Abstract
This study explores the social-ecological contexts of unauthorized immigrant families and their U.S.-born children, through examining how otherwise similarly low-income, urban, Latino immigrant families differ on the basis of the parents’ legal status and interactions with the immigration system. Drawing on social-ecological theory, variations based on parents’ legal vulnerability among exosystem-level experiences (e.g., parents’ occupational stress, discrimination experiences) and microsystem-level experiences (e.g., parents’ mental health, parenting stress) were explored. Structured interviews were conducted with 178 families with an immigrant parent from Mexico, Central America, and Dominican Republic, and a child (aged 7-10 years) born in the United States. Unauthorized parents reported statistically higher occupational stress, ethnicity-based discrimination, challenges learning English, immigration challenges, and legal status challenges, and lower use of social services, when compared with authorized parents. The groups did not differ on microsystem factors (e.g., parent mental health, and parenting, marital, and family stress).
Keywords
Introduction
Social-ecological developmental theorists have long argued for the importance of looking beyond the immediate family context to understand the trajectories of individuals. Scholars (e.g., Viruell-Fuentes, Miranda, & Abdulrahim, 2012) have also argued for moving away from individual and culture-based frameworks for understanding immigrant experiences, and toward more structural explanations that emphasize the social inequalities many immigrants face. The present data contribute to our understanding of the ways in which immigrant parent legal vulnerability, defined as legal status and a history of interactions with the immigration system, results in differing contexts and patterns of social inequality and exclusion for otherwise similarly low-income, urban, Latino immigrants in a northeastern state. Drawing on social-ecological theory (Bronfenbrenner, 1986), exosystem-level experiences, including parents’ occupational stress, experiences of discrimination, and social service use, as well as microsystem-level experiences, including parents’ mental health and parenting stress, were explored. Their variations based on the parents’ legal vulnerability are analyzed. The authors urge policymakers, advocates, and practitioners who endeavor to support immigrant family well-being to take seriously this more nuanced understanding of immigrant families’ within-group variability as they craft their policy and practice responses.
Review of Literature
Unauthorized 1 Immigrants in the United States: Demographics and Policies
The most commonly accepted estimate of the unauthorized population in the United States was approximately 11.7 million in 2012 (Passel, Cohn, & Gonzalez-Barrera, 2012), slightly lower than the peak of 12 million unauthorized immigrants in 2007 (Hoefer, Rytina, & Baker, 2012). Nearly 9 million of these 11.7 million are from Mexico, the Caribbean, and Central America (Hoefer et al., 2012). Children represent the smallest share of the unauthorized population (10%; Capps, Bachmeier, Fix, & Van Hook, 2013). However, approximately 5.5 million children have at least one unauthorized parent, and of those, at least 4 million were born in the United States as citizens (Suárez-Orozco, Yoshikawa, Teranishi, & Suárez-Orozco, 2011). The number of U.S.-citizen children with unauthorized parents has grown substantially, from 57% of children in immigrant families in 2000 to 79% in 2009 (Donato & Armenta, 2011). Thirty-two percent of unauthorized adults and 51% of unauthorized children live in families with incomes below the federal poverty level. Despite these levels of poverty, employment rates are high, with 87% of unauthorized men and 57% of unauthorized women participating in the labor force (Capps et al., 2013).
Many policies have affected unauthorized immigrants and their children since the intensification of anti-immigration policies over the past few decades. Such legislation includes the 1986 Immigration Reform and Control Act, which made it illegal for an employer to hire an unauthorized immigrant and imposed a required employer-based verification system (Haque, 2004). Additionally, the Illegal Immigrant Responsibility and Immigrant Reform Act (IIRIRA) and Anti-Terrorism Effective Death Penalty Act (AEDPA) of 1996 expanded the range of offenses for which a person could be deported, permitted retroactive deportation for previous offenses, restricted judicial review as a means to appeal deportation, and strengthened the cooperation between local and state enforcement to enforce immigrant laws (Donato & Armenta, 2011; Hagan, Eschbach, & Rodriguez, 2008).
