Analyses of Social Issues and Public Policy, Vol. 0, No. 0, 2020, pp. 1--25
Attitudes Toward Separating Immigrant Families
at the United States–Mexico Border
Wade C. Rowatt,* Rosemary L. Al-Kire, Hilary Dunn, and Joseph Leman
The primary aims of this project were to estimate levels of agreement or opposi-
tion with immigrant family separation at the U.S.–Mexico border and to identify
psychological variables that account for variability in attitudes toward immigrant
family separation. In Study 1, a sample designed to be representative of Americans
in the United States responded to a question about the zero-tolerance policy that
resulted in immigrant family separation. In Study 2, participants in two conve-
nience samples completed online surveys with measures of perceived threat posed
by immigrants, dehumanization, social dominance, political ideology, religiosity,
and immigrant family separation. Across samples, the majority of respondents
opposed separating immigrant families. Conservative political ideology was a
consistent correlate of support for immigrant family separation. Dehumanization
of immigrants and social dominance orientation also accounted for unique vari-
ability in support for immigrant family separation. Given the potentially harmful
effects of extended parent-child separation caused by the zero-tolerance policy, it is
important to understand the roles of dehumanization, social dominance, and ideol-
ogy in attitudes toward immigration policies. Predictors of support for restrictive
policies could be targets of individual or community-level interventions designed
to reduce immigrant prejudice. Directions for future research and implications for
U.S. communities and policies affecting immigrants are discussed.
∗Correspondence concerning this article should be addressed to Wade C. Rowatt, PhD, Professor,
Department of Psychology & Neuroscience, Baylor University, One Bear Place #97334, Waco, TX
76798–7334, phone 254-723-8227; fax 254-710-3033 [e-mail: firstname.lastname@example.org].
Acknowledgments: The authors thank the editor and two peer-reviewers for very constructive and
thoughtful recommendations for improving previous drafts of this manuscript. The data, syntax files,
and study materials are available here: https://osf.io/g6djt/. Study 1 was not preregistered. The materials
for Study 2 were available on OSF before data were collected. Please address correspondence regarding
this article to Wade C. Rowatt, Department of Psychology & Neuroscience, Baylor University, Waco,
DOI: 10.1111/asap.12198 C2020 The Society for the Psychological Study of Social Issues
2Rowatt et al.
In May 2018 the U.S. government implemented a controversial zero-tolerance
policy to prosecute all adults crossing the U.S.–Mexico border illegally (Kandel,
2018; Zero-Tolerance for Offenses, 2018). There was no exception for asylum
seekers or those traveling with children (who could not be detained in the federal
criminal facilities with their parent(s); Kandel, 2018). After a few months, 2,000 or
more immigrant children were separated from their parents and moved to different
facilities. Almost 500 children remained separated from their parents months later
(Kopan, 2018) and that was discovered to be a low estimate (see Alvarez, 2019;
Alvarez & Sands, 2019). More than a year later, many immigrant children had
still not been reunited with their parents. There were reports that some immigrant
children died in crowded facilities (Jordan, 2018; Jordan 2019); many were not
receiving adequate healthcare or schooling, and some may never be reunited with
Whether one favors or opposes the policy or implementation, research shows
parent-child separation is harmful to children and family systems (Dozier, Zeanah,
Wallin, & Shauffer, 2012; Noor, 2018; Wood, 2018). For example, immigrant
children separated from their mothers in the same detention centers reported more
emotional problems and total difficulties compared to the detained children not
separated from their mothers (MacLean et al., 2019a). Furthermore, the estimated
prevalence of post-traumatic stress disorder symptoms among detained immigrant
children (17%) was higher than the lifetime prevalence among adolescents in
the United States (4.7%; MacLean et al., 2019a). Distress and mental health
disorder risk was consistently elevated among detained children regardless of the
amount of time for which they were separated from their parents (MacLean et al.,
2019b). There are fewer studies about long-term effects of parent-child separation,
though traumatic experiences in childhood are typically associated with an array
of psychological symptoms in adulthood (Cloitre et al., 2009).
Given the negative effects of separating parents and children, our overarching
goals for this project were to identify and better understand factors associated
with support for (or opposition to) a quite harsh zero-tolerance policy that results
in immigrant family separation. We hope to bring attention to several factors
(such as dehumanization, perceived threat, and system-justifying ideologies; Jost,
2019) that play a role in anti-immigrant sentiment and support for a policy that
has potentially life-or-death consequences for refugees, asylum seekers, and their
families. Dehumanization of immigrants, for example, could be a common link
between support for a zero-tolerance policy and indifference to the negative effects
of family separation. To the degree that predictors of support are malleable, they
could be targets for interventions that humanize immigrants or reduce support for
zero-tolerance policies that lead to unhealthy immigrant family separation. Efforts
may also be directed to develop a more balanced or flexible immigration policy
that is sensitive and responsive to the diversity of family systems in other cultures
(McDonald-Wilmsen & Gifford, 2009).
Predictors of Immigrant Family Separation 3
Despite a reemerging interest among scholars to better understand factors
that predict (un)welcoming attitudes and policies toward immigrants and refugees
(see Jetten & Esses, 2018; Streib & Klein, 2018), there is currently very little so-
ciopsychological research about attitudes toward immigrant family separation in
the United States. Addressing this gap will identify correlates of support or oppo-
sition to immigrant family separation that could be targets of interventions aimed
at humanizing immigrants, increasing perspective-taking, or increasing prosocial
motivations to help refugees in one’s community. A 2018 report indicated almost
three-fourths of respondents in the United States disapproved of an immigration
border policy that separates children from their parents when they enter the coun-
try without permission (Palmer & Igoe, 2019). Respondents were divided along
political lines, as 53% of Republicans favored and 90% of Democrats opposed
the policy that resulted in immigrant family separation (Palmer & Igoe, 2019).
In the following studies we examine the level of support (or opposition) for the
zero-tolerance policy that caused immigrant family separation at the U.S.–Mexico
border (Study 1) and investigate potential correlates of support for immigrant
family separation (Study 2).
Previous research has identified several demographic predictors of anti-
immigrant attitudes, including political and evangelical Christian religious af-
filiation (McDaniel, Nooruddin, & Shortle, 2011), being male, but not age (see
Cowling, Anderson, & Ferguson, 2019). The primary aims of Study 1 were to
estimate levels of agreement/opposition with immigrant family separation at the
U.S.–Mexico border and to investigate whether participant’s gender, age, religious
affiliation, or political ideology account for variability in favoring or opposing
family separation. Given Cowling et al.’s (2019) meta-analytic findings, we hy-
pothesized being male, conservative political ideology, and Christian religious
affiliation would be associated with support for the specific zero-tolerance policy
that resulted in immigrant family separation. We also explored whether age, gen-
der, political ideology, and religious affiliation accounted for unique variance in
the immigrant policy attitudes.
Data for Study 1 were obtained from the Public Religion Research Institute’s
June 2018 Immigration Survey (see www.prri.org). The full survey included sev-
eral questions about respondents’ views about how to restrict or accommodate
immigrants or refugees (e.g., build a border wall, pass a law to prevent refugees
from entering the United States). Given our focus on immigrant family separation,
4Rowatt et al.
we limited our analyses to the item about the degree of support/opposition for
immigrant family separation.
