April 2018

Volume: 15, No: 2, pp. 147 – 166

ISSN: 1741-8984
e-ISSN: 1741-8992
www.migrationletters.com

Article history: Received 29 June 2017; Accepted 6 September 2017

Bianca E. Bersani
Adam D. Fine
Alex R. Piquero
Laurence Steinberg
Investigating the Offending Histories of Paul J. Frick
Undocumented Immigrants Elizabeth Cauffman

Abstract
This study investigates the association between undocumented immigration and crime among
youthful offenders. Using official record and self-reported offending measures collected across
seven-waves of data from the longitudinal Crossroads Study, the prevalence and variety of
offending are compared for undocumented immigrant, documented immigrant, and US-born
groups during the transition into young adulthood. Results suggest that, as compared to
documented immigrants and US-born peers, undocumented immigrants report engaging in less
crime prior to and following their first arrest. Conversely, official records reflect a marginally
higher level of re-arrest among undocumented immigrants, particularly in the months
immediately following the first arrest. This divergence in findings warrants focused consideration
to disentangle whether the difference is due to differential involvement in crime, differential
treatment in the justice system, or a combination of factors. Additional research is needed to test
whether the results found in this study generalize to other immigrant groups and contexts.
Keywords: undocumented immigrants; immigrants; offending; arrest; immigration-crime nexus.

Introduction
A quarter of the estimated 43.6 million immigrants who currently reside in the
United States are unauthorised, and of the undocumented population, over half
are Mexican (Passel & Cohn, 2016). Concerns about the immigrant population,
particularly those who are unauthorised, vary widely from job competition and


Bianca E. Bersani, PhD, Department of Sociology, University of Massachusetts Boston, United States. E-
mail: Bianca.bersani@umb.edu
Adam D. Fine, MA, Department of Psychology & Social Behavior, University of California, Irvine, United
States. E-mail: finea@uci.edu
Alex R. Piquero, PhD, Program in Criminology, University of Texas at Dallas, United States. E-mail:
apiquero@utdallas.edu
Laurence Steinberg, PhD, Department of Psychology, Temple University, United States and King
Abdulaziz University, KSA. E-mail: lds@temple.edu
Paul J. Frick, PhD, Department of Psychology, Louisiana State University, United States and Learning
Science Institute of Australia, Australian Catholic University, Australia. E-mail: pfrick@lsu.edu
Elizabeth Cauffman, PhD, Department of Psychology & Social Behavior, University of California, Irvine,
United States. E-mail: cauffman@uci.edu.
Acknowledgement: This research was supported by the John D. and Catherine T. MacArthur Foundation
and the Office of Juvenile Justice and Delinquency Prevention, and the County of Orange.
Copyright @ 2018 MIGRATION LETTERS | Transnational Press London
148 Offending Histories of Undocumented Immigrants
healthcare costs to criminal proclivities (Chavez 2008; Wang 2012). The anxiety
surrounding the ‘criminal immigrant’ resonates deeply despite a long history of
empirical evidence showing that the foreign-born are less involved in crime
than their US-born peers (Lee & Martinez 2009). In a clear contrast to extant
research, concerns voiced by immigration opponents to justify increasingly
punitive measures and exclusionary policies underscore the crime that flows
across the border owing to waves of immigration (Ewing, Martínez & Rumbaut
2015).
The persistence of the perception that immigrants are crime prone may be due,
in part, to the lack of studies distinguishing undocumented immigrants from
their documented immigrant peers. Consequently, the low levels of crime
observed among the foreign-born may mask the crime committed by a
subgroup of immigrants with unauthorized residence in the US.
Undocumented immigrants comprise a heterogeneous group and include those
who entered the country illegally and those whose legal residency expired or
was revoked. A nascent body of scholarship accounts for variation in immigrant
status1 and is limited to correctional population samples (Hagan & Palloni 1999;
Hickman & Suttorp 2008) or a focus on drug use (Katz et al., 2011). As a result,
important questions remain concerning the prevalence and pattern of offending
among undocumented immigrants, which hold import for current and
emergent policies aimed at immigration reform and the enforcement of
immigration laws.
This research fills an important void in the literature by examining the offending
patterns of immigrant youth distinguished by their legal residency in the US.
Using data from the Crossroads Study, a longitudinal investigation of male,
first-time juvenile arrestees followed for three years, we examine the prevalence
and variety of criminal behavior using official and self-reports of offending. The
longitudinal nature of the data allow for the study of offending patterns during
the precarious developmental period spanning adolescence into young
adulthood (Monahan et al., 2015), when criminal involvement is at its peak
(Hirschi & Gottfredson 1983; Piquero et al. 2003). This may be an especially
risky period for undocumented immigrants who, during the transition to
adulthood, confront abruptly an illegal identity (Gonzales 2016). Our focus is
on Latino immigrants, particularly Mexican immigrants, to the United States
for two reasons: Latinos characterize the vast majority of immigrants to the US
(Passel & Cohn 2016), and in terms of policy, they represent a group of

1 Variation in immigrant legal status has been measured and categorized in different ways. It is important to
note that there is an increasing fluidity in the categories used to capture variation in status. Take, for
instance, the criteria for those designated as removable or deportable that includes those who are
undocumented as well as permanent legal residents, and in some cases, naturalized citizens, and potentially
those with provisional executive residency authority. In this research, we distinguish undocumented from
documented immigrants.
www.migrationletters.com
 Bersani et al. 149
immigrants who have long been branded as being at particular risk for
detrimental outcomes, including crime, due to their persistent deficits and
disadvantages (Hagan & Palloni 1999; Portes & Rumbaut 2006; Telles & Ortiz
2008).

Immigration and Crime

Anxiety regarding the potential criminal proclivities of immigrants is
longstanding and dates back, at least, to the turn of the 20th century (see Simes
& Waters 2014). The extensive research on the immigrant-crime nexus yields
little empirical evidence that immigrants are more crime-prone than US-born
Americans (Lee & Martinez 2009). The finding that the foreign-born are
relatively less criminal holds in studies examining self-reported crime and risky
behavior in general population studies (Bersani 2014; Bui 2009; Butcher & Piehl
1998; Harris 1999; Vaughn et al. 2014), self-reported and officially recorded
crime in offender based samples (Bersani et al. 2014; Jennings et al. 2013), and
rates of recidivism and incarceration (Ibanez et al. 2016; Rumbaut et al. 2006).
The lower levels of offending are not attributable to a differential tendency for
the foreign-born to under-report their involvement in crime (see Bersani and
Piquero 2017). The finding that immigrants have relatively low levels of
criminal involvement, despite exposure to traditional criminogenic risks, is
referred to as the immigrant paradox.

