Sick Days: Logical Versus Survey Identification of the Foreign-Born Population in the United States

Abstract

The self-reported number of workdays missed due to injury or illness, or sick days, is a reliable measure of health among working-aged adults. Although sick days is a relatively underexplored health-related outcome in migration studies, it can provide a multidimensional understanding of immigrant wellbeing and integration. Current understandings of the association between migration status and sick days are limited for two reasons. First, in the United States, few nationally representative surveys collect migration status information. Second, researchers lack consensus on the most reliable approach for assigning migration status. We use the 2008 Survey of Income and Program Participation (SIPP) to examine sick days and draw comparisons between two methods for assigning migration status—a logical approach and a survey approach. The logical method assigns migration status to foreign-born respondents based on characteristics such as government employment or welfare receipt, while the survey approach relies on self-reported survey responses. Sick days among immigrants was correlated with and predicted by other health conditions available in the SIPP. Comparisons of sick days by migration status vary based on migration assignment approach. Lawful Permanent Residents (LPRs) reported more sick days than non-LPRs and appear less healthy when migration status is assigned using the logical approach. The logical approach also produced a gap in sick days between LPRs and non-LPRs that is not replicated in the survey approach. The results demonstrate that if migration status is not measured directly in the data, interpretation of migration status effects should proceed cautiously.

Keywords

sick days immigration SIPP legal status measurement

Introduction

A growing body of empirical evidence suggests that work, broadly conceptualized, is an important yet often overlooked contributor to an individual's health, as well as to health disparities across racial/ethnic groups (Ahonen et al. 2018; Lipscomb et al. 2006). Work is complex because it is intertwined with socioeconomic status, age, race/ethnicity, nativity, and gender (Ahonen et al. 2018). It is a source of income and may provide benefits such as health insurance, but it can also expose individuals to physical, environmental, and psychosocial risks (Lipscomb et al. 2006). A lack of understanding of the linkages between work and health may be particularly impactful in research focused on the foreign born, for whom migration decisions are often interrelated with work-related factors (Ahonen, Benavides and Benach 2007; Flynn et al. 2014; Massey et al. 1993; Massey and Espinosa 1997). Additionally, the sizeable foreign-born population (44.4 million) in the United States is essential to the (aging native-born) labor force (Budiman 2020). Currently, more than 17 percent of the US labor force is foreign born (US Bureau of Labor Statistics 2020), and nearly two out of three foreign-born adults participate in the labor force (US Bureau of Labor Statistics 2020).

A central, but often unexamined, work-related indicator of health is sick days, or time off for illness or injury (Ahonen, Benavides and Benach 2007; Ahonen et al. 2018). Among working-age adults, sick days—self-reported number of days of work missed due to illness or injury—is a reliable global measure that encompasses physical, social, and psychological dimensions of health and predicts mortality (Bambra and Norman 2006; Kivimäki et al. 2003; Marmot et al. 1995; Vahtera, Pentti and Kivimäki 2004). The number of sick days also serves as a proxy for the severity of a worker's illness or injury (Premji and Krause 2010). Yet to date, in the US-based health and immigration literature, sick days has been a relatively underexplored health-related outcome, despite the fact that it can provide a multidimensional understanding of immigrant wellbeing and integration.

A critical factor complicating the examination of sick days in the context of migration is accounting for an immigrant's legal and citizenship status (hereafter, migration status), given the diversification in recent decades of the foreign-born workforce in the United States (Budiman 2020; Congressional Budget Office 2020). Due to limited national-level US data and a lack of consensus as to the most reliable approach for assigning migration status, current understandings of the empirical association between migration status and the propensity to miss work due to illness is limited.

This article contributes to the growing body of empirical evidence examining differences in health and wellbeing across migration status groups (Altman et al. 2020a; Altman et al. 2020b; Altman and Bachmeier 2021; Hamilton, Hale and Savinar 2019; Spence et al. 2020; Waters and Pineau 2016; Young and Madrigal 2017) by concentrating on sick days. Our primary goal is to examine the prevalence of sick days among immigrants by migration status and to determine whether the prevalence of sick days varies by the approach used to assign migration status. To conduct our analyses, we use the 2008 Survey of Income and Program Participation (SIPP)¹ administered by the US Census Bureau. The SIPP allows us to not only examine our outcome of interest but, as importantly, compare two methods for assigning migration status to foreign-born respondents in survey microdata: a survey approach and a logical approach. The survey approach uses self-reported migration status via survey questions, while the commonly utilized logical approach assigns migration status to foreign-born respondents based on characteristics such as government employment or welfare receipt (Borjas 2017; Borjas and Slusky 2022). We, first, test the association between sick days and other health conditions and then use multivariate regression to predict sick days by migration status assignment method. Given the disproportionate representation of Hispanics in precarious migration statuses in the United States, we present results for all immigrants and separately by Hispanic ethnicity. We find that sick days is correlated with and predicted by other health conditions available in the SIPP among immigrants. Our regression results show that the reporting of sick days differed between Lawful Permanent Residents (LPRs) and non-LPRs based on migration status assignment method. Specifically, using the logical approach, we find that LPRs reported more sick days and appear less healthy than their non-LPR counterparts.

This article is guided by two broad motivations. The first is to add to social science understandings of how and the degree to which migration status—a variable most often omitted from studies of immigrant integration (Massey and Bartley 2005; Nee and Holbrow 2013)—is associated with various integration outcomes. To do so, we focus on a previously underexamined outcome, sick days, that relates to both immigrant health and immigrant socioeconomic position and mobility. The second motivation is to bring the importance of methodological decisions in migration status research into clearer focus. We address these goals through a novel research design, made possible by the SIPP, that tests two of the leading and most influential methodological approaches in a field fraught with methodological uncertainty and lacunae in data (Clark, Glick and Bures 2009; Glick 2010; Van Hook et al. 2015)—a survey approach and a logical approach.

In the sections that follow, we begin by discussing the existing literature on sick days, particularly as it pertains to the foreign-born population, highlighting why sick days is an informative indicator of immigrant health. We, then, consider whether and how sick days might vary by migration status, specifically whether immigrants with precarious migration statuses (i.e., unauthorized) will be more or less likely to report sick days. From there, we discuss and provide examples of studies using the logical and survey approaches and show that the two assignment approaches often produce conflicting estimates of migration status effects. The subsequent section introduces the data used, our measures, and analytical approach. We, then, present results from our tests of association between sick days and other health conditions, followed by a descriptive portrait of the SIPP sample, and provide estimates from our multivariate regressions predicting sick days by migration status assignment method. We conclude this article with a discussion of the implications of migration status assignment methods on health and wellbeing and the utility of sick days as an indicator of immigrant health and wellbeing.

Background

In the United States, federal laws do not mandate paid sick leave policies, and US state laws vary widely (Society for Human Resource Management 2021; US Department of Labor 2021). In most states, employers determine not only whether to offer sick days, paid or unpaid, but also employees’ eligibility based on tenure, how sick days accrue, and how these days can be used (National Conference of State Legislatures 2021; National Partnership for Women and Families 2021). Generally, sick-day benefits vary based on the employee's status and characteristics, such as full- or part-time employment, and by occupation and industry (US Department of Labor and US Bureau of Labor Statistics 2020). Most full-time, private-industry workers in 2020 had paid sick leave (86 percent), compared to only 45 percent of part-time workers (Ibid.). Workers in management or professional occupations have greater access to paid sick leave than do service, sales, and construction workers (Ibid.).

Several theoretical and empirical reasons exist to examine sick days among the foreign-born population. First, self-reported number of sick days does not rely on a doctor's diagnosis. Considering a health outcome that does not depend on a doctor's diagnosis is particularly imperative when studying a population, such as the foreign born, who have low rates of insurance coverage and for whom access to health care is closely linked to migration and citizenship status (Capps and Fix 2013; Derose et al. 2009; Goldman, Smith and Sood 2005; Marrow and Joseph 2015). Second, compared to the US-born population, immigrants are concentrated in service, construction, and production occupations and in part-time or contingent work, all characterized by low pay and limited access to benefits such as paid sick leave, overtime pay, and health insurance (US Bureau of Labor Statistics 2020; US Government Accountability Office 2015). Service workers in the bottom wage quartile and part-time employees are much less likely to receive sick pay benefits compared to managers or professionals in the highest wage quartile and full-time employees (US Bureau of Labor Statistics 2010). The work conducted by foreign-born residents in the United States often entails a high degree of occupational risk and a disproportionately high rate of occupational injuries (Byler and Robinson 2018; Hall and Greenman 2015; Orrenius and Zavodny 2009). Injuries often include physical injuries, musculoskeletal, repetitive motion injuries, falls, environmental exposures to toxic chemicals and noise, and psychosocial injuries (Orrenius and Zavodny 2009). Taken together, previous studies suggest that foreign-born workers in the United States, who are concentrated in dangerous jobs, are unlikely to receive benefits such as sick leave in the face of an occupational injury.

