The Podcast for Employers Who Are Hiring At Scale
Are you involved in the hiring of dozens or even hundreds of employees a year? If so, you'll know that the typical sourcing tools, tactics, and strategies just don't scale. This podcast features news, tips, case studies, and interviews with the world's leading experts about the good, the bad, and the ugly when it comes to high-volume hiring.
How Automated Hiring Systems Are Making the Worker Shortage Worse with Dr. Noelle Chesley University of Wisconsin Milwaukee
In this first, full episode of the High Volume Hiring Podcast, host Steven Rothberg of College Recruiter interviews Dr. Noelle Chesley, Associate Professor and Director of Undergraduate Sociology Studies at the University of Wisconsin Milwaukee.
In the July 1, 2022 UWM Report, Dr. Chesley wrote about the pros and cons of automated hiring systems, which include applicant tracking systems (ATS), job boards, assessment sites, and video interviewing platforms. She identified four primary problems: (1) algorithms may not use the best criteria for matching; (2) using automated hiring systems can create problems at scale; (3) more recruiters want "purple squirrels"; and (4) automated hiring systems can lead to alientation. In today's podcast, Dr. Chesley describes and discussed each of these in a balanced, nonjudgmental manner. That a hiring system is automated is neither good nor bad, but how they're designed and used can lead to positive or negative outcomes for candidates, employers, and others.
Steven Rothberg: Welcome to the High Volume Hiring Podcast. I'm Steven Rothberg, the Founder of College Recruiter job search site. At College Recruiter, we believe that every student in recent grad deserves a great career. This podcast features news tips, case studies, and interviews with the world's leading experts about the good, the bad, and the ugly when it comes to high volume hiring. Thanks for joining us.
Today's guest is Dr. Noelle Chesley, an Associate Professor and incoming Chair of the Department of Sociology at the University of Wisconsin-Milwaukee. Dr. Chesley is a social scientist who is passionate about identifying and solving complex problems at the intersection of data, technology, gender, health, and inequality. I first ran across Dr. Chesley's work when I read an article that she wrote about how automated hiring systems are making the worker shortage worse. Dr. Chesley, welcome to the show.
Dr. Noellen Chesley: Thank you. Great to be here.
Steven Rothberg: So, first off, before we get into the actual nuts and bolts, tell us a little bit about yourself and what you're up to.
Chesley: Yeah, sure. Well, in my professional life, I spent a lot of time researching the broader social implications of new and emerging technology. And what I found is when we were having all these broader discussions about the great resignation and how the pandemic was shifting work practices, it was raising a lot of questions for me that were linked to my broader research in this area. And really got me wondering about something that I didn't see people talking about much, which was how changes in hiring practices, particularly the use of new technological systems, might be impacting the experience of locating workers and workers getting jobs.
Steven Rothberg: And it was really interesting to me because I think there's so much of a belief that more is better when it comes to technology. And sometimes more technology is better, but it also really depends on the technology and how it's used. And I think we can get into that as we dive into this topic a little bit more. So the article that was in the July 1st, 2022 UWM Report, you first talk about what you're including when you reference automated hiring systems. So without getting into the problems yet, what kind of systems are you thinking of when you talk about automated hiring systems?
Dr. Noelle Ches...: Yeah, well, it's a whole amalgamation of things. Because we're as researchers, trying to come up with a phrase that captures an overall process. But I think it's really important to understand that we're talking about, as you mentioned already, pretty complex socio-technical systems that are bringing together more rudimentary online data forms. Coupled with, in some cases, platform technologies that use much more sophisticated algorithms or machine learning types of applications. And so it really can run the gamut.
And the other thing I'll say about that is that there's also pretty large variation in how different organizations use these systems in terms of the level of human involvement at different aspects of the hiring process. So we're really talking about a whole conglomeration of technologies and platforms and applications, and how those get coupled with a whole variety of human practices on the side of the organizations that use these things.
Steven Rothberg: Yeah. Awesome. So in my mind ... so we're including things like job boards, job aggregators. We're including asynchronous and bisynchronous video interviewing systems, like a HireVue. We're including sourcing tools like a LinkedIn, virtual career fairs, applicant tracking systems, assessments, all of that.
Dr. Noelle Ches...: Yes. Yes. Thank you for putting it more in the terminology of the hiring audience. That's perfect.