Legal versus illegal immigration

Illegal immigration and crime are inextricably linked; illegal entry is itself
classified as a federal offense. Undocumented residency includes not only those
lacking authorized entry to the US, but also individuals whose authorized entry
and residency in the US has been revoked or expired. Historically, violations of
immigration law were handled as administrative offenses, and the focus was
directed at detaining and deporting individuals deemed serious or violent
offenders (Meissner et al. 2013). Recent years have witnessed the growth of
‘crimmigration,’ or the criminalization of immigration (Chacón 2012; Stumpf
2006), where immigration law is increasingly punitive and broad in scope
(Ewing et al. 2015). Notably, while anxieties highlight “illegal” immigrants,
concerns levied at the immigrant-crime nexus are more widely shared and felt
by all immigrants (see Perea 1996; Rosaldo 1997).
The relation between illegal immigration and traditional criminal offending is a
contested one. On the one hand, those who enter the country without
authorization may comprise a select group of individuals whose criminal
involvement prohibits them from legal entry or whose entrance to the country
is for the purpose of engaging in crime. Similarly, if the revocation of one’s legal
status is the result of involvement with the criminal justice system, this may

Copyright @ 2018 MIGRATION LETTERS | Transnational Press London

150 Offending Histories of Undocumented Immigrants
signal a heightened criminal propensity.2 Criminal behavior is a key determinant
of revocation; an estimated seven and a half percent of undocumented
immigrants have a criminal conviction, three-percent of whom have committed
a felony offense (Rosenblum 2015). On the other hand, undocumented
immigrants may be less criminally involved than their peers as they actively
work to not draw attention to themselves and risk deportation (Gottfredson
2004). Belief in the obligation to obey the law among immigrants exceeds that
of US–born peers (Correia 2010; Kirk et al. 2012; Piquero et al. 2014; Rengifo
& Fratello 2015) even among detained populations (Ryo 2017). In this vein,
those who migrate to the US illegally are doing so, in large part, for economic
opportunity, family reunification, and to flee violence (Chavez 2012; Massey et
al. 2014).
Negligible empirical attention has been aimed at exploring differences in
offending for immigrants distinguished by (un)documented status.
Consequently, questions remain as to whether the lack of a relation between
immigration and crime holds among the undocumented, and whether the
undocumented population poses a unique criminal risk. The lack of research
into the undocumented-immigrant-crime nexus is due, in large part, to the rarity
in which datasets include information on offending and data on immigrants’
documentation status. Official data sources that record details such as
citizenship, legal and/or documented status are impacted by disparities in the
criminal justice system that, all else equal, can result in disproportionately more
severe outcomes for immigrants. For instance, immigrants are subject to
heightened policing (Davies & Fagan 2012), increased rates of pre-trial
detention (Reid et al. 2005), and use of deportation in lieu of criminal
proceedings (Hagan & Palloni 1999). Notably, Orrick and colleagues find that
nativity status (Orrick & Piquero 2015) did not influence sentencing disparities
and that deportation status (Orrick, Compofelice & Piquero 2016) was
associated with the receipt of shorter sentences. In short, it is unclear the extent
to which disparities in criminal justice system contact may affect the validity of
arrest and correctional data used to test for differences across subsamples.
The few empirical studies that distinguish immigrant status generally fail to find
systematic evidence that undocumented immigrants pose a unique crime threat.
Specifically, Katz and colleagues (2011) found that illegal immigrants (i.e.,
overstayed visa and undocumented entry) were less likely to use marijuana,
methamphetamine, and illicit drugs compared to legal US citizens, but were
more likely to report use of powder cocaine. In a series of studies, Hickman
and colleagues compared recidivism rates among a sample of non-deportable

2 Unfortunately, though “overstayers” or those who arrive to the US legally and let their visa expire are the
focus of heightened attention recently (see Passel & Cohn 2016), too little is known about those with
expired visas to speculate about whether these individuals would have heightened or lessened propensities
to crime.
www.migrationletters.com
 Bersani et al. 151
(i.e., legal immigrants and naturalized U.S. citizens) and deportable immigrants
(i.e., illegal entry, overstayed visa, or visa revocation) from the Los Angeles
County jail. Hickman and Suttorp (2008) found that deportation status was not
related to one-year reoffending risk, and overall re-arrest rates for the sample
were relatively low (38%) compared to recidivism rates within the general US
prison population (Durose et al. 2014). More recently, Wong and colleagues
(2015) and Hickman and colleagues (2016) found that authorization status did
not predict re-arrest risk in a 9-year period following release; however a
subsample of the unauthorized immigrant group (i.e. illegal entry, overstayed
visa, or visa revocation), namely previously-deported immigrants, were at
greater risk of recidivism than their peers.

Current Research
This study complements and advances limited existing research in a number of
notable ways. First, we examine the offending behavior of undocumented and
documented immigrants and compare their offending to similar justice-
involved US-born peers. Second, drawing on a sample of youth selected into
the study at their first arrest, with follow-up interviews occurring every 6
months for three years, the sample is comprised at baseline of individuals
without prior formal justice system involvement that may systematically affect
their offending. Moreover, the longitudinal nature of the data and the study’s
emphasis on the transition to adulthood allows us to examine changes in
offending during peak years of risk for antisocial behavior. If differences in
offending as a function of documentation status exist, it is reasonable to expect
that these data would identify them. Finally, we examine both self-reported
offending and official reports of (re)arrest to assess incidence (i.e., prevalence)
and severity (i.e., variety/frequency) of offending.

