Introduction

Undocumented immigrants (also known as “illegal immigrants”) and their undocumented children face the constant threat of deportation because they entered the United States illegally or entered legally with visas but remained in the US after the expiration of their entry permits. Many families with undocumented immigrant parents are known as “mixed-status” families, that is, they are undocumented parents, sometimes undocumented children who immigrated with their parents, and US-born citizen-children. The number of children in mixed status families is estimated to be around 9 million, about 4 million of whom are US-born citizens and most of the from Latin America (Taylor et al. 2011). US-born members of these families—referred to in this paper as “citizen-children”—are protected by US law but suffer from the deportability and actual deportation of their parents and siblings. In effect, citizen-children suffer from their parents’ precarious legal status and vulnerability to deportation. The constant dread of the possible arrest, detention, and deportation of their parents sets the context that places citizen-children at risk for negative psychological effects and disruption of their developmental trajectories (Brabeck and Xu 2010; Dreby 2012; Zayas and Bradlee 2015). Then, the actual arrest, detention, and deportation of parents serve only to complete the trauma, and the certain detrimental impact on the children’s mental health (Zayas 2015). This is a class of children that deserve our utmost care and attention. After all, children of Hispanic immigrants are at higher risk for psychosocial problems in view of their higher rates of poverty, discrimination, and other social conditions that marginalize them (Zayas 2015).

Lamentably, there has been a lack of public attention and empirical research about citizen-children at a time when the US has reached unprecedented levels of deportations (Lopez and Gonzalez-Barrera 2013). In the past decade, nearly 2 million persons were removed from the country, 81 % of them to Latin America. Of the estimated 11.7 million undocumented immigrants living in the United States, 52 % are from Mexico (Passel et al. 2013), placing them at greatest risk for deportation. With estimates showing that for every two adults that are deported, one citizen-child is directly affected (Capps et al. 2007), over 1 million US citizens may have been subjected directly to the impact of immigration enforcement over the past decade. There have been reports that describe the negative effects on children when their parents have been detained. For example, as a result of the workplace raids that occurred in 2006 and 2007, 900 undocumented workers were arrested and were separated from 500 of their children (many under age five) (Capps et al. 2007). Based on personal accounts from social workers, teachers, parents, and local advocates, the majority of these 500 children suffered emotional distress.

According to the Center for Policy Priorities’ report A Child Alone and Without Papers (Thompson 2008), children who are separated from their parents, are more likely to be affected by separation anxiety, depression and other psychological stressors. US-born children of undocumented immigrants experience many stressors, not least of which are relocations as parents seek work, separations from extended families, and fear of discovery by US Immigration and Customs Enforcement (ICE) officials (American Psychiatric Association 2008; Lamberg 2008). One limitation of the psychological literature is the scarcity of empirical studies on the mental health of citizen-children in mixed-status families. The present study on the psychological effects of parents’ detention or deportation on citizen-children joins other studies (e.g., Allen et al. 2013; Brabeck and Xu 2010; Dreby 2012) that explore this growing and much-needed area of research.

Parents’ undocumented status is strongly associated with poverty, discrimination, parental distress, and poor physical and mental health of their children (American Psychological Association 2012). Families of immigrants suffer economic hardships which affect their abilities to meet their housing costs and lead to overcrowded households and frequent moves (Ayón et al. 2012). Children of Mexican parents are more likely than non-Hispanic white children to live in families below the federal poverty level and children of undocumented parents are also more likely to suffer from food insecurity (i.e., reports of reduced quality, variety, or desirability of diet limited by lack of money and other resources; US Department of Agriculture 2014) than children of US citizens (Kalil and Chen 2008; Ortega et al. 2009). Along with food insecurity that compromises their children’s health growth and development, undocumented status influences profoundly the utilization of health care and clinical encounters with health providers of immigrant families and their citizen-children. Ortega et al. (2007) asserted that undocumented immigrants have fewer visits to health providers than citizens with authorized immigrant status. They also found that undocumented Mexicans are less likely to have a usual source of care and more likely to report having had negative experiences with their health care providers and institutions than US-born Mexicans.