Steven Rothberg: Hey, that's why I get paid the big bucks. So your article then discusses some of the pros and cons of the automated hiring systems. And for people who haven't read it, I definitely encourage them to. Because one of the things that I've seen with articles that are somewhat similar to this is that sometimes they seem to be written by the vendor or some cheerleader, a PR agency on your behalf. And so everything is perfect and glorious. And at other times, they're written by maybe extreme privacy advocates, where all data is bad data, and nobody should be collecting any kind of data whatsoever. And everything should be horse-and-buggy.
You definitely did neither. It's part of what I just loved about it. It was just very factual based. Here are some good things, here are some bad things. Whether you think it's good or bad, we need to discuss these things. And what is the outcome of them? So one of the problems that you outlined was you said that algorithms may not be the best criteria for matching. And what did you mean by that?
Dr. Noelle Ches...: Well, let me just zoom out for a minute and just talk about two big issues that come up when we talk about automation, and then I'll ... promise, I'll address your question. But I'm going to be wanting to refer back to these ideas probably, as we move through this. I think there's two big issues with automation. One of them has to do with opacity. So the fact that there's this, often we sometimes call it a black box component to a lot of these tools, where they start to get complicated. And we don't know exactly how they're working always, in every real world application.
And we use these tools because we want to do the second thing, which is scale up a process. And so the importance about a lot of the things that we're talking about now and scale, is that scale is what we want to improve problems of efficiency. So organizations who are trying to hire people have a problem right now, and it's a real problem, which is that they get a lot of applications for jobs. That's part of the access that the internet has provided. And certainly, today's job applicants expect to use online systems and things like that. So we can't just decide we're not going to use any of these things.
But the scale issue really means that if we get anything wrong when it comes to hiring, that we are going to be able to scale this up very, very quickly. Especially when you think about the fact that a lot of organizations are using third party vendors. You mentioned one, just HireVue ... I'm not endorsing them or saying they're bad. But just the point is, if there's a problem with something and we're adapting it across organizations in the economy, we're very, very quickly going to be able to scale up problems.
So you asked me a little bit about what are some of the problems that can come up with matching. And this really speaks to, if you want to think about it in terms of the hiring process, you can think about where a lot of people who need to hire or are wanting to get jobs start, which is how do we get information about what jobs are out there or what job candidates are out there? And there's lots of different places people go. You've already mentioned a few of them. We sometimes use job boards. We sometimes use platforms like LinkedIn. We sometimes use other social media sites like Facebook.
And all of those different tools are using different algorithms, different designs to try to match job applicants to open jobs. And there's a lot of research now, I think an emerging area of research just on targeted online job ads in general ... or I should just say targeted advertising. Because really, you can think of a job ad as a type of targeted advertising. And one of the things we're learning from that research is that it's possible that whole swaths of the population get missed. And it happens for a lot of different reasons. It doesn't have to be for any kind of nefarious reason. And we into the weeds on that if you want to.
But the basic idea here, just to give an example, is there's research like a study that I'm familiar with that looked at Google. And Google's algorithm will target job ads to more men than women in some cases, just because of the way people have their privacy settings set in their profiles. So that's important when we talk about worker shortages or any problems with large volume hiring, just because if you're missing a lot of women, they're a big part of the workforce, for one thing. So you don't want to be potentially filtering out potentially 50% of the workforce with something that you're doing with your job ad.
Steven Rothberg: I loved how you referenced the black box aspect, because some of the tools on the market that purport to do matching, they look at the job ad, they look at the candidate's resume or CV or other data and try to see, is there a good match? And then they typically will rank those for the employer. Here are the people who applied that are the best match for your job. And some employers are really good about seeing that and understanding that there's algorithms, there's a process that the software is following to determine are you ... are the keywords in your resume very similar to the keywords in the job ad. And they will rank certain candidates higher than others.
And so, correct me if I'm wrong. I mean, I think what you're you're saying is that the fact that it's automated, the fact that there's an algorithm does not mean that it's a good algorithm, that it's a good way of matching. Yeah. And then, you also touched on, I think the second problem that you outlined in the article. And that is that the automated hiring systems can create problems at scale. So, my words not yours, but if you're recruiting one or two people a year, whether you use automated systems or not, if you have a problem, it's a small problem. If you're hiring 200,000 people a year, if you have any problem, it's a big problem.
And so it seems to me like what you're talking about then is that for employers that are hiring at scale, doing high volume hiring, when they use these systems, any kind of a problem, big or small, just gets magnified. Can you speak to that?