After the terror attacks on September 11, 2001, immigration control was increasingly viewed as a matter of national security. In October 2001, the U.S. Congress enacted the Uniting and Strengthening America by Providing Appropriate Tools Required to Intercept and Obstruct Terrorism (USA PATRIOT) Act, thereby expanding the groups of immigrants eligible for deportation and increasing the authority to deport those who were deemed to be a threat (Hagan, Castro, & Rodriguez, 2010). Less than a decade later, the Secure Communities Strategy was developed by U.S. Immigration and Customs Enforcement (ICE) in 2009 to identify unauthorized people in U.S. jails who are eligible for deportation. It formed a partnership and database sharing agreement between the Department of Justice and Department of Homeland Security. Secure Communities contributed to increasing fears in many immigrant neighborhoods; following its inception, many unauthorized immigrants were identified by local law enforcement and turned over to ICE officials who detained or deported them, often for minor offenses. This program had a major impact on families, as nearly 40% of individuals arrested through Secure Communities had a U.S.-citizen spouse or child (Kohli, Markowitz, & Chavez, 2011). The program was discontinued in November 2014 (Johnson, 2014) but was replaced by the Priorities Enforcement Program. More recently, in the context of ongoing and multiple failed efforts to pass Comprehensive Immigration Reform, President Obama issued an Executive Order that would stay deportations for unauthorized parents of U.S.-citizen children who met certain requirements, and would expand the Deferred Action to Childhood Arrivals (DACA) program. However, this action was blocked by federal judicial decision in response to a law suit in which multiple states alleged that the President acted outside of correct procedure in issuing the order, and that injunction has been upheld in a recent decision by the U.S. Court of Appeals Fifth Circuit.
The Developmental Contexts of Unauthorized Immigrant Families: Patterns of Social Exclusion
For decades, developmental theory has emphasized the importance of the environment in shaping children’s development, with a particular focus on the family as a context. Developmental theorists primarily focused on intrafamilial processes (such as parent-child interactions) until Bronfenbrenner (1986) successfully argued that extrafamilial conditions could affect intrafamilial processes. Specifically, exosystem theories posited that the experiences parents have (e.g., at work or within their adult social networks) could indirectly influence the development of children. Suárez-Orozco, et al. (2011) argued that the developmental implications of unauthorized status should be examined through this type of social-ecological framework.
Microsystem contexts
Immediate postmigration contexts, experiences of proximal social settings, and family processes are just some of the contexts that may account for the effects of parental unauthorized status on child health, behavioral, and academic outcomes (Yoshikawa & Kalil, 2011). Following arrival in the United States, unauthorized parents may deal with extreme anxiety over the uncertainty of their situation and the risk of deportation, and parental stress can have negative effects on parenting practices (Crnic & Low, 2002). Research has also supported the effects of discriminatory experiences on stress among immigrant adults (Dawson, 2009). In the home context, overcrowding is often an issue since unauthorized immigrants are not eligible for subsidized housing. Children of immigrants are four times as likely as children of non-immigrants to live in crowded housing, defined as more than two people per bedroom (Capps, 2001), which can pose a health hazard (Krieger & Higgins, 2002). However, a strength for children in immigrant families is that they are more likely than children of U.S.-born parents to live in families with married parents (Landale, Thomas, & Van Hook, 2011), an advantage found to be beneficial for children’s cognitive and social-emotional outcomes.
The school setting is an additional microsystem, or immediate context of the child’s development, in which unauthorized status plays a role. Children from groups with high proportions of unauthorized immigrants are less likely to be enrolled in preschool (Hernandez, Denton, & Macartney, 2008). This is problematic, given the strong evidence that early childhood education is important for school readiness and cognitive development (Magnuson, Lahaie, & Waldfogel, 2006). Later on, immigrant parents may face barriers to involvement in their children’s schools and thus are less likely than non-immigrant parents to engage in events at the school or interact with their children’s teachers (Sibley & Dearing, 2014), placing their children at a higher risk for reduced achievement (Lahaie, 2008). Children in immigrant families are more likely than non-immigrant youth to attend underresourced schools with higher levels of segregation, a higher proportion of students in poverty, and with more safety problems at school (Crosnoe, 2005).
Exosystem contexts and social exclusion
Outside of the immediate family context, research on the developmental contexts of children of unauthorized parents reveals a pattern of social exclusion. Unauthorized migrants often pay large sums of money to enter the United States, incurring high levels of migratory debt (Suárez-Orozco, et al., 2011). This often forces them to take on inflexible (and often multiple) jobs to pay off this debt, which can limit the time they have to invest in their children’s education and upbringing. They may also live in neighborhoods with large numbers of impoverished immigrants, which generally have fewer community resources for families to utilize (Yoshikawa & Kalil, 2011). The employment challenges of unauthorized parents may also have indirect effects on the developmental context of their children (Yoshikawa & Kalil, 2011).
