ArticlePDF Available

Variations in Citizenship Profiling by Generational Status: Individual and Neighborhood Characteristics of Latina/os Questioned by Law Enforcement About Their Legal Status

Abstract

Although racial profiling is widely studied, the related issue of citizenship profiling by law enforcement has received little scholarly attention. In this study we begin to address citizenship profiling, which may be highly salient in light of the increasing policing of immigration in the United States through Secure Communities and other federal, state and local efforts to localize the enforcement of immigration laws. Using a sample of 563 Latina/o adults residing in 46 neighborhoods in El Paso County, Texas, USA, we assess the impacts of a variety of individual and neighborhood characteristics on the likelihood of being questioned about citizenship status by law enforcement. Results using hierarchical generalized linear models (HGLMs) show that, at the individual-level, first-generation Latina/o immigrants and second-Latina/os are more likely to be questioned about citizenship status than third- and later-generation Latina/os. At the neighborhood-level, living in a neighborhood with a mid-level of Latina/o immigrant characteristics increased the probability of being questioned. The implications of these findings for citizenship profiling are discussed.
Vol.:(0123456789)
1 3
Race and Social Problems (2018) 10:293–305
https://doi.org/10.1007/s12552-018-9235-3
Variations inCitizenship Profiling byGenerational Status: Individual
andNeighborhood Characteristics ofLatina/os Questioned byLaw
Enforcement About Their Legal Status
MariaCristinaMorales1· DeniseDelgado1· TheodoreCurry2
Published online: 17 July 2018
© Springer Science+Business Media, LLC, part of Springer Nature 2018
Abstract
Although racial profiling is widely studied, the related issue of citizenship profiling by law enforcement has received little
scholarly attention. In this study we begin to address citizenship profiling, which may be highly salient in light of the increas-
ing policing of immigration in the United States through Secure Communities and other federal, state and local efforts to
localize the enforcement of immigration laws. Using a sample of 563 Latina/o adults residing in 46 neighborhoods in El Paso
County, Texas, USA, we assess the impacts of a variety of individual and neighborhood characteristics on the likelihood of
being questioned about citizenship status by law enforcement. Results using hierarchical generalized linear models (HGLMs)
show that, at the individual-level, first-generation Latina/o immigrants and second-Latina/os are more likely to be questioned
about citizenship status than third- and later-generation Latina/os. At the neighborhood-level, living in a neighborhood with
a mid-level of Latina/o immigrant characteristics increased the probability of being questioned. The implications of these
findings for citizenship profiling are discussed.
Keywords Citizenship profiling· Latina/os· Immigrant neighborhoods· Generational status· Policing
Introduction
Mexicans have historically been systematically labeled as
‘criminals’, ‘bandidos (bandits)’, ‘foreigners’, and ‘illegals’
in the US (Sáenz and Morales 2015; Carter and Lippard
2015; Escobar 1999; Macias-Rojas 2016; Mirandé 1987;
Rios 2011). Such social constructions of criminality have
historically influenced law enforcement to ‘protect’ the pub-
lic from Mexican-origin people who are perceived as crimi-
nal and culturally flawed (Escobar 1999). The racial profiling
of Latina/os has occurred (see Miller 2011; Koch etal. 2016)
despite evidence of the Latina/o Immigrant Crime Paradox,
which shows that Latina/o immigrants have lower crime
than blacks, whites, and native-born Latina/os, in spite of
their low socioeconomic status (e.g., Burchfield and Silver
2013; Hagan and Palloni 1999; Kubrin and Ishizawa 2012;
Martínez 2015; Martínez and Stowell 2012; Martínez and
Valenzuela 2006; Sampson 2008; Sampson and Bean 2006;
Stowell etal. 2009).
Despite low crime rates among Latina/o immigrants,
immigration policing that enforces civil and criminal viola-
tions of federal immigration law has grown across the US.
The post 9/11 period marked the growth of the local policing
of immigration beyond patrolling the ‘line’ or the interna-
tional US Mexico boundary (Arriaga 2017; De Genova and
Peutz 2010; Leerkes etal. 2013; Ngai 2014; Nguyen and Gill
2016; Provine 2016; Provine and Sanchez 2011). The sur-
veillance of immigration also expanded under the 1996 Ille-
gal Immigration Reform and Immigrant Responsibility Act
via the expansion of state and local police functions under
287(g) policies where state and local law enforcement agen-
cies enter agreements with the Attorney General to perform
immigration law enforcement functions (see Armenta 2012;
Provine etal. 2016; Wong 2012). Under President Obama,
* Maria Cristina Morales
mcmorales@utep.edu
Denise Delgado
dndelgado@miners.utep.edu
Theodore Curry
trcurry@utep.edu
1 Department ofSociology andAnthropology, University
ofTexas atEl Paso, ElPaso, USA
2 Department ofCriminal Justice, University ofTexas atEl
Paso, ElPaso, USA
294 Race and Social Problems (2018) 10:293–305
1 3
287(g) was phased out for being too costly and inefficient
(Provine etal. 2016); however, it has been reinstated and
strengthened under the Donald Trump Presidency (2017).
Similarly, Secure Communities is a US federal admin-
istrative initiative facilitating local- and state-level law
enforcement to partner up with the US immigration and
customs enforcement (ICE) and the Federal Bureau of
Investigations (FBI) for immigration policing. Secure Com-
munities ended in November 2014 and was replaced with
Priority Enforcement Program (PEP). Both programs share
similar objectives to utilize state and local police and courts
for civil immigration enforcement but differ in that ICE no
longer requests that local law enforcement detain a migrant
unless he/she has been convicted of a crime or is consid-
ered a national security threat (García Hernández 2015).
The recent and unprecedented convergence of criminal and
immigration law at the levels of statute, policy, and imple-
mentation has been referred to as ‘crimmigration’ (Stumpf
2006). This large-scale restructuring, marked by a shift from
national to local immigration policy, gave rise to a new
regime in immigration enforcement in the US that makes
the distinctions of “illegality” between citizens/documented
and undocumented individuals more important than before
(Coutin 2011; Macias-Rojas 2016).
Given these developments, the overall objective of
research is to begin to develop what we call citizenship pro-
filing, or who is suspected of being undocumented and thus
questioned about their citizenship status by law enforcement.
We built the literature on the domestic policing of federal
immigration law by examining the likelihood of being asked
about citizenship status by law enforcement, statistically and
theoretically introducing the concept of citizenship profiling,
and by considering both individual- and neighborhood-level
factors. In particular, we provide a statistical profile of who
law enforcement suspects of being undocumented on the
basis of individual-level factors (generational status, sex,
and age) and by considering how contextual factors at the
neighborhood-level (residing in poverty and Latina/o immi-
grant neighborhoods) might affect who law enforcement
questions about citizenship. We base this study on primary
research consisting of 563 surveys collected in 2014 in El
Paso, Texas, US, a city along the US México that is 82.2%
Latina/o.
Due to the local policing of federal immigration law,
racial profiling among Latina/os is a growing concern
(Briggs 2014; Cox and Miles 2015). Racial profiling occurs
when Latina/os are stopped and questioned by law enforce-
ment due to suspicion of their criminality. This study builds
upon racial profiling to start to investigate what we call
citizenship profiling.’ Citizenship profiling extends the
concept of racial profiling in a number of important ways.
In accordance with intersectionality, citizenship status is
interpreted through racialization. In citizenship profiling,
race/ethnicity may be the initial indicator that law enforce-
ment uses for stops or searches but it is also connected to
cultural characteristics associated with ‘foreignness’ such as
English language fluency, accents, certain types of clothing,
and a lack of understanding mainstream norms in the US.
Therefore, similar to racial profiling, citizenship profiling is
connected to phenotypical characteristics associated with
people of color, but it is extended to perceptions of who may
be undocumented or not a legal resident.
Racial andCitizenship Proling
Law enforcement officers may have daily encounters with
immigrants even in locations across the US where state,
county, or local law enforcement do not have agreements
with the federal government to police immigration law.
Scholars have argued that the growth in immigration polic-
ing at the local-level is a response to the rapid increase in
immigrant populations and/or the browning of the popula-
tion (Armenta 2016a; Sáenz 2010), although this pattern is
less likely in traditional migration destinations (Wong 2012).
Yet, the ‘browning’ and immigrant growth across the US
suggests that even states, counties, and local law enforce-
ment departments without agreements with the federal gov-
ernment to police immigration law are interacting more than
ever with immigrants.