These results are consistent with Yoshikawa’s (2011) findings that undocumented parents are not only less likely to use health care services because they lack health insurance and fear being reported to immigration authorities, but also may not use public programs such as food stamps and child care subsidies, for which their citizen-children are eligible, for fear of having their undocumented status discovered. Research on fear and threat of deportation associated with parents’ legal status has been linked empirically to delay and avoidance in using health care (including mental health), social services and welfare benefits available to their US-born children (Xu and Brabeck 2012).

Ayón (2014) contended that immigrant mothers understand that their citizen-children are entitled to health and other public services but “scare tactics” such as being asked for their identification cards or documents despite the fact that the service is available for their US-born children, prevent them from accessing care for their children. As a result of fears of being deported and lack of access to health insurance, parents choose to pay in cash medical care, adding to their already tight financial situations (Ayón 2014; Cristancho et al. 2008; Kullgren 2003). Brabeck and Xu (2010) conducted a qualitative study with 132 immigrants from Guatemala, Colombia, Dominican Republic, El Salvador, Mexico, and Honduras, and their interviews revealed that more than half of the parents felt that immigration enforcement policies and practices caused them to worry about their capacity to provide financially for their children.

Citizen-children may miss important educational experiences that are critical to their development progression and mental health. For example, children of undocumented parents are less likely to be enrolled in public, preschool programs and they also have lower rates of positive development-promoting activities compared with children of citizens or legal permanent residents in the US (Crosnoe 2006; Hernandez et al. 2008; Kalil and Crosnoe 2009; Kalil and Chen 2008; Matthews and Ewen 2006; Ortega et al. 2009; Yoshikawa 2011). In short, children of undocumented parents experience limited access to educational opportunities during the most critical years of a child’s mental development. Out of fear of being discovered as unauthorized immigrants and reported to immigration enforcement, immigrant parents may simply not enroll their children in school programs.

Legal vulnerability to detention and deportation exerts a detrimental impact on the daily lives of parents and citizen-children in mixed-status households. The recent intensification of immigration enforcement activities by ICE has put undocumented parents and their citizen-children at increased risk for family separation, economic hardship, and psychological trauma (Henderson and Baily 2013; Lamberg 2008). The looming potential of arrest, detention, deportation, loss, and separation raises tension and stress, which in turn strains relationships between parents and between parent and children. Dreby (2012) reported that children’s daily lives are not only been directly altered by parental deportation but also indirectly affected by the immigration policies that criminalize their parents, relatives, and neighbors regardless of their citizenship status and actual involvement with immigration enforcements. For example, on the basis of interviews conducted with 40 families in north central Indiana, Chavez et al. (2012) found that the uncertainty of living in a mixed-status family raised parents and children’s stress levels.

Similar findings were reported by Brabeck and Xu (2010), who also found the undocumented parents’ legal vulnerability to be significantly associated with children’s well-being: as the parents’ level of deportability increases so too does their children’s stress levels. Yoshikawa (2011) noted that immigrant parents’ own state of worry and anxiety about the vulnerable legal status they occupy is transmitted to their young children through words and deeds. These are important findings because they show that increasing risk of family separation can have a long-lasting impact on children and lead to psychological trauma among other potential mental health problems.