In addition to sick days being an important indicator of immigrants’ health, there are reasons to expect a gradient in sick days by migration status. Immigrants in the United States who are unauthorized or who have a precarious status (i.e., those who lack legal and/or citizenship status) are concentrated in high-risk, low-paying occupations and in non-standard or contingent working arrangements that do not provide sick-day benefits (Capps et al. 2013; Feldman 2006; Hall and Greenman 2015; US Government Accountability Office 2015). Additionally, returns on human capital, job mobility, and any benefits (i.e., paid leave) that accrue from job mobility are stratified by legal status (Hall, Greenman and Farkas 2010; Hall and Greenman 2015; Hall, Greenman and Yi 2018). Therefore, foreign-born workers who also have a precarious status may experience disadvantages in terms of access to sick days compared to their foreign-born peers with non-precarious statuses.

Yet it is unclear whether immigrants with precarious migration statuses will report greater or fewer sick days than their foreign-born peers with less precarious statuses. On the one hand, ethnographic studies, geographically constrained surveys, and occupation-specific research suggest that immigrants generally work, even when sick or injured. For example, unauthorized or legally precarious immigrants have been shown to be less likely to take sick days relative to legally resident migrants (Virtanen et al. 2001). Immigrants with precarious statuses often work out of fear of losing their job or their migration status being revealed by employers (De Castro et al. 2006; Eggerth et al. 2011; Premji and Krause 2010). Foreign-born workers report being fired for reporting and/or asking for time off following an injury, even though all workers in the United States, regardless of migration status, are granted rights and protections under the federal Occupational Safety and Health Act (OSHA) against employer retaliation (De Castro et al. 2006; Flynn, Eggerth and Jacobson 2015; Gleeson 2012). Waldinger and Lichter (2003) report a strong preference among employers in Los Angeles for immigrant workers primarily because they are perceived to be a pliant and reliable source of labor, implying the expectation that such workers would infrequently miss work due to illness.

On the other hand, having a precarious migration status in the United States is often conceptualized as a risk factor for poor health (Gonzales and Chavez 2012; Pebley et al. 2006; Perreira and Pedroza 2019; Yoshikawa and Kalil 2011). Because immigrants with precarious migration statuses work in risky occupations, they may report more sick days than immigrants with other less exclusionary statuses. However, to date, limited national analyses of sick days have been conducted, and none, to our knowledge, disaggregate the foreign-born population into migration status groups. A few occupation-specific studies have found some evidence that immigrants with precarious statuses reported more sick days. For example, in a sample of hotel room cleaners working in Las Vegas, Nevada, Hispanic and non-English speakers reported more sick days than their coworkers (Premji and Krause 2010). Among construction workers, Hispanics were more likely to report work-related injuries, to take time off work following an injury, and to have longer durations of missed work compared to non-Hispanic whites (Anderson, Hunting and Welch 2000; Brown, Brooks and Dong 2021). Though not examined, it is possible that immigrants with precarious statuses are likely sicker when reporting a sick day than those with less precarious statuses (Virtanen et al. 2001). The evidence to date, for reasons not limited to small, non-representative or occupation-specific samples, is inconclusive regarding the migration status gradient in sick days. In this article, our aim is to build upon prior theorizing and research to establish basic empirical patterns with respect to the prevalence of sick days among immigrants and whether sick days varies by migration status. While important, testing the mechanisms that directly link migration status to sick days is beyond the scope of our current analyses.

Migration Status Assignment

Beyond the scant number of studies, discussed above, examining the association between sick days and migration status, recent work in the broader literature on immigration, legal status, and health highlights inconsistent associations between physical and mental health outcomes and migration status (Hamilton, Hale and Savinar 2019; Waters and Pineau 2016; Young and Madrigal 2017). For some health outcomes, particularly those related to maternal health and birth outcomes, immigrants with precarious migration statuses experience comparable or even more favorable health than their legal peers (Kelaher and Jessop 2002; Korinek and Smith 2011; Reed et al. 2005). For example, in a study of low-birthweight births to mothers in New York City, babies born to unauthorized immigrant Hispanic women were not significantly more likely to be of low-birthweight, compared to those born to US-born Hispanic mothers (Kelaher and Jessop 2002). Yet the association between migration status and other types of health outcomes is less clear.

Limited or conflicting patterns by migration status are often found for physical health outcomes like blood pressure, diabetes, and chronic pain (Bitler and Shi 2006; Hamilton, Hale and Savinar 2019; Young and Pebley 2017). Evidence from the Los Angeles Family and Neighborhood Survey (LAFANS) indicated no variation in blood pressure by migration status among Hispanics in models that did not adjust for duration of US residence. By contrast, using the National Agricultural Workers Survey (NAWS), Hamilton and colleagues found that unauthorized Hispanic immigrant farmworkers reported less pain and chronic conditions, compared to their authorized peer farmworkers, and that naturalized citizens reported the most chronic health conditions (2019). These studies, while informative, provide conflicting evidence about the association between migration status and health outcomes.

A frequently theorized explanation for the conflicting evidence on the association between migration status and health outcomes emphasizes the lack of consistency or standardization in measurement of legal and/or citizenship status (Hamilton, Hale and Savinar 2019; Young and Madrigal 2017). A key reason for this inconsistency is that currently, only one nationally representative survey in the United States contains information about both citizenship (customarily included in most major surveys, including those administered by the federal government) and legal status among non-citizens—the SIPP. While the SIPP is primarily used to study economic wellbeing, it has also fielded questionnaire modules focusing on the physical health and wellbeing of both children and adults.

In the SIPP, foreign-born respondents are asked a series of survey questions about citizenship, visa status upon first entry to the United States, and any adjustments to visa status, post-arrival. Researchers have used this series of SIPP questions to assign migration status and to examine its association with socioeconomic outcomes such as wages, occupational risks and mobility, and material hardship (Altman et al. 2020a; Hall, Greenman and Farkas 2010; Hall and Greenman 2015). In one of the only studies to use the SIPP to examine physical health by migration status, Ziol-Guest and Kalil (2012) found that children whose parents were nonpermanent residents experienced poorer health outcomes and restricted access to health care compared to children whose parents’ status was less precarious.

In addition to the SIPP, a few geographically restricted or sub-population quantitative surveys contain questions that can be used in combination to assign an immigrant's citizenship and/or legal status (i.e., California Health Interview Study (CHIS)) or directly ask about legal status (i.e., LAFANS, NAWS). Because these data are not nationally representative, however, their generalizability is limited. Other types of data including administrative records (e.g., medical or birth records, Medicaid) have been used to approximate migration status through the presence or absence of identifying information, such as a Social Security Number or driver's license (Young and Madrigal 2017). While useful, administrative data often lack detailed information about sociodemographic characteristics or health and wellbeing indicators. Other national surveys used to study the foreign-born population, such as the National Health Interview Survey (NHIS) and the Current Population Survey (CPS), contain questions about citizenship status but do not contain legal status information. These surveys have been used to study immigrants’ health but are often limited to citizen versus non-citizen comparisons (Antecol and Bedard 2006). However, focusing solely on the citizen-noncitizen comparison among immigrants leaves unmeasured legal status variation within the non-citizen population.