Empirical work has found that groups of immigrants with high proportions of unauthorized people tend to have the lowest levels of access to institutional resources, as indicated by their lower likelihood of having a checking or savings account, credit card, or driver’s license (Yoshikawa, Godfrey, & Rivera, 2008). Yoshikawa and colleagues (2008) found that parents with more access to institutional resources (typically, those who were more likely to be authorized) had less economic hardship, lower levels of psychological distress, and provided their children with more cognitive stimulation.
Although U.S.-citizen children are eligible for public assistance, many do not receive it, possibly due to a lack of English proficiency which makes it difficult for parents to access it, or due to parents’ fear about giving personal information that could identify their unauthorized status (Yoshikawa & Kalil, 2011). Approximately 71% of unauthorized adults had no health insurance coverage in 2011 (Capps et al., 2013) and 17% of children of unauthorized parents have no health insurance (Ku & Jewers, 2013). Additionally, those who are insured may receive lower quality services due to communication barriers (Lessard & Ku, 2003).
The present data contribute to our understanding of the social-ecological contexts of unauthorized immigrant families and their U.S.-born children, through examining how otherwise similarly low-income, urban, Latino immigrant families in a northeastern state differ on the basis of the parents’ legal status and interactions with the immigration system. Furthermore, in contrast to research that has focused on early childhood (e.g., Yoshikawa, 2011), this study focused on a subset of parents with U.S.-born children in the middle childhood age range (7-10).
Method
Participants and Recruitment
Results included in this article are drawn from the quantitative phase of the first author’s study on the influence of parents’ legal vulnerability on immigrant families and children. For inclusion in the study, participants had to be immigrants from the Dominican Republic, Mexico, or Central America, and parents of at least one child (aged 7-10 years) born in the United States. Participants were recruited from an urban ring comprised of three cities in a northeastern state, with a high density of Latino immigrants living predominantly in low-income or working-class neighborhoods. Recruitment was primarily accomplished via the three school districts in this urban ring. Upon approval from the superintendent and building-level principals, informational flyers were sent home with all students in second, third, and fourth grades in participating elementary schools (N = 26 schools), regardless of whether they met study criteria. Parents filled out a form indicating whether they would like to be contacted about the study for possible participation. Researchers additionally made presentations about the project and handed out flyers at immigrant community organizations, parent-teacher organization (PTO) meetings, local ethnic festivals, English for Speaker of Other Languages (ESOL) classes, a Spanish language radio program, and Spanish-speaking church services of several denominations (e.g., Catholic, Episcopalian, Evangelical). Parents who indicated they wanted follow-up were contacted by bilingual researchers, who screened to ensure that the family met criteria for participation, explained the study purpose and procedures, and reviewed informed consent.
Procedures
The first author’s Institutional Review Board (IRB) and the IRBs of all three participating school districts approved the study. Additionally, a Certificate of Confidentiality was obtained from the National Institutes for Health. If the family met the criteria and the parent wanted to participate, an appointment was made in the location of the family’s choice (e.g., parents’ home, researcher’s office, or public space, for example, a library community room). At the time of the meeting, the researcher again reviewed informed consent and obtained signed consent and permission from parents and assent from the child. Structured interviews that included several standardized measures were then orally conducted with one parent and one child. Specifically, the parent and child were interviewed sequentially (one at a time) during the same visit. Parents selected whether the adult or child would be interviewed first, and were given the option of remaining present while the child was interviewed. Because of the challenges interviewing both parents (e.g., because the partner was working), only one parent was interviewed. Interviews were conducted in the parents’ and child’s language of preference. Results from this article are drawn from the parent interviews. Data were collected between September 2013 and November 2014, and completed before Obama’s Executive Order in 2014.
Measures
All measures used had been previously utilized by other researchers with Latino immigrants, and adequate-to-good psychometric properties were reported. All measures and consent materials were piloted with a small sample (N = 6) and a focus group (N = 4) and minor revisions were made (e.g., tweaking translations, revising negatively worded statements) to ensure that they were valid and comprehensible for participants.
