DAN LICHTER: Why don't we try to get going here and start on time? My name is Dan Lichter. I'm the director of the [INAUDIBLE]. And on my right here is [INAUDIBLE], who is the new [INAUDIBLE] research, and he's going to say a few things, but I'm going to say a few things first. Let me start off by saying first it's a great time to be a social scientist.
SPEAKER 2: Yes, indeed.
DAN LICHTER: And I'm not being facetious. I'm saying that because almost everything today has a very large social science component whether we're talking about environmental change. We think about the behavioral consequence of that. We have a role, whether we're talking about technological change with respect to, let's say, engineering, on the automobile-- driverless automobile. There's the old questions about adoption and [INAUDIBLE]. Who uses these things? How do we get people to actually use that technology?
We think about bio-social. We think about how our environment shapes the expression of genes? I mean, there's just been a lot of technological change that we always respond to as social scientists. So the environment is changing rapidly, and so I just think it's a great time to study lots of different things. And I'm especially pleased as ISS director that, over the past several years, we've funded some, I think, really timely important projects that really reflect that.
Today we're going to talk about-- it was going to be about DACA originally, and we're going to talk about immigration here and sort of political movements associated with immigration. And that's going to be the focus here [INAUDIBLE] appropriate topic that is both in service to basic social science, but I think it also is in service to public outreach, to [INAUDIBLE] research, to other things that matter in the public sphere. So I'm very pleased with that.
We have two other projects that are ongoing right now. Another one is on mass incarceration, which, again, is a big topic that's important for [INAUDIBLE] minority communities, including inner city neighborhoods, where there's a lot of [INAUDIBLE] till recently towards doing something about the penal code [INAUDIBLE] incarceration more generally. And then we have a big project, too, on China, which what better topic than China today [INAUDIBLE].
So we've got lots of things going on, and in the next years we're starting two additional projects. One is on social media, including cyber bullying, fake news, these sorts of things, which I think is, again, a really timely thing that matters. Again, it's a response to technological change in the media. And Facebook is a good example of this. How are we, as social scientists, responding to these technological changes?
We also are funding a big new private that I think you might think about as radical collaboration since everybody is talking about radical collaboration. We're talking about big data and machine learning, and how that affects inequality and discrimination in hiring and other practices as a result of these decisions that are made by machines, not necessarily by people. So these are all really good topics, and I think are part of what we're trying to do at ISS, which is become kind of a unit that is not only an information clearing house, but also tries to represent the interests of the social scientists across the universe.
And so I think these are all good things. We have our other kinds of projects, as well, with our fellows program and our small grants program. And all these things, I think, are sort of putting a face on the kinds of things we do here at Cornell University. And again, one of the things that we're trying to do with projects like this is provide a platform going forward. How do we identify areas of excellence? How do we build on those areas in terms of faculty? How do we track faculty? How do we train students in these areas that we can put our stamp on these areas?
And I think in the case of immigration there is a ton of talent around the university, and that's become clear over the last 5 to 10 years. And I think ISS and the kinds of things we funded has been a big part of that. So thank you all very much for coming, and I also want to thank Emmanuel Giannelis, who is the new VP for research. He's already-- would I say beleaguered after six or eight months on the job? But I appreciate his support financially for ISS and the social sciences, but also I think he wants to learn about the social sciences, so we're really happy here to have him here to say a few words and maybe-- I don't know what else you want to say, but be a cheerleader like I'm trying to do. I always try to get everybody to be a cheerleader when I do this sort of thing, but Emmanuel, go ahead and say [INAUDIBLE].
EMMANUEL GIANELLIS: Thanks, Dan. Thank you. Thank you. [INAUDIBLE] I will echo what Dan said and say, yes, it is a great time to be a social scientist, but I'll modify it a little bit because I was not trained to be a social scientist. And since I've declared that I don't want to take another test in my life to become a social scientist, I would say it's a great time to collaborate with a social scientist.
So before I go on that, I would say, yes, I am the new vice provost for research. And of course, the good thing about that is that I have another five years or so to say that, if I said something stupid or naive, please excuse me. I'm just new to the job. And I just started, Dan, about seven, eight months, right? So I've got another four and change.
He gave me [INAUDIBLE] doing things that are already happening obviously as a faculty member-- no disrespect for the faculty members-- I take credit for those-- for the good things, but I had nothing to do. All these things were sort of running and going before I arrived, but I am a big supporter of the social sciences in [INAUDIBLE] way. As a physical scientist engineer myself, I think what really can distinguish us from our peer institutions is it that we can bring these disciplines together, and we can actually put programming that some [? of our other ?] peer institutions do not offer. And now in the new position what I find really interesting and exciting is being able to catalyze this intersection.
So we have some programming that I don't want to sort of say much about, but the idea is how do we actually bring people, especially graduate students from different disciplines to work together on a common project. And it's not to be able to write only better papers and get us better grants [INAUDIBLE] provide them with a better education-- projects and pieces that they can talk about, and things that can sort of make a difference and distinguish them from the competition. So a lot of interesting ideas to talk about, and I would encourage you to reach out to me and educate me in the different fields that you are working on.
On this particular topic, I'm really excited to sort of say a few words and open this conference for two reasons. Well, I'm an immigrant myself. I came here in 1980, and this country gave me a hospitable place that has allowed me to prosper and progress in my career. And I value that, and I still feel and act like an immigrant after all these years. So immigration is always at my heart, but the second important thing-- and not just the immigration alone, but some of the things that you're [? attacking-- ?] they come at the heart of our president now talks about doing research with impact. And [INAUDIBLE] of many other topics that really have impact in the societies and the communities that we live. And all of the topics that we mention-- they sort of have that direct impact that our president is talking about.