Participants, Procedure, and Measures
Participants from all 50 states in the United States (n=1,018; 495 men,
523 women) 18 years of age or older (M=53; SD =19; range =18–94) were
surveyed by telephone between June 20 and 24, 2018 (407 landline, 611 cell
phone). The sample was designed to represent the total U.S. adult population. As
shown in Table 1, the majority of participants were female (51%), white (64.6%),
and Christian (70.6%). Income was assessed categorically (see Table 1). Political
ideology was coded so that higher ratings indicate more conservative political
ideology: 1 =(very liberal)to5=(very conservative). To assess views about the
zero-tolerance policy which resulted in separating immigrant children from their
parents at the U.S.–Mexico border, respondents were asked to indicate their level
of agreement with the statement, “An immigration border policy that separates
children from their parents and charges parents as criminals when they enter the
country without permission” 1 =(strongly opposed)to4=(strongly favor).
Prior to subsequent analyses, data were omitted from participants who declined or
indicated they did not know how to answer the immigrant family separation item
(n=61, 6%). A small percentage of the entire sample did not answer questions
about their political ideology (n=53, 5.3%) or religious affiliation (n=45,
4.4%), which led to different sample sizes for some analyses. However, this size
sample (n=1,018) was more than sufficient to detect small correlations (r=.10)
and small mean differences (d=.20). For example, according to Cohen (1992)
783 participants were needed to detect a small correlation (p<.05, power =.80)
and 393 participants were needed to detect a small mean difference (p<.05,
The majority of participants surveyed opposed separating immigrant families
and charging the parent(s) as criminals (46.9% strongly opposed, 26.4% opposed,
12.4% favored, and 12.5% strongly favored; see Table 2). Women expressed
slightly more opposition (M=1.83, SD =1.05) than men (M=1.98, SD =1.03),
F(1, 956) =4.66, p=.03, d=.14. However, given the small effect size, we
caution readers not to overestimate the magnitude of this gender difference.
Next, we computed the correlations between political ideology, demograph-
ics, and degree of agreement with immigrant family separation. As shown in
Table 3, conservative political ideology was the strongest correlate of support
for the policy that resulted in immigrant family separation (r=.42, p<.001,
Predictors of Immigrant Family Separation 5
Tab le 1 . Sample characteristics
Measure PRRI sample MTurk sample College student sample
Sex % n%n%n
Male 48.6 495 57.9 106 27.1 39
Female 51.4 523 42.1 77 72.9 105
Total 100.0 1,018 100.0 183 100.0 144
White 64.6 658 74.3 136 67.4 97
Black 11.7 119 8.7 16 4.2 6
Asian 2.1 21 8.2 15 11.8 17
Hispanic 12.5 127 7.1 13 11.1 16
Native-American 2.2 22 1.1 2 .7 1
Other 4.1 42 .5 1 4.2 6
No response 2.3 23 0 0 .7 1
Total n=1,012 n=183 n=144
Religious Affiliation %n%n%n
None 24.0 196 51.4 93 9.7 14
Protestant 41.1 335 25.1 45 48.6 70
Catholic 28.2 230 14.2 25 22.2 32
Other .9 7 8.0 14 13.9 20
Jewish 2.3 19 .5 1 .7 1
Buddhist .9 7 .5 1 0 0
Don’t know 1.6 16 1.6 3 2.1 3
Total n=816 n=183 n=144
Measure PRRI sample
<$50,000 (unspecified) 1.1
$100,000 and over (unspecified) 1.6
$250,000 and above 2.2
n=911). Participant age, gender, and Christian religious affiliation were weakly
associated with support for immigrant family separation.
To explore the role of religious affiliation further, we computed an Analysis of
Variance to test the effect of religious affiliation (1 =Christian, 2 =religious non-
Christians, 3 =irreligious). Non-Christian religious persons opposed the policy
more (M=1.59, SD =0.90, n=37) than irreligious persons (i.e., none, atheist,
agnostic; M=1.70, SD =0.99, n=188) or Christians (M=1.93, SD =1.03,
n=533), F(2, 755) =4.91, p=.008, partial eta2=.013. A post hoc con-
trast indicated Christians opposed the policy which resulted in immigrant family
6Rowatt et al.
Tab le 2 . Descriptive statistics
PRRI (Study 1) MTurk (Study 2) College students (Study 2)
Measure M SD Range M SD Range M SD Range
Age (in years) 52.97 19.05 18–94 33.28 9.74 18–71 18.64 0.81 18–22
Conservative political ideology 3.10 1.24 1–5 3.27 1.87 1–7 4.22 1.54 1–7
Religiosity (1 =not at all, 4 =very
– – – 1.74 1.02 1–4 2.92 0.92 1–4
Favors immigrant family separation
(1 =strongly oppose, 4 =strongly favor)
Favors separating children from parents
seeking asylum (1 =strongly disagree,
– – – 2.34 1.91 1–7 2.52 1.70 1–7
Favors separating children from parents
crossing the border illegally (1 =strongly
disagree, 7 =strongly agree)
– – – 2.36 1.89 1–7 2.62 1.89 1–7
Predictors of Immigrant Family Separation 7
Tab le 3 . Correlates of support for immigrant family separation (Study 1)
Variable 1 2 3 4 5
1. Age –
2. Gender .21** –
3. Conservative political ideology .15** −.03 –
4. Religious affiliation (1 =Christian; 2 =non-Christian) −.32** −.17** −.25** –
5. Immigrant family separation .08*−.07*.42** −.11** –
Tab le 4 . Immigrant family separation regressed on demographic variables (Study 1)
Age .03 .70 .001
Gender −.06 −1.81 .004
Political ideology .40 11.24*** .147
Religious affiliation (1 =Christian; 2 =non-Christian) −.02 −.45 .000
Note. *denotes significant standardized beta. sr2represents squared semi-partial correlation. ***p<
separation less strongly than did irreligious individuals (Mdiff =.23, SE =.09,
95% CI: [.02, .44], p=.028, partial eta2=.010).
In another exploratory analysis, we regressed the immigrant family separa-
tion variable (1 =strongly oppose; 4 =strongly favor) on political ideology, sex
(1 =male, 2 =female), age, and religious affiliation (1 =Christian, 2 =non-
Christian) to identify any unique predictors. As shown in Table 4, conservative
political ideology remained the only significant predictor of support for immi-
grant family separation when age, participant sex, and religious affiliation were
statistically controlled [R2=.17, F(4, 713) =37.04, p<.001]. Because we did
not conduct a statistical power analyses to plan the sample size for this regression,
we provide a post hoc sensitivity analysis (Giner-Sorolla et al., 2019a, 2019b).
Using G*Power (version 22.214.171.124; Faul, Erdfelder, Buchner, & Lang, 2009), we
determined that a sample of 717 participants provided 80% power to detect effects
of sr2=.011 or higher and power of 90% to detect sr2=.015 or higher.
There are at least two notable findings in this study. First, a clear majority of
surveyed people(74%) oppose separating immigrant children from their parent(s),
which occurred because of the 2018 zero-tolerance policy. Second, there exists a
moderate, positive association between conservative political ideology and sup-
port for immigrant family separation (even when sex, age, and religious affiliation
are statistically controlled). Findings from the current study indicate conservative
8Rowatt et al.
political ideology predicts immigrant prejudice, superseding impacts from reli-
giosity affiliation. This is consistent with prior research showing religiosity’s asso-
ciation with immigrant prejudice is through political ideology (see Rowatt, 2019).