Dr. Noelle Ches...: Yeah. Well, that's what I was trying to get at, even when we were talking about the job matching problem, but you said it exactly right. I mean, the whole purpose of these automated tools that we're talking about is to scale up, to deal with the problem of having to hire a lot of people quickly. And make, hopefully good decisions across a lot of different job applications. And the reality is ... and I just want to emphasize this again. And you just described a situation that's probably beyond the capacity of human agents. So most of us don't have it in us to be able to look at a few hundred job applications, much less thousands.
But if you have anything pre-programmed into your system, whether we're talking about the way that the system matches to potential job applicants out there, how it filters through those applications to provide the ranking you just talked about. Or even, I mean, we could even get into other discussions about the use of some assessment tools and interviewing and things like that. And the kinds of automation that are out there. Yeah, anything that isn't working well, is scaled up. That's just by definition, what we're talking about. And if you've got one bad recruiter who isn't doing a good job, you can hopefully locate and fire that person. But when you're talking about something that's, in a few seconds handling hundreds of thousands of job applicants, that's a much harder problem.
Steven Rothberg: Yeah. And the third issue that you talked about in the article was that more recruiters want purple squirrels. Why is that a problem?
Dr. Noelle Ches...: Well, so just ... the purple squirrel thing is just another way of talking about unicorns, really. Which may be a term that your audience is more familiar with. And it comes from a research article. So it's not my term, by the way. But what this really gets at is that human automated system interaction. So I think there are some real interesting research studies out there now that are showing how interacting with these systems taps into certain kinds of human tendencies and is reshaping the behavior, actually, of hiring organizations and the recruiters within those organizations.
And I'll just say that what we see in this context is something that as a scholar who looks at the broader social implications of technology and innovation, is something I've seen in other contexts as well. And that is how the emergence of new technology reshapes, and often elevates expectations. And that's exactly what the purple squirrel research that I'm referring to is really getting at. Which is this idea that because we have all of these tools available that can track candidates across so many different characteristics ... that can rank them, that can filter them. There's some evidence that this is really impacting the expectations, especially of people in leadership in organizations, and filtering down to what they're asking their recruitment staff to do and look for.
And so it's changing a little bit, how ... it looks like, according to at least a little bit of limited research on this topic. It looks like it's changing a little bit, how the human recruiters involved are behaving, what kinds of passive candidates they're trying to source. And also just the level of difficulty it's creating as these much, much higher expectations are outstripping, maybe the realities of the job applicant pool.
Steven Rothberg: Well, it sounds like me on a daily basis. If I can think of something, then the developer should just do it within same amount of time as it took me to think about. Before we leave off, the last of the four problems that you referenced in the article was that automated hiring systems can lead to alienation. Let's talk about that too, because that's not something that I've seen much attention paid to.
Dr. Noelle Ches...: Well, and I'll just tell you ... so let's soften your comment a little bit and say, automated hiring systems may lead to worker alienation. This is actually something I'm really interested in as a researcher in this area. But I will tell you, I have been scouring the published research and there is not a ton of research on this. So I think this is something just to think about if you're an organization or if you're a recruiter. So some of this comes from, again, that same study of recruiters I was just talking about, their perceptions about the job seekers and job candidates they work with, and the sense that people are putting their applications into a black hole.
I've seen some accounts from newspapers. There was something published in The Guardian about this a while back. But I haven't seen a lot of systematic research apart from, there is some broad public opinion research in the United States to suggest that at least among US adults, if you ask them their opinions about using automated hiring practices, letting machines or algorithms make decisions about this, they're very nervous about it. Those aren't job candidates, those are just the general public, but there is a level of apprehension about it out there. And I would say that I've seen in a few limited studies, evidence that this could be a concern.
Steven Rothberg: Yeah. In a lot of ways when read that, that commentary, the discussion around that, to me, it's very much a fear of the unknown. That we all have reason to fear what technology is doing with our data, how it's being used. Is it going to be for our benefit? Is it going to be to our detriment? And when you just don't know, when you get into that black box situation, I think for a lot of people, their natural tendency is to go to the negative. They may be very justified by that.
Dr. Noelle Ches...: Well, certainly ... can I just speak to that one point?
Steven Rothberg: Sure, please.