And if you've been listening to her, she talks about research with impact, or impactful research. And I really take as an opportunity to emphasize how important these areas are for the university and for the president. And so that's pretty much what I wanted to say. So I'll finish, again, by thanking you for coming together. I want to thank Dan and the leadership team here for putting this platform together, for many people to come and talk about these important issues, and encourage you one more time to reach out to me with ideas of how I can help you with your research and your scholarship. So once again, thank you and good luck with the lecture [INAUDIBLE].
DAN LICHTER: Thanks, Emmanuel. I think what I'm going to do is I'm going to turn this over to Shannon Gleeson, who is sort of spearheading this with her colleagues here. And she's an associate professor in ILR, a sociologist type background. She does a lot of work on, I would say-- and I'm probably mischaracterizing you-- kind of political sociology in the sense of social movement. She's more of a qualitative researcher, but not entirely. So this quite a group with both a qualitative and a quantitative side. And so I'll let Shannon introduce the topic. She can provide the overview, and then she can introduce her colleagues.
SHANNON GLEESON: Great. Thank you, Dan. I want to jump off from the standpoint of saying this is a great time for social science, but not a great time for immigration. And so I think more than anything this is an important space for us to be thinking about what the academy needs to be doing in terms of thinking about these important policy issues and I think also trying to understand what can the precarious experience of immigrants tell us about other marginal populations that we, as sociologists [INAUDIBLE] political scientists, in the room are interested in learning about-- and economists.
Before I do that, though, I just want to extend my gratitude to the ISS, to Dan, and as well as to Kim Weeden, who was his predecessor and who was in charge when we were funded. Also [? Lori ?] [? Sanken, ?] who is no longer here, but was our excellent administrative assistant. And Megan Pillar, who's new and who's been really helpful. And of course, Analisa [INAUDIBLE], who's here. If you haven't already met Analisa, she's been really the backbone of a lot of what ISIS does and has been incredibly supportive.
DAN LICHTER: [INAUDIBLE]
SHANNON GLEESON: So thank you, Analisa.
And in addition, we've worked with a number of students both at the ILR school, and PAM, and in soc who have been really critical to helping us think through this. And as well, Kate and I have worked with some colleagues and students in New York City. So you're going to hear from us, but this is really the result of a number of different efforts.
I'll say a little bit about my colleagues, and then we'll move forward, but I want to first acknowledge Jordan Matsudaira, who's in PAM, but who also wasn't able to be here today, but Matt will be talking a little bit about the work he does. This started as a project that Matt and I agreed to spearhead. We got our arm twisted by this one over here, and I think it turned into a really fruitful interdisciplinary deep dive into questions around, on the one hand, the enforcement context for immigration, which even in the last three years has rapidly changed, as well as all of our interests in different types of immigrant incorporation.
So I guess I'll introduce you guys, which feels strange, but Matt Hall, who is my colleague here over in PAM. Steven Alvarado, who's over in sociology. And then my colleague, Kate Griffith, who's in ILR with me. We're both in the labor relations, law, and history department.
So I'll jump right in and to say a little bit about the motivations for this study. We started this program, as I mentioned, in a really different political moment at a time when there really still felt like there was a possibility for another major-- some sort of policy shift at the time during the Obama presidency had still been at the top of his policy priorities, but Congress had not moved, and there's a couple of things that had happened. So we haven't had a major legalization program since the 1986 Immigration Reform and Control Act, but during the Obama administration, there were two shifts that were consequential.
The first is DACA, which Dan mentioned-- the Deferred Action for Childhood Arrivals program, and which a lot of us on this campus have heard about because of the way it impacts our students, but the other was a program that didn't end up going anywhere, but being embattled in the courts, which is a deferred action for parental accountability, which would have impacted over 5 million-- I think is the estimate-- parents of US and legal permanent resident children. And we went into this trying to figure out what can we learn from that policy moment if that executive action were to go through.
And so we went into this a bit-- not on a whim, but knowing that this was an uneven policy conduct so we had conversations with Kim about whether or not it made sense to move forward. We ended up getting the support. The policy itself fell apart, and as you know, we are a very far place away from where we are in 2014, but we were interested in this because there's a dearth of data that really helps us examine the impact of immigration status.
Matt's going to talk a little bit more about what the challenges are methodologically for understanding the impact of immigration status on a number of different immigrant outcomes, but after 1986, there was a range of studies that helped us kind of pick apart what the impact was on this broad based amnesty, but we don't really have recent data since then. And most of what we have looked at has been cross-sectional. And so we went into this trying to take advantage of this particular policy moment and then had to really quickly retool.
The other main part of this, especially coming from Kate and I's interest, is that we were interested in understanding the experiences of workers, and you'll see also of students in the educational system. But in our case, we were interested in understanding how legal status impacted immigrant outcomes, but not simply all the various points of precarity which we are kind of well attuned to in terms of low wages, occupational hazards, and a number of other things that, for example, Matt has worked on and myself, as well. And Mary Jo Dudley here is from the farm worker program and can give you a really firsthand account of what that looks like for folks in our area, but we also wanted to know the role of legal mobilization.
And just to be clear about that, there's a number of different rights that are afforded to low wage workers regardless of immigration status, and we were really interested in how immigration status impacts their ability to actually mobilize those rights and to come forward and engage in claims making. And that's important not just because of those individual's ability to make good on those rights, but also because claims making is really at the foundation of how we enforce labor standards in this country.
And for undocumented workers, we have long been interested in how, on the one hand, there is the existence of these formal rights, but then also a number of different barriers to access. And I think there's parallels to how we've thought about this in the educational context, where there are formal rights available, but a lot of institutional challenges that remain.
