Content uploaded by Homero Gil de Zúñiga
Author content
All content in this area was uploaded by Homero Gil de Zúñiga on Oct 13, 2020
Content may be subject to copyright.
Content uploaded by Homero Gil de Zúñiga
Author content
All content in this area was uploaded by Homero Gil de Zúñiga on Jul 27, 2020
Content may be subject to copyright.
534
|
wileyonlinelibrary.com/journal/ejsp Eur J Soc Psychol. 2020;50:534–546.
© 2019 John Wiley & Sons, Ltd.
|
Although immigration has always been part of human history and
is today an important phenomenon for the development of mo-
dernity, it is also a complex and multifaceted process. In spite of
growing diversity and globalization, attitudes toward immigration
have become more restrictive in recent decades, particularly in
places like Europe. Previous research has highlighted the impor-
tance of complex systems of personal beliefs to the understand-
ing of attitudes toward immigrants, with studies considering
either Right Wing Authoritarianism (RWA), Social Dominance
Orientation (SDO), or human values. On the one hand, existing
empirical evidence suggests that RWA, SDO, and traditional/
normative and security values, which emphasize conservatism,
Received: 3 June 2018
|
Revised: 22 August 2019
|
Accepted: 19 September 2019
DOI: 10.100 2/ejsp.26 35
RESEARCH ARTICLE
Human values and ideological beliefs as predictors of attitudes
1| Magdalena Bobowik2|1|
1| Homero Gil de Zuñiga3,4 |5,6|
7|8
1School of Psychol ogy, Massey Univers ity,
Auckland, New Zealand
2Depar tment of S ocial Ps ycholog y
and Meth odolog y of Behavior
Science s, University of the Basque Country,
San Sebastián, Spain
3Department of Communication, Media
Innovation Lab, U niversity of Vie nna,
Vienna, Austria
4Faculty of Communication and
Literature, Die go Port ales Uni versit y,
Santiago, Chile
5Instit ute for Internatio nal and So cial
Studies, Tallinn University, Tallinn, Estonia
6Centre for Applied Cross-cultural
Research, Vic toria Uni versit y of Welling ton,
Wellington, New Zealand
7Higher School of Economics, Russian
National Research University, Moscow,
Russia
8Department of Psychology, Federal
University of Paraiba, J oao Pess oa, Brazil
Magdale na Bobowik, Unive rsity of the
Basque Country, San Sebastián, Spain.
Email: magdalena.bobowik@ehu.eus
Funding information
This research was s uppor ted by Grant
FA2386-15-1-0003 from the A sian Of fice of
Aerospace Research and Development.
Immigration is a worldwide subject of interest, and studies about attitudes toward
immigrants have been frequent due to immigration crises in different locations across
the globe. We aimed at understanding individual-level effects of human values and
ideological beliefs (Right-Wing Authoritarianism—RWA, and Social Dominance
Orientation—SDO) on attitudes toward immigrants, and whether country-level vari-
ables (perception of Islamic fundamentalism as a threat, perception of immigrants
as a threat, and international migrant stock) moderate these relations. With repre-
sentative samples from 20 countries (N = 21,362; the Americas, Europe, Asia, and
Oceania), and using Multilevel Bayesian regressions, results showed the negative ef-
fect of RWA, SDO, and existence values on attitudes toward immigrants, and the
positive effects of suprapersonal and interactive values. Cross-level interactions in-
dicated that the effects of RWA, SDO, and suprapersonal and existence values were
intensified in countries with societally high levels of perceiving Islamic fundamental-
ism as a threat. International migrant stock served as a country-level moderator for
the effects of SDO and RWA only. When country-level moderators were included
simultaneously, Islamic fundamentalism as a threat was the most consistent modera-
tor. Framing theory is offered as a plausible explanation of these results.
attitudes toward immigrant s, cross-level interaction, human values, Islamic fundamentalism,
multilevel analysis, right wing authoritarianism, social dominance orientation, threat
|
535
ATTITUDE S TOWARD IMMIG RANTS ACROSS 20 COU NTRIES
obedience to rules, and concern for the stability of society, have
a negative relation to perception of immigrants and outgroups
in general (Craig & Richeson, 2013; Cohrs & Stelzl, 2010; Danso,
Sedlovskaya, & Suanda, 2007; Davidov, Meuleman, Billiet, &
Schmidt, 2008; Davidov, Meuleman, Schwartz, & Schmidt, 2014;
Duckitt & Sibley, 2010; Esses, Dovidio, Jackson, & Armstrong,
2001; Pratto, Sidanius, & Levin, 2006; Vecchione et al., 2012). On
the other hand, values that represent more open-minded people
who strive for knowledge and maturity values, such as universal-
ism and the suprapersonal, are likely to predict positive attitudes
toward immigrants (e.g., Davidov et al., 2008; Vecchione et al.,
2012). In the present article, we integrate these individual-level
phenomena with country-level theories about how existential
threats to a society might influence individual-level relationships
between variables. More specifically, we examine three different
forms of societal threat (perception of Islamic fundamentalism as
a threat, perception of immigrants as a threat, and international
migrant stock) as moderators of the relationship between the indi-
vidual-level phenomena (SDO, RWA, and human values) and atti-
tudes toward immigrants.
Recently, communication theorists such as Powell (2011)
have theorized that media framing (see Reese, 2007) together
with fear mongering by political elites (e.g., Ettinger & Udris,
20 0 9) is resp onsib le fo r cons tru c ti ng Isl am ic te r ro r is m a s an exi s-
tential threat to Western civilization. According to Powell (2011),
“Terrorism and the way countries like Iran and Iraq ‘threaten us’
and our ‘way of life’ and speculation about possible future at-
tacks are a focus of media attention” (p. 91); Western media are
attracted to reporting terrorist acts because they afford the op-
portunity to provide a highly dramatic, over-simplified account
of a struggle between “our good” and “their evil”, using familiar
stereot ypes of “Muslims as Other” and tropes like the “War on
Terror” (Lewis & Reese, 2009). Ettinger and Udris (2009) elabo-
rate on this approach by describing how Swiss mass media have
for decades problematized immigrants, and more recently, one
political party has framed local Muslims as both immigrants im-
perfectly integrated into Swiss societ y and potential targets for
radicalization by international terrorists. This framing of a local
minorit y as possible allies of a feared international outgroup is
a powerful frame for positioning Muslim people as part of in-
ternational terrorism. It is almost certainly more prevalent in
Europe and the United States than in parts of the world where
Muslim immigration and DAESH-inspired terrorist attacks are
less salient. In these Western countries, the representation of
immigrants in general may have become conflated with that of
Muslim terrorists and local immigrants becoming radicalized
(Ettinger & Udris, 2009), thereby strengthening the impact of
threat-based predictors of attitudes toward immigrants (see
Claassen & McLaren 2019). On the other hand, in countries lack-
ing such frames, DAESH-inspired terrorism will be less conflated
with immigration, and a threat-based dynamic is less likely to be
activated. Therefore, it is theorized that fear of Islamic funda-
mentalism at the country-level should be a powerful moderator
of individual-level relationships between attitudes toward immi-
grants and its psychological predictors.
This is not the only or even the most plausible theory of how
media framing might moderate attitudes toward immigrants. As
noted by Ettinger and Udris (2009), many frames (e.g., failure to
integrate, petty criminality, competition for jobs, diluting tradi-
tional values, etc.) have been used by politicians to problematize
immigrants, long before terrorism became a dominant discourse
(see van Dijk, 200 0). In terms of historical trajectories (Liu, Fisher
Onar, & Woodward, 2014), some countries have had a welcom-
ing view of immigration, while others have had time to develop
strong opposition to specific outgroups, and consider that em-
bracing such groups might compromise their community (Brader,
Valentino, & Suhay, 2008; Landis & Albert, 2012; Mahroum, 2001;
Schwartz, Sagiv, & Boehnke, 20 00). For instance, the arrival of a
culturally distinct immigrant group can trigger feelings of threat,
activating authoritarian predispositions and galvanizing anti-im-
migrant at titudes. Stenner (2005) calls this ac tivation pattern the
“authoritarian dynamic”. Anti-immigration discourses regularly
flare up in many locations all over the world because immigrants
can be framed as threatening locals’ way of life, even without any
fear of terrorism (Brader et al., 2008). However, immigrant s are
not always constructed as a threat (van Dijk, 2000) in all coun-
tries, as the language used for framing immigrants can sometimes
be subtle or discreet (Liu & Mills, 20 06). Framing immigrants as a
threat may vary substantially, and therefore activate threat-based
dynamics predicting attitudes to immigrants to a different degree
across cultures (see Claassen & McLaren 2019). Thus, immigrants
as a threat might be a more general and more power ful country-
level moderator of individual-level predictors of attitudes toward
immigrants than Islamic fundamentalism.