We are grateful to the Public Research Religion Institute (PRRI) for access
to such a timely dataset from a sample designed to be representative of the United
States. This allows us to infer that the majority of people in the United States during
the summer of 2018 were opposed to the zero-tolerance policy that resulted in
immigrant family separation. However, a few methodological issues merit further
discussion. As is common in national surveys like this, to keep the survey brief,
a single-item was used to assess support for the policy which led to immigrant
family separation. Additionally, measures of other psychological factors were not
included—factors that could help explain why people are in favor of or opposed to
immigration policy that results in the separation of immigrant families. To build
on Study 1, and in attempts to understand the sociopsychological phenomena
more, we designed and administered a survey during the summer of 2018 with
measures of other constructs implicated in prejudices (e.g., dehumanization, social
dominance, perceived threat, and conservative ideologies).
To identify candidate psychological predictors of immigrant family separa-
tion, we turned to well-established literatures in social-personality and political
psychology about racial and ethnic prejudices (Dovidio, Hewstone, Glick, &
Esses, 2010; Nelson, 2016; Stephan & Stephan, 2000; Wagner, Christ, & Heit-
meyer, 2010). We hypothesized that agreement with immigrant family separation
could be predicted by the tendencies to dehumanize immigrants (Kteily, Bruneau,
Waytz, & Cotterill, 2015), perceive immigrants as realistic or symbolic threats
(Cowling et al., 2019; Esses, Dovidio, Jackson, & Armstrong, 2001; Jackson,
Brown, Brown, & Marks, 2001; Stephan, Renfro, Esses, Stephan, & Martin,
2005), and by indicators of conservative social, political, or religious ideology,
such as social dominance (Pratto, Sidanius, Stallworth, & Malle, 1994), right-wing
authoritarian personality (Sibley & Duckitt, 2008), and religious fundamentalism
(Altemeyer & Hunsburger, 1992).
Dehumanization of Immigrants
If one views an immigrant as less than fully human, it could be easier to support
a policy that includes separating parents from their children or detaining immigrant
children in cage-like environs (Merchant, 2018). Dehumanization attitudes appear
to have direct consequences for political policy attitudes. For example, among
Americans, there are negative correlations between measures of dehumanization
and support for Arab immigration (Kteily et al., 2015).
Predictors of Immigrant Family Separation 9
In 2018, when communicating about the implementation of the zero-tolerance
policy, the President of the United States allegedly compared immigration of dan-
gerous criminals at the U.S.–Mexico border to an infestation (Graham, 2018; Klein
& Liptak, 2018). Such language is dehumanizing. Further, misrepresented media
portrayals of immigrants can increase dehumanization and perceived threat (Esses,
Medianu, & Lawson, 2013). Humanization of immigrants, on the other hand, cor-
relates negatively with immigrant prejudice, and could promote a more inclusive
re-categorization of the out-group (Costello & Hodson, 2010) and opposition to
restrictive immigration policies.
Perceiving Immigrants as Threatening
When resources are perceived to be limited or threatened, people experience
realistic conflict and often engage in competition (Duckitt, 1994). People also
compete for, and seek to protect, jobs or access to services covered by various taxes
on their income. Negative sentiment emerges toward out-group members perceived
to pose threats, take local jobs, strain medical or education systems, or violate
sacred values or beliefs (Esses et al., 2001; Stephan & Stephan, 2000). Stephan
and his colleagues, for example, have found variability in negative prejudice toward
immigrants is due to perceived realistic or symbolic threats (Stephan et al., 2005;
Stephan, Ybarra, & Bachman, 1999). Among Australians, perceived threat posed
by refugees or asylum seekers predicted more restrictive social policies toward
immigrants (Hartley & Pedersen, 2015). Among Europeans, perceived threats
posed by immigrants (e.g., encroaching on one’s way of life) predicted increased
preference to send all immigrants back to their countries of origin (Jackson et al.,
2001). In Study 2, in the United States, we predicted perceived threat posed
by immigrants would correlate positively with support for separating immigrant
Ideological and Motivational Predictors of Immigrant Prejudice
In addition to dehumanization tendency and perceived threat, social-
personality and political psychologists have discovered several other patterns of
thinking and feeling about historically disadvantaged social groups (e.g., immi-
grants) that we posit will be associated with attitudes toward immigrant family
separation (Cohrs & Stelzl, 2010; Duckitt & Sibley, 2010). For example, social
dominance orientation (SDO) includes perceptions that a majority group is su-
perior to minority groups and includes endorsement of policies which support a
power hierarchy (Pratto et al., 1994). Right-wing authoritarianism (RWA) refers
to a blend of traditionalism, justification of harsh behaviors to those who violate
laws, and beliefs that people should submit to authorities in power (Altemeyer,
1981). Religious fundamentalism (RF) involves “the belief that there is one set
10 Rowatt et al.
of religious teachings that clearly contain the . . . essential inerrant truth about hu-
manity and the deity” (Altemeyer & Hunsberger, 1992, p. 118). RF correlates
positively with perceptions that immigrants pose threats (Rowatt, 2019). How-
ever, when political ideology was statistically controlled, the association between
general religiosity and immigrant prejudice appears to be negligible (see Cowl-
ing et al., 2019; Rowatt, 2019). Likewise, when the aggression facet of RWA
is statistically controlled, religiosity and attitudes toward value-violating groups
is negligible (Shen, Haggard, Strassburger, & Rowatt, 2013). As such, political
ideology and RWA are important to consider in the religion-prejudice link.
Study 2 Overview and Predictions
The primary aim of Study 2 was to identify social-personality predictors of
support for separation of immigrant families. We predicted immigrant dehuman-
ization, perceiving immigrants as threatening, conservative political ideology, and
measures of RWA, SDO, and RF would correlate positively with agreement about
separating immigrant families. Regression analyses were used to explore whether
these variables account for unique variability in support for immigrant family
Participants and Procedures
Data were collected from two samples. The first sample consisted of 183
adult respondents in the United States recruited from Amazon’s Mechanical Turk
during the week of July 17–23, 2018 (when it was widely publicized that many
immigrant children were separated indefinitely from their parents at the U.S.–
Mexico border). The second sample included 144 undergraduate students enrolled
at a private, mid-sized university in central Texas in the Fall 2018. College students
received research participation credit toward a course requirement. Demographic
characteristics for these samples are provided in Tables 1 and 2.
After consenting, each participant completed an online survey (https://
osf.io/48jtk/) with the following measures administered in a few randomized
blocks. To assess views about immigrant family separation, participants rated two
items: “How do you feel about separating children from parents seeking asylum?”
and “What do you think about separating children from parents who cross the bor-
der illegally?” 1 =(strongly agree)to7=(strongly disagree). Both items were
reverse-keyed so that higher values indicate greater agreement with immigrant
Predictors of Immigrant Family Separation 11
family separation. Because the item about immigrant family separation used in
Study 1 was double-barreled, we did not include the phrase about charging par-
ents as criminals who enter the country without permission. The measures detailed
below are well-validated and have been widely used by researchers.
The Realistic and Symbolic Immigrant Threat Scale (Stephan et al., 1999;
Stephan & Stephan, 2000) assessed attitudes toward immigrants or immigration
with specific items about perceived realistic threats like “Immigration has in-
creased the tax burden on Americans” and symbolic threats like “Immigrants
should learn to conform to the rules and norms of the American society as soon as
possible after they arrive” 1 =(very strongly disagree)to7=(very strongly agree).
Note that Stephan et al.’s (1999) subscales referenced specific target groups (e.g.,
Mexicans, Asians). We modified items to assess attitudes toward immigrants more
generally (see Rowatt, 2019). Because realistic and symbolic threat were strongly
correlated in both samples (MTurk r=.83; college students r=.73), we averaged
the variables to create a single measure of perceived immigrant threat.
Participants also completed the Ascent of (Hu)Man scale (Kteily et al., 2015),
which was designed to assess degrees of blatant dehumanization towards a target.