But at the crux of our inquiry was really, how do we start to think about other categories of non-citizens? To be sure, I think the main focus, both at the policy level and I would say even at in academia, has been primarily on the dichotomy between the undocumented and documented populations. And that's largely because there's 11 million undocumented individuals in this country. About 5% of the civilian workforce is undocumented, which is about eight million individuals. And as we can see, any time you turn on the news, there's some real important policy questions around the issues that shape the well-being of this population, as well as the many other immigrants who are in households with them.
And so the literature in the social sciences on immigrant incorporation has focused primarily at this intersection between documented/undocumented legal status. And that's the case in terms of how we understand wages, occupational hazards and mobility. Very little research out there has kind of looked at other non-citizen categories with some exceptions, including Matt's work. Educational achievement has primarily focused on this, as well. And then to some extent, especially in terms of civic and political engagements, there's a lot of questions about how undocumented populations can or cannot engage in civic life.
And I think that there's a lot of theoretical reasons about why we would be interested in the impact of legal status. On the one hand, there's a lot of instrumental value that comes of having legal status. So Steve [INAUDIBLE] here who's an attorney and who works with us here on campus and in the community can tell you all the different rights and benefits afforded to someone if they have legal status. So there's these instrumental values, but then there's a number of other constitutive effects, which in other words shapes our relationship to the law. So as an undocumented individual, for example, many social scientists have been interested in how your relationship to the rights that you do have keeps you from coming forward and making good on them.
And so we're interested in a range of different mechanisms for how legal status operates. And on the one hand, certainly the fear of deportation is a big part of this inquiry, and Matt's going to talk a little bit about what we can learn from the changing enforcement context, but then we're also interested in, what are the challenges that exist beyond just the deportation apparatus in other areas of the social safety net and the labor market?
So that was one of the other main impetuses for this-- was to think about what other categories of documentation status are relevant given the trajectory of immigration policy shifts. And so that led us to our primary inquiry on temporary immigration status. And again, this isn't because undocumented status doesn't remain a very important reality for many workers and a primary driver of precarity, but there are a number of different temporary statuses-- and again, Steve is our local expert here on this-- that are often overlooked.
And part of this has to do with the data restriction. So in existing analyses, you often see a lot of focus on how nativity affects outcomes-- so whether or not you're foreign born or native born-- whether or not you're a citizen. That's often because that's the limited data that we have, and you've heard the citizenship question on the census debate in the news quite a bit lately. And then to some extent you have survey data or other imputation mechanisms that allow you to identify what are likely undocumented populations, but these kind of interstitial statuses, which I won't get into the large alphabet soup of what they are have, for the most part, received far less attention, with some important exceptions, which I'll mention.
And I think we wanted to refocus there, one, because DAPA, the 2014 proposed policy that I mentioned, would have created something very similar, which would have been a two year temporary reprieve from deportation and would allow some sort of employment access. And even though we have a good deal of information on guest worker programs, which is another kind of form of temporary visa, the distinction from those guest workers are that, one, there are far fewer guest workers, and also their visa is tied to their employment relationship. And so there's a distinct situation for these other temporary immigrants.
And we thought that this was also important because in reality there's enormous prosecutorial discretion in the ability for the immigration bureaucracy to extend these temporary statuses, and we just don't know a good deal about them. The largest category of these individuals-- also which has been in the news-- is the category of individual's temporary protected status. Both DACA, DAPA, and TPS, and a range of other programs have been very embattled. TPS for some countries has now been canceled, but we do have a large community on the order of about 300,000 individuals who have TPS-- largely Central American and Haitian.
And we weren't interested in necessarily creating a hierarchy of this is better than or worse than being undocumented or even having legal permanent residence, but we wanted to understand what the particular sets of opportunities and challenges were that this community of workers and students were facing. And in some ways, it also allowed us methodologically to proxy the population that would have benefited from DAPA. Maybe not demographically, because it was a much greater range of individuals that would have fallen under DAPA, but in terms of the mechanisms of legal status, it provided an opportunity to do that.
And we were standing on the shoulders of other people who have started some really important work on TPS, including Cecilia Mejivar, Miranda Hallet, who's actually a graduate of Cornell, Lacy [INAUDIBLE], who's in UCLA. And so we wanted to kind of look at that in the particular institutional contexts. So I'll end with just one more thing before I hand it over to Matt, which is to say that, even though TPS is likely to be phased out-- and I don't know what Steve thinks about that prediction, but it's certainly subject to a lot of political volatility. I think that it's fair to say that the shift politically has been towards proposals that lack a pathway to citizenship.
And so if we are to see a shift toward some sort of legalization program in the future, we are likely to see something that looks more like DACA or DAPA, I think, than probably something that looks like the 1986 amnesty. And many of the proposals towards low wage worker visas, like the W visa, which would have created some sort of low wage worker visa, would still have this temporary aspect to it. And so in addition to the ongoing opportunities for discretion in the immigration bureaucracy.
So we wanted to try to understand, what can we learn from this? There's obviously assets attached to having any kind of immigration status, but we also want to understand what the challenges were for temporary folks. And it's also the case because people fall in and out of status we often think of this as a fixed category, but we wanted to understand how falling in and out of status and the precarity associated with not necessarily having a pathway to permanence-- or permanent residence and eventually citizenship impacted individual outcomes.
And so as you'll see our approach to this project was to think about this across institutional context. We're going to be talking primarily the workplace and educational space. In our project, we also are looking at this with an eye towards differences across place. So even within the New York City region, we find some really interesting differences between, for example, individuals in Long Island versus in Brooklyn. And then also, as you'll hear here, differences across the life course and other intersecting points of precarity, including race and gender. So this is the intellectual project, and I'll turn it over to Matt to talk a little bit about the methodological challenges attached to this.