Finally, because both of these forms of threat are subjective,
being constructed out of mass media frames interacting with po-
litical agendas, we hypothesize that they are likely to be more in-
fluential as countr y-level moderators than an objective measure of
the impact of immigration, like the percentage of migrant stock in a
country. This is an objective measure of impact that has produced
country-level moderation in other research (Cohrs & Stelzl, 2010),
but objec tive fac tors are not always subjectively perceived as threat,
and therefore have typically less impact on psychological factors like
attitudes (Cohrs & Stelzl, 2010; Gijsberts, Scheepers, & Coenders,
2004; Quillian, 1995; Ramos, 2011; Ramos, Pereira, & Vala, 2016;
Semyonov, Raijman, Yom-Tov, & Schmidt, 2004).
Most previous studies examining predictors of attitudes toward
Islamic fundament alism and immigration have used European sam-
ples only (Claassen & McL aren 2019; Davidov et al., 2014; Henrich,
Heine, & Norenzayan, 2010), which makes it difficult to assess
whether the reported findings are specific for Europe or whether
they have global implications. In the present article, we use samples
from 20 countries with various cultural backgrounds (e.g., Brazil,
China, Indonesia, Tukey, New Zealand, Russia, and the United States)
to examine the moderation effect of each of the aforementioned
country-level variables.
536
|
ARAÚJO et Al.
|
|
Values, as general beliefs that guide people’s behavior (Gouveia,
2003; Schwartz, 1992), have been commonly studied as determi-
nants of outgroup responses. According to the Functional Theory
of Human Values (for more details, see Gouveia, Milfont, & Guerra,
2014), there are two widely accepted value functions: (1) values as
guides of human actions, and (2) values as expressions of human
needs (see Figure 1). The combination of these two functions yields
a three-by-two framework (types of orientation: social, central and
personal; and types of motivators: materialistic and humanitarian)
with six basic subfunctions of values: existence, suprapersonal,
normative, interactive, promotion, and excitement (Gouveia, 2013;
Gouveia et al., 2014).
Like security in Schwartz’s model, existence represents the basic
needs of human beings in biological and psychological spheres, em-
bracing concerns with stability, health, and sur vival. On the other
hand, suprapersonal values embrace more universalistic and hu-
manitarian characteristics (like universalism), including the desire to
obtain knowledge, maturity, and to appreciate beauty. Normative
values (comparable to conformity/tradition values) express the im-
portance given to, and the willingness to preserve, conventional
norms and rules, emphasizing life in society and respect for one’s
culture and traditions. Interactive values, equivalent to benevolence,
express the need to live in a social environment, where the support
of one’s peers is utterly desirable, as well as belonging to specific
groups and showing social affectivit y. Promotion values are mostly
related to obtaining success, having high levels of prestige in society,
and the desire to obtain power, in line with the dimension of power
in Schwartz’s model. Finally, excitement values represent the desire
to live with emotion, obtaining pleasure and expressing one’s sexu-
ality (Gouveia, 2013).
Universalism values, which focus on acceptance and general
concerns with others, tend to be related to more positive at titudes
toward immigrants. On the other hand, conservation values, es-
pecially securit y, are related to opposition to immigration, coher-
ently with the characteristics of traditionalism and distaste for the
unusual or unconventional (Schwar tz, 2007). Empirical evidence
has confirmed the role values play in shaping people’s response to
immigration. Negative feelings toward outgroups in general and
immigrants in particular are typical of people who endorse conser-
vation (especially security, but also conformity; Beierlein, Kuntz, &
Davidov, 2016; Davidov et al., 2008; Vecchione et al., 2012; Wolf et
al., 2019), but also of those who ascribe low importance to the values
of self-transcendence (especially universalism; Beierlein et al., 2016;
Davidov et al., 2008; Leong & Ward, 2006; Sagiv & Schwartz, 1995;
Saroglou & Galand, 2004; Schwart z, 2006; Vecchione et al., 2012;
Wolf et al., 2019), and who show strong desire for power (Vecchione
et al., 2012). These effects were replicated across European coun-
tries (Davidov et al., 20 08; Vecchione et al., 2012). In addition, a
study by Van der Noll (2014) revealed that values also predict at-
titudes toward Muslims in particular, with universalism and self-di-
rection related to less opposition and conservatism values to more
opposition to wearing headscarves, building mosques, or Islamic ed-
ucation. Therefore, based on previous findings, the most relevant
value subfunctions to predict attitudes toward outgroups are exis-
tence, normative, suprapersonal, and promotion.
|
Like some human values, both RWA and SDO have been shown to be
important predictors of negative at titudes toward immigration, and
to a generalized tendency to dislike outgroups and to ethnocentrism
(Bourhis & Dayan, 2004; Craig & Richeson, 2013; Cohrs & Stelzl,
2010; Duckitt & Sibley, 2010; Pratto et al., 2006). RWA represents a
concern with adherence to ingroup norms, rules, and submission to
authorities in order to ensure security and social order (Altemeyer,
1981), and SDO a preference for social inequality and acceptance
of hierarchy, with a positive perception of dominant groups (Pratto
et al., 2006; Sidanius & Pratto, 1999). Taking these definitions into
account, RWA and SDO are distinctively related to negative out-
group responses due to the particular sets of values and dif ferent
motivational goals that guide each one of them (Danso et al., 2007;
Thomsen, Green, & Sidanius, 2008), as theorized by the dual-process
model of prejudice (Duckitt, 2001). In this view, the main aspect of
RWA that leads to prejudice is the motivation to achieve securit y
from thre ats, which is repre sented by high levels of importance given
to values of social order, stability, and tradition. On the other hand,
SDO leads to prejudice due to the view of the world as a competi-
tive place, where dominant individuals or groups are more likely to
FIGURE 1 The structure of human values according to the
Functional Theor y. Note: Continuous lines separate goals and the
dashed lines separate needs
Suprapersonal
Existence
Survival needs
Thriving needs
Social goals
Personal goals
Central goals
Affectivity
belonging
support
Obedience
religiosity
tradition
Emotion
pleasure
sexuality
Power
prestige
success
Beauty
knowledge
maturity
Health
stability
survival
|
537
ATTITUDE S TOWARD IMMIG RANTS ACROSS 20 COU NTRIES
have better chances to achieve their goals, being expressed by val-
ues of power, and the desire to dominate and to be superior to oth-
ers (Craig & Richeson, 2013; Duckitt & Sibley, 2010). Finally, Imhoff
and Recker (2012) showed that both RWA and SDO were related to
Islamoprejudice. Therefore, we can expect both RWA and SDO to
be particularly efficient predictors of negative attitudes toward im-
migrants, especially when perceptions of threat are activated.
|
Perceptions of threat are known to play a crucial role in inciting
negative responses toward outgroups (Stephan et al., 1998; Stephan
et al., 2005). Immigrants may pose different types of threat to the
host majority, from a realistic (material) threat to the wellbeing that
originates, for instance, from fear of competition over scarce re-
sources (e.g., employment or social services), referring to material,
economic, and welfare threats, to a more symbolic threat, which is
a result of perceiving outgroup values and ideologies as threatening
to ingroup beliefs and moralit y (Stephan et al., 2002, 2005). Group
threat theory (Blumer, 1958) proposes that ethnic threat occurs at
the collective level, where ethnic group composition at the societal
level may be a determinant of responses toward the minority group.
Indeed, most of the research based on this idea examines how col-
lective-level threat may shape responses toward ethnic minorities,
including immigrants (Gijsberts et al., 2004; Hjerm, 2007; Quillian,
1995; Ra mos, 2011). In su mma ry, country‐level discourses about im-
migration and threat are produced by a combination of fac tors, in-
cluding mass media, political leadership, demographics, and resource
struggles. These form the representational resources (Liu, Sibley, &
Huang, 2014) that c an be articulated as societal-level threats by
mass media and political leaders (Powell, 2011; van Dijk, 20 00).
Further, these threat s can produce a societal dynamic that moder-
ates how individual-level predictors impact on immigration attitudes
in a given society.
Human values could therefore be used to justify hos tile outgroup
responses under perception of threat to these values. Confirming
this idea, perceptions of violations to personal and group values
have been used as justifications for hostile attitudes and discrimina-
tory behavior (Crandall et al., 2001). In this line, Cook, Cottrell, and
Webster (2015) found that perceptions of threat to values resulted
in more anti-atheist prejudice. Based on these previous findings, we
can expect that when threat is activated, people may use the expres-
sion of certain values (e.g., existence/security or normative/tradi-
tional values) to justify their negative outgroup responses in general
or attitudes toward immigrants in par ticular (Claassen & McLaren
2019; Cook et al., 2015).
More is known about the role of threat in the link between
ideological beliefs and outgroup responses. Crawford and Pilanski
(2013) found that both political liberalism and conser vatism pre-
dict intolerance to the opposing group, and that such intolerance
is heightened when one group perceives the other as threatening.