To move beyond the citizen-noncitizen dichotomy in analyzing national survey data, scholars have progressively adapted alternative methodological approaches. One increasingly common method is a logical approach to the imputation of unknown legal status. Originally developed by demographer Jeffery Passel (Passel and Clark 1998; Passel 2006; Passel, Van Hook and Bean 2006; Passel and Cohn 2018), the logical approach identifies legal immigrants based on characteristics such as government employment or welfare receipt that, based on immigration laws, would make it unlikely that a foreign-born respondent is illegally residing in the United States (Borjas 2017; Borjas and Slusky 2022). After identifying immigrants who are ‘probably legal,’ the logical method assumes that the remainder are unauthorized. One attractive advantage of the logical method is that, if the approach accurately assigns migration status, it could be applied to any data set (assuming the data set has the requisite indicators) to examine an outcome of interest. As such, the logical approach has been used to examine migration status differences across a range of outcomes, including wages, labor force participation, and health insurance coverage (Borjas 2017; Borjas and Cassidy 2019; Cohen and Schpero 2018; Stimpson, Wilson and Su 2013).

To date, a limited number of studies have directly compared the logical approach to other methods, such as a survey approach. Recent comparisons of the logical and survey approaches for assigning migration status on socioeconomic and wellbeing outcomes find that the logical and survey assignment approaches produce conflicting results (e.g., Altman et al. 2020b; Spence et al. 2020). These studies, which use the SIPP, notably suggest that the logical method disproportionately relies on reported Medicaid or other public assistance receipt to identify persons as probable LPRs (Ibid.). This reliance on public assistance receipt, in turn, yields an aggregate LPR population that is likely artificially disadvantaged, thus biasing estimates of migration status effects. Research evaluating five different assignment approaches, using Monte Carlo simulations, similarly concluded that the logical approach produced the most biased coefficients when estimating the association between migration status and health insurance coverage (Van Hook et al. 2015). Additionally, researchers found no evidence that the SIPP immigration-related questions, which are the foundation for the survey approach, produced biased demographic profiles of migration status groups (Bachmeier, Van Hook and Bean 2014). Accordingly, we expect that using the immigration-related questions in the SIPP to infer migration status (henceforth, the survey approach) will produce reliable estimates of sick days by migration status. A principal motivating reason to compare the survey and logical migration status assignment approaches is to highlight the sensitivity of estimates to the method used to assign status.

A brief example further illustrates the complication of migration status assignment comparisons and the resulting conclusions. Hall and colleagues estimated the wage penalty of being unauthorized among Mexican immigrants to the United States, using SIPP survey questions to assign migration status (2010). Borjas and Cassidy (2019) also estimated the wage penalty associated with legal status by applying a logical assignment approach to data from the American Community Survey (ACS) and CPS. After adjusting for demographic and socioeconomic characteristics, both studies, unsurprisingly, found that unauthorized workers experienced a wage penalty relative to authorized workers. Yet, below the surface lie stark differences. Borjas and Cassidy estimated a 35 percent gross gap in wages between authorized and unauthorized immigrants, which reduced to 4 percent after socioeconomic and demographic adjustment. Hall and colleagues, however, found the gross gap to be only 17 percent, which reduced to 8 percent following adjustment.

In summary, the state of social scientific knowledge on the link between migration status and a host of immigrant integration outcomes is potentially muddied by the existence of various methods of assigning immigrants’ migration status in surveys that do not include migration status questions. The result is a literature on migration status “effects” that is rich in diversity of migration status assignment methods, but poor in comparing these methods. Moreover, the paucity of generalizable survey data with information about persons’ legal and citizenship status is a major obstacle to addressing pressing questions regarding the association between migration status and health. Specifically, the absence of methodological and scientific consensus and comparative research hinders current understandings of how migration status operates as a social determinant of health. In turn, this lack of methodological—and, therefore, empirical—consensus underlying research on migration status effects hamstrings public policy interventions. Therefore, this article makes notable contributions to this important and growing literature on both substantive and methodological fronts.

Data and Methods

Data

To evaluate the association between migration status and sick days, we use the 2008 SIPP.² The SIPP, conducted by the US Census Bureau, is a longitudinal household survey consisting of a set of core questions and topical modules that collect information on the economic and social wellbeing of the US population (US Bureau of the Census 2001). Households are followed for approximately three years and interviewed every four months. We use the SIPP to sort foreign-born non-citizens into two migration statuses: LPRs and non-LPR, non-citizens. We focus our analyses on the foreign-born working age population between ages 18 and 64.

We approximate the sensitivity of our empirical results to the method used to classify survey respondents as LPRs or non-LPRs by comparing results based on two specific assignment approaches. The SIPP is strategic here because it includes relatively direct survey questions about legal status and a robust set of indicators that can be used in logical imputation methods, allowing us to test two independent measurement strategies, while holding the data set constant. We constructed survey-based measures of migration status, using SIPP survey questions asked of all foreign-born respondents about their citizenship and visa status. All foreign-born respondents are queried about their visa status upon arrival in the United States. Noncitizen respondents self-identifying as non-LPR arrivals are, then, asked if they have subsequently adjusted their visa status. LPRs and a residual category of non-citizen, non-LPR foreign-born respondents are identified, using these self-reported immigration questions. The latter group is predominately unauthorized immigrants but also includes legal, non-immigrants on temporary visas.

The logical approach assigns immigrants to one of these two migration status groups, using other indicators in the data presumed to be highly determinative of one's migration status. In implementing the logical approach, we follow recent work by Borjas and colleagues (Borjas 2017; Borjas and Slusky 2022). Borjas provides a concise articulation and replication of the logical approach outlined by Passel (Passel and Cohn 2014). In sum, Borjas's logical approach categorizes a foreign-born national as a non-citizen LPR if they meet one or more of the following criteria: 1) arrived to the United States before 1980; 2) receives Social Security, SSI, Medicaid, Medicare, or Military Insurance; 3) is a veteran or currently in the military; 4) is employed in the government; 5) receives housing subsidies or is a spouse of someone residing in/receiving housing subsidies; 6) was born in Cuba; 7) has an occupation requiring licensing; or 8) is the spouse of a legal immigrant or citizen. Again, the motivating rationale and assumptions behind the indicators selected by the logical method are that any foreign-born non-citizen resident of the United States meeting any of these criteria is highly unlikely to be an unauthorized immigrant.

Measurement

Sick days

SIPP respondents are asked the number of days in the past 12 months that illness or injury, including hospitalization, kept them in bed more than half the day. In addition to the continuous measure, we created two dichotomous indicators of sick days to capture two or more sick days and five or more sick days, as done by other scholars (Marmot et al. 1995; Strong and Zimmerman 2005). We present results for two-plus sick days because the share of working-age immigrant adults reporting five-plus sick days is relatively small (8.8 percent). All analyses described below tested both dichotomous measures and found substantively similar results (available from the authors).

Health conditions

Though limited, the SIPP includes indicators of the following health conditions. Self-rated health captures respondents’ evaluation of their overall health, using a likert scale ranging from 1 to 5 with higher values indicating more favorable health. Functional limitation is a continuous indicator of limitations, ranging from 0 to 10. We constructed a sum scale indicating the number of emotional conditions a respondent reports, ranging from 0 to 6. The self-reported conditions include frequently feeling depressed or anxious; having trouble concentrating, getting along with others, coping with stressors, or managing everyday activities; or having any other mental or emotional condition. We also include a dichotomous indicator of whether the respondent reported having a long-lasting physical or mental condition that made it difficult to remain employed or find employment.

Control variables

Our multivariate models controlled for several important background variables likely associated with our outcome and migration status. We included standard demographic and family measures including age (measured in years), sex (reference = female), currently married (vs. not currently married as the reference), and binary measures of race (Black = 1), Hispanic ethnicity (Hispanic = 1), and the presence of one's own children in the household (children present = 1). In terms of human capital and socioeconomic position, we controlled for educational attainment (reference = less than high school, high school, some college, or college BA or beyond) and included dichotomous indicators for having total family income below the poverty line, homeownership, having health insurance coverage, and being employed. We also adjusted for important immigration-related variables such as English language proficiency (reference = “only English”, compared to those who speak English “very well,” “well,” “not well,” or “not at all”), birth region (reference = Europe or Canada, Asia, Africa, Caribbean, Mexico or Central America, and South America), and duration of US residence, measured in years—shown to be an important covariate of health among immigrants (Hamilton, Hale and Savinar 2019; Young and Pebley 2017). Finally, we included contextual variables, specifically US region of residence (reference = northeast) and metropolitan status (reference = not in metropolitan area).