MATTHEW HALL: Great, thanks. So as Shannon just described, there are a lot of nuance and complexity both in practical terms and in legal terms around what these different statuses mean, but there's also a lot of methodological and data concerns, and the challenges when doing sort of research on these questions. And so I want to work through some of those and give you a sense for some of the issues that we grappled with over the course of the last couple of years.
So as you might imagine, that access to any sort of internal data on immigrants or on immigrants' status is heavily protected and is virtually impossible to get access to. And so we had actually had expectations or aspirations that we would be able to work with the Department of Homeland Security and the Department of Labor in trying to secure information on DACA applications with the hopes of being able to leverage information that was provided on the applications themselves in order to assess some sort of impact of the status on some sort of outcome that we can assess quantitatively.
And we were somewhat hopeful under the last administration. We're less hopeful under the current administration. And so this is an issue that we had to steadily grapple with, but even in the absence of the administrative data, you still obviously have large problems when you're talking about people that have a status that makes them vulnerable to interaction with the state. And so you have these large statistical problems of under coverage that are pervasive in any sort of sample estimate of immigrant populations.
And so the problem, of course, is that, if a census taker comes to the door of an undocumented family and asks them to report information on when they came to the country, how much money they make, how they get to work, they're unlikely to provide complete information on those questions. And so this creates a statistical challenge, and it's very difficult to build complete samples of the unauthorized or immigrant populations in general and individuals that have statuses in between those two points.
And so the census and other organizations have actually done a good deal of assessments of this problem. The census estimates that they miss somewhere between 15% and 25% of the undocumented population, but what's important is that that estimate is highly variable across different categories, across kind of origin groups, across characteristics of the migrants themselves, and then importantly across regions and locations, as well. So trying to generate measures of coverage in border towns is very different than it is in upstate New York or in large cities. So these are some issues that we had to constantly deal with in trying to make these assessments.
The solutions to some of these problems require kind of sophisticated statistical techniques. So on the one hand, there is ways that you can basically impute data from smaller scale surveys and then upscale it to-- or scale it up to larger samples. So if you know something about the relationships between a certain status in a small sample of individuals, then we can use those to estimate the counts or the likelihood of people in larger samples of data, and there's some statistical ways to do that. And these data fusion techniques are something that we have given a lot of thought to.
And then on the other side, there's similar challenges when doing smaller scale or qualitative related research. And so you have to be very deliberate and focused in how you collect samples. And this might be through legal service providers. This might be through what's called respondent driven sampling. And so we work through a lot of these questions as part of this project, and Kate's going to talk maybe a little bit more about some of those sampling issues in a few minutes.
So those data problems can lead to what we call estimation problems, and these are basically problems that, if you have errors in your data, they're not necessarily problematic as long as they are randomly distributed. And this is the main concern-- is that those errors are not randomly distributed, that both these coverage areas-- that is, who's participating in surveys and who is excluded from surveys, or who's in the data and who's out of the data-- those groups are different. And so if those groups are different, then that's going to lead to bias in your understanding of some sort of relationship between these legal statuses, or these temporary legal statuses and some outcome.
And there's also some good evidence on who is being missed by these data, and the general consensus is that the most vulnerable migrants are the ones that are most likely to be excluded, and this isn't too surprising. The people that have the most to lose are less likely to engage with officials, whether they're from the government or academic. And so these create estimation challenges.
And then beyond those kind of data and those statistical issues, you have bigger research design problems. And due to the fact that migrants-- their kind of entry pathways might make them differ on characteristics that are hard for us to measure. So migrants-- they come to the United States or some other contexts-- are different not only in their experiences, but they're also different in their motivations for migration. And those motivations for migration might be related to the sorts of outcomes that we're trying to measure.
And so, again, this is an issue of statistical bias that, if you're not able to account for those differences, then we're not sure that the measures are valid. And so research design ends up being really important to trying to determine these questions. And as you'll see from this work and especially from the work that Steven will share, we've given a lot of thought to how we can deal with some of those issues by leveraging kind of variation across eligibility years in these different statuses.
So other ways of dealing with this, of course, are having kind of longitudinal data where you can watch people over time. You can watch people maybe before migration and after migration, but at first we don't have that sort of data. We don't have kind of the optimal data for these questions.
And then another part of the project that we're not going to be discussing about too much here and is a part that Jordan and I have been taking the lead on is leveraging variation in enforcement contexts. So even though we don't necessarily know good information on the people that live in those environments, we do know information about what sorts of communities launched programs that were designed to reduce the size of their immigrant or their unauthorized population. And we were able to coordinate with the federal government to gain access to these applications to coordinate with the federal government on deportation proceedings. And so we can leverage that geographic variation both across places, but also over time within places in order to try to assess some of the impacts of the rise and kind of the intensified rise in deportations and enforcement on a host of different outcomes.
And then the last thing I want to say before passing it off to Kate to talk about some of the project based research is just emphasizing how important ISS and various other centers on campus were to helping us engage in these sort of questions. So as both Shannon and I have tried to emphasize, there is a lot of really thorny and complex conceptual, and legal, and methodological issues that impact all these sorts of questions that we're trying to ask. And the solution to these problems has to be multidisciplinary. We had to get a group of people that understand the legal dynamics, and understand some of the statistical dynamics, and understand all these theoretical issues in order to come up with some innovative solutions to these problems. And so we're very thankful to ISS for helping us engage in those conversations and to facilitate this discussion.