Studies have shown that the lingering possibility of deportation of parents affects children, leaving them with constant anxiety and vigilance about the potential becoming real (De Genova 2010; Dreby 2012; Talavera et al. 2010). The findings of these studies have suggested that the impact of immigrant children’s separations from their parents, especially separations from mothers, has negative psychological effects not just for the children but for the mothers as well, and not just during the separation but even after they were reunited (Gindling and Poggio 2009). A small, community-based study of children with undocumented parents showed that, of the twenty children who participated in the project, thirteen had scores in the borderline and clinical range in at least one behavioral disorder (Delva et al. 2013). Eight of the children had scores that were in the borderline clinical range or in the clinical range on more than one mental health disorder. The most commonly seen problems were attentional deficits, withdrawal and depression, anxiety and depression, and some rule-breaking behaviors. Parents’ legal vulnerability and experiences of detention and deportation, in particular, were strongly associated with children’s depression, anxiety, fears of separation, social isolation, self-stigma, aggression, and withdrawal (Brabeck and Xu 2010; Chavez et al. 2012; Delva et al. 2013; Dreby 2012; Gonzales et al. 2013).

In a well-designed project that is of particular relevance to the present study, Allen et al. (2013) examined the mental health of three groups of children of undocumented immigrants in Texas based on measures completed by children’s parents or caregivers. Findings highlight that children with a deported parent were significantly more likely to display internalizing problems (e.g., depression, anxiety) and externalizing (e.g., aggression, conduct problems) than children whose parents were not deported or in the process of deportation, after controlling for demographic variables and trauma history. These findings provide strong early verification of the impact of parents’ deportability and deportation on children.

The present report adds unique elements to this body of literature in two ways. First, all data for this cross-sectional study were collected directly from citizen-children rather than their parents. Second, our project included a binational sample of citizen-children: a group living in Mexico with their deported parents, a group in the US who parents were deported or in the process of deportation, and a group in the US that was not undergoing any deportation procedures. Our conceptual approach to this study draws on developmental systems theory (Lerner 2001), which posits that human growth and maturation occur within social, cultural, physical and architectural, economic, political, and historical structures that are themselves changing. This theoretical approach fits the citizen-children we study since developmental systems theory has at its core the notion that the basic process of development is a relational one—between the person and the multiple ecological levels that the person grows in (Lerner 2001). Our research question was: How do the citizen-children whose parents have been deported or under deportation proceedings compare with citizen-children whose parents have not been deported and are not under any deportation orders?

Method

Participants

From 2012 to 2014, we recruited three groups of US-born citizen-children of undocumented Mexican immigrant parents between 8 and 15 years of age in two countries (US and Mexico), metropolitan areas in two US states (Sacramento, CA and Austin, TX), and several states in Mexico (e.g., Michoacán, Sinaloa, etc.). The first group consisted of citizen-children living in Mexico with deported parents. These children had accompanied one or both parents to Mexico after deportation proceedings were initiated or when deportation orders were affirmed by US immigration courts; some took voluntary deportation in anticipation of deportation. Eligible participants for the first group were recruited from different towns where they lived in Mexico through collaboration with researchers at the Instituto Nacional de Psiquiatría of Mexico. The second group consisted of citizen-children who remained in the US with a parent or guardian after one or both parents had been deported to Mexico, detained by immigration enforcement, in deportation proceedings, or returned to the US after being deported. The last group that served as a comparison was comprised of citizen-children whose undocumented Mexican immigrant parents had never been detained or deported and were not undergoing any deportation proceedings. This third group provided a profile of the typical psychological and social situation of citizen-children, those who live with the constant threat of parents’ arrest, detention, and deportation. The citizen-children in the latter two groups were recruited through social and health agencies in communities in the greater metropolitan areas of Sacramento, California, and Austin, Texas, locations with large undocumented immigrant populations accessible to the research team.

We selected children in the preadolescent and early adolescent years, corresponding to the late-elementary and middle-school period, as this stage in life is marked by major developmental milestones in cognition, emotional processing, and behavior, and also the ascendance in influence of peer groups, extra-familial systems, and social roles. The age range foreshadows adolescence when psychiatric problems typically have their onset. As the study focused on US-born children of undocumented Mexican immigrants, children in this age group would have completed at least 4 years of education in US schools.