Method

Participants
The sample comes from the Crossroads Study3 a prospective study of male,
first-time juvenile offenders in three locales (Pennsylvania, Louisiana, and
southern California). This study uses data from the southern California site (N
= 532) because the vast majority of immigrant youth in the sample (81%), and
all of the undocumented immigrant youth, resided in that state. Crossroads is a
longitudinal study designed to examine the effects of justice system
involvement on first-time male juvenile offenders. Adolescent males who were
first-time offenders (i.e., no prior arrests), were between the ages of 13 and 17
(Mage = 15.46, SD = 1.21), and had committed an eligible offense (e.g.,

3 For more information about the Crossroads Study, visit http://sites.uci.edu/crossroadsinfo/

Copyright @ 2018 MIGRATION LETTERS | Transnational Press London
152 Offending Histories of Undocumented Immigrants
vandalism, theft, assault) were eligible to participate in the Crossroads Study.
The charges selected for study inclusion were moderate offenses that had
similar probabilities of being diverted and formally processed in court, based
on our review of historical records over the 5-year period immediately prior to
the commencement of the study. We identified eligible participants by
monitoring all new arrest reports in one county in southern California.
Approximately 80% of the eligible participants participated. The most frequent
charge was vandalism (47.10%) and 18.98% of the sample had been arrested
for a person offense. Consistent with demographic characteristics of youth
involved in the juvenile justice system in southern California, the analytic
sample was: 80.71% Latino, 17.82% White, and 1.5% other. The
preponderance of youth identified as being of Mexican ancestry: 95.7% of
documented first-generation, 83.8% of undocumented first-generation, 85.7%
of second-generation, and 85.7% of native-born Latinos (see Appendix for sub-
sample descriptive statistics).
Youth with complete immigration data (N = 477) did not differ on any key
study variable from the 532 youth enrolled in the CA site. Of the 477 youth
who had complete immigration data, approximately 95.60% (N = 456) had
complete data on all official record and self-report study variables for all waves
of the study. Results of Little’s missing completely at random test indicated that
data were missing completely at random (Li 2013; Little 1988; Little 1992).

Procedures
The Institutional Review Board (IRB) at University of California, Irvine
approved the study procedures. Signed parental consent and youth assent were
obtained for all participants before interviews were conducted. Participants
were informed of the nature of the study and were told that there was no
penalty for not participating. Youth completed an interview within 6 weeks
after the disposition hearing for their first arrest, as well as follow-up interviews
approximately 6, 12, 18, 24, 30, and 36 months after their initial interview. Face-
to-face interviews with the youth ranged from 2–3 hours and were documented
using a secure computer-administered program. A Privacy Certificate issued by
the Department of Justice protects participants’ privacy by exempting their
identity and responses from subpoenas, court orders, or other types of
involuntary disclosures. Participants were given a detailed explanation of the
Privacy Certificate before beginning the interview and were reminded again
before sensitive questions, such as those about offending and legal residency,
were asked.

Measures
Immigration Status. Following convention in the literature (Pew Research
Center 2013; Rumbaut & Ewing 2007), immigrant status was determined using

www.migrationletters.com
 Bersani et al. 153
information on the country of birth of the youth and his parents as reported by
the youth. Respondents born outside the US self-identified as legal resident
(yes/no). First-generation immigrants include youth who have foreign-born
parents and who were themselves born outside the US (N = 60; 12.58% of the
analytic sample). Of the first-generation immigrants, 37 reported that they were
not “a legal resident of the United States” (7.76%) and 23 as documented or
holding legal residency (4.82%). Second-generation immigrants include youth
born in the US with at least one foreign-born parent (N = 266; 55.77%). Native-
born White youth were born in the US to US-born parents and self-identified
as White (N = 74; 15.51%). Native-born Latino youth were born in the US to
US-born parents and self-identified as Hispanic/Latino (N = 77; 16.14%).
Self-reported Offending. Involvement in crime during each recall period was
assessed using the Self-Report of Offending (SRO; Huizinga et al. 1991).
Participants reported their history (lifetime and preceding six-months) of
involvement in 24 criminal acts (ranging in severity from selling drugs to
homicide). Two offending indices were created. First, a dichotomous indicator
of offending (1 = yes) was created to capture prevalence. Second, responses
were summed to indicate the number of different types of crimes in which the
youth engaged. Variety scores are widely used in criminological research
because they are highly correlated with measures of seriousness and frequency
of antisocial behavior (see Monahan & Piquero 2009), are less subject to recall
bias than are frequency scores (see Hindelang et al. 1981; Osgood et al. 2002),
and account for heterogeneity in crime types (Sweeten 2012).
Official Arrest. The southern California probation department and juvenile
court provided access to official record arrest data. The official record data
indicate both whether (prevalence) and the number of times (frequency) the
youth had been re-arrested at each wave during the 36-months following his
first arrest.

Analytic Strategy
We describe longitudinal patterns of the prevalence and variety of offending
for five groups of youth: undocumented immigrant, documented immigrant,
second-generation, US-born White, and US-born Latino. We provide the US-
born youth as reference points and to situate this work in the empirical literature
base. We first compare the prevalence and variety of offending leading up to
and including study participants’ adjudication arrest. We then examine patterns
of offending using self-reported offending and officially recorded arrests for
three years following baseline. Given the small sample sizes, we report the
results of post hoc power analyses (Champely 2013; Faul et al., 2009), with
power (1 - β) set at 0.80 and α = .05, specifically to aid future research (see
Hoenig & Heisey 2001; Gilbert & Prion 2016; Goodman & Berlin 1994). We
also include 95% confidence intervals for all effects.
Copyright @ 2018 MIGRATION LETTERS | Transnational Press London
154 Offending Histories of Undocumented Immigrants
Results

Offending Prior to First Arrest

At their baseline interview, youth retrospectively self-reported both their total
history of involvement in crime prior to their first arrest as well as their
involvement in crime during the six months immediately prior to their first
arrest. The pattern of results is similar across both measures of prior offending
(i.e., ever versus 6 months). For instance, the results indicated that prior to
baseline, self-reported offending was greatest for US-born youth and lowest for
the foreign-born. Undocumented immigrants reported the lowest rates of
offending compared to their peers (see Table 1). Self-reported offending among
undocumented immigrant youth was significantly less than second-generation
youth (incidence rate ratio (IRR)4 = .71, SE = .11, p = .018, 95% CI = .53, .94;
1-β = .21) and US-born White youth (IRR = .75, SE = .13, p = .084, 95% CI =
.54, 1.04; 1-β = .17).5 Statistically significant differences in self-reported
offending were not observed comparing undocumented immigrants with
documented immigrants (IRR = .92, SE = .21, p = .707, 95% CI = .59, 1.43; 1-
β = .12) and US-born Latino youth (IRR = .82, SE = .14, p = .234, 95% CI =
.59, 1.14; 1-β = .17).