Diversity was also negatively related to positive attitudes toward
immigrants among high authoritarians and among people experi-
encing their immediate environment as threatening, whereas it was
positively related to outgroup attitudes among low authoritarian in-
dividuals and among people residing in more secure neighborhoods
(Van Assche, Roets, Dhont, & Van Hiel, 2014). Cohrs and Asbrock
(2009) found that the effects of RWA on prejudice were stronger
when the outgroup was socially threatening, but the effec ts of SDO
on prejudice did not increase when the outgroup was presented as
competitive. Interestingly, Costello and Hodson (2011) showed that
it was symbolic threat in particular that enhanced the effect of SDO
on prejudice (through infra-humanization). Altogether, this literature
supports the idea that ideological beliefs are sensitive to threat, and
their impact is boosted in a context where societal threat is salient.
Suppor ting the idea that country-level factors might boost or
attenuate the effect that human values or ideological beliefs have
on attitudes toward immigration, Davidov et al. (2014) showed that
cultural differences influenced the relationship between values and
attitudes toward immigrant s. They found that countries with low
scores on embeddedness provided contex ts that activated values
to predict immigration attitudes more strongly because “in less em-
bedded societies people are socialized and encouraged to cultivate
their unique preferences and ideas and to pursue their own personal
goals” (p. 280). However, it seems unlikely that values should always
exert a stronger influence on at titudes in low-embeddedness con-
texts; rather, it seems more logic al that they should be activated
as guides to behavior in societies where they have formed habitual
associations with the object which is being evaluated (for possible
mechanisms, see Liu, Sibley, et al., 2014; Claassen & McLaren 2019).
In the present research, we argue that in some countries, immigra-
tion is a more salient topic due to societal discourses emphasized in
mass media, describing concerns about immigration or threat related
to immigrant groups (Claassen & McLaren 2019; Esses, et al., 2001;
Ettinger & Udris, 20 09; Powell, 2011; Reese, 2007). In this context,
values and ideological beliefs may be regularly invoked to guide peo-
ple’s attitudes toward these groups.
For the current study, we examine whether objective (interna-
tional migrant stock) and subjective indicators of threat (perception
of immigrants as a threat) create an atmosphere where individual
systems of beliefs are activated to guide people’s attitudes toward
immigrants. We also highlight the perceived threatening aspects of
a minorit y group (Muslims; perception of Islamic fundamentalism as
a threat) that might be considered as controversial in some specific
countries (e.g., European) due to the migrator y crisis and previous
terrorist attacks. Depending on the location, Muslim immigrants
might be par ticularly salient in people’s minds when they think
about immigration, and more likely to trigger an authoritarian mind-
set among locals (Claassen & McLaren 2019). In Europe in particular,
there may exist stronger resistance toward Muslim immigrants due
to the history of terrorist attacks claimed by Islamist groups or be-
cause of a general feeling of threat as a consequence of a history of
warfare and violent conflict in the Middle East (Echebarria-Echabe
538
|
ARAÚJO et Al.
& Fernández-Guede 2006; González, Verkuyten, Weesie, & Poppe,
2008).
Thus, we expect that perceptions of countr y-level threat will
moderate the strength of the relationship between SDO, RWA,
and values with attitudes toward immigrants. We expect that
suprapersonal, normative, promotion, and existence values, as
well as RWA and SDO, will have stronger effects on attitudes to-
ward immigrants in countries where (1) the perception of Islamic
Fundament alism as a threat is high (hypothesis 1); and (2) where
the perception of immigrants as a threat is high (hypothesis 2). The
relative strength of moderation produced by these two construc-
tions of societal threat will help to inform us about how immigra-
tion threat is constructed around the world today. Finally, (3) we
hypothesize that these effects will also be stronger where the mi-
grant stock is high (hypothesis 3), based on more traditional notions
of objective or realistic threat (see Cohrs & Stelzl, 2010).
|
Besides the main goal of providing empirical evidence for the cul-
tural moderation of threat, the present study also contributes to
the literature by examining a model in which ideological beliefs
mediate the relationship between values and attitudes toward im-
migrants. This idea builds on the dual process model (Duckitt &
Sibley, 2010) in which RWA and SDO are derived from different
types of motivators: RWA has foundations in the perception of the
world as threatening, while SDO is informed by a view of the world
as a competitive place (Duckitt, 2001). Values were not included in
the original dual process model; however, the belief systems that
precede RWA and SDO have a close similarit y with values. For in-
stance, considering the characteristics of RWA, such as submis-
sion to authority, it is plausible that normative values (e.g.,
obedience and tradition) might serve as predictors of this ideol-
og y.1 Using the same rationale, promotion values (e.g., power and
prestige) are expected to inform SDO due to its focus on dominant
groups and competitiveness (Feather & Mckee, 2012; Livi, Leone,
Falgares, & Lombardo, 2014). In addition, suprapersonal values,
represented, for instance, by maturity and tolerance, held by more
open-minded people with humanitarian characteristics such as
being likely to care for the welfare of others, are expected to ser ve
as a negative predictor of RWA and SDO (Cohrs, Kielman, Maes, &
Moschner, 2005). The mediational mechanisms of ideological be-
liefs on the relation between values and attitudes has been previ-
ously reported in the literature (see Feather & Mckee, 2012; Livi et
al., 2014), and we aim at replicating them in the present study.
Even though this is not the main objective of this ar ticle, we show
how these mediating mechanisms might contribute to an under-
standing of attitudes toward immigrants.
|
|
Data for this study come from a cross-cultural project (World Digital
Influence Project) with two waves of panel data with representative
samples from 20 countries in the Americas, Europe, Asia, and South
Africa. The measures used in this study were embedded in a larger sur-
vey including multiple psychological and communications constructs
(for further details, see Gil de Zúñiga et al. (2017); Liu, Milojev, Gil de
Zúñiga, & Zhang, 2018). The project was based at Massey University
(New Zealand) and at the Media Innovation Lab (MiLab; University of
Vienna, Austria). Participants (N = 21,362) were recruited online on
September 14–24, 2015, and several steps were taken in order to
ensure the quality of the data. A large group of competent scholars
worked on the translations and back-translations of the surveys. We
made use of Qualtrics, an online poll survey platform for data collec-
tion, and employed Nielsen, a well-known media polling company,
which partnered with local providers to generate a stratified quota
sample (on gender, age, and region) drawn from a massive pool with
over 10 million potential participants. The average duration for partici-
pants to finalize the survey was 35 minutes, and the response rate was
relatively high for a survey of this length (77%). Besides the questions
detailed on the measures section, the survey also had questions about
topics that are out of the scope of the current research (e.g., trust in
institutions, political participation, mental health). Rewards received
for participation varied across countries, as they were determined by
the local commercial survey panel provider. For further details of the
sample and country-level characteristics, see Table 1.
|
|
Basic values survey (BVS; Gouveia, 20 03)
This measure includes 18 items placed on a scale varying from 1
(completely unimportant) to 7 (of the utmost importance), where par-
ticipant s must indicate the level of importance they give to each
one of the specific values as a guiding principle to their lives (see
Gouveia et al., 2014, and see Figure 1 for a graphical representa-
tion). A previous study using the same sample as ours found in-
variance for all the six values dimensions using the approximate
Bayesian invariance approach, except for the item “religiosity”,
which was non-invariant (Vilar, under review). In the current re-
search, the item “religiosity ” was also excluded from the normative
dimension. We highlight that the models still have an item about
religiosity that is not part of the values scale. This item is described
in the control variables section.
Right Wing Authoritarianism
A brief 4-item version of the Right-Wing Authoritarianism (RWA)
Scale (Altemeyer, 1998) was used. Participant s responded to the
question “How much do you agree or disagree with the following
1 However, in co ntrast to this du al process mod el formu lation, both Fel dman (2003) an d
Claass en and McLaren ( 2019) have included va lues as part of t heir measure of RWA .
|
539
ATTITUDE S TOWARD IMMIG RANTS ACROSS 20 COU NTRIES
statements?” followed by the items (1) “Obedience and respect for
authority are the most important virtues children should learn”,
(2) “Our country needs a powerful leader, in order to destroy the
radical and immoral elements”, (3) “Young people should be allowed
to challenge their parents’ ways, confront established”, and (4) “In
these troubled times, laws have to be enforced without mercy, es-
pecially when dealing”. The response scale ranged from 1 (Disagree
Completely) to 7 (Agree Completely).
Social dominance orientation
A reduced version (4 items) of the Social Dominance Orientation
(SDO) Scale developed by Pratto and Stewart (2012) was used.