We excluded from the study children who were not in our target age group; children with a severe mental illness, autism, developmental disability; or whose parents or guardian refused to give consent. We also excluded citizen-children whose parents were deported due to a criminal felony conviction other than having entered the country illegally or had overstayed their visas. We also excluded any citizen-child in foster care or child welfare. In families with more than one eligible child, we interviewed all of them. Eighteen children from 8 sibling groups were interviewed and included in our analysis. While we intended to include eligible children after the first 3 months of their parents’ deportation, the challenges of recruiting sufficient numbers of children whose parents would permit their participation in the study required that we admit children whose parents had been deported more than 3 months earlier but no longer than 1 year.

Research protocols for this project were approved by the institutional review boards (IRB) for the protection of human subjects at our respective institutions, The University of Texas at Austin, the University of California, Davis, and the Instituto Nacional de Psiquiatría Ramón de la Fuente Muñiz. A federal certificate of confidentiality was obtained for additional protection of the children and their parents. In light of the tenuous legal status of the undocumented immigrant parents whose children entered the study, our IRBs permitted parents to give oral consent for their children to participate and for children to give oral assent to participating in the study. A total of 83 citizen-children completed the survey and were included in the analysis.

Measures

For this project, we selected measures that would address the key psychosocial issues that citizen-children would be likely to present given the disruptions in their lives (e.g., depression, anxiety, internalizing and externalizing behaviors, trauma, and self-worth). Additionally, measures had to have been used successfully with Hispanic populations in both Spanish and English versions. To meet these criteria, we utilized the following instruments.

Child Behavior Checklist (CBCL) and Youth Self-Report (YSR) DSM-Oriented Scales

For a diagnostic profile of children’s emotional and behavioral problems, parents and children respectively completed the Diagnostic and Statistical Manual of Mental Disorders (DSM)-oriented CBCL and YSR scales (Achenbach and Rescorla 2001). Parents completed the CBCL for children under the age of 11, and youth over the age of 11 completed the YSR. Consistent with the DSM categories, the CBCL and YSR DSM-oriented scales feature six subscales: (1) affective problems, (2) anxiety problems, (3) somatic problems, (4) attention deficit/hyperactivity problems, (5) oppositional defiant problems, and (6) conduct problems. The CBCL and YSR DSM-oriented scales respectively comprise 53 items with three response format: 0 (not true), 1 (somewhat or sometimes true), or 2 (very true or often true). The higher scores indicate greater levels of symptoms of each disorder. We used raw scores for the current analysis in order to take into account the full range of variation in these scales. The validity and reliability of the CBCL and YSR DSM-oriented scales have been documented (Achenbach and Rescorla 2001), and the scales have been successfully used with Hispanic and low income minority youth (Domenech Rodríguez et al. 2006; Rescorla et al. 2007). After deleting cases that had more than 5 % of missing data in the CBCL and YSR, our final sample consisted of 52 children who completed the YSR and 20 parents who completed the CBCL. Internal consistency for the present sample was high (α = 0.87).

Children’s Depression Inventory 2nd Edition (CDI-2)

Depressive symptoms were measured with the full length CDI-2 (Kovacs 2003). The CDI-2, a revision of the CDI, is designed to provide a comprehensive assessment of affective and functional problems of depression in children and adolescents aged 7–17 years. It consists of 28 items that yield a total score (e.g., computed as raw total or as standardized T-score), two scale scores (e.g., emotional problems and functional problems), and four subscale scores (e.g., negative mood/physical symptoms, negative self-esteem, ineffectiveness, and interpersonal problems). Children were asked to indicate the level of symptomatology with a 3-point scale: 0 (absence of symptoms), 1 (mild or probable symptom), or 2 (definite symptom). Higher scores indicate greater levels of depressive symptoms. Standardized T-scores were used in the current analysis. Its reliability and validity were documented (Kovacs 2003; Saylor et al. 1984) with Hispanic children (Liberman et al. 2012). The internal consistency on the CDI-2 for our sample was initially very low (α = 0.14). However, upon closer scrutiny we found that on the sole question about suicidal ideation no child endorsed the answer, “I want to kill myself.” When we removed this item from analyses, the CDI-2 internal consistency jumped to α = 0.86.