Arresting Offense
We look next at the distribution of crimes for the arresting offense across
subgroups. The majority of arrest offenses for all groups are for property crimes
(see Table 1). Person-related crimes comprise roughly a quarter of arresting
offenses for documented immigrants (26.09%) and US-born Latino youth
(23.38%), with the lowest rates observed for undocumented immigrants
(18.92%), US-born Whites (17.57%), and second-generation youth (15.41%).
Arresting offenses for drug related crimes are rare in this sample, comprising
less than 5% for US-born Whites (4.1%), US-born Latinos (1.3%) and second-
generation youth (1.5%), and absent from documented and undocumented
immigrant groups.

Recidivism: Self-Reported Prevalence and Variety of Offending

Cumulative prevalence rates for involvement in any crime across the three years
are reported in Table 1. Following the first arrest, 65% of undocumented

4 Incidence rate ratios (IRRs) are the ratio of the incidence rate in one group divided by the incidence rate
of the comparison group. An IRR of 1 indicates that the outcome rate is equivalent between the groups, an
IRR larger than 1 would indicate that the rate of reoffending among the target group is larger than the
comparison group, and an IRR between 0 and 1 indicates that the rate of reoffending among that target
group is smaller than the comparison group.
5 The power analyses indicate that these effects were found even though the power to detect that effect was

low for each analysis (.21 and .17, respectively). Because of the low power, it is critical that studies replicate
these findings.
www.migrationletters.com
 Bersani et al. 155
immigrant youth self-report engaging in at least one crime, compared to 70%
of documented immigrants, 77% of second-generation and US-born White,
and 78% of US-born Latino youth in the sample. The cumulative re-arrest
prevalence across the three-year period is statistically similar. Cumulatively
across the three-year period, undocumented immigrant youth commit a smaller
variety of crimes than second-generation (IRR = .58, SE = .11, p = .003, 95%
CI = .41, .83; 1-β = .21), US-born White (IRR = .69, SE = .14, p = .067, 95%
CI = .46, 1.03; 1-β = .17), and US-born Latino (IRR = .67, SE = .14, p = .054,
95% CI = .45, 1.01; 1-β = .17) youth.
Table 1. Prevalence of Self-Reported Offending and Officially Recorded Re-
Arrest
Documented Undocumented Second Native- Native-
Full CA
First First Generat born born
Sample
Generation Generation ion White Latino
N N= N = 23 N = 37 N= N = 74 N = 77
477 266
% of Sample 100 5 8 56 16 16
Pre-Baseline Self-
Reported Offending
Total Variety Ever 3.77 3.09 2.84 4.03 3.80 3.47
M(SD) (3.24) (1.81) (2.60) (3.38) (2.94) (3.53)
Arresting Offense
Drug (%) 1.68 0 0 1.5 4.05 1.30
Person (%) 16.56 21.74 13.51 15.04 14.86 23.38
Property (%) 72.96 69.57 75.68 73.31 71.62 72.73
Other (%) 8.81 8.70 10.81 10.15 9.46 2.60
Cumulative Self-
Reported Offending
Across 3 Years
Prevalence Total (%) 75.58 69.57 64.86 76.69 76.39 77.92
Variety M(SD) 4.23 3.13 2.70 4.68 3.95 4.01
(4.21) (3.28) (2.37) (4.53) (3.76) (4.18)
Cumulative Re-arrests
across 3 Years
Prevalence of re-arrest 44.86 39.13 48.65 46.99 39.19 42.86
(%)
Frequency re-arrests 1.23 .57 1.54 1.35 .97 1.10
M(SD) (1.99) (.84) (2.22) (2.11) (1.82) (1.83)

The wave-specific average rate of self-reported offending prevalence and

offending variety across immigrant and US-born subsamples are reported in
Figure 1. Following the baseline arrest, all study youth evidence a decline in the
prevalence and variety of offending across the three-year study period. While
there is variation in the relative rank ordering of the immigrant and US-born
subsamples at each follow-up, undocumented first-generation immigrants
consistently report the lowest prevalence of offending and the lowest variety of
offending compared to their documented immigrant and US-born peers.

Copyright @ 2018 MIGRATION LETTERS | Transnational Press London

156 Offending Histories of Undocumented Immigrants
We report the statistical differences in offending prevalence and variety at each
follow-up wave; however, we again note that fluctuations in statistical
significance may be influenced by the sample sizes. Because our interest is in
examining involvement in crime among undocumented immigrants, we use this
group as the statistical comparison. Overall, when significant differences are
observed, they are in the direction of undocumented immigrants offending at
lower rates than their US-born peers.6 Regarding differences in the variety of
offending, the patterns suggest that undocumented immigrants are involved in
a smaller variety of offenses than native-born Latino or White youth at certain
waves, but not consistently throughout the three years. Undocumented
immigrants consistently committed a smaller variety of offenses than second-
generation youth.7

Recidivism: Officially Recorded Re-arrest

In contrast to the findings for the self-reported offending measures, relative to
their peers, undocumented immigrants are re-arrested at a higher rate and more
often in the three-years following their first arrest. The cumulative prevalence
of re-arrest for all groups falls between 40% and 50% (see Table 1). Nearly half
of undocumented immigrants reported a subsequent arrest, and of those re-
arrested, the average rate of re-arrest is 1.54. The lowest rates of re-arrest are
observed for documented immigrant (39.13%; mean number of re-arrests =
.57) and US-born White (39.19%; mean number of re-arrests = .97) youth.
Although undocumented immigrant youth are just as likely to be re-arrested as
documented immigrant youth (OR = 1.47, SE = .80, p = .472, 95% CI = .51,
4.24; 1-β = .16) and their US-born White peers (OR = 1.47, SE = .60, p = .343,
95% CI = .66, 3.26; 1-β = .21), undocumented immigrants are re-arrested more