Demographic and migration experiences
Demographic information included responding parents’ country of origin, gender, age, number in household, marital status, parents’ work status, total number of children, number of children born in the United States, parent and participating child education level, average family monthly income, and language practices. Additional information was gathered regarding participating parents’ life pre-migration (pre-migration housing conditions; reason for migration) and during migration (whether safety concerns were present, whether money was borrowed for migration, whether children were left in country of origin), and upon entering the United States (perceived stress of migrating, whether friends and family were already in the United States and, if so, amount of support received).
Legal vulnerability
Parent legal status was assessed following the method used by Perreira et al. (2008). Specifically, parents were asked three yes/no questions assessing whether they had (a) U.S. citizenship, (b) Legal Permanent Resident (green card) status, or (c) a visa authorizing them to reside or work in the United States. Participants who answered “no” to all three questions were categorized as unauthorized. Parent/family interactions with the immigration system were assessed following the format used by Brabeck and Xu (2010). Parent participants were asked a series of yes/no questions that assessed whether they or a family member has ever been (a) detained by immigration authorities or (b) deported from the United States. (Note that “family member” meant any member of their immediate or extended family, and therefore did not expose any specific individual’s experience.) Parent legal status and family history of interactions with the immigration system were then collapsed into a single, continuous variable called parent legal vulnerability. This was measured for the responding parent participant only; IRB concerns about the ethics of an individual disclosing another individual’s legal status precluded the collection of information regarding the partner’s legal status. This variable was developed by Brabeck and Xu (2010) and has five levels, where 1 = lowest vulnerability and 5 = highest vulnerability. Specifically, 1 = authorized with no history of interactions with immigration system personally or in family, 2 = authorized, with family history of interactions with immigration system, 3 = unauthorized with no history of interactions with immigration system, 4 = unauthorized with family history of interactions with the immigration system, and 5 = unauthorized with personal and familial history of interactions with immigration system.
Economic stress
Economic stress was assessed via a series of questions based on the format used by Yoshikawa (2011), with supplemental questions for food scarcity, taken from the Guide to Measuring Household Food Insecurity (Bickel, Nord, Price, Hamilton, & Cook, 2000). Parent participants were asked a series of yes/no questions to assess whether in the past six months there was a time when they (a) were without telephone service; (b) did not pay the full amount of rent or mortgage; (c) were evicted for not paying rent or mortgage; (d) lost service from the gas, electric, or oil company because they did not make payments; (e) had enough work to support their family; (f) worried about paying debts; (g) did not know where next meal would come from; (h) ate cheaper food to save money; or (i) skipped meals to save money. The questions were translated into Spanish by a bilingual native Spanish speaker and then back-translated into English. Reliability for this nine-item scale was adequate (α = .71).
Occupational stress
Occupational stress was measured via the Spanish translation of a subscale of the Hispanic Stress Inventory–Immigrant Version (Cervantes, Padilla, & DeSnyder, 1991). Participants were instructed to consider the previous three months and to rate whether the experience described in the item was stressful for them on a 5-point Likert-type scale, with 1 = not at all stressful and 5 = extremely stressful. Sample items include “I have been criticized for my work” and “I have been forced to accept low wage jobs.” Reliability of the 15-item scale was good (α = .81).
Social service use
Social service use was assessed via the format used by Yoshikawa (2011). Parent participants were asked a series of yes/no questions to assess whether they have received (a) food stamps (i.e., Supplemental Nutrition Assistance Program (SNAP); (b) welfare (i.e., Temporary Assistance to Needy Families (TANF); (c) childcare subsidies; (d) Medicaid or State Children’s Health Insurance Program (SCHIP) for parent or child; (e) Women, Infants, and Children (WIC); (f) Head Start; (g) early intervention; (h) job training subsidies; (i) free or reduced price school lunch for child; (j) Supplemental Security Income (SSI); and (k) unemployment benefits. Reliability for the 10-item scale was low (α = .64), but this was expected since these items do not represent an underlying psychological construct but rather a cumulative account of various social services.
Parent mental health
Parent mental health was assessed via the Spanish translation of the Kessler-6 (Kessler et al., 2002). This measure contains six items to assess the presence of psychiatric symptoms. Parents are asked to consider the last 30 days and to respond how frequently they experienced each item, on a scale of 0 = never to 4 = all the time. Sample items include “so depressed nothing could cheer you up” and “hopeless.” Reliability for the six-item scale was good (α = .87).