With the expansion of policing of immigration, citizen-
ship profiling arises as an issue of increasing concern, yet
scholarly work about who is profiled as an undocumented
immigrant is very sparse. This is where we come in. Most
of the literature on the policing of immigrants is qualitative
in nature and has focused on the racialization of detecting
undocumented immigrants that has led to the dispropor-
tionate deportation of Latina/os (e.g., Golash-Boza 2012;
Provine etal. 2016; Romero 2006) and on the dis-function-
ality of local and state laws aimed at policing immigration
(e.g., Armenta 2016a; Golash-Boza 2015; Motomura 2011;
Provine etal. 2016; Provine and Sanchez 2011; Varsanyi
2008). Other studies have documented how law enforcement
involvement in federal immigration law has compromised
perceptions of police (Nygun and Gill 2016; Theodore and
Habans 2016).
The practice of detecting undocumented immigrants
has disproportionately targeted Latina/os, especially poor
Mexicans and Central Americans (Arriaga 2017; Chavez
2008; Ngai 2014; Provine etal. 2016; Provine and Sanchez
2011). In particular, municipal policing practices aimed at
immigrant removal has led to ethno-racial profiling, hyper-
surveillance, and abusive stops (Provine and Sanchez 2011).
Though research on this issue is scarce, existing findings
show that undocumented immigrants perceive that the
police are actively profiling them (Aranda and Vaquera
295Race and Social Problems (2018) 10:293–305
1 3
2015). Moreover, Theodore and Habans (2016) found that
immigration enforcement negatively affected the percep-
tions of police of both undocumented and authorized Latina/
os, which compromises cooperation with police and pub-
lic safety. Yet, even before the rise in the local policing of
immigration, Latina/os, including citizens and legal perma-
nent residents, reported they are more frequently stopped
and have more negative perceptions of the police (Vidales
etal. 2009).
When it comes to actual stops, close to 10% of Mexicans-
origin individuals, both native- and foreign-born, have been
stopped by the police and asked about their immigration
status in the US (Sáenz and Morales2015). Therefore, who
is questioned about citizenship status is not solely restricted
to immigrants but also to US citizens. This is evident in
the 834 US citizens who were mistakenly detained under
Secure Communities (Provine etal. 2016). Thus, similar to
the influence of racial profiling on drug enforcement dispari-
ties (Koch etal. 2016), there are indications of racial and
citizenship profiling to enforce immigration laws.
Policing Discretion andImmigrants
To address who is questioned by law enforcement for sus-
picion of being undocumented is difficult in part due to
the wide discretion that police work entails. Provine etal.
(2016, p.105) argued that “law enforcement is intensely
individualistic work in which officers on the street are gener-
ally beyond the view of their supervising officers and enjoy
wide discretion regarding how to investigate, and whom to
question, stop, and arrest”. Moreover, officers can almost
always find probable cause to pull over a driver or question
a pedestrian. For instance, officers can follow a car to find
probable cause (i.e., not wearing seatbelts) and make a stop
for technical violations such as a broken taillight (Provine
etal. 2016). Under such circumstances the real reason for
a stop differs from the violation that someone is cited for.
Immigration policing can also occur in departments with
no policies aimed at enforcing federal immigration policies
through the ‘back door’ via ordinances on the use of public
space that directly or indirectly exclude undocumented indi-
viduals (Varsanyi 2008). Perhaps the most common ‘back
door’ enforcement of immigration law is through traffic
offenses. Indeed, the most serious charge for over half of
the immigrants deported in 2013 was a traffic violation (Pro-
vine etal. 2016). In a study of 287(g) in Nashville, Armenta
(2016a, b) found that department culture prioritizes traffic
stops which inevitably places local police in contact with
unauthorized immigrants who, because of ineligibility for
driver’s license and identification cards, are vulnerable to
arrest. This is a concern for undocumented immigrants (first-
generation individuals without authorization to reside in the
US) who, with a few exceptions, do not have access to a state
IDs and driver’s licenses, and thus are committing an infrac-
tion with the very act of driving. In contrast, individuals who
are second- and later generations are US citizens who have
access to legally attain a driver’s license.
Law enforcement may disproportionately question first-
generation immigrants under the assumption of their illegal-
ity. Specifically, law enforcement may question those who
display immigrant characteristics such as Latina/o pheno-
type, lack of acculturation to US driving norms, limited Eng-
lish skills, types of dress, or a combination of these charac-
teristics, and suspect they are undocumented. In the case
of second-generation Latina/os, even though they are US
citizens, they may still display foreign-born cultural char-
acteristics being that they are from immigrant families and
thus may culturally still resemble the first-generation (refer-
ence). Therefore, based on insights from the literature we
predict that Latina/os who are first-generation (immigrant)
and second-generation (children of immigrants), in contrast
to third- and later-generations, are more likely to be targeted
and questioned about their citizenship by law enforcement
net of demographic factors at the individual-level and the
influence of the neighborhood-level characteristics.
Neighborhood‑Level
Despite an individual’s race/ethnicity and immigration sta-
tus, the neighborhood context in which individuals reside
could provide important contextual information about who
is asked about their citizenship status by law enforcement.
To date, most of the literature on policing and Latina/os and
immigrants focuses on how individual-level characteristics
influence questioning and detention by police, rather than
characteristics associated with neighborhoods in which indi-
viduals reside. On the other hand, most of the literature on
policing and neighborhood context in the US has focused
on ethnographic scholarship of the experiences of poor
urban blacks and has highlighted punitive surveillance, a
disproportionate number of citations, and disrespectful treat-
ment (e.g., Desmond and Valdez 2013; Rios 2011), among
other negative impacts. Below we discuss neighborhood-
level characteristics that may influence citizenship profiling
among Latina/os.
The influence of poverty and Latina/o neighborhoods
on policing has focused largely on crime (Martínez 2008).
Blacks and Latina/os are often relegated to reside in geo-
graphical areas marked by concentrated disadvantage and
social isolation due to an array of economic and social
disparities (e.g., joblessness, welfare dependency, pov-
erty, family disruption, and residential instability) all
of which are associated with crime (Martin etal. 2011).
Moreover, residing in poor neighborhoods exposes blacks
296 Race and Social Problems (2018) 10:293–305
1 3
and Latina/os to a range of negative effects including
the disproportionate experiences with police surveil-
lance and stops (e.g., Brunson 2007; Brunson and Miller
2006; Rios 2011). Thus, we predict that residing in poor
neighborhoods will be associated with an increased likeli-
hood of being questioned about citizenship status by law
enforcement.
There are also some indications that Latina/o immi-
grant neighborhoods are targeted for the presumed ille-
gality of its residents. Adler (2006) documented federal
police raids in an immigrant enclave that was transitioning
from Italian to Latina/o in New Jersey. Similarly, while
referring to border communities generally, Macias-Rojas
(2016) argued residents get branded as ‘perpetrators’ and
‘criminals’ through stops and arrest and prosecution, sen-
tencing, imprisonment for immigrant-related offenses and
even wrongful deportation. Romero (2006) also found the
Chandler Roundups in Arizona targeted residents for their
‘Mexicanness’, speaking Spanish, and being in a Latina/o
neighborhood. Given that citizenship profiling dispro-
portionately intends to target immigrants, we predict that
those who reside in low or mid-level Latina/o immigration
neighborhoods are less likely to be questioned about citi-
zenship status than those who live in neighborhoods with
a high concentration of immigrants.
Summary ofHypothesis
H1 Latina/os who are first-generation are more likely to be
questioned about their citizenship status by law enforcement
than those who are third- and later-generations.
H2 Latina/os who are second-generation are more likely to
be questioned about their citizenship status by law enforce-
ment than those who are third- and later-generations.
H3 As poverty in neighborhoods increases the likelihood
of being questioned about citizenship status by law enforce-
ment increases.
H4 Residing in a neighborhood with a low concentration
of Latina/o immigrants, in contrast to a neighborhood high
in Latina/o immigrant concentration, is associated with a
decrease in the likelihood of being questioned about citizen-
ship status by law enforcement.
H5 Residing in a neighborhood with a mid-level Latina/o
immigrant concentration as opposed to a neighborhood
high in Latina/o immigrant concentration is associated with
a decrease increase in the likelihood of being questioned
about citizenship status by law enforcement.
