different explanations and explanatory frameworks for say part-time work or unemployment, and whether you can actually reduce some of the stigma and bias depending on the narrative that’s used. And so you know, I think it would be great for folks to kind of think about that as an avenue forward that could have real policy impact and offer some real guidance to job seekers as they think about approaching the labor market.

[00:31:55.21] LARRY BOBO: And what about on the race discrimination front, is there anything here that gives us leverage on kind of breaking down what seems a pretty potent pattern of bias against African-Americans in the workforce?

[00:32:10.21] DAVID PEDULLA: Absolutely. So in the book, I had less kind of traction on that set of issues. But in my other research and other work I’ve done, have really thought a lot about what levers exist to reduce discrimination and bias? And so with my former mentor and advisor, Devah Pager, who tragically passed away in 2018, before she passed, we actually organized a conference here at the Radcliffe Institute where we brought folks together with the sole goal of answering the question what works to reduce discrimination and bias?

[00:32:39.99] And out of that we kind of put together a report with different people who came to the conference with different sections on the tools that we have that exist to reduce discrimination and bias. And a few of them just that I’ll note quickly that I think are important are, one is for companies who care about issues of diversity, equity, and inclusion, and really kind of getting some traction on reducing bias and discrimination, having clear benchmarks and goals in terms of data. Figuring out what those metrics are going to be. And then the key point here is making those data available to key stakeholders so that those stakeholders can hold you accountable. And so that’s one thing that really seems to be quite important.

[00:33:25.26] A second area is as we’ve seen the increased use of technology in the hiring and recruiting process. To really get out ahead of using any technology, to make sure that how it was programmed and developed and the algorithms that are used, don’t include biased data and aren’t leading to bias upfront. And then having a continuous monitoring process to make sure that bias isn’t getting introduced into the data as you continue to use say an applicant tracking system or algorithmic screening process.

[00:33:53.32] And then the last– or one other thing that comes out of the report that I think is useful is the importance of having buy-in across different levels of the organization, and particularly from managers and front-line managers into whatever diversity, equity, and inclusion initiatives are going to be brought to the organization. So you don’t want to have top-level CEOs, and C-suite executives mandating what’s going to happen, and then kind of pushing that out to the rest of the organization where you might be grafting something onto the organization that isn’t a great fit. But if you can get buy-in early on from key stakeholders about what you’re doing, you’re going to have much more success over the long term. And so I think those are a few of the areas that I think companies would be well-suited to pursue and think about as they approach these set of issues.

[00:34:36.51] LARRY BOBO: I see. And let me step back a bit and do now a slightly more academic track for a second here on maybe two sides or three. One would be how do you feel as someone who works in this arena of work and discrimination about the character and caliber of the research? I mean, there’s been a pretty big evolution right, and change in how we go at these issues.

[00:35:05.52] I think about some of the work that would have been done in the past on things like let’s say, discrimination in wages, and what people were earning. And it was considered satisfactory to just point to the residual, you did a regression equation predicting earnings, put in education, put in age, time in the labor force, and maybe if you had any detail, other human capital attributes. And whatever was left distinguishing Blacks from whites, that was discrimination, that’s how much was out there. So where are we today kind of intellectually as a field, as an area of scholarly endeavor, with respect to work and employment?

[00:35:48.60] DAVID PEDULLA: I think we’ve moved in a direction where there’s a higher level of skepticism about making claims that something is discrimination or that something– really attributing something to discrimination. At this point, I feel like in the evaluative process in the academic community, going through peer review, presenting at conferences, these sorts of things, making strong claims about discrimination and bias often requires, or is often held to the standard of some sort of exogenous variation, either a true experiment where the researcher has control over manipulating the race of a target, and then estimating the effect based off of that manipulation like we do in these audit studies, or having some sort of quasi-experimental variation where you can estimate these sorts of effects.

[00:36:35.55] Pure residual-based models out of a regression framework, I think are facing increasing skepticism. And certainly, when I have students who want to work with me around these sets of issues, I try to dissuade them from kind of pursuing a project that relies solely on that approach at least, if they want to be thinking about these issues of bias and discrimination. Because I think it can be really hard to feel confident in that sort of interpretation without some sort of exogenous variation, either experimentally or quasi-experimentally.