The participants saw a sequence of five images starting with a semi-erect simian
(monkey/ape) and progressing to an upright human and indicated, using a sliding
scale, how human the average member of the following groups are: Americans,
Mexicans, and immigrants. Responses were converted toa0(not human)to
100 (fully human) scale and reverse-keyed so that higher values indicate more
The SDO scale (Pratto et al., 1994) measures the extent to which people
believe some groups have the right to marginalize other groups or may be more
deserving than inferior groups. Example items include “Some groups of people
are simply inferior to other groups” and “It’s OK if some groups have more of a
chance in life than others.” The measure of SDO (and RWA and RF below) were
rated using a 1 =(very strongly disagree)to8=(very strongly agree) response
The RWA scale (Smith & Winter, 2002) assessed authoritarian aggression,
submission, and conventionalism. Example items include “What our country really
needs is a strong, determined leader who will crush evil, and take us back to our
The RF scale (Altemeyer & Hunsberger, 2004) measures agreement with
religious beliefs such as religious scriptures are without error and (in)tolerance
and (lower) openness towards other religious perspectives. Example items: “There
is no body of teachings, or set of scriptures, which is completely without error
(reverse-keyed)”and “To lead the best, most meaningful life, one must belong to
the one, fundamentally true religion.”
To assess general religiosity (Rowatt, LaBouff, Johnson, Froese, & Tsang,
2009), we included a four-item measure with questions about the frequency of
12 Rowatt et al.
praying: 1 =(never)to11=(several times a day); reading sacred texts 1 =(never)
to 6 =(several times a day); attending religious services 1 =(never)to9=(several
times a week); and general religiousness: “How religious do you consider yourself
to be? 1 =(not at all religious)to4=(very religious).” Z-scores were computed
for each item and summed.
Throughout the survey we placed a few filler items to check whether partici-
pants were paying attention. These attention-check items asked about the color of
most grass (correct answer: green) and instructed the respondent to select “2” from
three other numerical options. Participants who answered these questions incor-
rectly were omitted from the datasets before analyses. The survey also included
several items about how one feels about immigrant children being separated from
their parents (e.g., empathy, distress), but because they were beyond the scope of
Study 2, we do not include analyses with these items.
Study 2 Data Analysis Overview
Prior to testing the predictions, we omitted MTurk cases with identical loca-
tions (n =46; see Dennis, Goodson, & Pearson, in press) and from those who
failed attention-check questions (n=5). In the college-student dataset, we omitted
participants who were 17-years old (n=4), had completely missing data for the
family separation item and a social-personality measure (n=6), or failed one
of the attention-check questions (n=6). The final samples included 183 MTurk
and 144 college student participants. Because MTurk participants are, on average,
less religious than college students (Table 1; see also Lewis, Djupe, Mockabee, &
Su-Ya Wu, 2015) and age ranges were so different between MTurk and college
student participants (18–71 vs. 18–22), we conducted analyses separately for each
sample, but present the results within one narrative.
Study 2 Results and Discussion
Consistent with Study 1, a majority of respondents sampled in Study 2 op-
posed separating immigrant children and parents who were seeking asylum or
alleged to have crossed the U.S.–Mexico border illegally (see Table 2). For ex-
ample, 62% in the college student sample and about 70% of the MTurk sample
disagreed or strongly disagreed with separating immigrant children from their
parents seeking asylum. Because the two items about separating immigrant chil-
dren from their parents seeking asylum or crossing the border illegally correlated
strongly (MTurk: r=.89; college students: r=.68), we summed them to create
an aggregate “immigrant family separation” variable.
Next, we examined the internal consistency of the multi-item measures. As
shown along the diagonals in Tables 5 and 6, the internal consistency of scales
was very good to excellent (with the exception of the RWA scale among college
Predictors of Immigrant Family Separation 13
Tab l e 5 . Correlations and reliabilities for MTurk adults (Study 2)
Variable 1 2 3 4 5 6 7 8 9 10
1. Age –
2. Gender −.01 –
3. Religiosity (standardized) .23** .10 .89
4. Dehumanization of immigrants −.01 −.09 .05 –
5. Conservative political ideology .13 −.11 .34** .26** –
6. Social dominance orientation −.03 −.08 .16*.41** .53** .93
7. Right-wing authoritarianism .05 .01 .54** .42** .63** .66** .83
8. Religious fundamentalism .09 .09 .73** .11 .41** .36** .69** .96
9. Threat .09 −.10 .27** .49** .59** .68** .70** .37** .94
10. Favors immigrant family separation .03 −.10 .26** .41** .51** .61** .55** .33** .62** –
Note. Cronbach’s alphas are listed on the diagonal. *p<.05. **p<.01.
Tab l e 6 . Correlations and reliabilities for college student sample (Study 2)
Variable 1 2 3 4 5 6 7 8 9 10
1. Age –
2. Gender −.01 –
3. Religiosity (standardized) .01 .17*.89
4. Dehumanization of immigrants .02 −.25** −.16 –
5. Conservative political ideology .02 −.19*.34*.08 –
6. Social dominance orientation .06 −.21*−.06 .34** .46** .85
7. Right-wing authoritarianism −.21*−.12 .42** .15 .56** .33** .59
8. Religious fundamentalism −.07 .09 .78** −.05 .40** .03 .65** .92
9. Threat .06 −.24** .14 .36** .60** .65** .56** .27** .89
10. Favors immigrant family separation −.03 −.30** −.01 .37** .55** .58** .38** .09 .57** –
Note. Cronbach’s alphas are listed on the diagonal. *p<.05. **p<.01.
students). Some RWA scale items tapped religious and moral sentiments in the
same item, which could lead to somewhat less consistent responses (e.g., “People
should pay less attention to the Bible and other old traditional forms of religious
guidance, and instead develop their own personal standards of what is moral and
immoral (reverse-keyed)” or “Everyone should have their own lifestyle, religious
beliefs, and sexual preferences, even if it makes them different from everyone
else”). Because low internal consistency affects the magnitude of correlations,
results with RWA should be interpreted with caution.
Predictors of Agreement with Immigrant Family Separation
In both samples, favoring immigrant family separation was predicted by
dehumanization, conservative political ideology, SDO, RWA, and perceived threat
posed by immigrants (see Tables 5 & 6). The magnitude of associations was
moderate to strong. RF and general religiosity correlated positively with support
for immigrant family separation in the MTurk sample (but not among college
14 Rowatt et al.
students). As in Study 1, agreement with immigrant family separation correlated
negligibly with age and increased slightly among men.
Unique Predictors of Support for Immigrant Family Separation
Next, we used a series of regression analyses to identify unique predictors
of support for immigrant family separation. Because we did not conduct a priori
statistical power analyses to plan sample sizes for regressions, we provide post hoc
sensitivity analyses (Giner-Sorolla et al., 2019a, 2019b). Using G*Power (version
126.96.36.199; Faul et al., 2009), we computed that our first sample of 183 MTurk
participants provided 80% power to detect effects of sr2=.042 or higher and
power of 90% to detect sr2=.055 or higher. Our second sample of 144 university
students provided 80% power to detect effects of sr2=.052 or higher and power
of 90% to detect sr2=.069 or higher.
We simultaneously entered significant predictors of immigrant family sep-
aration; then trimmed the model by removing, one-by-one, variables known to
be confounded (or too highly intercorrelated with other predictors in the model)
or with lower tolerance values. For transparency, we started by simultaneously
entering the demographics and bivariate correlates of immigrant family separation
in Model 1 (see Tables 7 and 8). However, we caution readers not to interpret the
relationships in Model 1 because RWA and RF are known to be confounded with
conventionalism (see Mavor, Macleod, Boal, & Louis, 2009). For this reason, we
omitted RF in Model 2 (and retained general religiosity).