In a scale varying from 1 (Strongly Oppose) to 7 (Strongly Support),
participants should indicate “How much do you support or oppose
the following ideas about groups in general?”. The selected items
were (1) “In setting priorities, we must consider all groups”; (2) “We
should not push for group equalit y”; (3) “Group equalit y should
be our ideal”; and (4) “Superior groups should dominate inferior
groups”.
Attitudes toward Immigrants (feelings thermometer)
Participants were asked to rate their feelings toward immigrants
using a “feelings thermometer” on a scale varying from 1 (least warm)
to 7 (most warm). For this measure, attitudes toward immigrants
would follow the item “Please rate your feelings toward the follow-
ing groups using the ‘feelings thermometer scale’, with ‘most warm’
indicating ‘most positive’ and ‘least warm’ indicating ‘least positive’”.
Control variables
Participants were also required to answer some demographic ques-
tions for descriptive purposes, such as age and gender. Religiosity
and subjective social status were also used to control the effects
of values and ideological beliefs on attitudes toward immigrants.
Religiosit y was measured by a single item (how important is religion
to your life?) var ying from 1 (not important) to 4 (extremely important).
Subjective social status was also measured by a single item varying
from 1 to 10 (“On a scale of 1 to 10, with 10 being people who are
the most well off in society, and 1 being the people who are the least
well off, where would you describe your own position?”).
|
International migrant stock
International migrant stock was retrieved from the United Nations
website, Department of Economic and Social Affairs (http://www.
un.org/en/devel opmen t/desa/popul ation/ migra tion/data/estim
NSD)♀
Argentina 1,145 40.8 (14 .3 ) 51. 7 4.69 2.05 2.086302
Brazil 1,086 35.9 (12.0) 49.8 4.84 2.86 713, 568
Chile 964 35.1 (13.1) 51. 3 4.26 2.40 469,436
China 1,005 38 .7 (12.0 ) 44.4 4. 59 3.79 978,046
Estonia 1,16 8 47.8 (17.2) 48.8 5.37 4.57 202,348
Germany 1,093 45.1 (15.1) 53.9 5.35 2.99 12.005690
Indonesia 1,080 32.8 (9.8) 39. 5 4.33 3.33 328,846
Italy 1, 041 41.7 (12 .2) 55.3 5.39 3. 24 5.788875
Japan 975 46.7 (12.9) 42 .2 4.62 2.90 2. 043877
Korea 943 38.9 (12.7) 46.7 4 .90 3. 47 1.327324
New Zealand 1,157 49.5 (17.3) 56.4 5.23 3.29 1.039736
Philippines 1,064 34.2 (11.0) 61. 2 4.70 3.49 211,862
Poland 1,060 42 .0 (14.5) 54.0 5 .69 4.68 619,4 0 3
Russia 1,145 38.2 (12.8) 50.9 4.80 3.73 11 .6 4 3276
Spain 1,019 41. 0 (12. 6) 52.6 5.35 2.79 5.852953
Taiwan 1,008 36.3 (10.9) 50.8 4.77 4.02 2.8386 65a
Tur ke y 961 33.9 (11.0) 44.0 3.80 2.94 2 .964916
Ukraine 1,223 33.9 (9.40) 44.0 4.21 3.17 4.834898
United
Kingdom
1,064 50.6 (15.6) 54.1 5.65 4.00 8.543120
United States 1,161 49.8 (16.4) 59. 5 5.00 3. 31 46.627102
Abbreviations: IMS, international migrant s tock; IMT, perception of immigrants as a threat; ISF,
Perception of Islamic fundamentalism as a threat.
aScore from Hong Kong, China.
TABLE 1 Summary of sample and
country-level moderators
540
|
ARAÚJO et Al.
ates2/ estim ates15.shtml ). This indicator is based on official statis-
tics of the foreign-born or the foreign population from 2015. We
used the percentage of migrants in the total population of a countr y
from this dataset.
Islamic fundamentalism and immigration as a threat
We collected data using a single-item measure to capture the extent
to which people believe that Islamic fundamentalism is the greatest
threat to world peace today, and another item to capture the extent
to which people believe that immigrants are the greatest threat to
world peace today. These two items var y from 1 (completely disagree)
to 7 (completely agree). To use this information as a country-level
variable, we calculated their means by country.
|
Bayesian multilevel-linear regressions were employed to accom-
modate the nested structure of our data (using Mplus 7.31). All the
regressions were estimated using the Bayesian approach because
of the small number of level-2 units (i.e., country-level units) in our
sample (Hox, 2010; Stegmueller, 2013). According to Muthén and
Muthén (2010), Bayesian estimation shows advantages over analyses
that use maximum likelihood estimators because it does not rely on
large-sample theories of normality, and also does not assume sym-
metric distribution of the data. Kruschke, Aguinis, and Joo (2012)
articulate nicely the benefits of Bayesian estimation over frequentist
approaches (involving a null hypothesis significance test). Some of
these benefits are: (a) Bayesian estimation provides complete and
accurate distributional information regarding the parameters in the
regression model, while the frequentist approach creates only a local
approximation; (b) the Bayesian approach is flexible regarding non-
normal data and unsusceptible to sample size, unlike the frequentist
approach, which can vary in significance according to sampling; and
(c) instead of using a fragile p-value, Bayesian estimation uses ro-
bust posterior distribution and credible intervals based on real data.
Bayesian estimation also allows the researcher to use prior accu-
mulated knowledge to generate parameters (Kruschke et al., 2012).
However, as knowledge about our individual-level variables and the
dependent variable is still being developed, we decided to use the
default broad prior distribution (0.001). In this case, our parameters
are close to what would have been obtained by a ML estimator, but
Bayesian estimation improves confidence in deciding whether the
results are credible or not.
Using a multilevel procedure allows us to examine whether the
pattern of individual variables in each culture is universal or influ-
enced by country-level features ( Tabachnick & Fidell, 2007). First,
we computed the intraclass correlation coefficient (ICC) by including
the random intercept only, to check whether attitudes toward immi-
grants showed variability between countries. We then included the
individual-level predictors. Control variables (age, gender, religiosity,
and subjective social status) were estimated with random intercepts.
Values and ideological beliefs (RWA and SDO) were estimated with
random intercepts and random slopes because the cross-cultural
variability on these predictions was the focus of interest in this re-
search. Results fo r the individual-level effec ts were assessed in three
different models with each construct being added at a time until we
had all the predictors included. This procedure was adopted to check
what new predictors would add over predictors introduced in previ-
ous steps. The coefficient s that were not statistically different from
zero (credibility interval falling below or above zero) were excluded
in the subsequent models. Finally, each of the country-level predic-
tors and cross-level interactions were added. We tested a model for
one countr y-level moderator at a time (Models 4, 5, and 6), and then
two models with all the moderators competing for variance in the
same analysis (Models 7 and 8). Model 8 included the level-2 vari-
ables of perception of immigrant s and Islamic fundamentalism as a
threat at the individual level as well (level-1). For these analyses, we
group-mean-centered individual-level variables and we z-standard-
ized country-level variables. Participants’ characteristics (age, gen-
der, religiosit y, and social class) and psychological variables (RWA,
SDO, and human values) were placed as level-1 (individual) variables.
Perception of immigrants and Islamic fundamentalism as a threat,
and international migrant stock were the level-2 (societal) variables.
|
Sample characteristics are reported in Table 1. Descriptive statistics,
reliabilities, and correlations per country are reported as Appendices
S1–S4. The Bayesian multilevel regression models are shown in
Table 2. Figure 2 shows cross-level interactions.
|
Before checking the influence of individual-level and country-level
variables on attitudes toward immigrants, we computed the Intra
Class Correlation (ICC). Results suggested that attitudes toward im-
migrants were different across nationalities (ICC = 0.129, 95%
CI = 0.076, 0.234). This variance justifies the further evaluation of
the individual and country-level predictors (Hox, 2010). Then, indi-
vidual-level predictors were tested in three models. Model 1 in-
cluded demographics, level of religiosity and subjective social st atus.
These variables were added to control the effects of our constructs
of interest (values and ideological beliefs). Human values were added
in Model 2, and SDO and RWA2 were added in Model 3.
2 As valu es and ideolog ical be liefs are h ighly correla ted construc ts, we tested wh ether
their inclusion in the same analysis would produce multicollinearity. Examining the
varian ce inflation fa ctors (VIF ) for each predic tor dispelle d this conc ern (VIF < 3.0) .