Screen for Child Anxiety Related Emotional Disorders (SCARED)

Anxiety disorder symptoms were measured with the SCARED: Child Version (Birmaher et al. 1999). The SCARED consists of 41 items with five subscales: Panic Disorder/Significant Somatic Symptoms, Generalized Anxiety Disorder, Separation Anxiety Disorder, Social Anxiety Disorder, and Significant School Avoidance. Children were asked to describe the degree to which statements (e.g., When I feel frightened, it is hard to breathe, I get scared if I sleep away from home, and I feel nervous with people I don’t know well) were true about them with a 3-point scale: 0 (not true or hardly ever true), 1 (somewhat true or sometimes true), and 2 (very true or often true). Higher scores indicate greater levels of anxiety disorder symptoms. The SCARED total score and five subscale scores can be obtained by summing relevant items, and we used raw scores for the current analysis. The SCARED has good psychometric properties (Birmaher et al. 1999) and is validated with Hispanic children (Vigil-Colet et al. 2009; Hale et al. 2011; Wren et al. 2007). Internal consistency for the present sample was satisfactory (α = 0.78).

Piers-Harris Children’s Self-Concept Scale 2 (Piers-Harris 2)

The Piers-Harris Children’s Self-Concept Scale 2nd Edition, subtitled “The Way I Feel About Myself”, is a 60-item self-report questionnaire designed to measure a child’s own self-concept and perception between the ages of 7 and 18 years (Piers et al. 2002). To prevent participant fatigue and redundancy of measures, we chose four of the six domain scales of the Piers-Harris 2 for this project: behavioral adjustment, intellectual and school status, freedom from anxiety, and happiness and satisfaction. Children were asked to indicate whether each statement (e.g., I am well behaved in school, I get nervous when the teacher calls on me, I like being the way I am, and I am easy to get along with) applied to them by choosing a yes or no. The total raw scores, derived from the number of items an individual answered indicating positive self-concept, were converted to standardized T-scores for the current analysis. Higher T-scores represent more positive self-concept of each domain. Validity and reliability of the scale have been documented with Hispanic children (Piers et al. 2002). Internal consistency on the Piers-Harris for our sample was satisfactory on all subscales: behavioral adjustment (α = 0.81), intellectual and school status (α = 0.72), freedom from anxiety (α = 0.79), and happiness and satisfaction (α = 0.86).

Procedures

Recruitment occurred through referrals from various community sources. When a prospective participant was identified, a representative from the participating community partner approached the parents of a citizen-child who met basic criteria for inclusion and shared with them the purpose and participation requirements and potential benefits of the study. In this manner, we protected the family’s privacy. If parents and children declined no further action was taken. If the parents and children expressed an interest in hearing more about the study, they were offered the option of calling a member of the research team or permitting the community representative to provide the research team with a contact telephone number to call the family. Most often, families preferred to share their telephone numbers, and a member from the research team contacted them. Research staff then followed up with a phone call to provide the details of the research and obtain parental consent.

When written parental consent and children assent was completed, the child was interviewed alone and privately, without others present. All measures were read to the children by the interviewers. Parents or guardians completed the CBCL on younger children and provided demographic data on the family. All data collection occurred during one visit. Children and parents were compensated for their time with department-store gift cards valued at $25.

Demographic Variables

Demographic information included age (in years), gender (0 = male, 1 = female), school enrollment (0 = no, 1 = yes), and current living arrangement (0 = living without parents, 1 = living with one of parents, 2 = living with both parents).

Analytic Strategy

For descriptive analyses, means and standard deviation were used to assess clinical status in the aggregate group as well as in each subgroup. First, the three groups of children were compared using analyses of variance (ANOVAs) for continuous variables (e.g., clinical measurements). In order to determine which groups differ from each other, a post hoc test was performed using Tukey HSD.