Data andMethods
Sampling
This study is based on primary data from the El Paso Neigh-
borhood Survey consisting of 46 neighborhood clusters and
563 individual surveys. Our sampling design follows Robert
Sampson’s and associates (1997) Community Survey Com-
ponent of the Project on Human Development for Chicago
Neighborhoods (PHDCN) dataset. We operationalize neigh-
borhoods by aggregating census tracts that were similar in
terms of geographic information (such as interstates, roads,
and other landmarks), demographic and economic indicators
from the American Community Survey (ACS) (2015) 5year
(2008–2012) census tract data on economic, foreign-born,
language, and Latina/o, and a rapid ethnographic assessment
where we gained information from street-level observations
about whether the census tracts should be considered stand-
alone neighborhoods or joined with neighboring census
tracts. We then randomly selected neighborhood clusters and
randomly selected 20 households within each neighborhood.
The response rate is 74.9%. It is likely that it was the less
vulnerable that refused to participate and as such the size
effects might be even larger than those reported in this study.
Once selected at random, we mailed a letter notify-
ing respondents that their household had been randomly
selected to participate in the study and asked if they vol-
unteer to participate. Respondents were given the option
to either call and schedule an appointment or a research
assistant will show up to their household in 3–4days to
conduct the interview or schedule an appointment. The
adult in the household with the most recent birthday was
selected to answer the survey. Respondents were paid
$20.00 for participating in the study. Interviews were
conducted in the preferred language of the interviewee
(English or Spanish). The survey consisted of 261 ques-
tions and lasted 45min on average.
Data were collected between March and August of 2014
in El Paso, Texas, US, by two of the faculty authors on this
paper and 50 graduate/undergraduate students. Specifically,
students enrolled in a two-semester methods course that
focused on methodological training and consequent data
collection for this project. Students became CITI certified
and we received IRB approval. 94% of the interviewers
were Latina/o. For this study, we focused solely on Latina/o
respondents (N = 563). According to the American Com-
munity Survey (2015) El Paso County is 82.2% Latina/o
in contrast to our sample that is 83% Latina/o. In regard to
generational status, our sample is 18% first-generation, 33%
are second-generation, and 47% third generation and later.
Similar to our sample ACS (2015) for El Paso County, Texas
is 25.9% foreign-born (first-generation).
297Race and Social Problems (2018) 10:293–305
1 3
Context
El Paso, Texas, is located along the US–Mexico border. El
Paso represents an ideal site to conduct a study on percep-
tions of citizenship illegality given that according to the US
Census estimates 25.5% of the population is foreign-born
and 82.2% Latina/o (of which 76.6% is Mexican-origin).
Moreover, El Paso is a heavily surveilled city given its
location along the geopolitical international boundary with
Mexico.
While residents are familiar with the presence of federal
police, particularly border patrol, most policing encounters
involve local law enforcement. The city of El Paso also has
no 287(g) agreements that allow the domestic policing of
federal immigration law but did practice Secure Communi-
ties, now PEP. It is important to emphasize that even in cit-
ies that resist cooperation with federal efforts to engage in
local policing of immigration, there is still exposure to the
national emphasis on enforcement (Provine etal. 2016). In
such instances, citizenship profiling may occur even depart-
ments that do not adhere to partnerships with the Depart-
ment of Homeland Security being that policing immigra-
tion occurs any time field officers encounter a foreign-born
individual (see Armenta 2012).
Level‑1 Dependent, Independent, andControls
The dependent variable in the analysis is Asked Citizenship
Status and is based on the question, ‘In the last five years,
other than at ports of entry or immigration checkpoints, how
often has law enforcement personnel ever asked you about
your citizenship status?’ Although the question allows for
individuals to indicate if they have been stopped more than
once, due to statistical concerns with missing data and low
frequencies of individuals that have been asked about their
citizenship status in over five different occasions, we opera-
tionalize the dependent variable as ‘1’ if respondents were
ever asked about their citizenship status, ‘0’ if otherwise.
There are some caveats about the dependent variable. We
do not know if the individuals are asked about citizenship
status by police, sheriff, or border patrol. Yet, it is highly
probable that the questioning about citizenship is done by
local (police department or sheriff) given that, in El Paso,
border patrol is largely restricted to enforcing the interna-
tional boundary. We also lack data on where individuals
were asked about their citizenship status and the activi-
ties that they were involved in at that moment (i.e., driv-
ing, working, hanging out, etc.…), yet characteristics of the
neighborhoods’ individuals reside in can reveal the impact
of residing in an immigrant neighborhood and residential
poverty on the likelihood of being asked about citizen-
ship status. Lastly, this is an inferential rather than a direct
measure of citizenship profiling. Profiling is a complex
“conglomeration of physical, behavioral, and psychological
components that increase the probability of apprehending a
suspect” (Higgins 2008, p.1). Yet, we built upon quantita-
tive measures of racial profiling in the literature that use dis-
parities in stops and arrest (2008 reference) and individual
perceptions of being profiled (e.g., reference) as measures of
racial profiling. We do capture if individuals are suspected of
being undocumented being that officers are directly asking
about citizenship status. Therefore, the dependent variable
does capture our overall objective to create a profile of who
is being targeted for migration policing.
The independent variable is generational status. We meas-
ure generational status according to the standards in this
field (see Kao and Tienda 1995). The first generation con-
sists of Latina/os who were born outside the United States.
The second generation consists of Latina/os who were born
in the United States and whose mothers are foreign-born.
The third and/or later generations include Latina/os who
were born in the United States and whose mothers were also
born in the United States. Generational status is measured
using two dummy variables (first and second generation),
with Latina/os who are third and/or later generations being
the reference group.
At the individual-level we also included controls for the
demographic variables of sex and age. Sex is entered as a
dummy variable (1 = male, 0 = female). Age is measured
as a continuous variable. Unfortunately, we are not able to
include a control for social class at the individual-level due
to a large number of missing values.
Level 2: Latina/o Immigrant andPoverty
Neighborhoods
To capture the degree to which a neighborhood has char-
acteristics of a Latina/o immigrant community we used a
combination of survey and the American Community Sur-
vey (2015)(ACS 2008–2012) (Table3 in Appendix). Items
for the scale were converted to z-scores and averaged to cre-
ate a scale measure of Latina/o immigrant neighborhood.
Cronbach’s Alpha for characteristics of Latina/o immigrant
neighborhoods is 0.824. Factor analysis extracted a single
factor where the lowest loading is 0.771. Preliminary analy-
sis determined that the influence of the degree to which a
neighborhood has immigrant characteristics on being asked
about citizenship status by law enforcement is not linear so
we constructed dummy variables representing low Latina/o
immigrant neighborhoods (1 = z-scores representing the bot-
tom 30th percentile; 0 = otherwise) and mid-level Latina/o
immigrant neighborhoods (1 = z-scores between 31st and
60th percentile; 0 = otherwise), with the reference category
representing neighborhoods with high-levels Latina/o immi-
grants (z-scores above the 61st percentile) (Table1).
298 Race and Social Problems (2018) 10:293–305
1 3
To capture neighborhood poverty, we used ACS (2015)
5-year (2008–2012) census data pertaining to each neighbor-
hood cluster. Specifically, the poverty variable is a continu-
ous variable based on ACS data on the percent of households
that earn less than 10,000 dollars that pertains to each neigh-
borhood cluster. For neighborhood clusters with more than
one tract, the data for these clusters were averaged across
tracts.
HGLM Analysis
Hierarchical linear modeling analysis is required since
individuals are nested in neighborhoods and not randomly
distributed (Bryk etal. 1992). Multilevel data violates the
assumption that low-level observations are independent
leading to biased standard errors, inflated Type I errors,
and even incorrect inferences, thus multilevel techniques
are suitable for this study (Hox 2010). Multilevel modeling
was applied to cluster structures, indicating that individuals
(Level 1) are nested within neighborhoods (Level 2) and
can exert cross-level interaction effects. Missing data were
handled with listwise deletion of cases when the MDM file
was made.
Since our dependent/outcome variable is binary, it is
appropriate to use multilevel logistic regression also referred
to as hierarchical generalized linear models (HGLMs) to
analyse the nonlinear structural models (Raudenbush and
Bryk 2002). Unlike HLM, HGLM uses a binomial level-1
sampling model and a logit link. We are interested in the
probability of law enforcement asking about citizenship (1 if
yes, 0 if no) and every level-1 record corresponds to a person
with a single binary outcome so the model type is Bernoulli
(Raudenbush etal. 2011). We used Bernoulli distribution
because the dependent variable is binary and EM Laplace
iterations produce a unit-specific model of EM Laplace esti-
mation output.