In Model 2 (Tables 7 and 8), unique variability in immigrant family separation
was explained by dehumanization of immigrants, conservative political ideology,
SDO, (and perceived immigrant threat in the MTurk sample). In Model 3, we
dropped RWA; in part because RWA was negligibly associated with support for
immigrant family separation in Model 2, less internally consistent, and the RWA
tolerance coefficients were lower in both samples. Also notice the lower tolerances
for RF, RWA, and perceived threat posed by immigrants in Models 1, 2, and 3 for
In the final model, in both samples, unique variability in immigrant family
separation was accounted for by dehumanization of immigrants, conservative
political ideology, and SDO (when the demographics were controlled; see Model
4). The tolerance values are higher for each variable in Model 4, which is a sign that
multicollinearity is of lower concern (Menard, 2002). In the regression analyses,
social dominance was the strongest unique predictor of agreement with immigrant
family separation in the MTurk sample. Conservative political ideology was the
strongest unique predictor of favoring immigrant family separation among college
students (see Model 4 in Tables 7 and 8). And, curiously, across Models 1 to 3, the
perception that immigrants pose threats was a weak predictor of immigrant family
separation among college students (when the other predictors were statistically
Predictors of Immigrant Family Separation 15
Tab l e 7 . Immigrant family separation regressed on psychosocial and ideological predictors (Study 2, MTurk sample)
Model 1 Model 2 Model 3 Model 4
Age −.03 −.47 .001 .91 −.02 −.36 .000 .92 −.02 −.33 .000 .93 ←.01 −.14 .000 .93
Gender −.02 −.34 .000 .94 −.02 −.26 .000 .94 −.03 −.51 .001 .96 −.05 −.86 .002 .96
Religiosity .16 1.78 .010 .41 .13 1.81 .010 .58 .10 1.62 .008 .81 .13 2.12*.014 .83
Dehumanization .13 1.92 .012 .68 .14 2.05*.013 .71 .12 1.89 .011 .74 .18 2.97** .028 .83
Political ideology .15 1.91 .012 .55 .15 1.99*.013 .54 .14 1.85 .011 .58 .20 2.80** .025 .63
SDO .32 3.66** .043 .44 .32 3.78** .045 .43 .30 3.82** .045 .50 .40 5.68** .102 .64
Threat .26 2.78** .025 .37 .25 2.77** .024 .37 .25 2.84** .025 .41 – – – –
RWA −.06 −.52 .001 .23 −.06 −.60 .001 .27 – – – – – – – –
RF −.02 −.19 .000 .33 – – – – – – – – – – – –
R2.49** .49** .48** .46**
Note. sr2represents squared semi-partial correlation. Tolerance values are represented by tol.*p<.05 **p<.01.
16 Rowatt et al.
Tab l e 8 . Immigrant family separation regressed on psychosocial and ideological predictors (Study 2, college student sample)
Model 1 Model 2 Model 3 Model 4
Age −.08 −1.07 .005 .85 −.08 −1.08 .005 .86 −.10 −1.49 .009 .99 −.06 −.97 .004 1.00
Gender −.05 −.67 .002 .87 −.05 −.71 .002 .87 −.06 −.85 .003 .87 −.10 −1.50 .008 .85
Religiosity −.11 −.92 .004 .32 −.09 −1.18 .006 .65 −.07 −1.01 .004 .78 −.07 −1.02 .004 .77
Dehumanization .15 2.00*.017 .75 .15 2.07*.017 .75 .14 1.95 .016 .76 .20 2.88** .031 .82
Political ideology .39 4.03** .070 .47 .38 4.15** .070 .48 .35 3.99** .065 .52 .40 5.16** .101 .62
SDO .24 2.65** .030 .52 .25 2.84** .033 .52 .27 3.06** .038 .53 .31 4.03** .062 .66
Threat .14 1.24 .007 .35 .13 1.21 .006 .37 .14 1.43 .008 .42 – – – –
RWA .01 .08 .000 .35 .02 .15 .000 .45 – – – – – – – –
RF .01.08.000.25– –––– –––– –––
R2.51** .51** .50** .50**
Note. sr2represents squared semi-partial correlation. Tolerance values are represented by tol.*p<.05 **p<.01.
Predictors of Immigrant Family Separation 17
controlled); probably because of the intercorrelations with SDO and RWA. Again,
as shown in Model 4 (Tables 7 and 8), SDO, dehumanization of immigrants,
and conservative political ideology uniquely predicted agreement with immigrant
family separation when the other variables in the model were controlled.
In 2018, a clear majority of participants in the United States disagreed with
the separation of immigrant parents from their children at the U.S.–Mexico bor-
der (Cox, 2018; Palmer & Igoe, 2019), which was the result of a zero-tolerance
policy toward refugees and asylum-seekers crossing the U.S.–Mexico border. In
three samples of participants in the United States, we found conservative political
ideology was a strong and consistent correlate of support for immigrant family
separation. In Study 2, at the bivariate level, SDO, RWA, perceived threats posed
by immigrants, and dehumanization of immigrants each correlated positively and
substantially with support for immigrant family separation. These patterns are
largely consistent with previous research about general attitudes toward immi-
grants (Cohrs & Stelzl, 2010; Cowling et al., 2019; Wagner et al., 2010) and a
system-justification theoretical framework in which conservative ideology pre-
dicts support for policies that justify the status quo (Jost, 2019).
When the bivariate correlates of immigrant family separation were entered
simultaneously in regression analyses, the most consistent predictors of agreement
with immigrant family separation were conservative political ideology, SDO, and
dehumanization of immigrants. These sociopolitical variables are among a suite
of ideological predictors of parochial attitudes and behaviors that tend to contract
moral circles (Waytz, Iyer, Young, Haidt, & Graham, 2019). More universalistic
thinking about our shared superordinate identities (i.e., a majority of Americans
are descendants of immigrants or that we are all humans worthy of dignity and
respect), could broaden moral circles (cf. Waytz et al., 2019) and help foster
Implications for Immigrants, Communities, and Public Policy
At this point, we discuss some thoughts about the interface between basic
and applied research aimed at prejudice reduction, implications for communities
accommodating immigrants, and policies affecting immigrants and their com-
munities. The significant predictors we found of support for immigrant family
separation offer some clues about where to aim interventions designed to re-
duce anti-immigrant sentiment or support for separation. For example, programs
devised to provide more accurate depictions of contributions and strains immi-
grants make or face in communities (vis-a-vis threats only) or that humanize
immigrants as persons of worth could be tested to see if they reduce prejudice
18 Rowatt et al.
toward immigrants, increase empathy or perspective-taking, or change levels of
agreement with immigration policies.
Parochially segregating into socially dominant groups based on nationality,
status, or tribal beliefs could fuel intergroup bias or create a context in which
inaccurate stereotypes about immigrants could develop. Exposure to counter-
stereotypes, on the other hand, can be an effective means for reducing dehumaniza-
tion (Haslam & Statemeyer, 2016). Even the language used to refer to immigrants
is impactful, where terms “illegal” or “aliens” could elevate threat perceptions
and prejudice (Pearson, 2010) in ways that stratify communities based on sta-
tus, nationality, or social power. As such, inaccurate stereotypes and pejorative
terms about immigrants should be not be used in discussions about immigrant and
refugee issues, especially by key influential leaders in the United States.