The other piece of it is, of course, that the funding environment for doing immigration research right now is very difficult. It's very hard to secure money to do research on these sorts of questions, particularly from the federal government. And so having opportunities to do this sort of work on campus and to work with students, and to work with colleagues, and to actually do data collection has been a pleasure for us to be able to access. And then I just want to also emphasize that ISS-- that this project has had a broader reach beyond just kind of strict academic outlets or kind of research endeavors-- that it's enabled us to do public engagement with the community both off campus, but also on campus, as well.
And the one thing I want to mention here is that there is a conference that partly came out of this group that was co-sponsored with the Center for the Study of Inequality in the Cornell Population Center that helped frame some of our thinking on these issues in the fall and will be published in a forthcoming issue in the American Behavioral Scientist later on this summer. So I'm going to pass it over to Kate now to talk about some of the project research.
KATE GRIFFITH: So I was telling my team members here that I typically like to stand when I talk, and I was going to talk at the podium, but I like all the sporty team language of ISS, and I'll be a team player and sit, even though I'd rather stand. So I'm going to talk about the sub project that Shannon and I are doing. So we're sort of the qualitative sub project. And this is a project that neither of us would have done if it weren't for the ISS money we've received as a seed grant. And so we're very appreciative of that.
And we went into this with a real interest. I'm a law scholar. She's a sociologist. We have very similar interests, but very different disciplinary training, but we're both really interested in how differences in the law-- immigration status-- might affect how low wage workers perceive and navigate their working conditions and their claims making against employers when they're not paid for wages or face certain health and safety problems in the workplace. But we want to also have an awareness that there are other dynamics of course at play, such as gender, race, and immigrant origin, and also different local political context, some being more anti-immigrant than others.
So up here I've put up our just over arching legal research-- not legal-- research question for this sub project. How did the following impact working conditions and claims making? And it's really these three dimensions-- immigration statuses-- so undocumented, not having any recognition under immigration law. Temporary statuses-- so that there is a temporary reprieve from deportation and authorization to work, but no pathway to citizenship or legal permanent residents versus a third immigration status of permanent residence, or citizenship. Intersectional identity such as gender, race, immigrant origin, and the local political context.
So here is just an overview of our methodology, which I will go over. We can, of course, talk more about this in questions and answers, but I'll preview that methodological question will go to my sociology colleague. I'll do my best. So here's our methodology. It's an interview based study. We are doing interviews with low wage workers and three different immigration statuses. So the goal is 90 with workers with temporary protected status-- and we've completed those-- 90 undocumented workers, and 90 citizens, or LPRs.
We are interviewing both Haitians and Central Americans. We started with temporary protected status, and these two groups are, as Shannon mentioned, the dominant groups-- Haitians and Central Americans-- and then an equal amount of men and women. All of our interviews are happening in the New York City area, but there are really three segments within that where our interviews are taking place. New York City, which has mostly been in Brooklyn, where we're doing interviews. Long Island, both the suburban and some more rural areas of Long Island. And New Jersey-- mostly the commuter areas of New Jersey. So it's a mix of urban, and suburban, and rural locations.
We recruited our interviewees-- and still are recruiting our interviewees-- through organizations, mostly non-governmental organizations, mostly legal service oriented organizations, and then through referrals. So the people that are being interviewed are asked to refer us to others. And so some of the interviews are coming that way.
The interviews are conducted in English, Spanish, and Haitian creole, and we have a team of researchers assistants in New York City. In total, five people have done interviews for us, but mostly three in particular that are in each of these locations. So far we have 120 interviews complete. So we're still collecting data, and we have about 50 of these interviews analyzed. And I should say we've begun to analyze them. So Shannon and I had a conference at UC-Davis last month and just started to work through the first 50 interviews that we've coded and look for some initial themes, and I'm going to present basically on what we see so far, but this is very much preliminary.
So one thing looking at these interviews that really comes across is, yes, temporary protected status is an improvement over undocumented status in some ways, but it does erase other forms of precarity in the work place. Low wage work is precarious. So all over these interviews are stories accounting of situations where workers are not getting paid for all of the hours that they're working. So huge or very, very clear trend in these interviews is wage theft.
Another is injuries and health and safety concerns. So some were acute, like fingers-- losing fingers in construction, or breaking bones. Others more repetitive stress type injuries coming from service work, standing long periods as cashiers, or other kinds of service workers. And also complaints about-- or actually reports that complaints are futile. So when there are problems, it doesn't help, and it actually could hurt to talk to a supervisor or try to get some sort of relief.
So just to give you an example, a flavor of how this plays out in an interview, Evelyn, somebody that we interviewed, was talking about not getting paid for all of her work. And when the interviewer said, well, did you talk to someone? Did you ask? And she said, no, you just tolerate it. If you don't, you'll be kicked out of the job. So this fear that there will be some retaliation if they say anything when things go wrong.
Other forms of workplace precarity not necessarily connected to immigration status was a lot of reports about sexual harassment and racial harassment in the workplace-- sometimes coworkers, sometimes supervisors, but also, again, given the service nature of a lot of these jobs, customers, patients, basically the people they're interacting with in their work. When we pull out TPS as an immigration status, we're trying to look through, well, is there any particularities about a temporary immigration status that might play out and effect somebody's workplace experience or at least their perceptions of that.
What we see is that certainly TPS does provide some benefits over undocumented status, and the things that we hear in the interviews are, well, I can get certified in certain areas, and that opens up still low wage work, but some better conditions. Or even driver's licenses-- having an access to a driver's license, which undocumented people don't have access to, can change the kind of job opportunities that TPS workers have. And then benefits from the welfare state-- better health care and other benefits. So certainly there are some benefits, but we have also seen some aspects of temporary immigration status and how it's enacted and administratively processes that does seem to affect workplace experiences.