6 At six months following the first arrest, undocumented first-generation youth reoffended significantly less
than US-born Latino youth (Odds ratio (OR) = .47, SE = .19, p = .064, 95% CI = .21, 1.04; 1-β = .72). At
12 and 18 months, undocumented first-generation youth did not differ from other groups of youth. By 24
months, undocumented first-generation youth reoffended less than the native-born White youth (OR =
.44, SE = .19, p = .058, 95% CI = .19, 1.03; 1-β = .43). At 30 months, there were no differences in
reoffending rates. Finally, at 36 months, undocumented first-generation youth reoffended significantly less
than the native-born White youth (OR = .41, SE = .19, p = .051, 95% CI = .16, 1.01; 1-β = .47) and less
than native-born Latino youth (OR = .46, SE = .21, p = .091, 95% CI = .19, 1.13; 1-β = .41).
7 Undocumented first-generation youth committed a smaller variety of offenses than US-born White youth

at 12 months (IRR = .46, SE = .18, p = .050, 95% CI = .22, .99; Cohen’s d = .41; 1-β = .17).
Undocumented first-generation youth also committed a smaller variety of offenses than US-born Latino
youth at both 18 months (IRR = .41, SE = .16, p = .025, 95% CI = .18, .89; Cohen’s d = .41; 1-β = .17) and
24 months (IRR = .47, SE = .20, p = .072, 95% CI = .21, 1.07; Cohen’s d = .32; 1-β = .17). Finally,
undocumented first-generation youth committed a smaller variety of offenses than second-generation
youth at six months (IRR = .60, SE = .17, p = .076, 95% CI = .35, 1.05; Cohen’s d = .30; 1-β = .21), 12
months (IRR = .42, SE = .15, p = .013, 95% CI = .21, .83; Cohen’s d = .45; 1-β = .21), 18 months (IRR =
.40, SE = .15, p = .012, 95% CI = .20, .82; Cohen’s d = .41; 1-β = .21), 24 months (IRR = .47, SE = .18, p
= .043, 95% CI = .23, .98; Cohen’s d = .34; 1-β = .21), and 36 months (IRR = .42, SE = .17, p = .031, 95%
CI = .19, .93; Cohen’s d = .35; 1-β = .21).
www.migrationletters.com
 Bersani et al. 157
Figure 1. Average Rate of Self-Reported Offending by Immigrant and
Native-born Status
A. Total Crime Offending Prevalence

0.7
Documented First Generation
Undocumented First Generation
0.6 Second Generation
Offending Prevalence

Native-born White
0.5 Native-born Latino

0.4

0.3

0.2
6 12 18 24 30 36
Months Since Baseline

B. Total Crime Offending Variety

Documented First Generation

2.1 Undocumented First Generation
Second Generation
Native-born White
Native-born Latino
Offending Variety

1.6

1.1

0.6

0.1
0 6 12 18 24 30 36
Months Since Baseline

Copyright @ 2018 MIGRATION LETTERS | Transnational Press London

158 Offending Histories of Undocumented Immigrants
Figure 2. Average Rate of Re-Arrest by Immigrant and Native-born Status
A. Re-arrest Prevalence

31 Documented First Generation

Undocumented First Generation
27
Second Generation
23 Native-born White
Rearrest Percentage

Native-born Latino
19

15

11

-1
6 12 18 24 30 36
Months Since Baseline

B. Re-arrest Means
Documented First Generation
0.6
Undocumented First Generation
0.5 Second Generation
Native-born White
0.4
Native-born Latino
Rearrest Means

0.3

0.2

0.1

0
6 12 18 24 30 36
-0.1
Months Since Baseline

www.migrationletters.com
 Bersani et al. 159
frequently than documented immigrant youth (IRR = 2.73, SE = 1.33, p = .041,
95% CI = 1.05, 7.11; 1-β = .53).
We examine wave-specific rates of re-arrest prevalence and frequency (see
Figure 2). Following the baseline arrest, all study youth evidence a general
decline in the prevalence and frequency of re-arrest across the three-years.
While there is variation in the relative rank ordering at each follow-up,
undocumented immigrants are generally re-arrested more frequently than are
their documented first-generation immigrant and US-born peers (results
available upon request).

Robustness Analyses
We assess the robustness of our findings in two ways. First, because the
majority of immigrants in the full three-site Crossroads sample reside in
California, the analyses thus far have focused on youth in California. As a
robustness check, we replicated all analyses using data from all three study sites.
The model results were largely consistent. One exception is that analyses
conducted using data from all three sites suggests that undocumented
immigrants engaged in less self-reported offending prior to baseline than
documented immigrants (IRR = .73, SE = .11, p = .043, 95% CI = .54, .99).
Second, because of the longitudinal nature of the data and the clustering of
observations within individuals, we conducted a series of multi-level models
(MLM; Fitzmaurice et al. 2012; Lui et al. 2012; Singer & Willett 2004) and
examined whether immigration status accounts for variability in self-reported
offending and re-arrest over time (all results available upon request). MLMs
conceptualize growth curves using several levels of analysis, with
time/observations as level 1 and individuals as level 2. We estimated growth
models accounting for age (centered on 13) and traditional correlates of
offending (i.e., socioeconomic status, parental criminality).8 While we present
these findings cautiously given the small sample sizes and large standard errors,
considering the correlations among repeated observations provides a more
precise estimate of the rate of change and have more power as compared with
traditional models (Fitzmaurice et al. 2012). The results from the MLM analysis
confirm those reported above: 1) re-offending rates are generally lowest for
undocumented immigrants, and 2) undocumented immigrants tend to be
rearrested more frequently, particularly in the first year following their initial
arrest.