Parent social support
Parent social support was assessed via a series of questions based on the format used by Yoshikawa (2011). Parents were asked a series of yes/no questions to assess whether they have people on whom they can depend to (a) help with childcare, (b) make ends meet, and (c) find a job. Reliability for this three-item scale was low (α = .56), which was not problematic since these items represent a cumulative count of support rather than an underlying psychological construct.
Marital, parenting, and family stress
Marital, parenting, and family stress were measured via three subscales of the Hispanic Stress Inventory–Immigrant Version (Cervantes et al., 1991). The author’s Spanish translation of the subscales was used. For each subscale, participants were instructed to consider the previous three months and to rate whether the experience described in the item was stressful for them on a 5-point Likert-type scale, with 1 = not at all stressful and 5 = extremely stressful. Sample items from the Marital Stress subscale include “I’ve felt my partner and I couldn’t communicate” and “My partner and I have had disagreements about which language our children should speak.” Reliability for the 16-item Marital Stress scale was adequate (α = .73). Sample items from the Parenting Stress subscale include “My children haven’t respected my authority in the way they should have” and “My children have received bad grades at school.” Reliability for this 13-item scale was good (α = .78). Sample items from the Family Stress subscale include, “Due to lack of family unity, I’ve felt isolated and alone” and “I’ve noticed my family members becoming too individualistic.” Reliability for this 13-item scale was good (α = .85).
Discrimination, English language, immigration, and legal status stress
Ethnicity-Based Discrimination, English Language Challenges, Immigration Challenges, and Legal Status Challenges were measured via items from the Hispanic Stress Inventory–Immigrant Version (Cervantes et al., 1991). Participants were instructed to consider the previous three months and to rate whether the experience described in the item was stressful for them on a 5-point Likert-type scale, with 1 = not at all stressful and 5 = extremely stressful. Sample items for the Ethnicity-Based Discrimination scale include, “I feel others don’t accept me because of my Latino culture” and “Because I’m Latino, people expect me to work harder.” Reliability for the five-item scale was good (α = .77). Sample items for the English Language Challenges scale include “Because of my bad English, people have treated me badly” and “Because of problems with English I’ve had problems in my child’s school.” Reliability for the five-item scale was good (α = .75). Sample items for the Immigration Challenges scale include, “I’ve felt that in this country I’ll never be able to restore the respect and position I had in my country” and “I feel guilty for having left my friends and family in my country.” Reliability for this four-item scale was adequate (α = .70). Sample items from the Legal Status Challenges scale include, “I’ve been afraid of the consequences of deportation” and “My legal status has limited my contact with friends and family.” Reliability for the seven-item scale was good (α = .84).
Results
Participant Legal Status and Demographic Information
Legal status
Among the 178 participating parents, 51% (N = 90) had some type of authorization, while 49% (N = 88) did not. More specifically, 25% (N = 44) were naturalized citizens, 17% (N = 31) were Legal Permanent Residents (green card holders), 9% (N = 17) held another type of Visa (e.g., U-Visa), and the remaining 49% (N = 88) were unauthorized. Among participants, 12% (N = 22) reported that they had previously been detained by immigration authorities and 5% (N = 9) reported that they were previously deported from the United States. Thirty-three percent (N = 58) of the entire sample reported a family member had been detained, and 24% (N = 42) of the entire sample reported a family member had been deported.
Parent demographics
Demographic information for the responding parent (frequencies and percentages for categorical variables and means, ranges, standard deviations for continuous variables) is summarized in Table 1. Information is presented for the overall sample and for authorized versus unauthorized participating parents. Chi-square analyses and t tests were run to determine whether significant differences existed on each variable between authorized versus unauthorized parents.
Demographics.
Note. Asterisks placed next to a variable name rather than a number indicates that the chi-square statistic was significant, but no standardized residuals had an absolute value ≥1.96; hence, significance could not be attributed to a specific cell. HS = high school; N/A = not applicable.
Indicates that documentation status is significantly associated with the variable at the p < .05 level.
Indicates that documentation status is significantly associated with the variable at the p < .01 level.
Unauthorized parents were significantly younger and more likely to have an eighth-grade education or less. Unauthorized parents were more likely to be from Guatemala and Mexico, and were also more likely to be cohabitating but not married, a characteristic typical of families in sending countries where marriage ceremonies are often costly. Participating parents’ work status was also significantly related to legal status; while the standardized residuals were not greater than 2.0, the overall chi-square was significant, with data revealing a greater percentage of unauthorized parents (69.3%) not working, as compared with authorized parents (47.8%). Unauthorized parents were also more likely to speak Spanish in their homes and to prefer Spanish over English.