Three models were used in this study. Model 1 is an
unconditional model to determine if the variation in law
enforcement asking about citizenship status varies across
neighborhoods. Model 1 also examined the utility of mul-
tilevel modellng (Hox 2010) and served as the benchmark
for comparison with the other models. We then ran a ran-
dom coefficients model (Model 2) with Level-l predictors to
assess whether any of the slopes were significantly different
from zero when allowed to vary across neighborhoods. This
model was used to explore whether the individual-level fac-
tors contribute to law enforcement asking about citizenship
status, net of other individual-level controls. In particular,
Model 2 predicts the natural log odds of law enforcement
asking about citizenship status as a function of all individ-
ual-level covariates, where rij is a random effect that deter-
mines if being asked about citizenship status significantly
varies across neighborhoods net of individual characteris-
tics. A population-average model with robust standard errors
(Model 3) examines the relations between neighborhood-
level factors and the likelihood of being asked about citi-
zenship status accounting for individual-level factors. The
population-average model allows for random variation of
the neighborhood-level error term and thus estimates the
expected change in the mean outcome across the population,
not just for neighborhoods that share error variance. This
feature is contingent on the nonlinear logit function. Model 3
constitutes a full multilevel model examining the probability
of being asked about citizenship status by law enforcement:
Level-1 Model
Prob(Y = 1|B) = P
log[P/(1-P)] = B0 + B1*(AGE) + B2*(MALE) + B3*(FI
RSTGEN) + B4*(SECONDGEN)
Level-2 Model
B0=G00 + G01*(LESSTHAN) + G02*(ENCLLOW) +
G03*(ENCLMED) + u0
B1=G10
B2=G20
B3=G30
B4=G40
Mixed Model
eta = G00 + G01*LESSTHAN + G02*ENCLLOW + G0
3*ENCLMED
+ G10*AGE
+ G20*MALE
+ G30*FIRSTGEN
Table 1 Description of variables in the analysis. Source El Paso
neighborhood survey
Variables Description NMean SD
Dependent variable
Asked citizenship status 1 = Yes
0 = No
563 0.19 0.40
Level 1
Age Continuous 563 40.62 15.35
Sex 1 = Male
0 = Female
563 0.48 0.50
First generation 1 = Yes
0 = No
563 0.30 0.46
Second generation 1 = Yes
0 = No
563 0.33 0.47
Third generation Reference
Level 2
Immigrant neighborhood
Low-level 1 = Yes
0 = No
46 0.33 0.47
Medium-level 1 = Yes
0 = No
46 0.35 0.48
Large-level Reference
Percent of households that earn
less than $10,000
Continuous 46 11.42 8.32
299Race and Social Problems (2018) 10:293–305
1 3
+ G40*SECONDGEN
+ u0
Results
Bivariate Analysis
The predictive probabilities of being asked about citizen-
ship status by law enforcement are calculated to provide an
initial bivariate association between independent variables
and controls on being asked about citizenship status. Fig-
ure1 provides the individual-level (Level 1) probabilities
of being questioned about citizenship status. The predicted
probability of being questioned about citizenship status is
highest for the second-generation (0.59), followed by the
first-generation (0.52), and lastly the third and later gen-
erations (0.39). Therefore, there is preliminary support for
Hypotheses 1 and 2 in that the first- and second-generation
are more liked to be profiled for citizenship. Being male, in
contrast to female, is associated with a predicted probabil-
ity of 0.56 of being questioned by law enforcement about
citizenship status. Lastly, to predict the influence of age on
the likelihood of being asked about citizenship the mean
value of age is used (Sweet and Grace-Martin 2012). Being
of average age resulted in a 0.49 probability of being ques-
tioned about citizenship status by law enforcement.
Figure2 presents the predicted probabilities of being
questioned about citizenship status by law enforcement
across neighborhood-level characteristics (Level 2). Those
residing in neighborhoods with an average number of fami-
lies earning less than 10,000 have a probability of 0.50 of
being questioned by law enforcement about citizenship sta-
tus. Thus, half of individuals asked about citizenship status
reside in poor neighborhoods. In regard to Latina/o immi-
grant neighborhoods, the predicted probabilities of being
profiled for citizenship is greatest for those who reside in
neighborhoods with a mid-level of Latina/o immigrant
characteristics (0.57), followed by high Latina/o immi-
grant neighborhoods (0.49), and lastly those who reside in
neighborhoods with a low concentration of Latina/o immi-
grants (0.42). Therefore, there is some mixed preliminary
support about the influence of neighborhood Latina/o immi-
grant concentration (Hypothesis 5 and 6).
Hierarchical Generalized Linear Modeling (HGLM)
To more fully examine the relationship between genera-
tional status (Level 1) and poverty and Latina/o immigrant
neighborhoods (Level 2) on being questioned about citi-
zenship status by law enforcement, hierarchical general-
ized linear modeling (HGLM) is used. Table2 presents
the HGLM multilevel models predicting being asked about
citizenship status by law enforcement. The unconditional
model (Model 1) is a preliminary model that determines
the expected log-odds of being asked about citizenship sta-
tus by law enforcement. Model 1 excludes individual-level
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
1st generaon2nd generaon 3rd plus
generaons
Male age
Fig. 1 Predicted probabilities of being asked about citizenship status
by law enforcement, individual-level factors. Source El Paso Neigh-
borhood Survey
0
0.1
0.2
0.3
0.4
0.5
0.6
Poverty Immigrant
Neighborhood Low
Immigrant
Neighborhood Medium
Immigrant
Neighborhood High
Fig. 2 Predicted probability of being asked about citizenship sta-
tus by law enforcement, neighborhood-level factors. Source El Paso
Neighborhood Survey
Table 2 HGLM analysis for odds of being questioned about citizen-
ship
Likelihood ratio test of change of deviance is 18.89 model 1
(1586.22) to model 2 (1567.33) and 23.4 (model 1 (1586.22) to
model 3 (1562.82))
Significance levels: *p < 0.05, **p < 0.01, ***p < 0.001
Model 1 Model 2 Model 3
Level-1
Intercept 0.240*** 0.139*** 0.110***
First-generation 1.829* 1.888*
Second-generation 1.782* 1.772*
Third plus (reference)
Male 1.249 1.238
Age 0.972*** 0.972***
Level-2
% Poverty 1.025
Immigrant neighborhood
Low-level 1.151
Medium-level 1.560*
Large-level (reference)
300 Race and Social Problems (2018) 10:293–305
1 3
and neighborhood-level predictors and shows how much
variance there is around the intercept (Bryk and Rauden-
bush 1992). The intercept was significantly different from
zero (p-value < 0.001) and thus there is significant varia-
tion among neighborhoods (Bryk etal. 1992). Therefore,
there is a nested effect that warrants a multilevel analysis
given that Level 2 (neighborhood) predictors had variation
to explain. In this case, the expected log-odds corresponds
to a probability of 1/ (1 + exp{− 1.426758}) = 0.81 of Latina/
os being asked about their citizenship status by law enforce-
ment. Therefore, Latina/os who law enforcement ask about
their citizenship status are significantly different from those
who did not get asked and that being asked about citizenship
status varies across neighborhoods.
Model 2 introduces individual-level variables predict-
ing who is asked about citizenship status and it allows for
intercepts and slopes to vary across Level 2. The deviance
statistic allows a comparison between nested models (Heck
etal. 2013) and our results indicate that adding the Level 1
individual-level parameters decreases the deviance statistic
indicating improved model fit from the unconditional model
(Table2). Results show that the between neighborhood dif-
ferences observed in Model 1 remain statistically signifi-
cant after accounting for individual attributes. Supporting
Hypotheses 1 and 2, being first- or second-generation has a
positive effect on the probability of law enforcement offic-
ers asking about citizenship status net of the influence of
sex and age. In particular, being first-generation, in contrast
to third-generation and later, is associated with an increase
in the likelihood of being asked about citizenship status by
law enforcement by 83%. Being second-generation increases
the likelihood of law enforcement asking about citizenship
status by 78%, in contrast to those who are third- and later-
generations. Among the demographic controls, only age is
statistically significant. Specifically, with an increase in age
there is a decrease in the likelihood of being asked about
citizenship status by law enforcement.