Then, we collapsed the sample into two groups: (1) those citizen-children affected directly by parental detention and deportation and who lived in Mexico and the US and (2) those citizen-children living in the US whose parents had not been detained or deported at any time. This approach dichotomized the children into those directly affected by detention and deportation and those who are not. The t-tests were conducted to evaluate two group-differences of clinical characteristics. Differences were considered statistically significant if p < 0.05. All analyses were performed using IBM SPSS Statistics 22.

Results

To restate the purpose of our project, we sought to explore how citizen-children whose parents have been deported or under deportation proceedings compare with citizen-children whose parents have not been deported and are not under any deportation orders. To this end, we recruited a sample of 83 participants. The sample was comprised of citizen-children who accompanied their deported parents to Mexico (n = 31), citizen-children who were affected by parental detention/deportation but remained in the US (n = 18), and citizen-children whose undocumented parents were not under removal proceedings (n = 34). They aged between 8 and 15, with an average age of 11.4 (SD = 1.90). More than 60 % of participants were females, and a vast majority (98.8 %) was currently enrolled in schools. With regard to current living arrangement, children living with both parents, with one of the parents, and with none of the parents were 67.5, 30.1, and 2.4 %, respectively. All children had at least one parent of Mexican origin. No significant differences across subgroups of children were found in any of the background characteristics (not shown in tabular format; available upon request).

Table 1 presents clinical status of children and comparisons among each subgroup. With respect to the CBCL and YSR DSM-oriented scales, none of subgroups were considered as being at the borderline or clinical range when the clinical cut-off scores were applied (Achenbach and Rescorla 2001). However, group differences were found to be significant for the attention deficit/hyperactivity problems [F(2, 69) = 3.66, p < 0.05]. A post hoc test using Tukey HSD showed that children who were affected by parental detention/deportation but remained in the US were more likely than children whose undocumented parents were not under removal proceedings to report the attention deficit/hyperactivity problems. With regard to depression measured by CDI-2, none of subgroups of children fell within category of probable depression when the clinical cut off T-scores (≥60) were applied (Kovacs 2003). Children who accompanied their deported parents to Mexico had the highest average scores of depression symptoms (Mean = 57.0, SD = 10.8). Differences of CDI-2 T-scores among the three subgroups were not statistically significant due to low statistical power. Of the two scales and the four subscales of CDI-2, significant differences were found only in emotional problems [F(2, 80) = 3.89, p < 0.05] and negative mood/physical symptoms [F(2, 80) = 3.46, p < 0.05]. A post hoc test demonstrated that children who accompanied their deported parents to Mexico were more likely to report emotional problems and negative mood/physical symptoms than children whose undocumented parents were not under removal proceedings. In terms of anxiety-related disorders, group differences were not observed at the statistically significant level. However, all three subgroups of children fell within the category of probable anxiety disorders, including separation anxiety disorder when the standard cut-off scores were applied (Birmaher et al. 1999). With regard to children’s self-concept and perception assessed by Piers-Harris 2, children whose undocumented parents were not under removal proceedings had the highest average scores of four subscales. Significant differences were found for freedom from anxiety [F(2, 80) = 4.47, p < 0.05] and happiness and satisfaction [F(2, 80) = 3.97, p < 0.05]. A post hoc test demonstrated that children whose undocumented parents were not under removal proceedings were more likely to report positive self-concept and perception than children who accompanied their deported parents to Mexico.