Pre-migration and migration
Table 2 summarizes data on migration-related variables for the entire sample, with a breakdown according to the current legal status of the participating parent. Again, chi-square and t-test analyses were used to examine the relationship between the variables and legal status. Unauthorized parents were significantly less likely to have running water and bathrooms in their homes pre-migration. They were also more likely to leave their children in the country of origin when they migrated and to borrow money to pay for migration. Unauthorized parents were more likely to fear for their safety during migration and were less likely to receive help from family members already in the United States at the time of arrival.
Migratory Experience.
Note. Asterisks placed next to a variable name rather than a number indicates that the chi-square statistic was significant, but no standardized residuals had an absolute value ≥1.96, hence significance could not be attributed to a specific cell.
Total does not equal 100% for this categorical variable, since respondents were asked to check all that apply and to leave blank if the event did not occur.
For these variables, the scale was 1 = none, 2 = a little, 3 = a lot.
For this variable, the scale was 1 = not stressful, 2 = somewhat stressful, 3 = very stressful.
Indicates that documentation status is significantly associated with the variable at the p < .05 level.
Indicates that documentation status is significantly associated with the variable at the p < .01 level.
Participant Legal Vulnerability and Family Contexts
Table 3 displays descriptive statistics for all dependent variables by parent legal vulnerability. To explore how parents’ legal vulnerability relates to variables within families’ contexts, a series of regressions were run with parent legal vulnerability as the predictor. For each model, the following variables were entered as covariates to control for their effects on the dependent variables: parent age, parent education level, parent marital status (married/unmarried), average monthly family income, preferred language, number of children, gender, pre-migration stressors, and migration stressors.
Descriptive Data on Family-Context Variables by Parent Legal Vulnerability.
Only participants currently working responded to items.
Higher scores indicate higher stress/challenge.
Higher scores indicate higher levels of support.
Only participants in relationships responded to items.
Significantly predicted by parent legal vulnerability at .05 level.
Significantly predicted by parent legal vulnerability at .01 level.
Significantly predicted by parent legal vulnerability at .001 level.
Economic and occupational indicators
To explore the ways in which families differed on economic and occupational indicators as a result of parents’ legal vulnerability, two regression models were run, using parent legal vulnerability as the predictor, and economic stress and occupational stress as the dependent variables, with the aforementioned covariates controlled for. Parent legal vulnerability was a significant predictor of occupational stress (F(10, 152) = 3.00, p < .01, R2 = .17), with parents with greater levels of legal vulnerability reporting higher levels of occupational stress. After controlling for covariates, parents’ legal vulnerability was not a significant predictor of economic stress. See Table 4 for a summary of regression models.
Regression Results.
p < .10. *p < .05. **p < .01. ***p < .001.
Social service use
To explore how families differed in their use of social services overall, a regression was estimated with parents’ legal vulnerability as the predictor and the social service use scale as the outcome, controlling for covariates. Parents with greater levels of legal vulnerability were less likely to access social services compared to parents with lower levels of legal vulnerability (F(10, 171) = 6.53, p < .001, R2 = .29). To understand how parents’ legal status related to use of specific help sources, a series of 2 × 2 tables were run. In light of non-significant t tests between Parent Legal Vulnerability Levels 1 and 2 (all authorized participants), and between Parent Legal Vulnerability Levels 3, 4, and 5 (all unauthorized particpants), Levels 1 and 2 were collapsed into one level, and Levels 3, 4, and 5 were collapsed into a second level, to create two categories: authorized and unauthorized. Thus, the chi-squares analyzed parent legal status (authorized/unauthorized) by specific social service (use/no use). Results indicated that authorized parents were significantly more likely to report use of government-subsidized job training (χ2(1, N = 178) = 22.49, p < .001), social security benefits (χ2(1, N = 178) = 8.70, p < .01), unemployment benefits (χ2(1, N = 178) = 42.60, p < .001), welfare benefits (χ2(1, N = 178) = 30.68, p < .001), childcare subsidies (χ2(1, N = 178) = 22.82, p < .001), Medicaid for adults (χ2(1, N = 178) = 74.49, p < .001), and Head Start/Preschool (χ2(1, N = 178) = 19.03, p < .001). There were no significant differences between families with unauthorized versus authorized parents with regard to the use of early intervention, reduced/free school lunch, food stamps, Medicaid for child, or WIC.