[00:37:07.00] LARRY BOBO: So let me think about the kind of field experiment that you’ve done. I know that in the past, there have been some fairly trenchant critiques of auditing studies, the kind of matched pair studies and those as somehow not being as strong a test as many of the initial proponents would have argued for. And somewhat relatedly, more of a kind of scholarly ethics question about sending out what are really not real resumes in response to real job postings. So on the one hand, does the field experiment get us out of some of the challenges to matched pair auditing studies, and are there any ethical concerns or qualms in sending out what are basically kind of fictitious resumes out to real employers?

[00:37:59.82] DAVID PEDULLA: Absolutely. So I think with the correspondence studies where you’re sending out resumes, have helped to address some of the methodological critiques around the early audits that were often in-person actors, where there were concerns about small differences in appearance or simply what someone’s wearing or slight differences in physical attractiveness could be confounding the race effects. And so having job applications and resumes I think is a way where you can much more carefully standardize the comparison between say, a white and a Black job applicant.

[00:38:31.92] Of course, that introduces a new set of concerns. In addition to the ethical things that you mentioned, which I’ll come to in a second, it also the signal of race that is often used in these studies is names. And names and their intersection with race introduce a whole different set of methodological issues around what might be confounding racialized names, might class signaling be part of the story here? And so there’s I would say, like a little cottage industry of methodological work that’s really trying to think through names and what they signal and how to do that well.

[00:39:02.13] On the ethical front, this is an issue that I take really seriously, and I think as we do this work, we have to be really cautious and careful that the benefits of the knowledge that are produced by these types of studies are outweighing any sort of risk or burden of time that we’re placing on employers. And I think that having a body of research that clearly documents racial discrimination and how it varies, and potential ways to reduce it, is of utmost importance right now.

[00:39:31.95] In the social sciences and the world, we’re experiencing significant racial upheaval and racial contestation, and thinking about the kind of long-lasting consequences of slavery and other forms of racial inequality in the United States. And I think as social scientists, we have an obligation to do what we can to contribute to those conversations and to really play an important role in documenting the persistence of racial inequality. And this is really one of the best tools we have to do it. And so I think we need to think about the ethics, we need to ensure that IRBs are signing off on our research and that they’re kind of vetted through experts in research ethics. But I do think that this work is important and necessary and that we need to pursue these kind of gold standard tests of discrimination to understand how they’re playing out across the economy.

[00:40:22.08] LARRY BOBO: Great. Thank you. Let me ask another kind of in the academy question, and that is when you think about research in this area of kind of modern employment decision-making processes, and especially discrimination, how does your work and approach as a sociologist say differ from or overlap with that being done by our friends and colleagues in economics? I mean, is there a significant difference here, or have the differences between the two disciplines in engaging these issues really narrowed substantially over the years?

[00:40:56.62] DAVID PEDULLA: My sense is that there’s actually a fair amount of overlap. So I don’t think there’s a need to kind of draw stark disciplinary boundaries. I have great friends in economics who are doing fantastic work on racial discrimination and gender discrimination. And I think those conversations can be really, really powerful and useful between the disciplines.

[00:41:13.60] I do think one area where there might be some differences, is in sociology, we use a kind of breadth of methodological approaches, everything from experiments, to survey analysis, but then we also have a whole qualitative toolset that’s very common in the discipline, from ethnographic work to end up interviewing. And so I do think that something sociologists can bring to the table is really thinking about the ways that qualitative research can be put into conversation with large-scale quantitative analyses and experimental work to help think about mechanisms and underlying processes that are driving inequality.

[00:41:47.59] And it’s one of the things that drew me to sociology as a discipline to kind of have different methodological approaches in conversation with one another about these key theoretical and empirical issues. And so I do think that’s an area where sociology may differ slightly from economics. Although again, I think the differences in a lot of the work are relatively minor and I’ve really enjoyed the conversations I’ve had with my friends in economics around these sorts of issues.

[00:42:15.37] LARRY BOBO: Great. So what has the reception been to the book? Now that it’s out there kind of in the world and other folks are digesting it and engaging it, what are you hearing? And how are you feeling about the feedback?

[00:42:27.85] DAVID PEDULLA: So it’s been great. I’ve gotten some really nice feedback from colleagues and from students. And it was really fun, actually this spring I had a student who was enrolled in my undergraduate research design class who showed up to office hours in the second week, and was like, I just got your book. I’m so excited to read it and came to talk about it. And I think that’s the best feeling in the world.