8Parent highest level of education is used as a proxy for socioeconomic status. Based on the distribution,
the variable was split into three categories for analyses: 40.9% had parents who had not finished high
school, 23.5% had parents who had finished high school or had received a GED, and 35.6% had at least
some exposure to higher education (e.g., trade, business, professional, or college). Parental criminality
measures the history of arrest for either parent. Across the sample, approximately 14.46% of the sample
had at least one parent who had been arrested.
Copyright @ 2018 MIGRATION LETTERS | Transnational Press London
160 Offending Histories of Undocumented Immigrants
Discussion
The American public consistently worries about illegal immigration, with nearly
half perceiving that immigrants make American society worse when it comes
to crime (Lopez, Passel and Rohal 2015). Yet, mounting empirical evidence
finds immigrants are less likely to commit crime than their US-born peers. The
research-rhetoric disconnect is due, at least in part, to the limited data and
analysis distinguishing undocumented immigrants. To our knowledge, this is
the first study to examine longitudinal patterns of youthful offending using self-
report and officially-recorded arrests distinguishing immigrants by their legal
residency status.
First, consistent with previous research, self-reported offending prior to
engagement with the justice system (i.e., first arrest) is relatively and consistently
low for immigrants generally and for undocumented immigrants specifically
compared to their US-born peers. Second, compared to their documented
immigrant and US-born peers, undocumented immigrants continue to report
low levels of offending in the three-years following first arrest. Third, and in
contrast to the findings for self-reported offending, undocumented immigrants
are more likely to be re-arrested and are re-arrested more in the three-years
following their first arrest. Differences in the risk of re-arrest appear to be most
pronounced immediately following their first arrest.
The divergent finding using the self-reported offending and officially recorded
arrest outcomes warrants further scrutiny. While it may be the case that this
divergence reflects systematic crime reporting bias, research by Bersani and
Piquero (2017) finds no evidence that immigrants accused of serious crimes
under-report their offending and may instead over-report crime compared to
their US-born peers. First arrest may trigger increased scrutiny and reduced
tolerance for deviance that increases the risk for secondary sanctioning (see
Doherty et al. 2016; Liberman et al. 2014; Morris & Piquero 2013), which may
be amplified for the undocumented. Analyses are needed to disentangle
whether the pattern of results observed here reflect differential treatment in the
process of the administration of the law, differential involvement in crime, or
some combination of the two (Piquero 2008).
Collectively, research on immigration and crime suggests that undocumented
immigration status does not distinguish a group of distinctly crime-prone
individuals. The current study represents a next step in the ever-developing
research base on the immigrant-crime nexus. It is critical to note, however, that
the sample sizes for the foreign-born group are small, particularly when
disaggregated by legal residency status. Although this is a clear limitation,
longitudinal data containing information on self-reported and official measures
of crime as well as immigrant (un)documented status are exceptionally rare.
Though the data demands are high and studies of youthful populations are

www.migrationletters.com
 Bersani et al. 161
difficult to conduct, researchers should replicate these analyses as data become
available. This research is also limited to examining offending among Latino
youth. Research is needed to replicate this study with larger, more diverse
samples.

Conclusion
Justification for heightened immigration enforcement and the criminalization
of immigration law is based on the assertion that undocumented immigrants
represent a distinctly dangerous class. With this research, we add to the limited
knowledge base and investigate crime among those lacking legal US residency
by assessing patterns of offending longitudinally, both prior to and following
first arrest, using self-reported offending and officially recorded arrest
measures. The nature of the sample allows for a thorough study of offending
patterns during precisely the developmental period when criminal involvement
is at its peak. Nonetheless, results show that undocumented immigrants
generally report the lowest rate of offending prior to and following their first
arrest. Yet despite offending less, undocumented immigrants are more likely to
be re-arrested in the three-years following their initial contact with the justice
system. Divergence in the pattern of results contingent upon the measure of
criminal behavior – self-report or official arrest – warrants focused
consideration to disentangle whether the difference is due to: systematic crime
reporting bias (e.g., under-reporting of crime among undocumented
immigrants), differential treatment in the criminal justice system (e.g.,
intensified policing), or a combination of factors.

References
Bersani, Bianca E. 2014. An Examination of First and Second Generation Immigrant
Offending Trajectories. Justice Quarterly 31:315-343.
Bersani, Bianca E., Thomas A. Loughran, and Alex R. Piquero. 2014. Comparing
Patterns and Predictors of Immigrant Offending Among a Sample of Adjudicated
Youth. Journal of Youth and Adolescence 43:1914-1933.
Bersani, Bianca E. and Alex R. Piquero. 2017. Examining Systematic Crime Reporting
31 Bias Across Three Immigrant Generations: Prevalence, Trends and Divergence
in 32 Self-Reported and Official Reported Arrests. Journal of Quantitative Criminology,
33(4): 835-857.
Bui, Hoan N. 2009. Parent-Child Conflicts, School Troubles, and Differences in
Delinquency across Immigrant Generations. Crime & Delinquency 55:412-441.
Butcher, Kristin F. and Anne Morrison Piehl. 1998. Cross-City Evidence on the
Relationship between Immigration and Crime. Journal of Policy Analysis and
Management 17:457-493.
Chacón, Jennifer M. 2012. Overcriminalizing Immigration. Journal of Criminal Law and
Criminology 102: 613-652.

Copyright @ 2018 MIGRATION LETTERS | Transnational Press London

162 Offending Histories of Undocumented Immigrants
Champely, Stephane. 2013. Package ‘pwr’. Available from http://cran.r-
project.org/web/pack ages/pwr/pwr.pdf. Last accessed: May 6, 2017.
Chavez, Leo R. 2012. Shadowed lives: Undocumented immigrants in American society. Cengage
Learning.
Chaves, Leo R. 2008. The Latino Threat: Constructing Immigrants, Citizens, and the Nation.
Stanford: Stanford University Press.
Correia, Mark E. 2010. Determinants of Attitudes toward Police of Latino Immigrants
and Non-Immigrants. Journal of Criminal Justice 38:99-107.
Davies, Garth and Jeffrey Fagan. 2012. Crime and Enforcement in Immigrant
Neighborhoods: Evidence from New York City. The ANNALS of the American
Academy of Political and Social Science 641:99-124.
Doherty, Elaine Eggleston, Jaclyn M. Cwick, Kerry M. Green, and Margaret E.
Ensminger. 2016. Examining the consequences of the “prevalent life events” of
arrest and incarceration among an urban African-American cohort. Justice
Quarterly 33: 970-999.
Durose, Matthew R., Alexia D. Cooper and Howard N. Snyder. 2014. Recidivism of
Prisoners Released in 30 States in 2005: Patterns from 2005-2010. Washington, DC: US
Department of Justice.
Ewing, Walter. A., Daniel E. Martínez, and Rubén G. Rumbaut. 2015. The
Criminalization of Immigration in the United States. Washington, DC: American
Immigration Council Special Report.
Fitzmaurice, Garrett, M., Nan M. Laird, & James H Ware. 2012. Applied longitudinal
analysis (Vol. 998). John Wiley & Sons.
Faul, Franz, Edgar Erdfelder, Axel Buchner, and Albergt-Georg Lang. 2009. Statistical
power analyses using G*Power 3.1: Tests for correlation and regression
analyses. Behavior Research Methods, 41, 1149-1160.
Gilbert, Gregory, E., and Susan Prion. 2016. Making Sense of Methods and
Measurement: The Danger of the Retrospective Power Analysis. Clinical Simulation
in Nursing 12.8: 303-304.
Gonzales, Roberto G. 2016. Lives in Limbo: Undocumented Coming of Age in America.
Oakland, CA: California University Press.
Goodman, Steven, and Jesse Berlin. 1994. The use of predicted confidence intervals
when planning experiments and the misuse of power when interpreting
results. Annals of internal medicine 121.33: 200-206.
Gottfredson, Michael R. 2004. Crime, Immigration, and Public Policy. Report prepared for
the Merage Foundation for the American Dream.
Hagan, John and Alberto Palloni. 1999. Sociological Criminology and the Mythology
of Hispanic Immigration and Crime. Social Problems 46:617-632.
Harris, Kathleen Mullan. 1999. The Health Status and Risk Behaviors of Adolescents
in Immigrant Families. In Children of Immigrants: Health, Adjustment, and Public
Assistance, D. J. Hernandez (ed.), Pp. 286-347. Washington, DC: National Academy
Press.
Hickman, Laura J., and Marika J. Suttorp. 2008. Are deportable aliens a unique threat
to public safety? Comparing the recidivism of deportable and nondeportable aliens.
Criminology and Public Policy 7:59–82.
Hickman, Laura J., Jennifer Wong and Marika Suttorp-Booth. 2016. Is Previous
Removal From the United States a Marker for High Recidivism Risk? Results from