Model 3, the population-average model with robust
standard errors, contains valuable information concerning
neighborhood effects (third column of Table2). The devi-
ance statistic for model 3 indicates that considering both
Level 2 and Level 1 parameters further improve the model fit
from the unconditional model (model 1 of Table2). As the
percentage of residents in poverty increases in the neighbor-
hood there is a corresponding increase in the odds of being
questioned about citizenship status, but those results are not
statistically significant (Hypothesis 3). In regard to the influ-
ence of residing in Latina/o immigrant neighborhoods on the
odds of being profiled for citizen our study revealed some
interesting results. There is no support for Hypotheses 4 and
5, stating that residents of neighborhoods with low or mid-
level of Latina/o immigrant concentrations are less likely
to be questioned about citizenship than those who reside
in neighborhoods with a high concentration of Latina/o
immigrants. Yet, living in a neighborhood that is considered
mid-ranged in terms of level of immigrants, as opposed to
neighborhoods with a high degree of Latina/o immigrants,
increases the odds of being asked about citizenship status
by law enforcement by 56%. Therefore, neighborhood-level
predictors of immigration do extend our understanding of
who is asked about citizenship status. The effect of gen-
erational status exerts a positive and significant effect on
the odds of law enforcement asking about citizenship status,
even after introducing individual- and neighborhood-level
variables.
Discussion andConclusion
Based on primary research on 563 individuals residing in
46 neighborhoods, this study used HGLM methods to inves-
tigate the likelihood of Latina/os being questioned about
citizenship status by law enforcement. In particular, we test
whether first- and second-generation Latina/os, in contrast to
those who are third- and later-generations, are more likely to
be questioned about their citizenship status by law enforce-
ment. This research question is important to consider given
that law enforcement officers are increasingly required to
make distinctions between citizens and non-citizens (Coutin
2011). Furthermore, we also examine neighborhood effects
on who is asked about citizenship status, thus investigating
the structural impact on law enforcement questioning about
citizenship status. In particular, we predicted that those who
reside in poverty neighborhoods and in neighborhoods high
in Latina/o immigrant concentration are more likely to be
asked by law enforcement about their citizenship status. The
answer to these questions begin to address patterns in “citi-
zenship profiling” or who law enforcement stops and ques-
tions on suspicion of being undocumented.
So, who does law enforcement question about their citi-
zenship status? Building upon qualitative research, our study
found statistical evidence that first- and second-generation
individuals are more likely to be questioned by law enforce-
ment about their citizenship status than Latina/os of third-
generation and later, regardless of sex and age, and the
neighborhood characteristics of poverty and immigration
concentration. Therefore, immigrants and their children are
more likely to be profiled for citizenship. Particularly telling
are second-generation Latina/os who are disproportionately
questioned about citizenship by law enforcement despite
being US citizens.
The implications of our finding for the first- and second-
generations are vast. First, both first- and second-generation
Latina/os are subjected to constricted freedom and more
prone to criminalization in contrast to third- and later-gen-
erations, since they are subjected to heightened levels of
301Race and Social Problems (2018) 10:293–305
1 3
surveillance. Second, in regard to policing, questions arise
about the implications of citizenship profiling for commu-
nity policing that involves the protection and collaboration
of all residents, including immigrants. Third, being that first
generation are disproportionately questioned about their
citizenship status and that among them are undocumented
migrants it is important to evaluate policing procedures on
what type of identification is accepted and considered given
the consequences of arrest that can lead to deportation. This
is especially a concern given that most undocumented immi-
grants are deported for minor offenses including traffic viola-
tions (see Golash-Boza 2012).
The effects of neighborhood-level factors of poverty and
level of Latina/o immigration on being questioned about
citizenship status were also accessed. Residential poverty
does not significantly impact whether individuals are ques-
tioned by law enforcement about their citizenship status.
To assess the structural impact of residing in immigration
neighborhoods we used an array of indicators based on our
survey and American Community Survey data on trans-
nationalism (survey), acculturation to México (survey),
percent foreign-born, percent foreign-born not naturalized
(ACS), percent Spanish only at home (ACS), and percent
Mexican-origin (ACS) (Table3 in Appendix). We found
that those who reside in neighborhoods with mid-levels
of Latina/o immigration are more likely to be questioned
by law enforcement about their citizenship status than
those living in neighborhoods with high concentrations
of Latina/o immigrants. This advances our understanding
of policing and residing in Latina/o immigrant neighbor-
hoods. We found that the impact of residing in immigrant
neighborhoods is not linear when it comes to policing.
Residents in neighborhoods with high concentrations of
Latina/o immigrants are less likely to be questioned about
citizenship status than those who reside in neighborhoods
that exhibited less Latina/o immigrant features (mid-level
of Latina/o immigrant neighborhoods). We have a few
insights on this interesting outcome. Findings from the
literature suggest that the foreign-born tend to live in eth-
nic and immigrant enclaves (e.g., Portes and Bach 1985;
Portes and Stepick 1985) and that Latina/o immigrants’
fear of police may contribute to their isolation (Theodore
and Habans 2016). Latina/o immigrants also may restrict
their driving to avoid interactions with law enforcement
(see Armenta 2016a; Provine etal. 2016). Additionally,
immigrant isolation is connected to fear of restrictive
immigration policing including deportation (Menjívar and
Bejarano 2004; McDowell and Wonders 2009), which may
restrict their mobility outside of their communities and
thus also limit interaction with the police. Building upon
this knowledge, we suspect that those who live in neigh-
borhoods with mid-levels of Latina/o immigrants may be
more integrated into the city and more mobile than those
who live in high Latina/o immigrant whose mobility may
be more restricted to these segregated neighborhoods. As
such, there may be fewer opportunities for interactions
with law enforcement. Another possibility is that there is
less policing in Latina/o immigrant concentrated neighbor-
hoods given the Latina/o Crime Paradox where foreign-
born are less likely to commit crimes than native-born
(Burchfield and Silver 2013; Hagan and Palloni 1999;
Kubrin and Ishizawa 2012; Martínez 2015; Martínez
and Stowell 2012; Martínez and Valenzuela 2006; Samp-
son 2008; Sampson and Bean 2006; Stowell etal. 2009)
and that immigration revitalizes neighborhoods which
decreases crime (Martínez 2015; Martínez and Valen-
zuela 2006). Unfortunately, data limitations on where law
enforcement questioned individuals did not allow for us to
further explore these possibilities but we see these follow-
up questions as important for future research. Additionally,
there was no difference between how residents in low and
high immigrant neighborhoods were questioned by law
enforcement about citizenship status. While we outlined
above our rationale for why residents in high immigrant
neighborhoods are less likely to be questioned about citi-
zenship, we suspect that residents of neighborhoods low in
immigrant characteristics are less likely to be asked about
citizenship status because they have fewer immigrants, less
transnational activity with Mexico, and less acculturation
to Mexico.
Limitations of this study suggest avenues for future
research. First, we did not information on where and the
types of activities that individuals were involved in when
they were questioned by law enforcement. Having this
information could strengthen our analysis of neighborhood
effects. We do not know if individuals were questioned in
their neighborhoods or elsewhere, what we do know is that
where you live matters and there is neighborhood variation
in who gets targeted for questioning about their citizenship
status. Another limitation is not having accurate informa-
tion on undocumented status. While we did try to capture
undocumented status in our survey, we could not use this
variable because of missing data. Additionally, we do not
know the law enforcement agency that questioned individu-
als about their citizenship status. However, in El Paso it is
highly likely that the questioning is done by local (police
department or sheriff) being that border patrol concentrate
on enforcing the international boundary. The specificity of
the sample discourages generalization to other cities. Yet,
given the growth of the Latina/o population in cities across
the US (Saenz and Morales 2015) and the extension in the
policing of migration from the border to the interior (Leerkes
etal. 2013) we encourage future research in other locations
to examine variations and law enforcement disparities in the
degree that Latina/os are questioned about citizenship status.
Another limitation is the lack of data on Latina/os and skin
302 Race and Social Problems (2018) 10:293–305
1 3
color which is associated increase in the odds of police stops
and arrest. Lastly, a direct measure of language use may also
have impacted results.
In sum, our primary data was designed to conduct a mul-
tilevel analysis on who law enforcement questions about
citizenship along the US México border that allowed for
an investigation of individual and neighborhood effects. To
date, we have gained important insights on the implications
of policing immigration at the individual-level from qualita-
tive and policy analyses that revealed an array of negative
consequences including racial profiling and criminaliza-
tion of Latina/os (Heyman 2010; Longazel 2013; Provine
etal. 2016; Provine and Sanchez 2011; Sáenz etal. 2011),
removal and deportation (Armenta 2016a; Golash-Boza
2012; Motomura 2011; Provine etal. 2016) and compro-
mising community policing or cooperation with police
(Nygun and Gill 2016; Theodore and Habans 2016). At the
neighborhood-level there has been less attention given to the
policing of immigration. An important exception is Mary
Romero’s (2006) qualitative analysis of the Chandler Round-
ups in Arizona which found residents of Latina/o neighbor-
hoods being disproportionately targeted. We contribute to
this scholarship by using HGLM methods to establish that
first- and second-generation Latina/os and those who reside
in neighborhoods characterized by mid-level of Latina/o
immigration are disproportionately questioned about their
citizenship status by law enforcement.