And one just has to do with a real cumbersome administrative process. So temporary protected status-- it's often granted to a country for 18 months, or 6 months, and then there's a question about whether it's renewed. And then when it's renewed, there's paperwork to be done to update the employment authorization documentation. It's a real pain for employers, because these workers don't-- they have end dates to their work. And so when they're renewed and they need to update that, it can be a difficult process.
And so some workers talk about how they think some employers didn't give them a job opportunity and might have even preferred an undocumented worker who had fake papers with no end date, but looked real, or just paying under the table and not dealing with the paperwork. And also interruptions-- so a human resource department saying, well, this is an end date. I need updated work authorization, and then the difficulty for low wage workers to find a lawyer to help them have the right paperwork to know what to show the HR department to make sure they could get back to work.
And the other has to do with fear about TPS termination. So we see in the interviews a lot of concern, especially since the Trump election, about the end of TPS. And as Shannon mentioned, some of these countries have been terminated in the recent past because termination of temporary protected status, of course, makes this population deportable. And this population, unlike undocumented workers who have never been picked up in the system, the government has a lot of information about them. They're registered with the government. There's biometrics. There's addresses, and so some of the interviewees talked about feeling like they are under surveillance, and they're actually fearing that they will be deported immediately as soon as the program terminates.
So Michaela, one of our interviewees, a Haitian TPS holder, said along these lines-- this was just a few months after the Trump election-- with TPS it's like you live under fear. You don't know what's going to happen. I live with a lot of stress because of that. And so this, of course, affects their whole lives, not just their work lives.
And just to wrap up before handing it on to Steven, Shannon and I have started to publish some of the conceptual and theoretical legal dynamics. And I've put it up here-- the Comparative Labor Law and Policy Journal. And then we're working on this book chapter. And as soon as we get further in terms of analyzing our data, we hope to continue to publish articles, and if all goes well, ultimately a book, but we'd need some funding for that. And so we're actively seeking external funding along with, of course, internal Cornell grant money. And we've received a small Russell Sage Foundation grant, which will be helpful for us to move to the next step of getting a comparable sample of undocumented Haitian low wage workers and Central American low wage workers. And we've also submitted a larger proposal to the National Science Foundation.
SHANNON GLEESON: Fingers crossed.
STEVEN ALVARADO: Great. I'll just--
KATE GRIFFITH: Oh, you've got it.
STEVEN ALVARADO: I was going to stand, but--
KATE GRIFFITH: You're a team player.
STEVEN ALVARADO: You set the legal precedent the other way now to sit. Hi. I'm Steven. I just want to note and give thanks to ISS. This has been the first time I've ever been a part of such a great team of experts across methodological and substantive background. And being able to try to tackle this tough question and crack this tough nut of, what is the impact of something like a temporary protected status on anything? Really requires, I think, this type of collaboration that is so unique to this program that ISS has given us.
Being able to examine some data and draw a conclusion based on some statistical techniques is one thing, but figuring out why it is that you have that association in the first place really requires a qualitative perspective. And there, I think, is really where the meat of policy really lies, but in any case, it's been a real joy. And so I'll take a few moments to talk about the paper that I've been leading with Matt, and I want to give credit to our other co-author Alex [INAUDIBLE], who's a graduate student in sociology currently, as well.
So our focus really was on and is on this paper. Actually we're about to send the revisions back pretty soon hopefully, and hopefully it'll be out pretty soon at the journal of demography once we do so. And it's really focusing on the impact of TPS on students-- on undocumented young peoples' outcomes or educational outcomes, specifically on dropping out of school.
And I'll just jump right in. The basic question that we're really asking is, is there some sort of effect of having TPS on the outcomes of kids, of adolescents. And secondarily, are there differential impacts given a family's economic background? Does affect grow or dissipate as you move up or down the economic spectrum of your family?
We build off of previous research and invoke a quasi experimental set up to try to answer these questions. We take advantage of a catastrophic and very sad event that occurred in 2001-- a series of earthquakes in El Salvador in January of that year. That, as I'll talk about in a little bit, really gave us some purchase on being able to try to isolate the impact of TPS on these kids' outcomes independent of other factors.
We also are fortunate, given the data set that we use, to be able to examine this question within families where there are some kids who are undocumented and others who are us born. And that really gives us an even better sort of account, as opposed to some previous scholars who have delved into similar types of questions, into isolating causal effects. We also run a series of falsification tests [INAUDIBLE] robustness checks to try to see if our effects for Salvadoran shows up for other groups who are not eligible for TPS. And I'll show the results as we go forward.
So as I mentioned, we use the case of immigrants coming from El Salvador. Just quickly, Salvadoran immigrants comprise the fifth largest immigrant group of all immigrant groups. However, they are the second largest of the unauthorized immigrant groups in the United States. They are also the largest group of TPS holders, far outpacing other nations with about 195,000 TPS holders in 2016. TPS was first granted in 1990 as a result of the Civil War that was happening in El Salvador during the decade immediately preceding that and has been implemented-- folks have been able to reapply and come in and out of TPS ever since. And also, as I mentioned, the US government reacted to this earthquake in 2001 with this policy of granting temporary protected status to immigrants coming during this time. The majority of these folks remain unauthorized, but we do try to capitalize on the folks who are eligible for TPS for this analysis.
So what do we do? We look at census data from the ACS and PUMS, which is a large-- very large nationally representative sample from which we can gather information about place of birth and year of arrival for immigrants. And in that way, we're able to gain a sample of Salvadorans who were able to come right around this time period surrounding this earthquake. We restrict our sample to teenagers, primarily those ages 13 to 18 who have at least one parent of Salvadoran origin.