Analysis

While sick days may serve as an encompassing indicator of health among working-age adults, the literature on sick days for immigrants and by migration status is limited. Before examining sick days by migration status assignment method, we tested the association between sick days and other health conditions—self-rated health, functional limitation, emotional conditions, and work limiting conditions. Following the lead of other scholars who have examined sick days among all adults (Marmot et al. 1995), we present correlations and multivariate models predicting continuous sick days, using the other four health conditions.

We examined sick days by migration status assignment method, using bivariate and multivariate analyses among the foreign born. To aid in interpreting the results, we estimated the adjusted predicted probability of sick days by migration status assignment method. The regression models included all controls described above. We examined models for Hispanic immigrants separately because of their overrepresentation in disadvantaged migration statuses and socioeconomic statuses. All analyses were conducted in Stata 16 and weighted using the person weights provided by the Census Bureau.

Results

Tables 1–3 examine sick days as an encompassing measure of health. Table 1 provides univariate statistics for the mean number of sick days and the percentage of respondents reporting 2 + sick days and 5 + sick days for all foreign born in the sample and among the Hispanic foreign born. On average, working-age foreign-born adults reported 2.7 sick days. More than one in five foreign-born adults reported 2 + sick days (21.5 percent), and 8.2 percent reported 5 + sick days. Descriptive patterns are similar for Hispanic immigrants, though Hispanic immigrants reported on average only 2.5 sick days.

Table 1.

Descriptive Analysis of Sick Days among the Foreign-Born.

	All Foreign-born	Hispanic Foreign-born
Sick days (mean)	2.7	2.5
2 + Sick days (%)	21.5%	18.4%
5 + Sick days (%)	8.2%	7.0%
N	6,266	2,578

Source: Survey of Income and Program Participation (SIPP) 2008.

Note: Means and percentages are estimated using SIPP person and replicate weights.

Table 2.

Correlations Between Continuous Sick Days and Other Health Conditions among the Foreign-Born.

	All Foreign-born	Hispanic Foreign-born
Self-rated Health (1-5)	-0.191	-0.164
Functional Limitations (0-10)	0.272	0.249
Emotional Conditions (0-6)	0.221	0.218
Work Limiting Condition (0-1)	0.214	0.213
N	6,266	2,578

Source: Survey of Income and Program Participation (SIPP) 2008.

Note: Correlations are weighted using SIPP person weights.

Table 3.

OLS Regression Models Predicting Continuous Sick Days among the Foreign-Born.

	All Foreign-Born		Hispanic Foreign-Born
	b	SE	b	SE
Self-rated Health	−1.91***	0.291	−1.67***	0.342
Functional Limitations	2.50***	0.709	1.96*	0.342
Emotional Conditions	2.83*	1.314	2.96	0.667
Any Work Limiting Condition	1.75	3.786	4.07	2.069
Constant	8.32***	1.200	7.22***	1.289
N	6,266		2,578

Source: Survey of Income and Program Participation (SIPP) 2008.

Note: *p < .05, ***p < .001 Models are weighted using SIPP person and replicate weights.

Table 2 presents the correlations between sick days and other health conditions. Prior research suggests that sick days is a global measure of health that includes both physical and psychological dimensions (Marmot et al. 1995; Strong and Zimmerman 2005). Few prior analyses, however, test this assertion among immigrants. We, thus, estimate the correlation between the continuous indicator of sick days and four measures of health conditions, including self-rated health, functional limitations, emotional conditions, and any work limiting condition for all foreign-born and Hispanic foreign-born respondents. Among all foreign-born adults, self-rated health was negatively correlated with continuous sick days. Consistent with Marmot et al. (1995), we find that sick days was positively associated with the other health conditions among all foreign-born and Hispanic foreign-born adults.

The results from a linear regression model of continuous sick days, using the health conditions, are displayed in Table 3. Supporting the results of Table 2, Table 3 shows that health conditions are generally significant predictors of sick days in the expected directions for all foreign-born adults and for Hispanic foreign-born adults. The measure of any work-limiting health conditions is not predictive of sick days, and while emotional conditions is a significant predictor in the model for all foreign born, it is not significant in the Hispanic foreign-born model. In sum, these results suggest that sick days was associated with other health conditions among foreign-born respondents.

The next step of the analysis, shown in Table 4, compares descriptive statistics for sick days by migration status, using the logical and survey assignment methods. Both migration status assignment approaches suggest that LPRs reported more sick days and that a larger percentage of LPRs reported 2 + and 5 + sick days, compared to non-LPRs. Importantly, the differences between LPRs and non-LPRs in mean sick days and percentage reporting 2 + and 5 + sick days are only significant using the logical assignment method, not the survey method. LPRs assigned using the logical method reported an average of 2.8 sick days, and 23 percent reported 2 + sick days. In comparison, LPRs assigned using the survey method reported over one-half a day less of sick leave (2.2), and only 19 percent reported 2 + sick days. Non-LPRs assigned using the logical method reported 1.5 sick days, and only 16 percent reported 2 + sick days, relative to 1.9 sick days and 19 percent reporting 2 + sick days using the survey approach.

Table 4.

Descriptive Analysis of Sick Days among the Foreign-Born by Migration Status and Assignment Method.

	Logical		Difference (LPR-nonLPR)	Survey		Difference (LPR-nonLPR)
	LPR	non-LPR	Difference (LPR-nonLPR)	LPR	non-LPR	Difference (LPR-nonLPR)
Sick days (mean)	2.8	1.5	1.3*	2.2	1.9	0.3
2 + Sick days (%)	23.3%	16.1%	7.2%***	19.4%	19.1%	0.3%
5 + Sick days (%)	10.0%	5.2%	4.8%***	7.5%	6.9%	0.6%
N	1,822	2,212		2,727	1,307

Source: Survey of Income and Program Participation (SIPP) 2008.

Note: *p < .05, ***p < .001. Means and percentages are estimated using SIPP person and replicate weights.

We find several noteworthy comparisons by migration status assignment method. First, the zero-order difference in mean sick days between LPRs and non-LPRs is larger, and only statistically significant, using the logical assignment (difference of 1.3 sick days) relative to the survey assignment (difference of 0.3 sick days). Moreover, logically assigned LPRs (2.8) reported more sick days than survey-assigned LPRs (2.2), while logically assigned non-LPRs (1.5) reported fewer sick days than survey-assigned non-LPRs (1.9). Insofar as sick days approximate the health of individual immigrants, the simple comparisons presented in Table 4 suggest that the association between migration status and health varies, depending on the method used to infer an individual's migration status. The criteria constituting the logical assignment method “produce” a non-LPR population that appears healthier than the one defined by survey responses to questions about visas and legal status in the SIPP.

Next, we estimated separate multivariate regression models predicting sick days, using the logical approach and the survey approach measures of migration status. All models were weighted and fully adjusted. We tested OLS models for continuous sick days, separate logit models for 2 + and 5 + sick days, and zero-inflated negative binomial models. The results were consistent regardless of the specification of the dependent variable and model (available from the authors). Table 1A in the Online Appendix presents the logit results for 2 + sick days. Using the logical approach, LPRs had significantly increased log odds of reporting 2 + sick days compared to non-LPRs (about 37 percent higher odds, exp(0.317) = 1.37). In contrast, using the survey approach, LPRs did not significantly differ from non-LPRs in their log odds of reporting 2 + sick days. It is noteworthy that the LPR status coefficients are in opposite directions in the logical and survey approaches, positive and negative, respectively. Regarding the controls in the logit models for both approaches, men were significantly less likely to report 2 + sick days compared to women, and with each additional year of age, the log odds of reporting 2 + sick days significantly increased, and Blacks were more likely than non-Blacks to report 2 + sick days.

To aid in comparison and interpretation of the logistic regression results, we produced adjusted predicted probabilities of 2 + sick days by migration status assignment, using the margins command in Stata (Figure 1). The predicted probability of an LPR reporting 2 + sick days was larger using the logical approach than the survey approach (0.20 vs. 0.17, respectively). For non-LPRs, the predicted probability of 2 + sick days was smaller using the logical approach than the survey approach (0.16 vs. 0.19, respectively).

Figure 1.