www.migrationletters.com
 Bersani et al. 163
a 9-Year Follow-Up Study of Criminally Involved Unauthorized
Immigrants. Criminal Justice Policy Review 27:378-401.
Hindelang, Michael J., Travis Hirschi, and Joseph G. Weis. 1981. Measuring delinquency.
Beverly Hills: Sage Publications.
Hirschi, Travis, and Michael Gottfredson. 1983. Age and the Explanation of
Crime. American Journal of Sociology 89.3: 552-584.
Hoenig, John M. and Heisey, Dennis M. 2001. The Abuse of Power: The Pervasive
Fallacy of Power Calculations for Data Analysis. The American Statistician 55.1: 1-6.
Huizinga, David, Finn-Age Esbensen, and Anne Wylie Weiher. 1991. Are there
multiple paths to delinquency? Journal of Criminal Law and Criminology, 82, 83-118.
Ibañez, Gladys E., Michelle Agudo, Steve S. Martin, Daniel J. O’Connell, Rehab Auf,
and Diana M. Sheehan. 2017. Offending Behavior, Drug Use, and Mental Health
Among Foreign-Born versus US-Born Latino Criminal Justice Clients. Journal of
Immigrant and Minority Health, 19(3): 674-685.
Jennings, Wesley G., Kristen M. Zgoba, Alex R. Piquero, and Jennifer M. Reingle. 2013.
Offending Trajectories among Native-Born and Foreign-Born Hispanics to Late
Middle Age. Sociological Inquiry 83:622-647.
Katz, Charles M., Andrew M. Fox, and Michael D. White. 2011. Assessing the
Relationship between Immigration Status and Drug Use. Justice Quarterly 28:541-
575.
Kirk, David S., Andrew V. Papachristos, Jeffrey Fagan, and Tom R. Tyler. 2012. The
Paradox of Law Enforcement in Immigrant Communities: Does Tough
Immigration Enforcement Undermine Public Safety? The ANNALS of the American
Academy of Political and Social Science 641:79-98.
Lee, Matthew T. and Ramiro Martinez, Jr. 2009. Immigration Reduces Crime: An
Emerging Scholarly Consensus. In Immigration, Crime and Justice., vol. 13, Sociology of
Crime, Law and Deviance, W. F. Mcdonald (ed.), Pp. 3-16. Bingley, UK: Emerald
Group Publishing Limited.
Li, Cheng. 2013. Little’s Test of Missing Completely at Random. The Stata Journal, 795.
Liberman, Akiva M., David S. Kirk, and Kideuk Kim. 2014. Labeling effects of first
juvenile arrests: Secondary deviance and secondary sanctioning. Criminology 52: 345-
370.
Lui, Siwei, Michael J. Rovine and Peter Molenaar. 2012. Selecting a Linear Mixed Model
for Longitudinal Data: Repeated Measures Analysis of Variance, Covariance Pattern
Model, and Growth Curve Approaches. Psychological Methods 17:15-30.
Little, Roderick JA. 1988. Missing-data Adjustments in Large Surveys. Journal of Business
and Economic Statistics 6:287–296.
Little, Roderick JA. 1992. Regression with Missing X’s: A Review. Journal of the American
Statistical Association 87:1227–1237.
Lopes, Mark Hugo, Jeffrey Passel, and Molly Rohal. 2015. Modern Immigration Wave
Brings 59 Million to U.S., Driving Population Growth and Change Through 2065: Views of
Immigration’s Impact on U.S. Society Mixed. Washington, DC: Pew Research Center.
Massey, Douglas S., Jorge Durand, and Karen A. Pren. 2014. Explaining
Undocumented Migration to the US. International Migration Review 48.4:1028-1061.
Meissner, Doris, Donald M. Kerwin, Muzaffar Chishti, and Claire Bergeron. 2013.
Immigration Enforcement in the United States: The Rise of a Formidable Machinery.
Washington, DC: Migration Policy Institute.