Asses sing values and i deological b eliefs togeth er is relev ant because th ey are related, ye t
indep endent c onstr ucts t hat repre sent beliefs (C ohrs, Moschn er, Maes, & Kielma nn,
2005) . Althou gh some author s suggest an over lap of content amo ng values, SDO, a nd
RWA (e.g., vo n Collan i & Grumm , 2009), the mod est corr elations bet ween these
constr ucts sugge st that on e can add increm entally to the ot her when predi cting
depen dent measures . A similar appro ach has been ado pted in re cent lite rature o n
perso nality and val ues. Even thoug h person ality and valu es are thought t o repres ent an
integr ative model of cha racterist ics of the indivi dual, the mode st correlati ons bet ween
these t wo construc ts have driven sch olars to invest igate how they mig ht conjointly
influe nce external o utcome s (Parks & G uay, 2009; Park s‐Leduc , Feldman, & Bar di, 2015;
Volk, Tho ni, & Ruigrok, 2011).
|
541
ATTITUDE S TOWARD IMMIG RANTS ACROSS 20 COU NTRIES
Results showed that females, young people, and people with
high levels of religiosity and high self-reported social status had a
more positive view of immigrants. Considering values and ideologi-
cal beliefs, results showed that RWA, SDO, and existence values
predicted attitudes toward immigrants negatively, while supraper-
sonal and interactive values were positively associated with atti-
tudes toward immigrants. Promotion values also showed relations
statistically different from zero when predicting attitudes toward
immigrants, but this relation became non-statistically different from
zero when ideological beliefs were added in the model. The effects
of interactive and suprapersonal values decreased with the
inclusion of SDO and RWA, but results remained statistically differ-
ent from zero.3
3 An impo rtant concer n to be addresse d is whether peo ple high in R WA, SDO a nd
existe nce values woul d be cold toward any o utgrou p, or only toward sp ecific group s (e.g.,
immigr ants). We conduc ted extra anal yses with att itudes t oward blacks an d Americans
as depe ndent va riable s to address this i ssue. Results f or attitudes t oward immigra nts and
black s, usual ly seen as less ad vantag ed group s, showed largel y identical pa tterns of
relati ons with both ide ological bel iefs and h uman values. H owever, attitude s toward
Ameri cans, a domina nt group in t he internatio nal arena, show ed a disti nctly diffe rent
patte rn: RWA ser ved as a positive pr edictor, and SDO s howed only small n egative
correl ations. Supr apers onal and e xistence valu es were also stat ically diff erent from zero
for att itudes toward A mericans, an d in the same dire ction as attit udes toward
immigr ants. However, in co ntrast with at titudes towar d negative refer ence groups,
promot ion values were a lso positivel y and stro ngly associat ed with at titudes towa rd
Ameri cans. These re sults are avail able upon requ est to the autho rs.
TABLE 2 Bayesian multilevel regression models on attitudes toward immigrants
Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7 Model 8
Intercept 4.044a4.067a4.073a4. 065a4.062a4.065a4.058a4.068a
Individual level
Age −0.008a−0.010a−0.006a−0.006a−0.006a−0.006a−0.006a−0.0 06a
Gender(1=M, 2=F) 0.170 a0.125a0.114 a0.121a0.121a0.121a0.121a0.101 a
Religiosity 0.069a0.043a0.099a0.099a0.099a0.099a0.099a0.097a
Social class 0.071a0.069a0.067a0.066a0.066a0.066a0.066a0.071a
Interactive 0.164 a0.116 a0.113 a0.111 a0.113 a0.114 a0.102 a
Normative −0.049 ——————
Suprapersonal 0.267a0.165a0 .154 a0.158a0.156a0.153a0.169a
Existence −0.10 3a−0 .101a− 0.101a−0.10 2a−0 .103a−0.101a−0.138a
Excitement −0.039 ——————
Promotion −0.075a0.021 —————
SDO −0.233a−0.236a−0.233a−0.236a−0.236 a−0.165a
RWA −0 .167a−0.171a−0 .169a−0.171a− 0.169a− 0.10 0a
ISF −0.015a
IMT −0.264a
Cross-level interactions
ISF × Interactive −0.016 0.003 −0.003
ISF × Suprapersonal 0.071a0.092a0.079a
ISF × Existence −0.085a−0.097a−0.078a
ISF × SDO −0.096a−0.088a−0.070a
ISF × RWA −0.079a−0.083a−0.055
IMT × Interactive −0.019 −0.018 −0.010
IMT × Suprapersonal 0.005 −0.039 −0.035
IMT × Existence −0.023 0.022 0.042
IMT × SDO −0.004 0.042 0.027
IMT × RWA −0.004 0.037 0.029
IMS × Interactive −0.021 −0.021 −0.025
IMS × Suprapersonal 0.037 −0.006 −0.019
IMS × Existence −0.046 −0.001 0.017
IMS × SDO −0.098a−0.056a−0.037a
IMS × RWA −0.066a−0.026 −0.003
[Correc tion made on 17 Januar y 2020 after firs t online publication: Table 2 was replaced with the current version which now shows the values for
Cross-level interac tions in the correct columns.]
Note.: A t able with credibility intervals of all the parameters is available in the Supplementar y materials.
Abbreviations: IMS, International Migrant Stock; IMT, Immigrants as a threat; ISF, Islamic Fundamentalism as a threat.
aResults with the 95% C.I. falling below or above zero.
542
|
ARAÚJO et Al.
|
Cross-level interactions for all country-level moderators were
checked (1) with each moderator being examined separately (one
at a time), and (2) with all the moderators included in the analysis
at the same time. Correlations between the three types of threat
were moderate or low (r < .485). In line with our hypotheses, re-
sults showed moderation effects for the perception of Islamic
fundamentalism as a threat (hypothesis 1), and also international mi-
grant stock (hypothesis 3) on predictions of values and ideological
beliefs on attitudes toward immigrant s.
Analyses revealed that in countries where the Islamic funda-
mentalism was perceived as more threatening at the country-level,
the effects of SDO, RWA, and suprapersonal and existence values
on attitudes toward immigrants were intensified (Model 4). For
countries where international migrant stock was high, the effect of
SDO and RWA on attitudes toward immigrants was higher than in
countries where the international migrant stock was low. However,
there were no country-level moderation effects for percentage of
migrant stock on the relation of values and at titudes toward immi-
grants (Model 6). With regard to the countr y-level moderation ef-
fect of perception of immigrants as a threat (hypothesis 2), none of
the results was statistic ally dif ferent from zero (Model 5). We then
examined a model in which all the cross-level interactions were
added simultaneously (Model 7). Results showed that the percep-
tion of Islamic fundamentalism as a threat was the most consistent
cultural-level moderator. Additionally, as the perception of Islamic
fundamentalism as a threat and the perception of immigrants as a
threat were collected at the individual level (these country-level
moderators were country means of items collected in the survey),
we examined a final model adding these variables at both the indi-
vidual and country-levels (Model 8). Result s showed lower cross-
level interactions for the perception of Islamic fundamentalism
and immigrants as threats, but most of the effects identified pre-
viously remained statistically different from zero. These findings4
provide the basis for rejecting hypothesis 2, but provided partial
support for hypotheses 1 and 3. Figure 2 shows cross-level inter-
actions for Islamic fundamentalism as a moderator (results are
from Model 4).
|
In addition to the previous analyses examining country-level mod-
eration of the effects of values and ideological beliefs on attitudes
toward immigrants, we provide complementary information about
the mediational mechanisms involving values, ideological beliefs,
and attitudes toward immigrants. We aimed at contributing to the
literature reporting that ideological beliefs mediate the relationship
between values and prejudicial outcomes (e.g., Duckitt & Sibley,
2010; Feather & Mckee, 2012; Livi et al., 2014). Results showed that
RWA and SDO fully mediated the effec t of promotion values on at-
titudes toward immigrants, but only partially mediated the effects of
interactive, suprapersonal and existence values on attitudes toward
immigrants (these results are available in the Appendices S1–S4). As
4 In the Ap pendices S1–S4 we inclu de analy ses using Human D evelopmenta l Index as a
countr y-level moder ator. We did this to exam ine whet her moderati on effects wo uld also
be more co mmonplace in de veloped count ries, since the se countries ar e more commonly
a migrat ion destinat ion. The p attern of count ry-level mode ration for HDI w as simila r to
the one fo r the percepti on of Islam ic fundament alism as a threat . These two var iables
were mod erately highl y correlated. T he latter was a st ronger predic tor when all
moderators were included simultaneously.
FIGURE 2 Cross-level interactions. Note: Each figure shows
the variation of strength of the individual-level regressions
(Y-axis = unstandardized regressions slopes of individual-level
predictors) when the country-level moderator is accounted (X-
axis = standard deviations of Islamic fundamentalism). Black line is
the regression for dif ferent levels of the moderator and gray line is
the 95% credibility interval. DV = Attitudes toward immigrants. This
figure is referent to Model 4 of Table 2.