Table 1 Comparison of mental health profiles among the study groups
Full size table

Table 2 summarizes comparisons between children directly affected by parental deportation/detention (n = 49) and children not directly affected by parental deportation/detention (n = 34). With regard to the CBCL and YSR DSM-oriented scales, no significant difference across the two groups was observed. In terms of the CDI-2, a significant difference in depression was found between the two subgroups; children affected directly by parental deportation/detention had higher levels of depressive symptoms (t(81) = 2.25, p < 0.05) than their counterparts. In particular, children affected directly by parental deportation/detention were more likely to report their emotional problems, including negative mood/physical symptoms and negative self-esteem, than their counterparts. Although no significant differences in anxiety related disorders were observed, both groups fell within the category of probable anxiety disorders. With regard to self-concept and perception, children affected directly by parental deportation/detention had lower mean scores of freedom from anxiety (t(81) = –2.88, p < 0.01) and happiness and satisfaction (t(81) = −2.84, p < 0.01) than those of children not directly affected by parental deportation/detention.

Table 2 Comparison between citizen-children affected directly by parental deportation/detention and citizen-children not directly affected
Full size table

Discussion

No country, but especially one with a tradition of the rule of law and a rich history of civil and human rights laws, should formulate its deportation policies and practices without knowledge from scientific evidence. Our findings are consistent with other studies (e.g., Allen et al. 2013; Ayón 2014; Brabeck and Xu 2010; Chavez et al. 2012; Delva et al. 2013; Dreby 2012; Yoshikawa 2011) in building an empirical base on how current immigration policy and deportation practices affect the mental health and well-being of citizen-children and their families. Among the three group citizen-children that we studied—a group in Mexico whose parents were deported, a group in the US whose parents had been deported or were undergoing deportation procedures, and a comparison group of citizen-children of undocumented immigrant parents with neither detention nor deportation proceedings—there were few differences that reached statistical significance. However, children with a parental history of detention or deportation reported possible attention deficits. In addition, citizen-children in Mexico with deported parents displayed more depressive symptoms than other children. All three subgroups of children fell within the category of probable anxiety disorders. Children whose undocumented parents were not under removal proceedings had more positive self-concepts and perception than children who accompanied their deported parents to Mexico. When groups were collapsed into two categories–children affected by parental deportation or detention regardless of location (n = 49) and children not directly affected by parental deportation or detention (n = 34), the differences remained. Children affected directly were more likely to report higher levels of depressive symptoms and emotional problems (e.g., negative mood, physical symptoms, and negative self-esteem) and lower levels of freedom from anxiety and happiness and satisfaction than their counterparts. A glimmer of hope was evident in our data when we discovered that none of the 83 children in our sample reported suicidal ideation or intent. We explored age and gender differences in the mental health profile of our sample but found no significant differences. Future research with larger samples may yield different results.

These findings point to the probable disruptive effects that parents’ detention and deportation can have on the psychosocial functioning of children. Even living under the cloud of the deportability of their parents has a negative effect on children. There is the constant sense of vulnerability to losing a parent and a home if parents are arrested, detained, and deported. The high level of anxiety across all groups of children in our study appears to support this point. As Allen et al. (2013) point out, the apprehension of a parent’s removal, or the actual loss of a parent for immigration law violations, causes a level of stress that can lead to aberrant developmental trajectories in otherwise healthy children. Our findings align with the concern voiced by other researchers that detention or deportation of a parent can have significant emotional and behavioral consequences and detrimental educational outcomes.

Several issues, however, limit our findings. The sample size (83 citizen-children) which was divided into three groups reduced statistical power, particularly in the CBCL and YSR in which a number of cases had to be dropped due to missing data. Further, the broad developmental and age range (8–15 years of age) and the social and geographic heterogeneity of the sample may have contributed to low internal validity and reliability in some of our measures. Parents’ delicate legal status and their general suspicion and mistrust of large institutions affected our recruitment and, consequently, the representativeness of the citizen-children who ultimately participated. Undocumented parents do not disclose their legal status easily, and even the trust they placed in the community partners that referred them to us was not sufficient to encourage more families to participate in the study. And among those parents who completed measures on their children and who permitted their children to participate, there may have been a level of circumspection operating in their responses and their children’s responses to our questions. In Mexico, the difficulty of locating larger number of children was related to the dispersed settlement patterns of families in a country of 761,600 square miles (one fifth the size of the US) and our inability to get any official numbers as to where these repatriated parents were going. Many returned without notifying the government.