And we're looking primarily at what's called the 1.5 generation. That is kids who arrived younger than 14 into the United States, but have continuously lived in the United States and have acclimated to the educational institutions in the US. We are looking at school dropout-- generally, whether or not a child is out of school. We play around with some alternative definitions, as well, but we focus on dropping out in general. And we also are able to look at other measures-- other key measures such as parents' education, family income, et cetera-- to add into our model to try to, again, isolate this impact of TPS itself.
As I alluded to, we are identifying TPS eligibility. We don't necessarily know if they had TPS, but we're trying to get a strong educated guess on whether or not these kids were eligible based on the policy eligibility window that the US government put in place, which was essentially anybody who was here prior to March of 2001 from El Salvador was eligible. And so these kids who have arrived-- who note that they arrived before 2001 we are classifying as likely TPS recipients, and we're classifying those who came just after this window after 2001 as likely unauthorized, so folks without TPS or not eligible for TPS.
And so we limit the samples to those who are arriving in the two years before and after 2001 to try to maintain some sort of semblance of similarity among these immigrants so we can plausibly claim that these are similar types of folks that are coming to the US, just some before this eligibility cutoff-- and you can't see it. It doesn't come out well-- and those who are coming just after it. It's green. I didn't know that. Sorry.
MATTHEW HALL: It's green here.
STEVEN ALVARADO: Yeah, it's green here. It's not green there. Anyway, so we have this cut off window, this nice policy cut off window that's given us this sort of natural experiment. You have this exogenous shock, this hand of God experience, this earthquake that came down and is sort of creating a similar population that is either eligible or not for TPS.
We used two sources of variation. That is eligibility for TPS based on the arrival year, and then we also look at within family differences. As I mentioned before, we're looking at differences [? now ?] between siblings in the same family. And this is possibly-- this is possible because there's a high prevalence of families of mixed status. A lot of immigrant families are composed of some kids who-- older kids usually, of course, naturally who came to the United States undocumented perhaps with their parents, and then subsequent younger siblings who were born in the United States and are US citizens.
I'll skip through some of the details of this, and what I want to focus on here is just simply that we're looking at essentially the difference in the differences of the children who are unauthorized on the one hand and are not eligible for TPS, and the students who were unauthorized, but are eligible for TPS, who initially were unauthorized immigrants who were eligible for TPS, and the outcomes for US born kids. And that's just the essential difference-- the essential estimate that we're after.
And what we find, essentially, is that in our ultimate model in our sibling fixed effects model, where we also control for a host of, as I mentioned, family characteristics, as well as year fixed effects and state fixed effects to try to capture temporal or spatial confounding, we find that this model is showing us that the interaction between being foreign born and TPS-- that is, the difference between those kids who are unauthorized who came after the 2001 window and those who came before who likely had TPS-- having TPS is reducing the kids' odds or chances of dropping out of school.
Now, as I mentioned before, we run a series of falsification tests to check the robustness. And the idea here is that, if we see a similar result for kids coming from other countries who are not eligible for TPS, then perhaps our TPS effect really isn't real. And as you can see here, we don't find any significance in these difference in differences in these interaction terms for any of these other groups that we checked, which include Mexicans, Cubans, and Chinese, who, as I said before, were not eligible for TPS.
We also are interested in figuring out, as I mentioned, whether or not there is some sort of variation in heterogeneity in these effects across a family background. The first indicator that we use is looking at parents' education. What we find is that it appears that this impact of TPS seems to be isolated for children of parents who have at most a high school education. And the second indicator that we look at is poverty, and we see that this effect seems to be isolated among kids who are coming from families in poverty. Well, so, the general story here is this effect seems to be found among lower socioeconomic status families-- immigrant families.
So to sum up, we're finding consistent evidence that TPS reduces dropout. This effect is large enough to erase the immigrant disadvantage if you're looking at the main effects there. The placebo test, the robustness checks also confirm that these effects are limited to Salvadorans-- that is, a group that was eligible for TPS and is not showing up for the other groups that we tested for-- the other countries. Also we find that the benefits are conditioned by family SES. So children from poor and less educated families seem to be benefiting the most.
And we extend our findings to thinking about DACA overall-- really is our larger sort of puzzle that we're trying to get at. And we conclude that extending at least these provisional documented statuses to undocumented kids has a beneficial effect on these educational outcomes that are realized even before students are graduating from high school. So with that, we'll finish and open it up for questions. Thank you very much for your time.
DAN LICHTER: I want to be mindful of everyone's time. How much time do we have here, Analisa? We're already out, right? But so feel free for those of you who have to depart-- we understand that. But for those of you who can stick around, maybe we have a little time for a few questions here [INAUDIBLE]. Questions, anyone?
SHANNON GLEESON: Maybe we'll collect a few. I said maybe we'll collect a few [INAUDIBLE].
STEVEN ALVARADO: Comments, questions?
SHANNON GLEESON: [INAUDIBLE].
AUDIENCE: Yeah, I have a [INAUDIBLE]. This is just a question which I don't know if you've thought about at all, but since your category is people with TPS, and as we see the disintegration of TPS and, in fact, the targeting of TPS people for deportation, do you have any follow up with that community? Do you have any-- assuming that there is some building of relationships with that community.
STEVEN ALVARADO: I think this is question perhaps you guys can answer.
SHANNON GLEESON: Maybe we'll collect a few, and then we'll come at each one.
AUDIENCE: Yes, I had a question for Shannon and for Kate, regarding the study-- the interviews that you're doing. Given the concerns of low wage workers, are you also interviewing this non-movement or sort of native workers in terms of are there similar concerns related to precarious work [INAUDIBLE] and other related issues?
SHANNON GLEESON: Any questions? [INAUDIBLE] take a moment [INAUDIBLE].