–0.5
–0.3
–0.1
0.1
0.3
0.5
–2
–1.8
–1.6
–1.4
–1.2
–1
–0.8
–0.6
–0.4
–0.2
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
Slope of suprapersonal on DV
–0.5
–0.3
–0.1
0.1
0.3
0.5
–2
–1.8
–1.6
–1.4
–1.2
–1
–0.8
–0.6
–0.4
–0.2
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
Slope of existence on DV
–0.5
–0.3
–0.1
0.1
0.3
0.5
–2
–1.8
–1.6
–1.4
–1.2
–1
–0.8
–0.6
–0.4
–0.2
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
Slope of SDO on DV
–0.5
–0.3
–0.1
0.1
0.3
0.5
–2
–1.8
–1.6
–1.4
–1.2
–1
–0.8
–0.6
–0.4
–0.2
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
Slope of RWA on DV
Perception of Islamic fundamentalism as a threat
(country-level moderator)
|
543
ATTITUDE S TOWARD IMMIG RANTS ACROSS 20 COU NTRIES
expected, RWA was mainly informed by normative and supraper-
sonal values, but it was also moderately related to promotion values,
which was expected to be correlated with SDO only. The expected
relations between SDO and the values of promotion and supraper-
sonal were observed.
|
The current research examined the role of RWA, SDO, and human
values in shaping attitudes toward immigrants when considering
the moderating effects of objective and subjective country-level
indicators of threat. We examined one objec tive measure of
potential threat, which was the percentage of international migrant
stock in one’s country, and t wo subjective measures of threat, which
referred to country-level perceptions of threat regarding immigration
and Islamic fundamentalism. There were cross-level interactions
statistically different from zero only for Islamic fundamentalism and
international migrant stock as moderators. Of these two, the threat
of Islamic fundament alism was far more pervasive: The predictive
links bet ween SDO, RWA, suprapersonal values, and existence
values intensified in relation to attitudes toward immigrants in
countries where Islamic fundamentalism was considered more
threatening to society.
This pattern of threat-based moderation is difficult to explain
using traditional cross-cultural theory, which relies on enduring fea-
tures of a society, such as individualism or collectivism, to produce
moderation (Bond, 2019). Following the logic of traditional cross‐cul-
tural psychology, immigration as a threat together with percentage
of migrant stock would be more plausible candidates for producing
moderation than the threat posed by Islamic fundamentalism. This is
because the former index long-term features of culture (e.g., some
countries have long been receiving nations for immigration, others
not), whereas Islamic fundamentalism would only have been per-
ceived as a major threat after 9/11 in most countries (e.g., Powell,
2011).
A plausible account for this pattern of result s can be provided
by the idea of framing in communications theory. According to
Reese (20 07), “Frames are organizing principles that are socially
shared and persistent over time, that work symbolically to mean-
ingfully structure the social world” (p. 150). Ettinger and Udris
(2009), Lewis and Reese (2009), and Powell (2011) (among others)
have put together compelling accounts of how mass media and
political discourse post‐9/11 have combined to produce a frame
where the “War on Terror”, portraying Muslims as terrorist s,
and conflating international terrorist s with local Muslims serve
as organizing principles that structure ordinary citizens’ under-
standings of Islam and make salient the idea that immigration is a
threat to the national way of life. This threatening frame activates
prejudicial predispositions (like SDO, RWA, and existence values),
making them more influential on attitudes toward immigrant s (and
Muslims; see Strabac & Listhaug, 2008). These findings are coher-
ent with previous studies (e.g., Claassen & McLaren 2019), that
state that (a) the potential threat to conformity and the status quo
activates authoritarian behavior, and (b) an increase in authoritar-
ianism is particularly observed when immigrants have an Islamic
background.
However, this does not explain why the link between supraper-
sonal values and more positive feelings toward immigrants was also
streng thened in the moderation analyses. We suspect that this link
might be the by-product of the high level of human development
(HDI) in these countries (that co-varied at the country-level with the
perception of Islamic fundamentalism as a threat; see Appendices
S1–S 4 ).
With regard to international migrant stock as a countr y-level
variable, our results showed that cross-level interac tions were sta-
tistically different from zero only for the relationship bet ween ideo-
logical beliefs (RWA and SDO) and feelings toward immigrants (but
not for values). When all the cross-level interactions were included
simultaneously, only the interaction between international migrant
stock and SDO remained statistically different from zero. This lack
of significant cross-level interactions for an objective country-level
measure goes in line with the idea that there is a subjective bias re-
lated to a specific group (Muslims) in countries with high perception
of Islamic fundamentalism as a threat (Claassen & McLaren 2019).
In countries like Europe and the United States, values appear to be-
come implicated in attitudes toward immigrants because of subjec-
tive salience rather than objective factors. Complementing framing
theory, Esses, Medianu, and Lawson (2013) argued independently
that the media in Western countries tend to spread a negative view
of immigrants and refugees by dehumanizing them. Esses et al.
(2013, p. 518) state that “the media portrays immigrants and refu-
gees as ‘enemies at the gate’ who are attempting to invade Western
nations”.
The consistency of the effects of Islamic fundamentalism as a
moderator shows the strength that a subjective perception of threat
has on people’s thought s and feelings toward immigrants. This goes
in line with the studies developed by Esses et al. (2013), and espe-
cially by Betz and Meret (2009), which revealed how much Islamic
immigration is influencing political mobilization and decision-making
regarding policies in European countries. Such conflation leads to
the idea that one way to reduce barriers between immigrants and
the host population would be to focus attention on and develop
deeper understanding of specific immigrant groups.
Finally, the mediational model examined sug gested an inter-
esting path in which ideological beliefs mediate the relationship
between values and attitudes toward immigrants. Although this
mediational mechanism has already been tested to predict attitudes
toward women and patriotism (Feather & McKee, 2012; Livi et al.,
2014), this is the first time that it has been examined in a large global
dataset, with a focus on attitudes toward immigrants. The path
model contributes to expanding the range of the dual process model
by showing that RWA and SDO are informed by different values
(e.g., normative values predic t RWA only), amplifying the conceptual
independence of RWA and SDO (Duckitt, 2001; Duckit t & Sibley,
2010). Nevertheless, there was an unanticipated effect of promotion
544
|
ARAÚJO et Al.
values on RWA, crossing the two theoretically hypothesized medi-
ational paths. Power and prestige were expected to drive the pref-
erence for group inequality/SDO (Cohrs, Kielman, et al., 2005; Livi
et al., 2014), but our findings showed a stronger effect of promotion
values on RWA than SDO.
|
While this research documented cross-cultural moderation
effects of threat on the predictors of feelings toward immigrants,
it has two major limitations as well. The first limitation is that we
did not measure feelings toward different groups of immigrants.
This prevents more fine-grained analyses of what predicts
people’s feelings toward different immigrant groups, and what
determines the salience of these different groups in producing
overall feelings toward immigrants. Secondly, as this was our first
foray into examining cross-level moderation effects, we did not
take independent measures of media salience or media framing
to measure objective societal threat. Essaying such measures is
a major and difficult undertaking that requires a consideration of
both qualitative and quantitative approaches to empirical data.
According to Reese (2007), a frame is an organizing principle
that should be measured by more than just text counts. It is an
interpretive resource that warrants further examination in future,
perhaps interdisciplinary, research.
|
This stu dy examined the m oderating role of p erceptions of subje ctive
and objective indicators of threat in the relationship between human
values and ideological beliefs (SDO and RWA) and attitudes toward
immigrants. Threat derived from Islamic fundamentalism and thus
a specific outgroup (Muslims) mattered at the countr y-level: The
link between SDO, RWA, suprapersonal, and existence values and
attitudes toward immigrants intensified in countries where Islamic
fundamentalism was considered more threatening. This suggests
that the polarizing discourse in Western countries on immigration
from Islamic countries as a threat to the integrity and security of
Western societies serves as a basis for endorsing beliefs related
either to security (existence values) or to tolerance (suprapersonal
values and low levels of SDO) to activate hostile versus favorable
attitudes toward immigrant s, respectively. That is, societal threat
may modify the way human values or ideological beliefs shape
responses toward immigrants. These findings may serve as a basis
for future research oriented at deconstructing imagined threats
(Kinvall, 2017), construed both as personal fears about one’s security
and wellbeing and in media discourses as a threat to national security
or people’s ways of life.
The authors confirm they have no conflict of interest to declare.
Authors also confirm that this article adheres to ethical guidelines
specified in the APA Code of Conduct and the authors’ national eth-
ics guidelines.
The data is publically available at https ://osf.io/jwkmt/ .
Rafaella de C. R. Araújo https://orcid.
org/0000-0001-8548-6238
Magdalena Bobowik https://orcid.org/0000-0002-2341-0665
Roosevelt Vilar https://orcid.org/0000‐0002‐9414‐6080
James H. Liu https://orcid.org/0000‐0001‐9520‐5727
Homero Gil de Zuñiga https://orcid.org/0000-0002-4187-3604
Nadezhda Lebedeva https://orcid.org/0000‐0002‐2046‐4529
Valdiney V. Gouveia https://orcid.org/0000-0003-2107-5848
Altemeyer, B. (1981). Right‐wing authoritarianism. Winnipeg, MB:
University of Manitoba Press.