Acknowledgements We want to thank Harmon Hosch for his involve-
ment in the data collection and Josiah Heyman for his thoughtful
comments.
Funding The data used in this study is based on the El Paso Neigh-
borhood Survey funded by the National Science Foundation (Award
1251897, PI-Theodore Curry, Co-PI Maria Cristina Morales, and Co-PI
Harmon Hosch) examining the immigration crime paradox.
Appendix
See Table3.
303Race and Social Problems (2018) 10:293–305
1 3
References
Adler, R. H. (2006). “But they claimed to be police, not la migra!” The
Interaction of Residency Status, Class, and Ethnicity in a (Post–
PATRIOT Act) New Jersey Neighborhood. American Behavioral
Scientist, 50(1), 48–69.
American Community Survey (ACS). (2015). Five-year 2008–2015
census tract data for El Paso County, Texas. https ://www.censu
s.gov/progr ams surve ys/acs/. Accessed 2 March 2 2015.
Aranda, E., & Vaquera, E. (2015). Racism, the immigration enforce-
ment regime, and the implications for racial inequality in the lives
of undocumented young adults. Sociology of Race and Ethnicity,
1(1), 88–104.
Table 3 Operationalization of immigrant enclave neighborhoods, El Paso neighborhood survey (EPNS) and American community survey (ACS)
Variable and question(s) Operationalization
Transnationalism (EPNS)
Because of a co-residence in Juarez
For work or to make money
To visit your children or other relatives
For entertainment, fun or recreation
For medical care (such as to visit a doctor,
dentist, or pharmacy)
To buy things or go shopping
To participate in civil activities or com-
munity events
Because I own a business in Mexico
For some other reason
Next we want to ask about questions about visits to Juarez, Mexico, that you or any current member
of your household might have done in the last 12 months. So, how often have you or any current
member of your household gone to or visited Juarez because of the following reasons in the last
12 months?
Write in number of times, such as 0 times, 15 times, 38 times, etc. 999 = not applicable because
hasn’t been to Juarez in past 12 months
Acculturation to México (EPNS)
I speak Spanish
I enjoy speaking Spanish
I associate with Méxicans and/or Méxican
Americans
I enjoy listening to Spanish language
music
I enjoy Spanish language TV
I enjoy Spanish language movies
I enjoy reading in Spanish (e.g., books,
magazines, and newspapers)
I write in Spanish (e.g., letters, notes,
emails, text messages)
My thinking is done in the Spanish
language
My contact with people in México has
been…
My father identifies or identified himself
as “Méxicano”
My mother identifies or identified himself
as “Méxicana”
My friends, while I was growing up, were
of Méxican-origin
My family cooks Méxican foods
My friends now are of Méxican-origin
I like to identify myself as a Méxican
American
I like to identify myself as a Méxican
ARSMA-II (revised)
Cuellar, Arnold and Maldonado (1995)
1 = not at all, 2 = very little or not very often, 3 = moderately, 4 = much or very often, 5 = extremely
often or almost always
Percent foreign-born ACS Where was this person born? Outside of the US (specify)
Census tract/neighborhood cluster mean
Percent foreign-born not naturalized ACS Is this person a citizen of the United States? Yes, US citizen by naturalization. Census tract/neigh-
borhood cluster mean
Percent Spanish-speaking only homes ACS Does this person speak a language other than English at home? Yes
What is this language? Spanish
How well does this person speak English? Not at all
Census tract/neighborhood cluster mean
Percent Mexican-origin ACS What is this person’s ancestry or ethnic origin? Mexican
Census tract/neighborhood cluster mean
304 Race and Social Problems (2018) 10:293–305
1 3
Armenta, A. (2012). From sheriff’s deputies to immigration officers:
Screening immigrant in a Tennessee jail. Law and Policy, 34(2),
191–210.
Armenta, A. (2016a). Between public service and social control: Polic-
ing dilemmas in the era of immigration enforcement. Social Prob-
lems, 63(1), 111–126.
Armenta, A. (2016b). Racializing crimmigration: Structural racism,
colorblindness, and the institutional production of immigrant
criminality. Sociology of Race and Ethnicity, 3(1), 82–95.
Arriaga, F. (2017). Relationships between the public and crimmigration
entities in North Carolina: A 287(g) program focus. Sociology of
Race and Ethnicity, 3(3), 417–431.
Briggs, C. (2014). The reasonableness of a race-based suspicion: The
Fourth Amendment and the costs and benefits of racial profiling
in immigration enforcement. Southern California Law Review,
88(2), 379–412.
Brunson, R. K., & Miller, J. (2006). Gender, race, and urban policing:
The experience of African American youths. Gender and Society,
20(4), 531–552.
Brunson, R. K. (2007). ‘Police don’t like black people’: African-Amer-
ican young men’s accumulated police experiences. Criminology
and Public Policy, 6(1), 71–101.
Bryk, A. S., & Stephen, W., & Raudenbush (1992). Hierarchical
linear models: applications and data analysis. London: Sage
Publication.
Burchfield, K. B., & Silver, E. (2013). Collective efficacy and crime
in Los Angeles neighborhoods: Implications for the Latina/o
paradox collective efficacy and crime in Los Angeles neigh-
borhoods: Implications for the Latina/o paradox. Sociological
Inquiry, 83(1), 154–176.
Carter, S. J., & Lippard, C. (2015). Group position, threat, and immi-
gration: The role of interest groups and elite actors in setting the
“Lines of Discussion’. Sociology of Race and Ethnicity, 1(3),
394–408.
Chavez, L. R. (2008). The Latina/o threat: Constructing immigrants,
citizens, and the nation. Stanford: Stanford University Press.
Coutin, S. B. (2011). The rights of noncitizens in the United States.
Annual Review of Law and Social Science, 7, 289–308.
Cox, A. B., & Miles, T. J. (2015). Legitimacy and cooperation:
Will immigrants cooperate with local police who enforce fed-
eral immigration law? Coase-Sandor Working Paper Series in
Law and Economics No. 734, 2015. Retrieved July 3, 2018
from https ://chica gounb ound.uchic ago.edu/cgi/viewc onten
t.cgi?refer er=https ://schol ar.googl e.com/&https redir =1&artic
le=2438&conte xt=law_and_econo mics.
Cuellar, I., Arnold, B., & Maldonado, R. (1995). Acculturation rat-
ing scale for Mexican Americans-II: A revision of the original
ARSMA scale. Hispanic Journal of Behavioral Sciences, 17(3),
275–304.
De Genova, N., & Peutz, N. (2010). The deportation regime: sov-
ereignty, space, and the freedom of movement. Durham: Duke
University Press.
Desmond, M., & Valdez, N. (2013). Unpolicing the urban poor: Con-
sequences of third-party policing for inner-city women. Ameri-
can Sociological Review, 78(1), 117–141.
Escobar, E. J. (1999). Race, police, and the making of a politi-
cal Identity: Mexican Americans and the Los Angeles Police
Department, 1900–1945. Berkeley: University of California
Press.
García Hernández, C. C. (2015). PEP Vs. secure communities. CRIM-
MIGRATION: The intersectionality of criminal law and immi-
gration law. http://crimm igrat ion.com/2015/07/07/pep-vs-secur
e-commu nitie s/. Accessed 7 July 2015.
Golash-Boza, T. (2012). Due process denied: Detentions and deporta-
tions in the United States. New York: Routledge.
Golash-Boza, T. (2015). Deported: Immigrant policing, disposable
labor and global capitalism. New York: New York University
Press.
Hagan, J., & Palloni. A. (1999). Sociological criminology and the
mythology of Hispanic immigration and crime. Social Problems,
46(4), 617–632.
Heck, R. H., Thomas, S., & Tabata, L. (2013). Multilevel modeling of
categorical outcomes using IBM SPSS. Routledge.
Heyman, J. M. (2010). Human rights and social justice briefing 1:
Arizona’s immigration law S.B.1070. Society of Applied Anthro-
pology, 21(3), 1–9.
Higgins, G. E. (2008). Racial profiling. Journal of Ethnicity in Crimi-
nal Justice, 6(1), 1–2.
Hox, J. J. (2010). Multilevel analysis: Techniques and applications.
New York: Routledge.
Kao, G., & Tienda, M. (1995). Optimism and achievement: The educa-
tional performance of immigrant youth. Social Science Quarterly,
76(1), 1–19.