There are a number of suggestions that emerge for future research. Binational studies such as this one need larger samples to raise statistical power. Focusing on a narrower age group, such as adolescents only, would insure greater comparability among the sample. We endorse a multi-informant approach, and in the future investigators should widen the scope of data collected to include more family-centered data, such as information on parents, siblings—both citizen and undocumented—and school information. Expanding the use of different instruments that measure other psychological and social constructs will enhance future research.

Notwithstanding these limitations, there are several strengths of the project that make it a useful addition to the burgeoning literature on children and immigration enforcement. First, while we succeeded in recruiting vulnerable and difficult-to-reach families and children, the challenges of conducting this kind of research with such a population are substantial. The fact that we were able to engage and recruit the children and parents in two countries (US and Mexico), metropolitan areas in two US states (Sacramento, CA and Austin, TX), and several states in Mexico (e.g., Michoacán, Sinaloa, etc.) who lived under difficult circumstances is a strength of our study. Our experience points to the feasibility of conducting binational research on such populations and the need for focused team work. It also speaks to the rich possibilities that exist for studying an important segment of the population directly affected by immigration enforcement. Other strengths are related to methodological factors. For example, unlike other reports that have surveyed parents about their children’s psychological adjustment in regard to immigration status, we collected data directly from the children using standardized measures previously used with Hispanic populations. Another advantage is that our sample included US-based children with and without parental detention or deportation and Mexico-based children following a parent’s or parents’ deportation. Other researchers have studied the children of undocumented immigrants from different Latin American countries, thus introducing potential differences in the experiences of immigration and deportation. While no single project can be expected to correct for all these limitations, it is vital that research continue and improve on past efforts.

There are some practice and policy implications that can be drawn from our findings. For practitioners, this study points to the need for close assessment of citizen-children not just whose parents are under deportation proceedings, or have been deported, but clinical attention to citizen-children who live under the fear of parents’ deportability. Children in all of these categories present complex clinical and family pictures. Those practitioners working with this population in Mexico can help children and families with the adjustment to the new cultural and linguistic and social context. Issues of self-worth and self-image are evident when children who were enculturated to a US context are uprooted by government action and forced to undergo a process of acculturation to Mexican society.

There are some implications for immigration enforcement that can be derived from the findings of the detrimental impact that deportation has on the mental health of US-born children of undocumented immigrants. First and foremost, immigration enforcement policies and practice should be concerned with the circumstances and wellbeing of citizen-children during the detention and deportation of their parents. These children are, after all, citizens who deserve all the protections to which they are entitled. Prospectively, social, health, and immigration enforcement policies must look at the impact that living under the threat of deportation and the actual deportation process has on citizen-children. Before workplace raids occur or immigrants are detained, immigration officials should consult with local child welfare authorities and mental health professionals to determine what is in the best interest of the citizen-children who will be affected by their enforcement actions. The process can begin with providing children access to their parents during the detention process. Preparations must be made for children’s care and counseling to minimize the trauma that they suffer when parents are arrested, detained, deported, and repatriated. Federal immigration officials and local child welfare and social services must work together to minimize the disruption to children’s attachment. Most undocumented immigrant parents present very little threat to public safety and, therefore, supervised release after an arrest will insure that they can provide care for their citizen-children’s health and development. Careful planning for the care and future needs of citizen-children should be undertaken well before a parent is deported.

While these implications are prospective, there is the question of what we do about the children, such as those in our study whose parents are already deported. Some of the children have been left in the care of others in the US and some of the children moved to other countries with their deported parents. Again, action is needed to provide the protections that they as citizens would enjoy in the US One initial solution is to insure access to consulates and to the services that are provided for expatriates or even tourists. Policies should also be proactive such that efforts are undertaken to seek out US citizen-children and insure their health, well-being, and educational progress.