Altemeyer, B. (1998). The other “authoritarian personality”. Advance s in
Experimental Social Psychology, 30, 47–92. ht tps ://doi.org/10.1016/
S0 0 65-2601( 08)60 382-2
Beierlein, C., Kuntz, A., & Davidov, E. (2016). Universalism, conservation and
attitudes toward minority groups. Social Science Research, 58, 68–79.
Betz, H., & Meret, S. (2009). Revisiting Lepanto: The political mobili-
zation against Islam in contemporar y Western Europe. Patterns of
Prejudice, 43, 313–334.
Blumer, H. (1958). Race prejudice as a sense of group position . Pacific
Sociological Review, 1, 3–7.
Bond, M. H. (2019). Travelling from the past into the future of cross‐cul-
tural psycholog y: A personal-scientific journey. In D. Matsumoto &
H. C. Hwang (Eds.), The handbook of culture and psychology (pp. 45–
66). New York, NY: Oxford University Press.
Bourhis, R. Y., & Dayan, J. (2004). Acculturation orientations towards
Israeli Arabs and Jewish immigrant s in Israel. International Journal of
Psychology, 39, 118–131.
Brader, T., Valentino, N. A., & Suhay, E. (20 08). What triggers public
opposition to immigration? An xiety, group cues, and immigr ation
threat. American Journal of Political Science, 58, 959–978.
Claassen, C ., & McLaren, L . (2019). Do threats galvanize authoritarians or
mobilize non‐authoritarians? Experimental tests from 19 European soci‐
eties. Retrieved from SSRN http://dx.doi.org/10.2139/ssrn.3278685
Cohrs, J. C., & Asbrock, F. (2009). Right‐wing authoritarianism, social dom-
inance orientation and prejudice against threatening and competitive
ethnic groups. European Journal of Social Psychology, 39, 270–289.
Cohrs, J. C., Moschner, B., Maes, J., & Kielmann, S. (2005). Personal val-
ues and at titudes toward war. Peace and Conflict: Journal of Peace
Psychology, 11, 293–312.
Cohrs, J. C., & Stelzl, M. (2010). How ideological attitudes predic t host
society members’ attitudes toward immigrants: Exploring cross-na-
tional differences. Journal of Cross‐Cultural Psychology, 49, 789–810.
Cohrs, J. C., Kie lman, S., Maes, J., & Moschner, B. (2005). Ef fects of Right-
Wing Authoritarianism and threat from terrorism on re stric tion of
civil liberties. Analyses of Social Issues and Public Policy, 5, 26 3–2 76 .
|
545
ATTITUDE S TOWARD IMMIG RANTS ACROSS 20 COU NTRIES
Cook, C . L., Cottrell, C. A., & Webster, G. (2015). No good without God:
Antiatheist prejudice as a function of threat s to morals and values.
Psychology of Religion and Spirituality, 7, 2 17–2 26 .
Costello, K., & Hodson, G. (2011). Social dominance-based threat reac-
tions to immigrant s in need of assistance. European Journal of Social
Psychology, 41, 2 20 –231.
Craig, M . A., & Ri cheson , J. A. (2013). N ot in my back yard! Aut horita rianism,
social dominance orientation, and support for strict immigration poli-
cies at home and abroad. Political Psychology, 35, 417–429.
Crandall, C., D'Anello, S., Ugurlu, N. S., Lazarus, E., Wierzbińska, G. W., &
Feather, N. T. (2001). An attribution-value model of prejudice: Anti-fat at-
titudes in six nations. Personality and Social Psychology Bulletin, 27, 30–37.
Crawford, J. T., & Pilanski, J. M. (2013). Political intolerance,
right and left. Political Psychology, 35, 841–851. https ://doi.
org/10.1111/j.1467‐9221.2012.00926.x
Danso, H. A., Sedlovskaya, A., & Suanda, S. H. (2007). Perceptions of immi-
grants: Modifying the attitudes of individuals higher in social dominance
orientation. Personality and Social Psychology Bulletin, 33, 1113–1123.
Davidov, E., Meuleman, B., Billiet, J., & Schmidt, P. (2008). Values and
support for immigration: A cross-country comparison. European
Sociological Review, 24, 583–599.
Davidov, E., Meuleman, B., Schwartz, S. H., & Schmidt, P. (2014).
Individual values, cultural embeddedness, and anti-immigration sen-
timents: Explaining differences in the ef fect of values on attitudes
toward immigration across Europe. Kölner Zeitschrift für Soziologie
und Sozialpsychologie, 66, 263 –28 5.
Duckit t, J. (20 01). A dual process cognitive-motivational theory of ide-
ology and prejudice. Advances in Experimental Social Psychology, 22,
41–1 13 .
Duckit t, J., & Sibley, C. G. (2010). Right-wing authoritarianism and social
dominance orientation differentially moderate intergroup effects on
prejudice. European Journal of Personality, 24, 583–601.
Echebarria-Echabe, A., & Fernández-Guede, E. (20 06). Effects of ter-
rorism on attitudes and ideological orientation. European Journal of
Social Psychology, 36, 259–265.
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, 518–536.
Ettinger, P., & Udris, L. (2009). Islam as a threat? Problematisation of
Muslims in the mass media and ef fects on the political system. In L.
Marsden & H. Savigny (Eds.), Media, religion, and conflict (pp. 59–80).
Farnham, UK: Ashgate Publishing.
Feather, N. T., & Mckee, I. R. (2012). Values, right-wing authoritarian-
ism, social dominance orientation, and ambivalent attitudes toward
women. Journal of Applied Social Psychology, 42, 2479–2504.
Feldman, S. (2003). Enforcing social conformit y: A theory of authorit ari-
anism. Political Psychology, 24, 41–7 4.
Gijsberts, M ., Scheepers, P., & Coenders, M . (200 4). Exclusion of legal
migrant s in Wester n Europe. In M. Gijsberts, L. Hagendoom, & P.
Scheepers (Eds.), Nationalism and exclusion of migrants: Cross‐national
comparisons (pp. 41–60). Aldershot, UK: A shgate.
Gil de Zúñiga, H., Diehl, T., Huber, B., & Liu, J. H. (2017). Personality traits
and social media use in 20 countries: How personality relates to fre-
quency of social media use, social media news use, and social media use
for social interaction. Cyberpsychology, Behavior, and Social Networking,
20(9), 540–552. https ://doi.org/10.1089/cyber.2017.0295
González, K. V., Verkuyten, M., Weesie, J., & Poppe, E. (2008). Prejudice
towards Muslims in The Netherlands: Testing integrated threat the-
or y. The British Psychological Society, 47, 6 67– 685 .
Gouveia , V. V. (2003). A Natureza Motivacional dos Valores Humanos:
Evidências acerca de uma Nova Tipologia. Estudos de Psicologia
(Natal), 8, 431–443.
Gouveia , V. V. (2013). Teoria funcionalista dos valores humanos:
Fundamentos, aplicações e perspectivas futuras [Functional theory of
human values: Fundaments, applications, and future perspectives]. São
Paulo, Brazil: Casa do Psicólogo.
Gouveia , V. V., Milfont, T. L., & Guerra, V. M. (2014). Functional theory
of human values: Testing its content and structure hypotheses.
Personality and Individual Differences, 60, 41– 47.
Henrich, J., Heine, S. J., & Norenzayan, A. (2010). The weirdest people in
the world? Behavioral and Bran Sciences, 33, 6 1–1 35 .
Hjerm, D. (2007). Do numbers really count? Group threat theory revis-
ited. Journal of Ethnic and Migration Studies, 33, 1253–1275.
Hox, J. J. (2010). Multilevel analysis: Techniques and applications. New
York, NY: Routledge.
Imhoff, R ., & Recker, J. (2012). Differentiating Islamophobia: Introducing
a new scale to measure Islamoprejudice and secular Islam critique.
Political Psychology, 33, 811–824.
Kinnvall, C. (2017). R acism and the role of imaginar y others in Europe.
Nature Human Behaviour, 1. https ://doi.org/10.1038/s41562-017-012
Kruschke, J. K., A guinis, H., & Joo, H . (2012). The time has come:
Bayesian methods for data analysis in the organizational sciences.
Organizational Research Methods, 15, 722–752.
Landis, D., & R. D. Albert (Eds.) (2012). Handbook of ethnic conflict:
International perspectives. New York, NY: Springer.
Leong, C . H., & Ward, C. (2006). Cultural values and attitudes toward
immigrants and multiculturalism: The case of the Eurobarometer sur-
vey on racism and xenophobia. International Journal of Intercultural
Relations, 30, 799–810.