Koch, D. W., Lee, J., & Lee, K. (2016). Coloring the war on drugs:
Arrest disparities in black, brown, and white. Race and Social
Problems, 8(4), 313–325.
Kubrin, C. E., & Ishizawa, H. (2012). Why some immigrant neighbor-
hoods are safer than others: Divergent findings from Los Angeles
and Chicago. The Annals of the American Academy of Political
and Social Science, 641(1), 148–173.
Leerkes, A., Bachmeier, J. D., & Leach, M. A. (2013). When the bor-
der is ‘Everywhere’: State-level variation in migration control
and changing settlement patterns of the unauthorized immigrant
population in the United States. International Migration Review,
47(4), 910–943.
Longazel, J. G. (2013). Moral panic as racial degradation ceremony:
Racial stratification and the local-level backlash against Latina/o/a
immigrants. Punishment and Society, 15(1), 96–119.
Macías-Rojas, P. (2016). From deportation to prison: The politics of
immigration enforcement in post-civil rights America. New York:
New York University Press.
Martin, M. J., Carthy, B., Conger, R. D., Gibbons, F. X., Simons, R. L.,
Cutrona, C. E., & Brody, G. H. (2011). The enduring significance
of racism: Discrimination and delinquency among black Ameri-
can youth. Journal of Research on Adolescence, 2(3), 662–676.
Martínez, R. Jr. (2015). Latina/o homicide: Immigration, violence, and
community. New York: Routledge.
Martínez, R. Jr., & Stowell, J. I. (2012). Extending immigration and
crime studies national implications and local settings. The Annals
of the American Academy of Political and Social Science, 64(1),
174–191.
Martínez, R. Jr., & Valenzuela, A. Jr. (2006). Immigration and crime:
race, ethnicity, and violence. New York: NYU Press.
Martínez Jr, R. (2008). Latina/o crime and delinquency in the U.S. In
H. Rodriguez, R. Saenz & C. Menjivar (Eds.), Latinas/os in the
U.S.: Changing the face of America (pp. (114–126). New York:
Springer.
McDowell, M. G., & Wonders, N. A. (2009). Keeping migrants in
their place: Technologies of control and racialized public space
in Arizona. Social Justice, 36(2), 54–72.
Menjívar, C., & Bejarano, C. (2004). Latino immigrants’ perceptions
of crime and police authorities in the United States: A case study
from the Phoenix metropolitan area. Ethnic and Racial Studies,
27(1), 120–148.
Miller, J. G. (2011). Search and destroy: African American males in the
criminal justice system. New York: Cambridge University Press.
Mirandé, A. (1987). Gringo justice. Notre Dame: University of Notre
Dame Press.
Motomura, H. (2011). The discretion that matters: Federal immigration
enforcement, state and local arrests, and the civil-criminal line.
UCLA Law Review, 58, 1819.
305Race and Social Problems (2018) 10:293–305
1 3
Ngai, M. M. (2014). Impossible subjects: Illegal aliens and the making
of modern America. Princeton: Princeton University Press.
Nguyen, M. T., & Gill, H. (2016). Interior immigration enforcement:
The impacts of expanding local law enforcement authority. Urban
Studies, 53(2), 302–323.
Portes, A., & Bach, R. L. (1985). Latin journey: Cuban and Mexican
immigrants in the United States. Berkley: University of California
Press.
Portes, A., & Stepick, A. (1985). Unwelcome immigrants: The labor
market experiences of 1980 (Mariel) Cuban and Haitian refugees
in South Florida. American Sociological Review, 50(4), 493–514.
Provine, D. M., & Sanchez, G. (2011). Suspecting immigrants: Explor-
ing links between racialised anxieties and expanded police powers
in Arizona. Policing and Society, 21(4), 468–479.
Provine, D. M., Varsanyi, M., Lewis, P. G., & Decker, S. H. (2016).
Policing immigrants: Local law enforcement on the front lines.
Chicago: The University of Chicago Press.
Raudenbush, S. W., & Bryk, A. S. (2002). Hierarchical linear models:
Applications and data analysis methods. Thousand Oaks: Sage
Publications.
Raudenbush, S. W., Bryk, A. S., Cheong, Y. F., Congdon, R., & Toit,
Du, M (2011). Hierarchical linear and nonlinear modeling
(HLM7). Lincolnwood: Scientific Software International.
Rios, V. M. (2011). Punished: Policing the lives of black and Latina/o
boys. New York: New.
Romero, M. (2006). Racial profiling and immigration law enforcement:
Rounding up of usual suspects in the Latina/o community. Critical
Sociology, 32(2–3), 447–473.
Sáenz, R. (2010). Latina/os, whites, and the shifting demography of
Arizona. Population Reference Bureau. http://www.prb.org/Artic
les/2010/usari zonaL atina /os.aspx. Accessed 24 April 2014.
Sáenz, R., Menjivar, C., & Garcia, S. J. E. (2011). Arizona’s SB 1070:
Setting conditions for violations of human rights here and beyond.
In J. Blau & M. Frezzo (Eds.), Sociology and human rights: A Bill
of rights in the twenty-first century (pp.155–178). Newbury Park:
Pine Forge Press.
Sáenz, R., & Morales, M. C. 2015. Latinos in the United States: Diver-
sity and Change. Polity Press.
Sampson, R. J. (2008). Rethinking crime and immigration. Contexts,
7(1), 28–33.
Sampson, R. J., & Bean, L. (2006). Cultural mechanisms and killing
fields: A revised theory of community-level racial inequality. In
R. D. Peterson, L. Karvo & J. Hagan (Eds.), The many colors
of crime: Inequalities of race, ethnicity, and crime in America
(pp.8–36). New York: New York University Press.
Stowell, J. I., Messner, S. F., McGeever, K. F., & Raffalovich, L. E.
(2009). Immigration and the recent violent crime drop in the
United States: A pooled, cross-sectional time-series analysis of
metropolitan Areas. Criminology, 47(3), 889–928.
Stumpf, J. P. (2006). The crimmigration crisis: Immigrants, crime,
and sovereign power. American University Law Review, 56, 367.
Sweet, S., & Grace-Martin, K. (2012). Data analysis with SPSS: A first
course in applied statistics. Boston: Allyn and Bacon.
Theodore, N., & Habans, R. (2016). Policing immigrant communities:
Latino perceptions of police involvement in immigration enforce-
ment. Journal of Ethnic and Migration Studies, 42(6), 970–988.
Trump, D. J. (2017). Executive Order: Border Security and Immigra-
tion Enforcement Improvements. The White House, Office of
the Press Secretary. https ://www.white house .gov/the-press -offic
e/2017/01/25/execu tive-order -borde r-secur ity-and-immig ratio
n-enfor cemen t-impro vemen ts. Accessed May 1 2017.
Varsanyi, M. W. (2008). Immigration policing through the backdoor:
City ordinances, the right to the city and the exclusion of undocu-
mented day laborers. Urban Geography, 29(1), 29–52.
Vidales, G., Day, K. M., & Powe, M. (2009). Police and immigra-
tion enforcement: Impacts on Latina/o (a) residents perceptions
of police. Policing: an International Journal of Police Strategies
and Management, 32(4), 631–653.
Wong, T. K. (2012). 287 (g) and the politics of interior immigration
control in the United States: Explaining local cooperation with
federal immigration authorities. Journal of Ethnic and Migration
Studies, 38(5), 737–756.
... Unlike research on racial profiling, studies of the impact of citizenship profiling on police legitimacy are sparse. The concept of "citizenship profiling" refers to incidents where stops, searches, and questioning by law enforcement are motivated by a suspicion of the individual's lack of citizenship (Morales, Delgado, and Curry 2018). As such, citizenship profiling is used by law enforcement officers to identify immigrants whom they suspect are undocumented and thus in violation of federal immigration law. ...
... trust) and behavioural (i.e. cooperation) aspects of police legitimacy; (2) we collected 691 face-to-face interviews in 46 neighbourhoods and employ a multilevel statistical approach that allows for generalization and an assessment of both individual and contextual neighbourhood characteristics; and (3) most of the focus in this literature is on immigrants' perceptions of police, while here we study the policing of immigration that has implications for both the foreign-and native-born (Morales, Delgado, and Curry 2018;Provine et al. 2016). ...
... Additionally, when contact with police is viewed as procedurally just by immigrants, it led to more positive assessments of legitimacy (Murphy and Mazerolle 2018;Pryce, Johnson, and Maguire 2017). While we can gain some important insights from previous research on immigrant-police relations, the policing of immigration is a separate question that has implications for both foreign-and native-born (Morales, Delgado, and Curry 2018;Provine et al. 2016). ...