Seeing positive behaviors could change the way we think about immigrants in
our midst. For example, Choi, Poertner, and Sambanis (2019) found that bystanders
in German train stations assisted a native German more than a Muslim immigrant,
but when the Muslim immigrant (confederate) enforced a social norm against
littering, assistance rates were comparable. When immigrants share social norms
held by those in a host country or community, they are more likely to receive aid.
Positive, first-person experiences and interactions with immigrants could have
humanizing effects. Seeing immigrants as fully human could help engender more
compassion as refugees flee unsafe conditions and navigate the complicated and
changing legal immigration process. Community-based programs that increase
cooperative intergroup contact could reduce negative perceptions of immigrants
as threatening (cf. Lemmer & Wagner, 2015). Citizens who choose to become
involved in efforts to help struggling refugees or asylees in their communities
could experience reduced anxiety about immigration and increased empathy for
immigrants (cf. Pettigrew & Troop, 2008). Some community religious organiza-
tions with missions to help the poor are already assisting immigrants (Garland,
2019) and could be models for others.
Religiosity and political ideology are also important variables to consider
together when trying to better understand attitudes toward immigrants and im-
migration policy. A somewhat paradoxical role of religion in prejudices is well-
documented (Johnson, Rowatt, & LaBouff, 2010; Rowatt, Carpenter, & Haggard,
2013) and is also found with regard to immigrant prejudice. When political ide-
ology is not statistically controlled, Christians are less opposed to immigrant
family separation than non-Christians (including religious unaffiliated persons;
see Study 1). This finding fits with a recent meta-analysis in which there was a
small effect of categorically measured religious affiliation (e.g., Christian vs. un-
affiliated) on negative attitudes toward immigrants across 21 studies (see Cowling
et al., 2019). However, we also found that the association between religious af-
filiation and immigrant family separation was negligible when political ideology
was statistically controlled, which is consistent with studies using dimensional
Predictors of Immigrant Family Separation 19
measures of religiosity (see Rowatt, 2019). That political ideology is a more con-
sistent predictor of prejudices than measures of religiousness could also be due to
the multidimensional nature of religion (as measured by categorical group mem-
bership, frequency of religious behaviors, orthodoxy of beliefs, or the rigidity or
flexibility of how one holds religious beliefs as immutable or open to change).
Perceived immigrant threat is also an important variable to consider when
trying to better understand attitudes toward immigrants and immigration policies.
Perceiving immigrants as dangerous or threatening could make it easier for one to
justify tightening boundaries and exclusionary measures to protect an in-group or
maintain the status quo (cf. system justification theory; Jost, 2019). When negative,
inaccurate stereotypes of immigrants as threatening are perpetuated throughout
the media; some could perceive a stronger need for stricter border security or zero-
tolerance policies that lead to immigrant family separation. By viewing immigrants
as less than human, one could more easily think that it is acceptable to separate
parents and children for extended periods of time. Challenging stereotypes, even
confrontationally, is an effective way to reduce bias (Czopp, Monteith, & Mark,
2006). Coming to conversations about immigrants or with a refugee with an open
mind (i.e., xenosophia; Streib & Klein, 2018) could engender wisdom that comes
from interpersonal experiences with immigrants. Policies that balance the security
of host nations and their citizens with the rights of refugees also seem prudent.
Strengths, Limits, and Directions for Future Research
In June and July 2018, the majority of Americans were opposed to the sep-
aration of immigrant children from their parents (which was a consequence of
a zero-tolerance policy). Use of data from a large sample of participants in the
United States is a notable strength of Study 1. And, even though the unrepresen-
tative MTurk and college student samples varied substantially with regard to age
and religiousness, agreement with immigrant family separation was predicted at
the bivariate level by the same psychological variables (i.e., conservative politi-
cal ideology, SDO, RWA, perceived threat, and dehumanization of immigrants).
Exploratory regression analyses reveal SDO, conservative political ideology, and
dehumanization of immigrants are unique predictors of agreement with immigrant
family separation among MTurk workers and college students.
Several limits merit a mention. First, because we used convenience samples
in Study 2, it is important not to overgeneralize the findings from Study 2 to
all people in the United States. Second, it is important to study these phenom-
ena in more culturally diverse or less “WEIRD” samples (see Henrich, Heine, &
Norenzayan, 2010). Third, many participants were surveyed during the summer
of 2018 when immigrant family separation was being widely publicized in the
United States, which could affect responses. As time passes, expressed views
could become less extreme or people could become apathetic or habituate to the
20 Rowatt et al.
level of need of refugees fleeing violence or uninhabitable homes due to natural
disaster. Fourth, like many researchers, we relied solely on self-report measures,
which can be subject to desirable responding or other motivations. Fifth, note that
in Study 2 we observed marginal internal consistency of the RWA measure in the
college student sample. We urge cautious interpretations of the RWA findings,
and call for researchers to strengthen measurement tools of the RWA construct.
Sixth, to assess dehumanization of groups we used discrete categories (American,
Mexican, Immigrant) which are not mutually exclusive for some individuals (i.e.,
Mexican American, Mexican immigrant). Future research should investigate the
intersectionality of social identity (i.e., gender, ethnicity, nationality) nested within
the self. Finally, we cannot conclude from simple correlations that political ideol-
ogy, social dominance, or dehumanization cause support for separating immigrant
families. Studies with longitudinal or experimental designs (cf. Bansak, Hain-
mueller, & Hangartner, 2016) are needed to investigate how ideology, dehuman-
ization, and other factors influence attitudes toward immigrants and immigration
In future studies, it is important to attempt to replicate and extend studies like
these in countries around the world where refugees and immigrants are seeking
asylum, safe haven from persecutors or unsafe living conditions, or to begin a
new life. For example, Hartley and Pedersen (2015) found that perceived threat
increases restrictive social policies toward refugees and asylum seekers among
Australians. Bansak et al. (2016) also showed how multiple personal facets of a
refugee (e.g., religions, reason for migrating, profession) affect European voters’
acceptance of asylum seekers. Streib & Klein (2018) document a recent rise in
xenophobia among Germans and report some dimensions of religiousness predict
negative attitudes toward immigrants. Understanding how threat, dehumaniza-
tion, or social dominance emerge and interact, for example, could be crucial in
efforts to better understand intergroup attitudes of citizens toward refugees and
As mentioned at the beginning of this article, whether one agrees or disagrees
with the zero-tolerance policy or implementation, there is mounting evidence that
immigrant parent-child separation has deleterious psychological and health effects
for the children and family systems involved (MacLean et al., 2019a; Stange &
Stark, 2019; Wood, 2018). Furthermore, it appears many immigrant children are
not being treated in a manner consistent with pediatricians’ recommendation to
“treat all immigrant children and families seeking safe haven who are taken into
U.S. immigration custody with dignity and respect to protect their health and
well-being” (Linton, Griffin, & Shapiro, 2017).
Since we collected and analyzed this data, there have been additional reports
of thousands of unaccompanied children migrating across national borders in and
around Europe and reports of poor treatment of immigrant children separated from
their parents at the southern U.S border. As immigration policies are debated,
Predictors of Immigrant Family Separation 21
refined, and hopefully improved, more research translating basic psychological
research on immigrants and refugees could be incorporated (Okazaki, Guler,
Haarlammert, & Liu, 2019; Silka, 2018). More qualitative methods to understand
immigrants and their plight are also needed (cf. McDonald-Wilmsen & Gifford,
2009). Although remarkably useful for uncovering patterns, the online survey
methods and statistical approaches are less personal and may not fully capture
the dignity of the whole person.