Lewis, S. C., & Reese, S. D. (2009). What is the war on terror? Framing
through the eyes of journalists. Journalism & Mass Communication
Quarterly, 86(1), 85–102.
Liu, J. H., Fisher Onar, N., & Woodward, M. (2014). Symbologies, technol-
ogies, and identities: Critical Junctures Theory and the multi-layered
Nation-State. International Journal of Intercultural Relations, 43, 2–12.
https ://doi.org/10.1016/j.ijint rel.2014.08.012
Liu, J. H., Milojev, P., Gil de Zúñiga, R., & Zhang, J. R. (2018). The global
trust i nventory as a “Prox y Measure” for So cial Capital: M easurement
and impact in 11 democratic societies. Journal of Cross‐Cultural
Psychology, 49, 789–810.
Liu, J. H., Sibley, C. G., & Huang, L. L. (2014). History matters: The impact of
culture-specific symbols on political attitudes and intergroup relations.
Political Psychology, 35, 57–79. https ://doi.org/10.1111/pops.12027
Liu, J. H., & Mills, D. (20 06). Modern racism and market fundamentalism:
The discourses of plausible deniability and their multiple functions.
Journal of Community and Applied Social Psychology, 16, 83–99.
Livi, S., Leone, L ., Falgares, G., & Lombardo, F. (2014). Values, ideological
attitudes and patriotism. Personality and Individual Differences, 64,
141–146.
Mahroum, S. (2001). Europe and the immigration of highly skilled labour.
International Migration, 39, 27–4 3.
Muthén , L. K., & Muthén, B. O. (2010). MP lus statistical a nalysis with latent
variables user’s guide. Los Angeles, C A: Muthén & Muthén.
Parks, L., & Guay, R. P. (2009). Personalit y, values, and motivation.
Personality and Individual Differences, 47, 675–684.
Parks-Leduc, L ., Feldman, G., & Bardi, A . (2015). Personalit y traits and
personal values: A meta-analysis. Personality and Social Psychology
Review, 19, 3–2 9.
Powell, K. A. (2011). Framing Islam: An analysis of U.S. media coverage of
terrorism since 9/11. Communication Studies, 62(1), 90–112. https ://
doi.org /10.1080/10510 974.2011. 533599
Pratto, F., & Stewart, A . L. (2012). Group dominance and the half-
blindness of privilege. Journal of Social Issues, 68(1), 28–45. https ://
doi .or g/10.1111/ j.15 40-4560.2011.01734. x
Pratto, F., Sidanius, J., & Levin, S. (2006). Social dominance theory and
the dynamics of intergroup relations: Taking stock and looking for-
ward. European Review of Social Psychology, 17, 271–320.
546
|
ARAÚJO et Al.
Quillian, L. (1995). Prejudice as a response to perceived group threat:
Population composition and anti-immigrant and r acial prejudice in
Europe. American Sociological Review, 60, 586 –611 .
Ramos, A . (2011). Human values and opposition towards immigration in
Europe. PhD thesis, Institute of Social Sciences, University of Lisbon.
Ramos, A ., Pereira, C. R ., & Vala, J. (2016). Economic crisis, human values
and attitudes towards immigrants. In M. Voicu, I . C. Mochmann, & H .
Dülmer (Eds.), Values, economic crisis and democracy (pp. 104–137).
London, UK and New York, NY: Routledge.
Reese, S. D. (2007). The framing project: A bridging model for media re-
search revisited. Journal of Communication, 57(1), 148–154.
Sagiv, L., & Schwartz, S. H. (1995). Value priorities and re adiness for out‐
group social contact. Journal of Personality and Social Psychology, 69,
43 7– 4 48.
Saroglou, V., & Galand, P. (2004). Identities, values, and religion: A study
among Muslim, other immigrant, and native Belgian young adults
after the 9/11 attacks. An Internatio nal Journal of Th eory and Resea rch,
4, 97–132.
Schwar tz, H. S ., Sagiv, L., & Boehnke, K. (20 00). Worries and values.
Journal of Personality, 68, 309–346.
Schwar tz, S. H. (1992). Univer sal in the content and structure of val-
ues: Theoretical advances and empirical tests in 20 countries. In
M. P. Zanna (Ed.), Advances in experimental social psychology (Vol. 25),
(pp. 1–65). New York, NY: Academic Press.
Schwar tz, S. H . (2006). A theor y of cultur al value orientations: explica-
tion and applications. Comparative Sociology, 5, 1 32 –187.
Schwartz, S. H. (2007). Value orientations: Measurement, antecedents and
consequences across nations. In R. Jowell, C. Roberts, R. Fitzgerald,
& G. Eva (Eds.), Measuring attitudes cross‐nationally. Lessons from the
European Social Survey (pp. 169–203). London, UK: Sage.
Semyonov, M., R aijman, R., Yom-Tov, A., & Schmidt, P. (2004). Population
size, perceived threat, and exclusion: A multiple-indicators analysis
of attitudes toward foreigners in Germany. Social Science Research,
33, 6 81–7 01.
Sidanius, J., & Pratto, F. (1999). Social dominance: An intergroup theory of
social hierarchy and oppression. New York, NY: Cambridge Universit y
Press.
Stegmueller, D. (2013). How many countries for multilevel modeling?
A comparison of frequentist and bayesian approaches. American
Journal of Political Science, 57, 748–761.
Stenner, K. (2005). The Authoritarian dynamic. New York, NY: Cambridge
University Press.
Stephan, W. G., Boniecki, K. A ., Ybarra, O., Bettencourt, A., Er vin, K. S.,
Jackson, L. A ., … Renfro, C. L. (2002). The role of thre ats in the ra-
cial attitudes of Blacks and Whites. Personality and Social Psychology
Bulletin, 28, 1242–1254.
Stephan, W. G., Renfro, C . L., Esses, V. M., Step han, C. W., & Martin, T.
(2005). T he effects of feeling threatened on attitudes toward immi-
grants. International Journal of Intercultural Relations, 29, 1–19.
Stephan , W. G., Yba rra, O., Martí nez, C. M., Schw arzwald, J. , & Tur-Kaspa,
M. (1998). Prejudice toward immigrants to Spain and Israel: An inte-
grated threat theory analysis. Journal of Cross‐Cultural Psychology, 29,
559–5 76.
Strabac, Z., & Lis thaug, O. (2008). Anti-Muslim prejudice in Europe: A
multilevel analysis of survey data from 30 countries. Social Science
Research, 37, 268–286.
Tabachnick, B . G., & Fidell, L. S. (2007). Using multivariate statistics (5th
ed.). New York, NY: Allyn and Bacon.
Thomsen, L., Green, E. G. T., & Sidanius, J. (200 8). We will hunt them
down: How social dominance orientation and right-wing authori-
tarianism fuel immigrant persecution of immigrants in fundamen-
tally different ways. Journal of Experimental Social Psychology, 44,
1455–146 4.
Van Assche, J., Roets, A., Dhont, K ., & Van Hiel, A. (2014). Diversity
and out-group attitudes in the Netherlands: The role of authoritar-
ianism and social threat in the neighbourhood. Journal of Ethnic and
Migration Studies, 40, 1414–1430 .
Van der Noll, J. (2014). Religious toleration of Muslims in the German
Public Sphere. International Journal of Intercultural Relations, 38,
60–74 .
Van Dijk, T. A. (2000). New(s) racism: A discourse analytic al approach.
In S. Cottle (Ed.), Ethnic minorities and the media (pp. 33–49).
Buckingham, PA: Open University Press.
Vecchione, M ., Caprara, G ., Schoen, H., Castro, J. L . G., & Schwart z, S.
H. (2012). The role of per sonal values and basic traits in perceptions
of the conse quences of immigration: A three-nation study. British
Journal of Psychology, 103, 359–377.
Volk, S., Thoni, C., & Ruigrok, W. (2011). Personality, personal values and
cooperation preferences in public goods games: A longitudinal study.
Personality and Individual Differences, 50, 810–815.
Von Collani, G., & Grumm, M. (2009). On the dimensional structure of
personality, ideological beliefs, social attitudes, and personal values.
Journal of Individual Differences, 30, 107–119.
Vilar, R. (2019). Basic Values Survey (BVS): A 20‐nation validation study.
Manuscript submitted for publication.
Wolf, L., Weinstein, N., & Maio, G. (2019). Anti‐immigrant prejudice:
Understanding the roles of (perceived) human values and value dis-
similarity. Journal of Personality and Social Psychology, 117, 925–953.
https ://doi.org/10.1037/pspi0 000177
Additional suppor ting information may be found online in the
Supporting Information section.
Araújo RCR, Bobowik M, Vilar R, et
al. Human values and ideological beliefs as predictors of
attitudes toward immigrant s across 20 countries: The
country-level moderating role of threat. Eur J Soc Psychol.
2020;50:534–546. https ://doi.org/10.1002/ejsp.2635