Dr. Noelle Ches...: So certainly we know from psychology research that we tend to ... the negative is more salient than the positive. But before we just leave it at that, I do want to say that I think there's pretty good evidence, from my perspective as a sociologist, that a lot of the automation technologies that we have in place in hiring right now, very much favor the needs of the hiring organization over the needs of the job applicant. And there are ... it's fairly anecdotal. But I think starting to be much more widespread complaints on the parts of job seekers about being ghosted by hiring organizations. About not getting any communication back about filling out all these applications and never knowing what happens to them. And never getting a job from it.
Steven Rothberg: Yeah. And especially with the applicant tracking systems. The very earliest ones that I ran into, I mean, way back in the 90s ... so I think Abraham Lincoln was still president. But a lot of them literally did not have an option to email the candidate to acknowledge receipt of a resume. And so the recruiters had to manually do that with each and every application. Now, if you get 12 applications a month because you're not hiring at scale, then that should be pretty manageable. If you're getting thousands a day, you'd have to have a whole team doing that. But every applicant tracking system on the market now has an option to automate that.
And so any employer that I hear of any kind of size, medium, large, that's using an ATS that doesn't at least kick out an automated email, thank you for applying. The next step takes us two weeks. We should be back to you within that period of time. And then when you reject a candidate or they don't move forward in the application process, you can automate that too. I feel like it's really not a technology problem, it's a willingness problem for a lot of employers.
Dr. Noelle Ches...: I think the problem's deeper than what you're talking about. So we're talking about communication, and you're right. You can automate responses. But I think there's the potential for a bigger disconnect that has more to do with the automatic filters people are using, probably as part of their automated tracking system. And I'll just use a little anecdote involving my 18 year old son, just to illustrate this. And again, I don't know how prevalent this is, or ... again, it's just an anecdote. It's not research. But my 18 year old, he's really into computers. He's built his own computer. He needed some computer parts. And he is also been looking for a job.
And he went to Best Buy to pick up some stuff, and had such a detailed conversation with the guy in their computer department. The guy was so impressed with everything that he knew, all the questions he was asking. He said, you know what? We are just desperate for people in our computer area here. Would you be interested in applying for a job? And my son said, I'm totally interested. I'd love to work here. And the guy said, okay, you have to go through our online system, but I'm going to be keeping an eye out for your application. And as soon as you submit it, I'll be trying to get back to you. So my son diligently went back and submitted his application through the online system, and he never heard anything.
Now, here's what I think could have happened there. It's not that he didn't get the email saying his application was accepted. It's that there was a disconnect between information he put into this online system, and how his information was filtered. I bet you that guy at Best Buy never even saw his application, just probably figured he didn't do it. So that's more what I'm talking about there. And it speaks, again, to the black box potential here, that we think we've done ... we think it's so simple. It's just, we'll automate these emails and get this communication flowing. But I think that there's a lot more potential for problematic ranking, filtering, all those kinds of things, in ways that could both alienate people, but also leave employers with limited access to qualified people.
Steven Rothberg: Yeah. It's one thing if there are some cracks in the hiring process. We all expect that it's not going to be perfect. But when those cracks are wide and frequent, then that's a problem.
Well, Dr. Chesley, thank you so much for joining us today. If listeners want to get more information from you, how should they reach out to you?
Dr. Noelle Ches...: They can contact me at my university email address, which is my last name, Chesley, @uwm.edu.
Steven Rothberg: That is easy. Thank you for joining us today on the High Volume Hiring Podcast, we appreciate your support. Please go to www.highvolumehiringpodcast.co... to subscribe for free on your favorite app. Review it. Five stars are always nice. And recommend it to a couple of people who want learn more about how best to hire dozens, or even hundreds of people. Today's podcast has been a co-production of Evergreen Podcasts and College Recruiter. A special thanks to our producer and engineer, Ian Douglas, and the rest of the team at Evergreen.
I'm your host, Steven Rothberg, the Founder of College Recruiter. Each year, we help more than seven million students and recent graduates find great new part-time, seasonal, internship, apprenticeship, and other entry level jobs. Our customers are primarily Fortune 1000 companies, government agencies, and other employers who are hiring at scale. They advertise their jobs that require zero to three years of experience on our niche global job search site. For more information, go to www.collegerecruiter.com/advertising, or email me at [email protected] Cheers.
Hear More From Us!
Subscribe Today and get the newest Evergreen content delivered straight to your inbox!