In closing, our research is a preliminary first step toward better understand-
ing the prevalence and correlates of immigrant attitudes and support for a zero-
tolerance policy that led to immigrant family separation at the U.S.–Mexico border.
Much more research is needed about immigration attitudes, activities that could
promote more understanding and compassion toward immigrants, and transla-
tion of research findings in efforts to develop immigration policies that are more
sensitive and responsive to the diversity of family systems.
Open Research Badges
This article has earned Open Data and Open Materials badges. Data and
materials are available at https://osf.io/g6djt/.
Altemeyer, B. (1981). Right-wing authoritarianism. Winnipeg: University of Manitoba press.
Altemeyer, B., & Hunsberger, B. (1992). Authoritarianism, religious fundamentalism, quest, and
prejudice. The International Journal for the Psychology of Religion,2, 113–133.
Altemeyer, B., & Hunsberger, B. (2004). A revised religious fundamentalism scale: The short and
sweet of it. The International Journal for the Psychology of Religion,14, 47–54.
Alvarez, P. (2019, November 6). Trump administration separated an additional 1,500 migrant families at
the US-Mexico border. CNN. Retrieved from https://www.cnn.com/2019/11/06/politics/family-
Alvarez, P., & Sands, G. (2019, January 17). ‘Thousands’ more children were separated than govern-
ment admitted, and it doesn’t know how many. CNN. Retrieved from https://www.cnn.com/
Bansak, K., Hainmueller, J., & Hangartner, D. (2016). How economic, humanitarian, and religious
concerns shape European attitudes toward asylum seekers. Science,354(6309), 217–222.
Choi, D. D., Poertner, M., & Sambanis, N. (2019). Parochialism, social norms, and discrimination
against immigrants. Proceedings of the National Academy of Sciences of the United States of
Cloitre, M., Stolbach, B. C., Herman, J. L., Kolk, B. V. D., Pynoos, R., Wang, J., & Petkova, E.
(2009). A developmental approach to complex PTSD: Childhood and adult cumulative trauma
as predictors of symptom complexity. Journal of Traumatic Stress,22, 399–408.
Cohen, J. (1992). A power primer. Psychological Bulletin,112, 155–159.
Cohrs, J. C., & Stelzl, M. (2010). How ideological attitudes predict host society members’ attitudes
toward immigrants: Exploring cross-national differences. Journalof Social Issues,66, 673–694.
Costello, K., & Hodson, G. (2010). Exploring the roots of dehumanization: The role of animal-human
similarity in promoting immigrant humanization. Group Processes & Intergroup Relations,13,
22 Rowatt et al.
Cowling, M. M., Anderson, J. R., & Ferguson, R. (2019). Prejudice-relevant correlates of attitudes
towards refugees: A meta-analysis. Journal of Refugee Studies,32, 502–524
Cox, D. (2018, June 26).. Growing divide on immigration and America’s moral leadership. Public
Religion Research Institute. Retrieved from https://www.prri.org/research/growing-divide-on-
Czopp, A. M., Monteith, M. J., & Mark, A. Y. (2006). Standing up for a change: Reducing bias through
interpersonal confrontation. Journal of Personality and Social Psychology,90, 784–803.
Dennis, S. A., Goodson, B. M., & Pearson, C. (in press). Online worker fraud and evolving threats
to the integrity of MTurk data: A discussion of Virtual Private Servers and the limitations
of IP-based screening procedures. Behavioral Research in Accounting, https://aaajournals.org/
Dozier, M., Zeanah, C. H., Wallin, A. R., & Shauffer, C. (2012). Institutional care for young children:
Review of literature and policy implications. Social Issues and Policy Review,6, 1–25.
Dovidio, J. F., Hewstone, M., Glick, P., & Esses, V. M. (2010). The SAGE handbook of prejudice,
stereotyping and discrimination. London, United Kingdom: Sage Publications.
Duckitt, J. (1994). The social psychology of prejudice. Westport, CT: Praeger.
Duckitt, J., & Sibley, C. G. (2010). Personality, ideology, prejudice, and politics: A dual-process
motivational model. Journal of Personality,78, 1861–1894.
Esses, V. M., Dovidio, J. F., Jackson, L. M., & Armstrong, T. L. (2001). The immigration dilemma:
The role of perceived group competition, ethnic prejudice, and national identity. Journal of
Social Issues,57, 389–412.
Esses, V. M., Medianu, S., & Lawson, A. S. (2013). Uncertainty, threat, and the role of the media
in promoting the dehumanization of immigrants and refugees. Journal of Social Issues,69,
Faul, F., Erdfelder, E., Buchner, A., & Lang, A.-G. (2009). Statistical power analyses using G*Power
3.1: Tests for correlation and regression analyses. Behavior Research Methods,41, 1149–1160
Garland, J. (2019, June 18). Fleeing north in the full armor of God. Christianity Today. Re-
trieved from https://www.christianitytoday.com/ct/2019/june-web-only/migrant-san-antonio-
Giner-Sorolla, R., Aberson, C. L., Bostyn, D. H., Carpenter, T., Conrique, B. G., Lewis, N. A., Jr.,
& Soderberg, C. (2019a). Powered to detect what? Considerations for planning and evaluating
sample size. Retrieved from https://osf.io/9bt5s/
Giner-Sorolla, R., Aberson, C. L., Bostyn, D. H., Carpenter, T., Conrique, B. G., Lewis, N. A., Jr., &
Soderberg, C. (2019b). Techniques and solutions for sample size determination in psychology:
A critical review. Retrieved from https://osf.io/psyjn/
Graham, D. A. (2018, June 19). Trump says Democrats want immigrants to “infest” the U.S.
The Atlantic. Retrieved from https://www.theatlantic.com/politics/archive/2018/06/trump-
Hartley, L., & Pedersen, A. (2015). Asylum seekers and resettled refugees in Australia: Predicting
social policy attitude from prejudice versus emotion. Journal of Social and Political Psychology,
Haslam, N., & Stratemeyer, M. (2016). Recent research on dehumanization. Current Opinion in
Henrich, J., Heine, S. J., & Norenzayan, A. (2010). The weirdest people in the world?. Behavioral and
Brain Sciences,33, 61–83.
Jackson, J. S., Brown, K. T., Brown, T. N., & Marks, B. (2001). Contemporary immigration policy
orientations among dominant-group members in Western Europe. Journal of Social Issues,57,
Jetten, J., & Esses, V. M. (2018). The reception of immigrants and refugees in Western countries: The
challenges of our time. Journal of Social Issues,74, 662–673.
Johnson, M. K., Rowatt, W. C., & LaBouff, J. (2010). Priming Christian religious concepts increases
racial prejudice. Social Psychological and Personality Science,1, 119–126.
Jordan, M. (2018, December 25). 8-Year-old migrant child from Guatemala dies in U.S. custody.
The New York Times. Retrieved from https://www.nytimes.com/2018/12/25/us/guatemalan-
Predictors of Immigrant Family Separation 23
Jordan, M. (2019 August, 20). The history of migrant children protection in America started
with two girls in Los Angeles. The New York Times. Retrieved from https://www.nytimes.
Jost, J. T. (2019). A quarter century of system justification theory: Questions, answers, criticisms, and
societal applications. British Journal of Social Psychology,58, 263–314.