KATE GRIFFITH: So we're not interviewing non-immigrants, but you raise a good and important point. So right now our focus is really looking at foreign born low wage workers with different immigration statuses. That said, given the literature on low wage work, I'd expect a lot of similarities in terms of the wage theft and the retaliation concerns I talked about, but these other areas where the immigration bureaucracy is creating some precarity we still think we can get at by having the differences between the three immigration statuses.
SHANNON GLEESON: I'll just address Mary Jo's question. Yeah, I think in an ideal world-- and this is something that was raised when we were thinking about getting external funding is in the ideal world you would do follow ups with individuals who have TPS, be able to look at them after they essentially fall out of status. And we've basically chosen not to do that for two reasons. We kind of out of a research ethics perspective collected as minimal data as possible-- information on these individuals, just enough to get that initial interview. And in order to secure that initial interview also to not delve into asking individuals if we could follow up with them. So it's a trade off.
In some of my other work, I've done follow ups with people long term, and you get a lot of attrition, and it also creates a set of methodological issues of who remains in your sample and who doesn't, but I think that that's just kind of a call we made because we didn't want to hang on in that kind of way to people's information. And we felt that especially with the current political moment it would essentially just freak people out to follow them up and say, I remember you had TPS, and now you probably don't. Want to talk to me?
But I think that the question that service providers are having-- and this is something that we're also doing, is talking to service providers and people like yourself, who are dealing directly with these folks. We're using that as an opportunity to hear about kind of what the impacts are to then be able perhaps later to interview folks and talk retrospectively about what that looked like, but this is one of these in an ideal world great data to have, but not probably the best approach to keep people's information safer.
DAN LICHTER: Go ahead, we can ask another one.
AUDIENCE: [INAUDIBLE] thinking about also maybe the next stage looking at other dimensions besides just in school out of school, like criminal justice involvement?
STEVEN ALVARADO: That would be an amazing project. We've definitely been interested in doing that, but we run up against data constraints essentially, where the census just doesn't have enough data. And other data sets don't have as good of data to try to proxy for TPS. And so no matter where we look-- that cross-sectional data and large administrative populations or survey data sets-- it just doesn't quite seem to have the-- we don't quite seem to be able to find the right combination of data to be able to look at some of those other super important questions of criminal justice, which would be, obviously, very important, and other health outcomes, as well, that we thought about. Yeah, unfortunately.
SHANNON GLEESON: But there are a lot of folks who are doing work on this, especially in the DACA population, in the intersection with policing generally, partly because it affects eligibility and people's ability to reapply, but also because we see a lot of DACA folks being picked up even when they have that status. So I think in New York City there are some good examples of folks who have been doing that work. I don't know if you want to add anything.
MATTHEW HALL: No, I think that's good. Grab another question here.
AUDIENCE: So because I'm privileged enough to work for Salvadoran refugees every day, I get to hear a lot of their narratives. And I understand that these people are coming in currently, and you are analyzing people who came in before a certain date, but do you document the narrative when they talk about the structure of the government, or the country, or their experience? Do you qualitatively analyze any of the narrative and/or document in that interview? And then what do you with those qualitative analyses or the narrative that they give in the interview?
SHANNON GLEESON: About their migration story?
AUDIENCE: Right, but their experience in-- because I understand now when I listen to their narratives it's a lot about the corruptive government or the government structure in the way that-- its relation to poverty and/or violence, but prior to the people that you interviewed that have come in at a different time, I'm wondering about their narrative and how do you qualitatively analyze that? And then what do you do with that information?
SHANNON GLEESON: Sure. It's interesting. We've had conversations with our RAs who've been doing this work, and you go into it with one thing you want to talk about, and you sometimes come out with an interview that's really a cathartic debrief about the trauma of having gone through a massive natural disaster and/or the migration process itself if you're coming through the southern border. And so I think there's two things that we, as analysts, have to be able to do. One is to think about how that broader context of the personal experience beyond your specific outcome variable, for example-- how does that shape the analytical focus that you have?
So for example, we have to think really hard about what the literature calls the dual frame of reference. So people-- how do they think about, for example, coming forward and making claims on their rights? And whether or not they understand themselves to have those rates in comparison not only to others in the United States, but what they had back home. And also the risk that they're going to assume in coming forward and doing that if it means potentially having to be deported or removed to go back to that. So that's one piece of it. So how do we think about how that impacts your analytical model, but then I think the other thing is we've had to think a lot about how to do justice to these individual stories and not just treat it as kind of a covariate.
And so I think that is something-- we have a really great research assistant who is an anthropologist who's been doing work with the Haitian community. And I think we have to think through how do we maybe work with her to create other outlets or other pieces to tell their story. And so yeah, it's a challenge.
DAN LICHTER: So we're going to have to [INAUDIBLE] here before everybody leaves and there's no one here to clap. So [INAUDIBLE].
Thank you all for your questions.
We've received your request
You will be notified by email when the transcript and captions are available. The process may take up to 5 business days. Please contact firstname.lastname@example.org if you have any questions about this request.
The Institute for the Social Sciences' Deportation Relief Project (2015-2018) brought together a group of interdisciplinary scholars to examine the impact of temporary deportation relief and work authorization on two key institutions for immigrants: schools and the workplace. Drawing on expertise from sociology, demography, economics, and the law, the team employs quantitative and qualitative methods to analyze these pressing issues for immigration scholarship and the current public debate.
Team members Shannon Gleeson (Labor Relations, Law, & History), Matthew Hall (PAM), Jordan Matsudaira (PAM), Kate Griffith (Labor Relations, Law, & History) and Steven Alvarado (Sociology) outlined the progress of their research and future plans at a capstone event on April 12, 2018.