Predicted probability of reporting two + sick days for LPRs and non-LPRs by assignment approach. Notes: Error bars show 95% confidence intervals. Each logit model includes controls for sex, age, race, Hispanic origin, region of birth, years in the US, limited English proficiency, education level, poverty status, employment status, insurance coverage, marital status, home ownership, state-region, metro status, and living with own child. Predicted probabilities were generated with covariates at mean values.

Discussion

We contribute to the literature assessing health outcomes by migration status by focusing on an underexplored health outcome, sick days. We used the 2008 SIPP to generate migration status categories, using both the logical approach outlined by Borjas (2017, 2022) and SIPP survey questions. To date, the association between migration status and health (and other) outcomes potentially lacks consensus because of the inconsistency in migration status measurement. Consequently, some studies suggest that immigrants with precarious statuses experience favorable health outcomes relative to their authorized peers (Kelaher and Jessop 2002; Korinek and Smith 2011; Reed et al. 2005), while other studies report inconsistent or poorer health for those with precarious statuses (Hamilton, Hale and Savinar 2019; Waters and Pineau 2016; Young and Madrigal 2017). Because few studies have compared survey-based results to logical-based results, empirical questions remain as to how sensitive estimates are to the method used to assign status. As we demonstrate in the analyses presented (and the comparisons we made in the literature review), if migration status is not asked or measured directly in the data, interpretation of migration status effects should proceed cautiously.

We began by examining sick days as a measure of health. Among all immigrants and in a Hispanic subsample, sick days was correlated with other health conditions available in the SIPP. Consistent with previous research (Marmot et al. 1995), these analyses indicate that sick days is an indicator of health status for immigrants. The foreign-born population participates in the labor force at a high rate but also faces systemic obstacles to accessing and utilizing health care. As a result, this population may be less likely to receive a diagnosis for objective health measures such as type 2 diabetes or heart disease (Bacon, Riosmena and Rogers 2017; Barcellos, Goldman and Smith 2012; Hsueh et al. 2020). Thus, an encompassing health indicator such as self-reported number of sick days may provide immigration and health researchers an additional and alterative measure of health.

Our empirical comparisons of sick days by migration status yielded important patterns. The logical approach implies that logically assigned LPRs experienced more sick days compared to survey-assigned LPRs; additionally, the logical approach suggests that non-LPRs reported fewer sick days relative to the survey approach. Overall, the logical approach produced a striking gap in sick days between LPRs and non-LPRs that is not replicated in the survey approach. If researchers rely exclusively on the logical method to assign migration status and then examine sick days, they may conclude that LPRs are in poorer health than non-LPRs. On the other hand, using survey questions researchers may conclude that LPRs and non-LPRs have comparable health statuses, based on reported sick days.

Why might we expect differences in mean sick days and 2 + sick days by migration status assignment approach, especially differences indicating that LPRs are less healthy when migration status is assigned using the logical approach? One plausible explanation raised in this article is measurement of migration status—notably, the criteria the logical approach uses for assigning LPR status. Recall that the logical approach assigns any foreign-born respondent LPR status if they report the receipt of SSI, Medicaid, Medicare, or Military Insurance. Receipt of these forms of public health insurance benefits may serve as a proxy for latent, underlying poor health or chronic conditions. Therefore, the dependent variable, health status, may implicitly be used as a criterion for the logical assignment of migration status and lead to biased conclusions regarding LPRs’ health status. Our multivariate models provide tentative evidence of this confounding when using public health insurance receipt as a criterion of migration status assignment and predicting sick days. In the model using the logical approach for migration status, health insurance coverage is not a significant predictor of 2 + sick days, but in the multivariate model using the survey approach, health insurance coverage is significant and positively associated with the log odds of taking 2 + sick days. The latter finding supports prior evidence of the positive association between insurance status and sick days, that is, those who are insured are more likely to take a sick day (DeRigne, Stoddard-Dare and Quinn 2016).

Furthermore, the logical approach counts anyone in an occupation requiring licensing as an LPR. Given that at least 11 states have extended professional and occupational licenses to unauthorized immigrants, including those who are Deferred Action for Childhood Arrivals (DACA) and Temporary Protected Status (TPS) recipients, the logical approach may incorrectly assign LPR status to individuals who are non-citizen non-LPRs (Catholic Legal Immigration Network 2019; Moriarty 2019). Additional detailed and recent data about DACA and/or TPS status are needed to elucidate these patterns.

It is plausible that the difference in sick days between LPRs and non-LPRs is due to the occupational distribution of these two groups, rather than (or in addition to) the method used to assign migration status. In other words, the gap in sick days reported by LPRs and non-LPRs may be attributable to the fact that LPRs and non-LPRs are employed in different types of occupations (Passel and Cohn 2016). LPRs are concentrated in professional and related occupations, as well as service, while unauthorized immigrants are disproportionately represented in service, production, and construction (Passel and Cohn 2016). If occupational distribution explained (or partially explained) the sick day gap between migration status groups, we would expect LPRs and non-LPRs to have significantly different probabilities of sick days, regardless of the migration status assignment approach. However, the gap in 2 + sick days is only statistically significant when using the logical approach to assign migration status. Using the survey approach yields no statistically significant gap in sick days reporting between LPRs and non-LPRs. Furthermore, while the multivariate models do not adjust for occupation directly, they do include factors that are highly correlated with occupation including employment status, educational attainment, English language proficiency, race/ethnicity, and birth region.

Additionally, latent, non-work-related daily stressors of living in the United States as an immigrant likely impact both LPRs and non-LPRs and, thus, sick days. It is possible that, due to their precarious status, non-LPRs experience more intense or more frequent stressors than LPRs. In this light, the logical method's results that LPRs report more sick days are surprising. However, we caution that further research, which may necessitate qualitative data, is needed to examine the role of these stressors on wellbeing outcomes, such as sick days by migration status.

The results presented here corroborate recent methodological work that similarly reports discrepancies in health and wellbeing outcomes between LPRs assigned using logical and survey approaches. For example, using public-use SIPP data, some scholars find that LPRs assigned using the logical approach had higher predicted probabilities of being in poverty and using public health insurance relative to survey-assigned LPRs (Altman et al. 2020b; Spence et al. 2020). Recently, Ro and Van Hook (2021) utilized the restricted-use SIPP to compare four assignment methods for identifying the unauthorized population among Asian and Hispanic non-LPRs. Their results “did not provide strong support for the logical edit approach as a preferred imputation approach” and point to how the criteria used in the logical approach are often part of the dependent variable (Ro and Van Hook 2021:10)

Accumulating evidence indicates that because the logical approach relies on public benefit program participation as a component of the method and because such participation is associated with health and wellbeing outcomes, the logical approach may bias conclusions about migration status effects. In addition to keen attention to migration status assignment and its constituent criteria, more comparative studies of assignment approaches, using both public and restricted-use data, are needed. Additionally, researchers must develop innovative methods and identify novel data sources or linkages to evaluate migration status assignments approaches. Ideally, researchers could identify an independent data source or estimate against which to evaluate the assignment approaches.

Despite the important implications of our research, this article has notable limitations. First, because of the absence of necessary visa status adjustment questions at the time of the survey for more recent SIPP panels (i.e., 2014 and 2018), we use data from 2008. There may be notable compositional differences between the foreign-born population surveyed in 2008 compared to those surveyed more recently. For instance, the number of lawful immigrants in the labor force grew from 18.1 to 19.5 million between 2009 and 2014 while the number of unauthorized immigrants in the labor force declined from 8.1 to 8.0 million in the same time period (Passel and Cohn 2016). Second, in the public-use SIPP data, we are unable to disaggregate the unauthorized fully from the legally resident non-immigrants (LNIs). In other words, in the public-use SIPP, the non-LPR group consists of foreign-born respondents who lack legal status and citizenship status (i.e., the unauthorized), as well as foreign-born respondents who are in the United States legally but not as an LPR or naturalized citizen (i.e., LNI). The inability to disaggregate the non-LPRs in public-use data may mask important within-group differences in their likelihood of reporting sick days. Finally, the SIPP has a lower response rate than other national surveys such as the American Community Survey. Response rates may also vary by migration status: immigrants with precarious statuses may be fearful of participating in a federal survey. Additionally, prior research suggests that the SIPP undercounts immigrants with precarious statuses at a higher rate than other surveys such as the American Community Survey (Bachmeier, Van Hook and Bean 2014). Yet sample sizes of non-LPRs in the SIPP, which predominantly includes unauthorized immigrants, were sufficient to estimate multivariate models. It is noteworthy that numerous recently published studies have utilized the SIPP to examine socioeconomic wellbeing outcomes by migration status (Altman et al. 2020a; Altman et al. 2020b; Ro and Van Hook 2021; Spence et al. 2020).