Immigration-related stressors
To explore the ways in which families who were different on the basis of parents’ legal vulnerability differed in parents’ experiences of immigration-related stressors (i.e., ethnicity-based discrimination, English language challenges, immigration-related challenges, and legal status challenges), four regression models were run with parent legal vulnerability as the predictor and these four scales as the dependent variables, controlling for background characteristics. Results indicated that parents’ legal vulnerability significantly predicted greater rates of ethnicity-based discrimination (R2 = .12, F(10, 158) = 2.18, p < .05); immigration-related challenges (R2 = .21, F(10, 159) = 4.30, p < .001); and legal status challenges (R2 = .51, F(10, 160) = 16.92, p < .001). After controlling for covariates, parents’ legal vulnerability did not predict English language challenges.
Parent mental health
To explore how parent psychological well-being varied on the basis of legal vulnerability, a regression was estimated with parents’ legal vulnerability as the predictor and parent mental health as the dependent variable, after including the covariates. Parents’ legal vulnerability was not a significant predictor of parent mental health, although the trend in data (see Table 3) suggests that parents with higher levels of legal vulnerability reported higher symptom levels.
Social relationships
To explore the ways in which families differed with regard to social relationships on the basis of parents’ legal vulnerability, four regression models were run, controlling for covariates, with parents’ legal vulnerability as the predictor, and social support, marital stress, parenting stress, and family stress as the dependent variables. Parent legal vulnerability was a significant predictor of Social Support Scale (R2 = .11, F(10, 161) = 2.06, p < .05), with parents with greater levels of legal vulnerability reporting fewer social supports. In light of the non-significant t tests between Parent Legal Vulnerability Levels 1 and 2 (all authorized), and between Parent Legal Vulnerability Levels 3, 4, and 5 (all unauthorized), the variable was rescored into a binary variable represented authorized and unauthorized. Results indicated that unauthorized parents were significantly less likely to have individuals who might help them with childcare (χ2(1, N = 178) = 4.09, p < .05), financial assistance (χ2(1, N = 178) = 7.26, p = .01), and assistance finding work (χ2(1, N = 178) = 8.77, p < .01). Neither marital stress, parenting stress, nor family stress was significantly associated with parents’ legal vulnerability after including covariates.
Discussion
This article describes some of the ways in which immigrant families’ parental legal status and history of interactions with the immigration system are associated with differing contexts that frame and influence experiences of immigrant families in the United States. While the participating families shared some struggles in common as a result of their low-income status and residence in poor or working-class urban neighborhoods, a distinct pattern emerged that differentiated families on the basis of parents’ legal vulnerability. The data reported here support a social exclusion theory of legal vulnerability and suggest ways in which legal status and interactions with the immigration system create a systematic pattern of marginalization and disadvantage. Although not significant, level of legal vulnerability is also correlated with greater mental health symptoms, including depressive and anxiety symptoms. Yet, contrary to what might be predicted, the families did not differ with regard to family stress and processes. The data do not support assertions that parents differ on the basis of legal vulnerability with regard to their marital relationships, parenting experiences, or family stress.
Despite differences in age of their children at time of study, the unauthorized parents in this sample were similar to unauthorized immigrants in other research (e.g., Yoshikawa, 2011) with regard to pre-migration and migration experiences, education levels, work status, and language use. Compared with authorized parents, they were significantly more likely to have experienced pre-migration poverty indicators, to have left children in their country of origin, to have borrowed money to migrate, and to have endured stressful incidents and fear during migration. Upon coming to the United States, they reported being less likely to have had support from their family members.
Unlike other samples in studies of unauthorized adults, the participants in this sample were less likely to be employed, but their partners were reported to be employed at a higher rate than authorized parents. Although average monthly income did not differ, parents with higher levels of legal vulnerability were more likely to experience job-related stress and, although not significant, scored higher on indicators of economic stress, including items related to food scarcity. This may be because they are working more jobs, under more stressful conditions, for less pay, and experience fear related to losing their job because of their legal status. It may also be related to the finding that they were less likely to participate in several social services, including Medicaid for themselves, and many additional social services for which they are ineligible due to their legal status (e.g., unemployment benefits, childcare subsidies, TANF). While this is consistent with previous research, it also counters a popular stereotype that unauthorized immigrants migrate to “free load” off of the social safety network in the United States. Importantly, this sample also reported lower levels of enrollment of their children in early childhood or preschool programs. This is important, given the well-established relationship between preschool participation and school readiness and success (Magnuson et al., 2006).