Kandel, W. A. (2018). The Trump administration’s “zero tolerance” immigration enforcement pol-
icy (Congressional Research Service Report No. R45266). Washington, DC: Retrieved from
Klein, B., & Liptak, K. (2018, June 19) Trump ramps up rhetoric: Dems want ‘illegal immi-
grants’ to “infest our country.”CNN Politics. Retrieved from https://www.cnn.com/2018/
Kopan, T. (2018, August 30). Hundreds of immigrant kids remain separated from parents. CNN Pol-
itics. Retrieved from https://www.cnn.com/2018/08/30/politics/family-separations-hundreds-
Kteily, N., Bruneau, E., Waytz, A., & Cotterill, S. (2015). The ascent of man: Theoretical and empirical
evidence for blatant dehumanization. Journal of Personality and Social Psychology,109, 901–
Lemmer, G., & Wagner, U. (2015). Can we really reduce ethnic prejudice outside the lab? A meta-
analysis of direct and indirect contact interventions. European Journal of Social Psychology,
Lewis, A. R., Djupe, P. A., Mockabee, S. T., & Su-Ya Wu, J. (2015). The (non) religion of Mechanical
Turk workers. Journal for the Scientific Study of Religion,54, 419–428.
Linton, J. M., Griffin, M., & Shapiro, A. J. (2017). Detention of immigrant children. Pediatrics,139,
MacLean, S. A., Agyeman, P. O., Walther, J., Singer, E. K., Baranowski, K. A., & Katz, C. L. (2019a).
Mental health of children held at a United States immigration detention center. Social Science
& Medicine,230, 303–308.
MacLean, S. A., Agyeman, P. O., Walther, J., Singer, E. K., Baranowski, K. A., & Katz, C. L.
(2019b). Characterization of the mental health of immigrant children separated from their moth-
ers at the U.S.-Mexico border. Psychiatry Research. https://doi.org/10.1016/j.psychres.2019.
Mavor, K. I., Macleod, C. J., Boal, M. J., & Louis, W. R. (2009). Right-wing authoritarianism, funda-
mentalism and prejudice revisited: Removing suppression and statistical artefact. Personality
and Individual Differences,46, 592–597.
McDaniel, E. L., Nooruddin, I., & Shortle, A. F. (2011). Divine boundaries: How religion shapes
citizens’ attitudes toward immigrants. American Politics Research,39, 205–233.
McDonald-Wilmsen, B., & Gifford, S. M. (2009). Refugee resettlement, family separation
and Australia’s humanitarian programme. UNHCR Research Paper No. 178. Retrieved
Menard, S. (2002). Applied logistic regression analysis (Vol. 106). Thousand Oaks, CA: Sage.
Merchant, N. (2018, June 18). Immigrant kids seen held in fenced cages at border facility. AP News.
Retrieved from https://apnews.com/6e04c6ee01dd46669eddba9d3333f6d5
Nelson, T. D. (2016). In T. D. Nelson (Ed.), Handbook of prejudice, stereotyping, and discrimination
(2nd ed.). New York, NY: Psychology Press.
Noor, M. (2018). The caged children generation: When governments authorize the practice of
child maltreatment. Journal of Social Issues, Retrieved from https://spssi.onlinelibrary.wiley.
Okazaki, S., Guler, J., Haarlammert, M., & Liu, S. R. (2019). Translating psychological re-
search on immigrants and refugees. Translational Issues in Psychological Science,5, 1–3.
Palmer, J. J., & Igoe, M. (2019, June 19) How Americans feel about DREAM, the
Wall, and family separation. Public Religion Research Institute. Washington, D.C.
24 Rowatt et al.
Retrieved from https://www.prri.org/spotlight/how-americans-feel-about-dream-the-wall-and-
Pearson, M. R. (2010). How “undocumented workers” and “illegal aliens” affect prejudice toward
Mexican immigrants. Social Influence,5, 118–132.
Pettigrew, T. F., & Tropp, L. R. (2008). How does intergroup contact reduce prejudice? Meta-analytic
tests of three mediators. European Journal of Social Psychology,38, 922–934.
Pratto, F., Sidanius, J., Stallworth, L. M., & Malle, B. F. (1994). Social dominance orientation: A
personality variable predicting social and political attitudes. Journal of Personality and Social
Rowatt, W. C. (2019). Associations between religiosity, political ideology, and attitudes toward im-
migrants: A mediation path-analytic approach. Psychology of Religion and Spirituality,11,
Rowatt, W. C., Carpenter, T., & Haggard, M. (2013). Religion, prejudice, and intergroup relations. In
V. Saroglou (Ed.), Religion, personality, and social behavior, (pp. 170–192). New York, NY:
Rowatt, W. C., LaBouff, J., Johnson, M., Froese, P., & Tsang, J. A. (2009). Associations among
religiousness, social attitudes, and prejudice in a national random sample of American adults.
Psychology of Religion and Spirituality,1, 14–24.
Shen, M. J., Haggard, M. C., Strassburger, D. C., & Rowatt, W. C. (2013). Testing the love thy
neighbor hypothesis: Religiosity’s association with positive attitudes toward ethnic/racial and
value-violating out-groups. Psychology of Religion and Spirituality,5, 294–303.
Sibley, C. G., & Duckitt, J. (2008). Personality and prejudice: A meta-analysis and theoretical review.
Personality and Social Psychology Review,12, 248–279.
Silka, L. (2018). Adding a community focus to the psychological investigation of immigration issues
and policies. Journal of Social Issues,74, 856–870.
Smith, A. G., & Winter, D. G. (2002). Right-wing authoritarianism, party identification, and attitudes
toward feminism in student evaluations of the Clinton-Lewinsky story. Political Psychology,
Stange, M., & Stark, B. (2019). The ethical and public health implications of family separation. The
Journal of Law, Medicine & Ethics,47(2_suppl), 91–94.
Stephan, W. G., Renfro,C. L., Esses, V. M., Stephan, C. W., & Martin, T. (2005). The effects of feeling
threatened on attitudes toward immigrants. International Journal of Intercultural Relations,
Stephan, W. G., & Stephan, C. W. (2000). An integrated threat theory of prejudice. In S. Oskamp (Ed.),
Reducing prejudice and discrimination (pp. 23–46). New York, NY: Psychology Press.
Stephan, W. G., Ybarra, O., & Bachman, G. (1999). Prejudice toward immigrants. Journal of Applied
Social Psychology,29, 2221–2237.
Streib, H., & Klein, C. (Eds.). (2018). Xenosophia and religion. Biographical and statistical paths for
a culture of welcome. Berlin, Germany: Springer International Publishing.
Wagner, U., Christ, O., & Heitmeyer, W. (2010). Anti-immigration bias. In J. F. Dovidio, M. Hew-
stone, P. Glick, & V. M. Esses (Eds.), The SAGE handbook of prejudice, stereotyping, and
discrimination (pp. 361–376). London, U K: Sage Publications.
Waytz, A., Iyer, R., Young, L., Haidt, J., & Graham, J. (2019). Ideological differences in the expanse
of the moral circle. Nature Communications,10, 1–12.
Wood, L. C. N. (2018). Impact of punitive immigration policies, parent-child separation and child
detention on the mental health and development of children. BMJ Pediatrics Open,2, 1–6.
Zero-Tolerance for Offenses (2018). 8 USCS § 1325(a).
WADE C. ROWATT, PhD, is Professor of psychology at Baylor University where
he teaches undergraduate and PhD courses in social-personality psychology, pos-
itive psychology, and the psychology of religion.
Predictors of Immigrant Family Separation 25
ROSEMARY “MARAH” AL-KIRE is a second-year psychology PhD student at
HILARY DUNN is a second-year psychology PhD student at Baylor University.
JOSEPH LEMAN is a postdoc at the Institute for Studies of religion at Baylor