[00:42:47.69] LARRY BOBO: Absolutely.

[00:42:47.89] DAVID PEDULLA: To have one of your students find your book and want to talk about it. So that was really exciting. And it’s really interesting. I think one interesting thing about the book is it came out in April 2020, right as the pandemic happened. And so everyone’s lives were kind of turned on their head as the book was coming out. And in some ways, it was not front and center on my mind at that moment either.

[00:43:11.11] I do think that because the book is interested in the effects of these different types of employment histories that there’s real room for the conversation now and in the coming months as the economy is recovering, as employers are ramping up hiring, and thousands and thousands of workers have been pushed into these types of positions that I studied in this book, moved into part-time jobs, had to take jobs below their skill level to make ends meet, haven’t been able to find work and are unemployed. And so I’m hopeful that potentially some of the findings here will be useful to the conversation moving forward as the economy is recovering. And I think they do open up kind of an interesting set of questions about in the context of the pandemic and the recession where so many workers have experienced these positions. Might some of the findings actually look different moving forward if so many more workers–

[00:44:00.58] LARRY BOBO: You’ve actually anticipated what was going to be one of my two last questions. So how does the COVID-19 era, and our hopes for coming out of it, either amplify or change the story that the book tells?

[00:44:12.18] DAVID PEDULLA: And I think that’s exactly the question, and something I’m hoping to pursue some work on. And I’m hoping other folks will continue to do some work in that area because it will be really interesting, certainly empirically just to know does the negative effect of unemployment look different in a post-COVID world? Does taking a job below your skill level mean something different?

[00:44:33.04] But I think it also provides a case of some really interesting or insight into some really interesting theoretical questions about when these positions become more normative. When they are more easily disentangled from the individual worker, and more easily attributed to a structural feature of– the pandemic affected so many people’s lives in so many ways and the economy in such a broad way, that it’s much more– it would be much easier for an employer to attribute a spell of unemployment to that rather than to something’s wrong with this individual worker. And I think it provides some opportunity to get some traction on those questions as well.

[00:45:04.14] LARRY BOBO: Yeah. So closely related to that observation, what’s next for David Pedulla then in this arena? Is it really to focus on the recovery from COVID-19 and its impacts on the economy or was there some other intellectual project that you had next on the burner?

[00:45:23.50] DAVID PEDULLA: So the project– I’m hoping to do some work around COVID and the recovery in the coming year. But actually as a next step to the book project, and leading up to– before the pandemic had finished collecting data for a project that was really trying to think about racial discrimination and gender discrimination, not as a kind of individual decision-making process but rather situating it within the organizations where those decisions are made.

[00:45:47.47] And so as sociologists and social scientists who have studied racial discrimination, we’ve been really good at generating estimates of racial discrimination effects in these field experiments. Folks in psychology and social psychology have done fantastic work getting at the underlying cognitive and social psychological mechanisms driving those processes. And we have some great macro-level work looking at state and national level variation in racial and ethnic discrimination. But it’s been really hard to look at how organizational context and organizational policies and practices shape racial discrimination in hiring, in large part because the data just doesn’t exist. You need a field experiment, and then detailed organizational data about those companies that were in the field experiment.

[00:46:26.88] And so a large project that I conducted over a bunch of years was to actually collect that data, matching a field experiment with a detailed survey of employers, where we can begin to ask questions like do more formalized and bureaucratic organizations discriminate less against African-Americans at the hiring interface? Or when organizations have more generous work-family policies, are they less or more discriminatory against women at the hiring interface? And so that’s really the project that I’ve been focused on and kind of am in the process of analyzing the data and writing up the results and am hopeful that the first set of papers from that project will be out in the coming year or two.

[00:47:01.14] [MUSIC PLAYING]

[00:47:01.64] LARRY BOBO: That sounds fantastic and really exciting and we will all be looking forward to reading that work as it begins to come out. Thank you so much, David, for spending this time with us on Upon Further Review. And we can all recommend to our fellow scholars around the division of social sciences reading, Making the Cut– Hiring Decisions, Bias and the Consequences of Non-standard Mismatch and Precarious Employment by David Pedulla, published by Princeton University Press. Thank you so much.

[00:47:30.80] DAVID PEDULLA: Thank you so much, Larry. This was a real pleasure.

[00:47:32.85] [MUSIC PLAYING]