In addition to higher levels of occupational stress and lower rates of participation in social service programs, these data reveal higher levels of stress related to ethnicity-based discrimination, English language learning, and legal status. Adults who are working several jobs and who fear discovery because of legal status may experience additional obstacles to attending ESOL classes. Additionally, legal status may provide an additional layer of discrimination based on race, ethnicity, and income. In the context of these multiple social structural disadvantages, we see higher rates of psychological distress (although the relationship is not statistically significant). These data support the stance of scholars for an emphasis on structural inequalities (instead of individual cultural explanations) in understanding immigrant health outcomes (see, for example, Viruell-Fuentes et al., 2012).
Theoretically, one might expect that families under the kinds of stress created by legal vulnerability and described herein would also report higher levels of strain related to marital relationships, parenting, and the family. The data from this sample did not support this hypothesis. Instead, families did not differ based on parents’ legal vulnerability with regard to their marital, parenting, and family stress. Consistent with some previous findings (e.g., Yoshikawa, 2011), they did differ with regard to the material forms of support that people in their social milieu could provide them: help with childcare, finding a job, and with finances. Yet amid the myriad of structural stressors and the relative absence of social support, the families headed by unauthorized parents report, as a group, generally intact partnerships, parent-child relationships, and family processes. This finding suggests extraordinary resiliency and strength in the context of multiple adversities, and underscores the potential for immigrants, including those without documents, to be an asset to the United States, a country that espouses support for family values and cohesiveness.
In summary, these data support a social exclusion theoretical explanation for a series of contextually based outcomes that constrain the lives of families with legally vulnerable immigrant parents. The participants with greater levels of legal vulnerability, as a group, started out with a relative disadvantage: less education, greater poverty, and greater stress during migration. Within the United States, they also exist within a structure that systematically disadvantages them: higher job-related stress, lower access to/use of social services, lower pragmatic forms of social support, greater obstacles to learning English, higher experiences of discrimination, and the fear of discovery and deportation. At the same time, the families demonstrate remarkable strength in the bonds of their relationships with partners, children, and other family members.
Implications
These data have significant implications for those seeking to develop social interventions as well as shape policies to more effectively address the lived experiences of immigrant families, particularly those who are unauthorized. All participants in this study have limited access to a host of traditional social services that are designed and have been found to enhance individual and familial well-being. Yet those who are legally threatened are significantly more challenged. These data suggest that despite a national or cultural tendency to attribute unemployment and a host of attendant factors (e.g., lack of health care, low educational attainment) to individual effort or failure, structural explanations may be more appropriate. These data also highlight the important within-group variability among immigrants; those affected by additional structural injustices (e.g., discrimination, lack of legal status) are more disadvantaged than their authorized counterparts. Regularizing status through naturalization, permanent residency or visas, could, therefore, contribute importantly to reducing parental and familial legal vulnerability and its associated costs.
Legal scholar David Thronson (2010) has suggested granting young children the right to apply for citizenship for their parents, a policy presently available only once they have achieved adulthood and only to parents outside the United States. Second, despite the significant negative effects of the exosystems discussed herein, these legally threatened families are more likely to be two-parent households and do not exhibit more significant familial or partner stress than the authorized comparison group. A strengths-based approach to working with them and their children—in school and/or community-based contexts—would build on these findings. Finally, translating these findings into popular discourse might contribute to improving the general public’s knowledge of this population and/or to reducing racialized stereotypes that represent these families as “illegal” and/or their children as “anchor babies.”
Limitations and Future Research
As with any research, this study had several limitations. The most notable limitations include the reliance on self-report data, the use of a convenience sample, and the reliance on parent-level data for analyses and conclusions. Future research may be aimed at better understanding the ways in which family resiliency buffers against structural sources of exclusion and marginalization, exploring how unauthorized parents’ social exclusion impacts their children’s development, and employing mixed-methods and longitudinal designs to understand the impact of legal vulnerability on immigrant families’ lives.
Footnotes
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was funded by the Foundation for Child Development (RIC-13-1), the Rhode Island College Faculty Research Fund, and the Rhode Island College Feinstein School’s Dean’s Office.