Article
Full-text available
The U.S. has witnessed increases in the local policing of federal immigration law. This practice raises concerns about citizenship profiling, where local law enforcement questions individuals on suspicion of being undocumented and, in turn, its influence on police legitimacy. We examine whether citizenship profiling influences two measures tapping into legitimacy, perceptions of trust in police effectiveness and cooperation with local police, among individuals and in Latina/o-immigrant neighbourhoods. Based on 691 surveys collected in 46 neighbourhoods in a county along the U.S.-Mexico border, multilevel analyses shows that citizenship profiling reduces both trust in policing effectiveness and cooperation with police among individuals. Moreover, while residents in Latina/o-immigrant neighbourhoods reported more cooperation with police, being citizenship profiled and residing in these neighbourhoods significantly reduced cooperation with law enforcement. In conclusion, there is strong evidence that the local policing of immigration compromises police legitimacy, raising concerns about perceptions of procedural justice.
Article
Full-text available
We develop and test an extension of differential coercion and social support (DCSS) theory focused on Latinx individuals that employs measures of coercion and social support, both at the individual and environmental levels, that may be especially salient to this population. Data come from Latinx adults (n=863) randomly sampled from El Paso County, Texas and were analyzed using logistic and OLS regression analyses and employed the PROCESS Macro for assessing mediating relationships. Findings show moderate support for hypotheses. Specifically, a structural measure of social support (neighborhood social support) was inversely associated with family violence and mediated relationships between three separate measures of coercion and family violence, supporting predictions. However, two cultural measures of social support (enculturation to Mexico and familismo) did not demonstrate such relationships, failing to support predictions. Additional findings show that two of the three measures of coercion (neighborhood crime and disorder and police harassment) were inversely associated with family violence as predicted, but that a third measure, acculturation to the U.S., was not. Implications for DCSS theory and criminal offending for Latinx are discussed.
Article
Full-text available
The consequences of racial microaggressions are most often discussed at an interpersonal level. In this review, we contend that microaggressions play an important role in maintaining systems of racial oppression beyond the interpersonal context. Specifically, we illustrate how microaggressions establish White superiority in the U.S. by othering people of color (e.g., treating people of color as if they are not true citizens) and communicating that they are inferior (e.g., environmental exclusions and attacks, treating people of color as second-class citizens). We also present evidence that microaggressions play a role in protecting and reinforcing systemic racism. By obscuring systemic racism (e.g., false colorblindness, denial of individual racism) and promoting ideas that maintain existing systemic inequalities (e.g., the myth of meritocracy, reverse racism hostility), microaggressions provide cover and support for established systems of oppression. Overall, we find considerable evidence—from both empirical studies and real-world examples—that microaggressions contribute to the maintenance of systems of racial oppressions in the U.S. We conclude with a discussion of how we might begin to challenge this cycle by increasing awareness of systemic racism and the microaggressions that aid in its perpetuation.
Book
From Deportation to Prison traces the punitive turn in immigration and border policy to the Department of Homeland Security’s Criminal Alien Program (CAP), originally designed to purge noncitizens from jails and prisons and ushering in enforcement priorities that process immigrants according to criminal history and risk. Macías-Rojas argues that new enforcement priorities under the Criminal Alien Program, rooted in the post–civil rights era of mass incarceration and prison overcrowding, fundamentally transformed detention and deportation in ways that merged the immigration and criminal justice systems. Deportation and immigrant detention, then, are no longer merely vehicles to purge noncitizens from jails and prisons, as was CAPS’s original mission; they are now the chief mechanisms driving federal criminal prosecution and imprisonment for immigration offenses. From a political analysis of policymaking at the congressional level, Macías-Rojas turns to a street-level ethnographic account of how new enforcement priorities take hold on the Arizona-Mexico border, capturing the ways in which border agents, local law enforcement, activists, border residents, and migrants themselves contend with criminal enforcement priorities that distinguish between rights-bearing “victims” and rightsless “criminals.” Combining history and ethnography, this book shows how, when implemented on the U.S.-Mexico border, the Department of Homeland Security’s criminal enforcement priorities have created an enforcement context that recognizes rights for some undocumented migrants deemed “worthy” of state protection, while aggressively punishing and criminally branding others. In this post–civil rights enforcement context, criminalization goes hand in hand with “humanitarianism” centered on “victims’ rights.”
Article
This study examines the determinants of earnings among two groups of recent immigrants—Cubans and Mexicans—interviewed at the moment of arrival in the United States and reinterviewed three years later. The specific goal is to examine the applicability to these results of causal variables suggested by recent alternative theoretical perspectives on income attainment.
Book
The United States currently is deporting more people than ever before: 4 million people have been deported since 1997 -twice as many as all people deported prior to 1996. There is a disturbing pattern in the population deported: 97% of deportees are sent to Latin America or the Caribbean, and 88% are men, many of whom were originally detained through the U.S. criminal justice system. Weaving together hard-hitting critique and moving first-person testimonials, Deported tells the intimate stories of people caught in an immigration law enforcement dragnet that serves the aims of global capitalism. Tanya Golash-Boza uses the stories of 147 of these deportees to explore the racialized and gendered dimensions of mass deportation in the United States, showing how this crisis is embedded in economic restructuring, neoliberal reforms, and the disproportionate criminalization of black and Latino men. In the United States, outsourcing creates service sector jobs and more of a need for the unskilled jobs that attract immigrants looking for new opportunities, but it also leads to deindustrialization, decline in urban communities, and, consequently, heavy policing. Many immigrants are exposed to the same racial profiling and policing as native-born blacks and Latinos. Unlike the native-born, though, when immigrants enter the criminal justice system, deportation is often their only way out. Ultimately, Golash-Boza argues that deportation has become a state strategy of social control, both in the United States and in the many countries that receive deportees.
Article
How does local law enforcement, with the aid of city and county governments, respond to racialized immigrant threat through policy implementation, namely, through adoption of intergovernmental agreements? More specifically, how is this response tailored for Latino immigrant communities, particularly in new destination communities? Across the country, scholars, activists, and politicians note the increasing use of local law enforcement to implement federal immigration enforcement measures through intergovernmental agreements, emphasizing the disproportionate impact on the Latino, more specifically the foreign-born Mexican, population. One such intergovernmental partnership is the 287(g) agreement between local law enforcement and Immigration and Customs Enforcement. Yet little is known about the process through which counties adopt, implement, and maintain such intergovernmental partnerships and the state actors that make it all possible.
Article
Solving crimes often requires community cooperation. Cooperation is thought by many scholars to depend critically on whether community members believe that law enforcement institutions are legitimate and trustworthy. Yet establishing an empirical link between legitimacy and cooperation has proven elusive, with most studies relying on surveys or lab experiments of people’s beliefs and attitudes, rather than on their behavior in the real world. This Article aims to overcome these shortcomings, capitalizing on a unique natural policy experiment to directly address a fundamental question about legitimacy, cooperation, and law enforcement success: do de-legitimating policy interventions actually undermine community cooperation with the police? The policy experiment is a massive federal immigration enforcement program called Secure Communities. Secure Communities was widely criticized for undermining the legitimacy of local police in the eyes of immigrants, and it was rolled out nationwide over a four-year period in a way that approximates a natural experiment. Using the rate at which police solve crimes as a proxy for community cooperation, we find no evidence that the program reduced community cooperation — despite its massive size and broad scope. The results call into question optimistic claims that discrete policy interventions can, in the short run, meaningfully affect community perceptions of law enforcement legitimacy in ways that shape community cooperation with police.
Article
The current immigration enforcement regime embodies a colorblind racial project of the state rooted in the racial structure of society and resulting in racism toward immigrants. Approaching racism from structural and social process perspectives, we seek to understand the social consequences of enforcement practices in the lives of undocumented immigrant young adults who moved to the United States as minors. Findings indicate that although legal discourse regarding immigration enforcement theoretically purports colorblindness, racial practices such as profiling subject immigrants to arrest, detention, and deportation and, in effect, criminalize them. Further, enforcement practices produce distress, vulnerability, and anxiety in the lives of young immigrants and their families, often resulting in legitimate fears of detention and deportation since enforcement measures disproportionately affect Latinos and other racialized immigrant groups in U.S. society. We conclude that policies and programs that exclude, segregate, detain, and physically remove immigrants from the country reproduce racial inequalities in other areas of social life through spillover effects that result in dire consequences for these immigrants and their kin. We argue that immigrant enforcement practices reflect the nation’s racial policy of our times.