Conclusion

This article contributes to a growing literature, spanning multiple disciplines, examining whether and how migration statuses are associated with health and wellbeing (Bitler and Shi 2006; Hamilton, Hale and Savinar 2019; Waters and Pineau 2016; Young and Madrigal 2017). Our research contributions are both empirical and methodological. Our empirical results call attention to the utility of sick days as an indicator of health for the foreign-born population. Methodologically, the results underscore the importance of migration status measurement. Empirically and methodologically, the results highlight how the logical migration status approach disproportionately represents LPRs as sick, impoverished, and reliant on public benefits. In the US context, this representation may lead to misdirected policies and interventions.

More broadly, the results raise conceptual issues for studying immigrant health, such as what health outcomes are analyzed and disaggregating the diverse foreign-born population by migration status. The results, especially the finding that survey-assigned LPRs and non-LPRs were not significantly different in their probability of 2 + sick days, point to conceptual puzzles about what it means for an LPR versus a non-LPR to take a sick day, particularly in a US context. If the LPR has paid sick leave and/or health insurance coverage, are they afforded time and compensation to recover? Do they take sick days when they are less sick than non-LPRs who may not have paid sick leave and/or health insurance? In other words, a sick day may be a latent indictor of illness severity or underlying disability for immigrants with precarious migration statuses. Further, researchers must consider if migration statuses are assigned using latent or observed components of the outcome of interest. If migration status is assigned with a component of the outcome, what are the magnitude and/or direction of the bias? In sum, the results compel researchers to assess immigrant integration using complex indicators of wellbeing such as sick days and careful assignment of migration status.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Claire E. Altman

Notes

Table 1A.

Logistic regression model predicting log odds of reporting 2 + sick days by legal status imputation approach

	Logical Approach		Survey Approach
	Coef.	se	Coef.	se
LPR status (ref = non-citizen, non-LPR)	0.317**	0.097	-0.148	0.103
Male	-0.535***	0.094	-0.572***	0.094
Age	0.016***	0.005	0.018***	0.005
Black (ref = non-black)	0.445*	0.205	0.429*	0.205
Hispanic (ref = non-Hispanic)	-0.005	0.179	0.024	0.178
Region of Birth (ref = Europe/Canada)
Asia	-0.053	0.166	-0.089	0.167
Africa	-0.078	0.299	-0.07	0.297
Caribbean	0.3	0.263	0.398	0.261
Mexico/Central America	-0.251	0.235	-0.331	0.237
South America	-0.081	0.243	-0.097	0.243
Years in US	-0.003	0.006	0	0.006
Limited English proficient	-0.085	0.107	-0.094	0.107
Below federal poverty line	0.087	0.109	0.108	0.108
Education (ref = less than high school)
High school/GED	0.043	0.126	0.04	0.125
Some college	0.131	0.135	0.144	0.135
Bachelor's or more	0.254	0.149	0.231	0.15
Employed	-0.185	0.097	-0.192	0.097
Own home	-0.133	0.098	-0.121	0.098
Has insurance	0.149	0.103	0.235*	0.101
Married	-0.232*	0.105	-0.164	0.104
Has own child	0.047	0.096	0.07	0.096
Sample size	4,034		4,034
Pseudo R2	0.057		0.055

Note: Model also includes the following covariates: US region of residence and metropolitan status. Weighted using SIPP person and replicate weights. *** p < .001, ** p < .01, *p < .05.

References

Ahonen

E. Q.

Benavides

F. G.

Benach

. 2007. “Immigrant Populations, Work and Health—a Systematic Literature Review.” Scandinavian Journal of Work, Environment & Health 33(2):96–104. http://www.jstor.org/stable/40967631.

Ahonen

E. Q.

Fujishiro

Cunningham

Flynn

. 2018. “Work as an Inclusive Part of Population Health Inequities Research and Prevention.” American Journal of Public Health 108(3): 306–11. https://doi.org/10.2105/AJPH.2017.304214 .

Altman

C. E.

Bachmeier

J. D.

. 2021. “The Weight of Being Unauthorized? Legal Status Variation in the Association Between Us Exposure and Obesity among Hispanic Immigrants in Los Angeles.” Journal of Immigrant & Minority Health 23: 936–45. https://doi.org/10.1007/s10903-021-01210-x.

Altman

C. E.

Heflin

Jun

Bachmeier

J. D.

. 2020a. “Material Hardship among Immigrants in the United States: Variation by Citizenship, Legal Status and Origin in the 1996–2008 SIPP.” Population Research and Policy Review 40: 362-99. https://doi.org/10.1007/s11113-020-09588-6.

Altman

C. E.

Spence

Hamilton

Bachmeier

J. D.

. 2020b. “Health Insurance Coverage: Logical Versus Survey Identification of the Foreign-Born.” Journal of Immigrant & Minority Health 23:606-14. https://doi.org/10.1007/s10903-020-01045-y.

Anderson

J. T.

Hunting

K. L.

Welch

L. S.

. 2000. “Injury and Employment Patterns among Hispanic Construction Workers.” Journal of Occupational and Environmental Medicine 42(2): 176–86. https://doi.org/10.1097/00043764-200002000-00016 .

Antecol

Bedard

. 2006. “Unhealthy Assimilation: Why Do Immigrants Converge to American Health Status Levels?” Demography 43(2): 337–60. https://doi.org/10.1353/dem.2006.0011.

Bachmeier

J. D.

Van Hook

Bean

F. D.

. 2014. “Can We Measure Immigrants’ Legal Status? Lessons from Two U.S. Surveys.” International Migration Review 48(2): 538–66. https://doi.org/10.1111/imre.12059 .

Bacon

Riosmena

Rogers

R. G.

. 2017. “Does the Hispanic Health Advantage Extend to Better Management of Hypertension? The Role of Socioeconomic Status, Sociobehavioral Factors, and Health Care Access.” Biodemography and Social Biology 63(3): 262–77. https://doi.org/10.1080/19485565.2017.1353407 .

10.

Bambra

Norman

. 2006. “What Is the Association Between Sickness Absence, Mortality and Morbidity?” Health & Place 12(4): 728–33. https://doi.org/10.1016/j.healthplace.2005.02.008 .

11.

Barcellos

S. H.

Goldman

D. P.

Smith

J. P.

. 2012. “Undiagnosed Disease, Especially Diabetes, Casts Doubt on Some of Reported Health ‘Advantage’of Recent Mexican Immigrants.” Health Affairs 31(12): 2727–37. https://doi.org/10.1377/hlthaff.2011.0973.

12.

Bitler

Shi

. 2006. Health Insurance, Health Care Use, and Health Status in Los Angeles County. San Francisco, CA: Public Policy Institute of California.

13.

Borjas

G. J

. 2017. “The Labor Supply of Undocumented Immigrants.” Labour Economics 46: 1–13. https://doi.org/10.1016/j.labeco.2017.02.004 .

14.

Borjas

G. J.

Cassidy

. 2019. “The Wage Penalty to Undocumented Immigration.” Labour Economics 61 (S1): 101757.

15.

Borjas

G. J

David

J. G.

. 2022. “Health, Employment, and Disability: Implications from the Undocumented Population.” American Journal of Health Economics 8 (1): 1–29. https://doi.org/10.1086/716465.

16.

Borjas

G. J.

Slusky

D. J. G.

. 2022. “Health, Employment, and Disability: Implications from the Undocumented Population.” American Journal of Health Economics 8 (1): 1–29. https://doi.org/10.1086/716465.

17.

Brown

Brooks

R. D.

Dong

X. S.

. 2021. “Injury Inequalities among Us Construction Workers.” Journal of Occupational and Environmental Hygiene 18 (4-5): 159–68. https://doi.org/10.1080/15459624.2021.1888958.

18.