Copyright @ 2018 MIGRATION LETTERS | Transnational Press London

164 Offending Histories of Undocumented Immigrants
Monahan, Kathryn C. and Alex R. Piquero. 2009. Investigating the Longitudinal
Relation between Offending Frequency and Offending Variety. Criminal Justice and
Behavior 36:653-673.
Monahan, Kathryn C., Laurence Steinberg and Alex R. Piquero. 2015. Juvenile Justice
Policy and Practice: A Developmental Perspective. In M. Tonry (Ed.), Crime and
Justice: A Review of Research (vol. 44, pp. 577-619). Chicago: University of Chicago
Press.
Morris, Robert G., and Alex R. Piquero. 2013. For whom do sanctions deter and
label?. Justice Quarterly 30: 837-868.
Orrick, Erin A., and Alex R. Piquero. 2015. Assessing the impact of Mexican nativity
on sentence length. Criminal Justice Policy Review 26:643-664.
Orrick, Erin A., Kiersten Compofelice, and Alex R. Piquero. 2016. Assessing the
Impact of Deportable Status on Sentencing Outcomes in a Sample of State
Prisoners. Journal of Crime and Justice 39: 28-40.
Osgood, D. Wayne, Barbara J. McMorris, and Maria T. Potenza. 2002. Analyzing
Multiple-item Measures of Crime and Deviance I: Item Response Theory
Scaling. Journal of Quantitative Criminology 18:267-296.
Passel, Jeffrey S. and D’Vera Cohn. 2016. Overall Number of U.S. Unauthorized Immigrants
Holds Steady Since 2009. Pew Research Center. Washington, DC.
Passel, Jeffrey S. and D’Vera Cohn. 2016. Homeland Security Produces First Estimate of
Foreign Visitors to U.S. who Overstay Deadline to Leave. Pew Research Center.
Washington, DC.
Perea, Juan F. 1996. Introduction. Pp. 1-10 in Immigrants Out! The New Nativism and the
Anti-Immigrant Impulse in the United States, edited by J. F. Perea. New York, NY: New
York University Press.
Pew Research Center. 2013. Second-generation Americans: A portrait of the adult children of
immigrants. Washington, DC: Pew Research Center.
Piquero, Alex R. 2008. Disproportionate Minority Confinement. The Future of Children
18:59-79.
Piquero, Alex R., Bianca E. Bersani, Thomas A. Loughran and Jeffrey Fagan. 2014.
Longitudinal Patterns of Legal Socialization in First-Generation Immigrants,
Second-Generation Immigrants, and Native-Born Serious Youthful Offenders.
Crime & Delinquency 62:1403-1425.
Piquero, Alex R., David P. Farrington and Alfred Blumstein. 2003. The Criminal Career
Paradigm: Background and Recent Developments. In M. Tonry (Ed.), Crime and
Justice: A Review of Research (vol. 30, pp. 359-506). Chicago: University of Chicago
Press.
Portes, Alejandro and Rubén G. Rumbaut. 2006. Immigrant America: A Portrait (3rd
edition). Berkeley, CA: University of California Press.
Reid, Lesley Williams, Harald E. Weiss, Robert M. Adelman, and Charles Jaret. 2005.
The immigration–crime relationship: Evidence across US metropolitan areas. Social
Science Research 34: 757-780.
Rengifo, Andres F., and Jennifer Fratello. 2014. Perceptions of the Police by Immigrant
Youth: Looking at Stop-and-Frisk and Beyond Using a New York City Sample.
Youth Violence and Juvenile Justice 1-19.
Rosenblum, Marc R. 2015. Understanding the Potential Impact of Executive Action on
Immigration Enforcement. Washington, DC: Migration Policy Institute

www.migrationletters.com
 Bersani et al. 165
Rosaldo, Renato. 1997. Cultural Citizenship, Inequality, and Multiculturalism. In Latino
Cultural Citizenship: Claiming Identity, Space, and Rights: 27-38.
Rumbaut, Ruben G., and Walter A. Ewing. 2007. The myth of immigrant criminality and the
paradox of assimilation: Incarceration rates among native and foreign-born men. Washington,
DC: Immigration Policy Center.
Ryo, Emily. 2017. Legal Attitudes of Immigrant Detainees. Law & Society Review 51:99–
131.
Simes, Jessica T. and Waters, Mary C. 2014. The Politics of Immigration and Crime,
pp. 457-483 in Sandra M. Bucerius and Michael Tonry, eds., The Oxford Handbook of
Ethnicity, Crime, and Immigration. New York, NY: Oxford University Press.
Stumpf, Juliet. 2006. The Crimmigration Crisis: Immigrants, Crime, and Sovereign
Power. American University Law Review 56:367-419.
Sweeten, Gary. 2012. Scaling Criminal Offending. Journal of Quantitative Criminology
28:533-557.
Telles, Edward E., and Vilma Ortiz. 2008. Generations of Exclusion: Mexican Americans,
Assimilation, and Race. New York: Russell Sage Foundation.
Vaughn, Michael G., Christopher P. Salas-Wright, Matt DeLisi, and Brandy R.
Maynard. 2014. The Immigrant Paradox: Immigrants are Less Antisocial than
Native-Born Americans. Social Psychiatry and Psychiatric Epidemiology 49:1129-1137.
Wang, Xia. 2012. Undocumented Immigrants as Perceived Criminal Threat: A Test of
the Minority Threat Perspective. Criminology 50:743-776.
Wong, Jennifer S., Laura J. Hickman, and Marika Suttorp-Booth. 2015. Examining
Recidivism among Foreign-born Jail Inmates: Does Immigration Status Make a
Difference over the Long Term? Global Crime 16:265-287.

Copyright @ 2018 MIGRATION LETTERS | Transnational Press London

166 Offending Histories of Undocumented Immigrants
Appendix. Sample Descriptive Statistics

Documented First

First Generation
Full CA Sample

Undocumented
Generation

Native-born

Native-born
Generation
Second

Latino
White
N N = 477 N = 23 N = 37 N = 266 N = 74 N = 77
% of Sample 100 5 8 56 16 16

Age at baseline 15.46 15.35 15.22 15.48 15.69 15.34

M(SD) (1.20) (1.30) (1.18) (1.20) (1.10) (1.31)

Latino (%) 80.71 100 100 93.23 0 100

Age at migration 0.67 (2.06) 4.83 (3.11) 5.13 (3.09) --- --- ---
Myears(SD)

Parental
Education (%)
No High 40.88 40.91 65.71 55.24 1.35 21.06
School Diploma
High School 23.52 27.27 25.71 22.18 21.62 27.63
Diploma
More than 35.6 31.82 8.57 22.58 77.03 51.32
High School

Parental Arrest 14.46 0.00 8.10 15.04 16.21 18.18

(%)

www.migrationletters.com
Copyright of Migration Letters is the property of Transnational Press London Ltd. and its
content may not be copied or emailed to multiple sites or posted to a listserv without the
copyright holder's express written permission. However, users may print, download, or email
articles for individual use.