Budiman

2020. “Key Findings About U.S. Immigrants.” Pew Research Center. https://www.pewresearch.org/fact-tank/2020/08/20/key-findings-about-u-s-immigrants/.

19.

Byler

C. G.

Robinson

W. C.

. 2018. “Differences in Patterns of Mortality Between Foreign-Born and Native-Born Workers Due to Fatal Occupational Injury in the USA from 2003 to 2010.” Journal of Immigrant and Minority Health 20(1): 26–32. https://doi.org/10.1007/s10903-016-0503-2.

20.

Capps

Bachmeier

J. D.

Fix

Van Hook

. 2013. A Demographic, Socioeconomic, and Health Coverage Profile of Unauthorized Immigrants in the United States. Washington, DC: Migration Policy Institute.

21.

Capps

Fix

. 2013. “Immigration Reform: A Long Road to Citizenship and Insurance Coverage.” Health Affairs 32(4): 639–42. https://doi.org/10.1377/hlthaff.2013.0187 .

22.

Catholic Legal Immigration Network. 2019. “Professional and Occupational Licenses for Immigrants.” file:///C:/Users/altmanc/Downloads/Professional-Occupational-licenses-for-immigrants2019.pdf.

23.

Clark

R. L.

Glick

J. E.

Bures

R. M.

. 2009. “Immigrant Families Over the Life Course: Research Directions and Needs.” Journal of Family Issues 30(6): 852–72. https://doi.org/10.1177/0192513x09332162 .

24.

Cohen

M. S.

Schpero

W. L.

. 2018. “Household Immigration Status Had Differential Impact on Medicaid Enrollment in Expansion and Nonexpansion States.” Health Affairs 37 (7): 394–401. https://doi.org/10.1377/hlthaff.2017.0978NaN.

25.

Congressional Budget Office. 2020. The Employment of Foreign-Born People. Congress of the United States. www.cbo.gov/publication/56357.

26.

De Castro

A. B.

Fujishiro

Sweitzer

Oliva

. 2006. “How Immigrant Workers Experience Workplace Problems: A Qualitative Study.” Archives of Environmental & Occupational Health 61(6): 249–58. https://doi.org/10.3200/AEOH.61.6.249-258.

27.

DeRigne

L-A.

Stoddard-Dare

Quinn

. 2016. “Workers Without Paid Sick Leave Less Likely to Take Time Off for Illness or Injury Compared to Those with Paid Sick Leave.” Health Affairs 35(3): 520–27. https://doi.org/10.1377/hlthaff.2015.0965.

28.

Derose

K. P.

Bahney

B. W.

Lurie

Escarce

J. J.

. 2009. “Review: Immigrants and Health Care Access, Quality, and Cost.” Medical Care Research and Review 66(4): 355–408. https://doi.org/10.1177/1077558708330425.

29.

Eggerth

D. E.

DeLaney

S. C.

Flynn

M. A.

J. Jacobson

. 2011. “Work Experiences of Latina Immigrants: A Qualitative Study.” Journal of Career Development 39(1): 13–30. https://doi.org/10.1177/0894845311417130 .

30.

Feldman

D. C

. 2006. “Toward a New Taxonomy for Understanding the Nature and Consequences of Contingent Employment.” Career Development International 11:28–47. https://doi.org/10.1108/13620430610642363.

31.

Flynn

M. A.

Carreón

Eggerth

D. E.

Johnson

A. I.

. 2014. “Immigration, Work, and Health: A Literature Review of Immigration Between Mexico and the United States.” Revista de Trabajo Social 6:129–49.

32.

Flynn

M. A.

Eggerth

D. E.

Jacobson

C. J.

Jr . 2015. “Undocumented Status as a Social Determinant of Occupational Safety and Health: The Workers’ Perspective.” American Journal of Industrial Medicine 58(11):1127–37. https://doi.org/10.1002/ajim.22531.

33.

Gleeson

. 2012. “Leveraging Health Capital at the Workplace: An Examination of Health Reporting Behavior among Latino Immigrant Restaurant Workers in the United States.” Social Science & Medicine 75(12): 2291–98. https://doi.org/10.1016/j.socscimed.2012.08.031 .

34.

Glick

J. E

. 2010. “Connecting Complex Processes: A Decade of Research on Immigrant Families.” Journal of Marriage and Family 72(3):498–515. https://doi.org/10.1111/j.1741-3737.2010.00715.x.

35.

Goldman

D. P.

Smith

J. P.

Sood

. 2005. “Legal Status and Health Insurance among Immigrants.” Health Affairs 24(6):1640–53. https://doi.org/10.1377/hlthaff.24.6.1640.

36.

Gonzales

R. G.

Chavez

L. R.

. 2012. “Awakening to a Nightmare” Abjectivity and Illegality in the Lives of the Undocumented 1.5-Generation Latino Immigrants in the United States." Current Anthropology 53(3): 255–81. https://doi.org/10.1086/665414.

37.

Loretta

Hirsh

A. T.

de Groot

Mather

K. J.

Stewart

J. C.

. 2020. “Undiagnosed Diabetes among Immigrant and Racial/Ethnic Minority Adults in the United States: National Health and Nutrition Examination Survey 2011–2018.” Annals of Epidemiology 51: 14–9. https://doi.org/10.1016/j.annepidem.2020.07.009 .

38.

Hall

Greenman

. 2015. “The Occupational Cost of Being Illegal in the United States: Legal Status, Job Hazards, and Compensating Differentials.” International Migration Review 49(2): 406–42. https://doi.org/10.1111/imre.12090.

39.

Hall

Greenman

Farkas

. 2010. “Legal Status and Wage Disparities for Mexican Immigrants.” Social Forces 89(2):491–513. https://doi.org/10.1353/sof.2010.0082.

40.

Hall

Greenman

. 2018. “Job Mobility among Unauthorized Immigrant Workers.” Social Forces 97(3):999–1028. https://doi.org/10.1093/sf/soy086.

41.

Hamilton

E. R.

Hale

J. M.

Savinar

. 2019. “Immigrant Legal Status and Health: Legal Status Disparities in Chronic Conditions and Musculoskeletal Pain among Mexican-Born Farm Workers in the United States.” Demography 56(1):1–24. https://doi.org/10.1007/s13524-018-0746-8.

42.

Kelaher

Jessop

D. Jones

. 2002. “Differences in Low-Birthweight among Documented and Undocumented Foreign-Born and Us-Born Latinas.” Social Science & Medicine 55(12):2171–75. https://doi.org/10.1016/S0277-9536(01)00360-4.

43.

Kivimäki

Head

Ferrie

J. E.

Shipley

M. J.

Vahtera

Marmot

M. G.

. 2003. “Sickness Absence as a Global Measure of Health: Evidence from Mortality in the Whitehall Ii Prospective Cohort Study.” BMJ 327(7411): 364. https://doi.org/10.1136/bmj.327.7411.364 .

44.

Korinek

Smith

K. R.

. 2011. “Prenatal Care among Immigrant and Racial-Ethnic Minority Women in a New Immigrant Destination: Exploring the Impact of Immigrant Legal Status.” Social Science & Medicine 72(10):1695–703. https://doi.org/10.1016/j.socscimed.2011.02.046.

45.

Lipscomb

H. J.

Loomis

McDonald

M. A.

Argue

R. A.

Wing

. 2006. “A Conceptual Model of Work and Health Disparities in the United States.” International Journal of Health Services 36(1): 25–50. https://doi.org/10.2190/BRED-NRJ7-3LV7-2QCG .

46.

Marmot

Feeney

Shipley

North

Syme

S. L.

. 1995. “Sickness Absence as a Measure of Health Status and Functioning: From the Uk Whitehall Ii Study.” Journal of Epidemiology and Community Health 49(2):124–30. http://doi.org/10.1136/jech.49.2.124.

47.

Marrow

H. B.

Joseph

T. D.

. 2015. “Excluded and Frozen Out: Unauthorised Immigrants’(Non) Access to Care After Us Health Care Reform.” Journal of Ethnic and Migration Studies 41(14):2253–73. https://doi.org/10.1080/1369183X.2015.1051465.

48.

Massey

D. S.

Arango

Hugo

Kouaouci

Pellegrino

E. Taylor

. 1993. “Theories of International Migration: A Review and Appraisal.” Population and Development Review 19(3):431–66. https://doi.org/2938462.

49.

Massey

D. S.

Bartley

. 2005. “The Changing Legal Status Distribution of Immigrants: A Caution.” International Migration Review 39(2):469–84. https://doi.org/10.1111/j.1747-7379.2005.tb00274.x.

50.

Massey

D. S.

Espinosa

K. E.

. 1997. “What's Driving Mexico-U.S. Migration? A Theoretical, Empirical, and Policy Analysis.” American Journal of Sociology 102(4):939–99. https://doi.org/10.1086/231037.

51.

Moriarty

. 2019. “It's Time to Allow Immigrants to Earn Professional Licenses.” fwd.us.

52.

National Conference of State Legislatures. 2021. “Paid Sick Leave.” https://www.ncsl.org/research/labor-and-employment/paid-sick-leave.aspx.

53.

National Partnership for Women & Families. 2021. “Paid Sick Day Statutes.” https://www.nationalpartnership.org/our-work/resources/economic-justice/paid-sick-days/paid-sick-days-statutes.pdf.

54.

Nee

Holbrow

. 2013. “Why Asian Americans Are Becoming Mainstream.” Daedalus 142(3):65–75. https://doi.org/10.1162/DAED_a_00219.

55.

Orrenius

P. M.

Zavodny

. 2009. “Do Immigrants Work in Riskier Jobs?” Demography 46(3):535–51. https://doi.org/10.1353/dem.0.0064.

56.

Passel

J. S

. 2006. The Size and Characteristics of the Unauthorized Migrant Population in the U.S. Washington, DC: Pew Hispanic Center.

57.

Passel

J. S.

Clark

R. L.

. 1998. Immigrants in New York: Their Legal Status, Incomes, and Taxes. Vol. Washington, DC: The Urban Institute.

58.

Passel

J. S.

Cohn

D'V.

. 2014. Unauthorized Immigrant Totals Rise in 7 States, Fall in 14: Decline in Those from Mexico Fuels Most State Decreases. Washington, DC: Pew Research Center's Hispanic Trends Project.

59.

Passel

J. S.

Cohn

D’V.

. 2016. Size of U.S. Unauthorized Immigrant Workforce Stable After the Great Recession. Washington, DC: Pew Research Center.

60.

Passel

J. S.

Cohn

D'V.

. 2018. US Unauthorized Immigrant Total Dips to Lowest Level in a Decade. Vol. Washington, DC: Pew Research Center.

61.

Passel

J. S.

Van Hook

Bean

F. D.

. 2006. Narrative Profile with Adjoining Tables of Unauthorized Migrants and Other Immigrants, Based on Census 2000: Characteristics and Methods. Vol. Washington, DC: US Census Bureau and Sabre Systems.

62.

Perreira

K. M.

Pedroza

J. M.

. 2019. “Policies of Exclusion: Implications for the Health of Immigrants and Their Children.” Annual Review of Public Health 40: 147–66.

63.

Premji

Krause

. 2010. “Disparities by Ethnicity, Language, and Immigrant Status in Occupational Health Experiences among Las Vegas Hotel Room Cleaners.” American Journal of Industrial Medicine 53(10): 960–75.

64.

Prentice

J. C.

Pebley

A. R.

Sastry

. 2006. “Collection of Legal Status Information: Caution! Prentice Et Al. Respond.” American Journal of Public Health 96(3): 399–a-400. https://doi.org/10.2105/AJPH.2005.079871.

65.

Reed

M. M.

Westfall

J. M.

Bublitz

Battaglia

Fickenscher

. 2005. “Birth Outcomes in Colorado's Undocumented Immigrant Population.” BMC Public Health 5(1):100. https://doi.org/10.1186/1471-2458-5-100.

66.

Van Hook

. 2021. “Comparing the Efectiveness of Assignment Strategies for Estimating Likely Undocumented Status in Secondary Data Sources for Latino and Asian Immigrants.” Population Research and Policy Review 40(2): 1–16. https://doi.org/10.1007/s11113-021-09658-3.

67.

Society for Human Resource Management. 2021. “State by State: Paid Sick Leave.” https://www.shrm.org/resourcesandtools/legal-and-compliance/employment-law/pages/state-local-paid-sick-leave-chart.aspx.

68.

Spence

Bachmeier

J. D.

Altman

C. E.

Hamilton

. 2020. “The Association between Legal Status and Poverty among Immigrants: A Methodological Caution.” Demography 57(6):2327–35. https://doi.org/10.1007/s13524-020-00933-0.

69.

Stimpson

J. P.

Wilson

F. A.

. 2013. “Unauthorized Immigrants Spend Less Than Other Immigrants and Us Natives on Health Care.” Health Affairs 32(7): 1313–18. https://doi.org/10.1377/hlthaff.2013.0113.

70.

Strong

L. L.

Zimmerman

F. J.

. 2005. “Occupational Injury and Absence from Work among African American, Hispanic, and Non-Hispanic White Workers in the National Longitudinal Survey of Youth.” American Journal of Public Health 95(7):1226–32. https://doi.org/10.2105/AJPH.2004.044396.

71.

US Bureau of Labor Statistics. 2010. “Paid Sick Leave in the United States.” In Program Perspectives: National Compensation Survey, Vol. 2. Washington, DC: U.S. Department of Labor.

72.

US Bureau of Labor Statistics. 2020. “Foreign-Born Workers: Labor Force Characteristics −2019.” USDL-20-0922 https://www.bls.gov/news.release/pdf/forbrn.pdf.

73.

US Bureau of the Census. 2001. Survey of Income and Program Participation Users’ Guide. Washington, DC: U.S. Department of Commerce.

74.

US Department of Labor. 2021. “Sick Leave.” https://www.dol.gov/general/topic/workhours/sickleave.

75.

US Department of Labor and US Bureau of Labor Statistics. 2020. “National Compensation Survey: Employee Benefits in the United States, March 2020.” https://www.bls.gov/ncs/ebs/benefits/2020/employee-benefits-in-the-united-states-march-2020.pdf.

76.

US Government Accountability Office. 2015. Contingent Workforce: Size, Characteristics, Earnings, and Benefits. GAO-15-168R Contingent Workforce.

77.

Vahtera

Pentti

Kivimäki

. 2004. “Sickness Absence as a Predictor of Mortality among Male and Female Employees.” Journal of Epidemiology and Community Health 58(4): 321–26.

78.

Van Hook

Bachmeier

J. D.

Coffman

D. L.

Harel

. 2015. “Can We Spin Straw into Gold? An Evaluation of Immigrant Legal Status Imputation Approaches.” Demography 52(1):329–54. https://doi.org/10.1007/s13524-014-0358-x.

79.

Virtanen

Kivimäki

Elovainio

Vahtera

Cooper

C. L.

. 2001. “Contingent Employment, Health and Sickness Absence.” Scandinavian Journal of Work, Environment & Health 27(6):365–72. http://www.jstor.org/stable/40967151.

80.

Waldinger

Lichter

M. I.

. 2003. How the Other Half Works: Immigration and the Social Organization of Labor. Berkeley: University of California Press.

81.

Waters

M. C.

Pineau

M. Gerstein

, eds. 2016. The Integration of Immigrants into American Society. Washington, DC: National Academies Press. http://www.nap.edu/download.php?record_id=21746 .

82.

Yoshikawa

Kalil

. 2011. “The Effects of Parental Undocumented Status on the Developmental Contexts of Young Children in Immigrant Families.” Child Development Perspectives 5(4):291–97. https://doi.org/10.1111/j.1750-8606.2011.00204.x.

83.

Young

M-E.

Madrigal

D. S.

. 2017. “Documenting Legal Status: A Systematic Review of Measurement of Undocumented Status in Health Research.” Public Health Reviews 38(26):1–25. https://doi.org/10.1186/s40985-017-0073-4.

84.

Young

M-E.

Pebley

A. R.

. 2017. “Legal Status, Time in the USA, and the Well-Being of Latinos in Los Angeles.” Journal of Urban Health 94:764–75. https://doi.org/10.1007/s11524-017-0197-3.

85.

Ziol-Guest

K. M.

Kalil

. 2012. “Health and Medical Care among the Children of Immigrants.” Child Development 83(5):1494–500. https://doi.org/10.1111/j.1467-8624.2012.01795.x.