Multiculturalism in dominant ethnic populations: A transnational profile analysis

Multiculturalism in dominant ethnic populations: A transnational profile analysis Alessia Rochira a , * , 1 , Serena Verbena b , 2 , Erica Briozzo c , 3 , Evelyn De Simone d , 4 , Francesca Esposito e , 5 , Rocío Garrido f , 6 , Manuel Garcia Ramirez g , 7 , Virginia Paloma f , 8 , Maria Vargas-Moniz h , 9 , Terri Mannarini a , 10 a Department of Human and Social Science, Applied Psychology Lab, University of Salento, via di Valesio, 73100 Lecce, Italy b Department of Education, Languages, Literatures and Psychology (FORLILPSI), University of Florence, Italy c Applied Psychology Research Center Capabilities, and Inclusion, ISPA-Instituto Universit´ ario, Rua Jardim do Tabaco, 34, 1149–041 Lisboa, Portugal d Department of Human and Social Science, University of Salento, via di Valesio, 73100 Lecce, Italy e Instituto de Ciˆ encias Sociais da Universidade de Lisboa; School of Social Sciences University of Westminster, 32-38 Wells street, W1T 3UW, London, United Kingdom f Department of Social Psychology, Universidad de Sevilla, Calle Camilo Jos´ e Cela, s/n, 41018 Seville, Spain g CESPYD, Department of Social Psychology, Universidad de Sevilla, Calle Camilo Jos´ e Cela, s/n, 41018 Seville, Spain h Applied Psychology Research Center Capabilities, and Inclusion, School of Psychology, Ispa-Instituto Universit´ ario, Rua Jardim do Tabaco, 34, 1149–041 Lisboa, Portugal ARTICLE INFO Keywords: Multiculturalism Dominant ethnic population Latent profile analysis National identity Universalism ABSTRACT Following a person-centered approach, the present study investigates support for multiculturalism of dominant ethnic members. Using Latent Profile Analysis, the current investigation aimed to identify subgroups of dominant ethnic participants and examine potential difference in their endorsement of multiculturalism towards subaltern ethnic groups. Based on the existing literature, subgroups of dominant members were identified along with the combination of multiple obstacles (i.e., national identity and intolerance) and resources (i.e., positive intergroup contact, psychological resilience and universalism) towards the endorsement of multiculturalism. Findings from a transnational sample of individuals (N. 636) across three Southern European Countries (i. e., Italy, Portugal and Spain) yielded five distinct profiles (i.e., Cosmopolitans, Glocals, Parochials, Resilient Intolerants and Disengaged). Also, they indicated that the identified subgroups differed on * Correspondence to: Department of Human and Social Science, Via di Valesio, 73100 Lecce, Italy. E-mail address: [email protected] (A. Rochira). 1 https://orcid.org/0000–0003-4240–5076 2 https://orcid.org/0000–0001-6053–8599 3 https://orcid.org/0000–0001-5787–0654 4 https://orcid.org/0000–0002-9367–7365 5 https://orcid.org/0000–0003-1698–5046 6 https://orcid.org/0000–0002-2238–0222 7 https://orcid.org/0000–0002-6703–4841 8 https://orcid.org/0000–0002-0707–1982 9 https://orcid.org/0000–0002-4875–9118 10 https://orcid.org/0000–0003-3683–8035 Contents lists available at ScienceDirect International Journal of Intercultural Relations journal homepage: www.elsevier.com/locate/ijintrel https://doi.org/10.1016/j.ijintrel.2024.102051 Received 15 May 2023; Received in revised form 24 August 2024; Accepted 25 August 2024 International Journal of Intercultural Relations 103 (2024) 102051 Available online 12 September 2024 0147-1767/© 2024 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY license ( http://creativecommons.org/licenses/by/4.0/ ). patterns of key psychosocial variables and degree of endorsement of multiculturalism. The majority of the sample (Cosmopolitans, Glocals and Disegnaged) show high level of multiculturalism, despite interesting differences characterized the distribution of the profiles across the three countries: the more intolerant profiles (Parochials and Resilient Intolerants) were prevalent in Italy. Contrarily to our expectations, national identity was not exclusionary per se, as has been reported elsewhere. Rather, depending on its combination with other key variables, it worked either as an obstacle or as a resource towards the support for multiculturalism of dominant ethnic members. Potential contributions of Latent Profile Analysis for an in depth understanding of multiculturalism and implications for interventions are discussed. Introduction Above all during the past two decades, European societies have become progressively more diverse as a result of both significant intra-European Union (EU) mobility and non-EU immigration flows (Trenz & Triandafyllidou, 2017), with differences across European countries. According to the European Commission (European Commission, 2023), at the beginning of 2021, 8.4 % of the entire EU population was born outside Europe (i.e., 37.5 million people) and 5.3 % (i.e., 23.7 million people) was made up of non-EU citizens. The increasing diversity of Europe brings both challenges and opportunities as well as raising important questions to which multiculturalism is a possible answer (Arasaratnam, 2013). Overall multiculturalism is discussed as a way modern societies can deal with the increasing cultural heterogeneity resulting from the ongoing migration movements. In a broader perspective, multiculturalism assumes diversity as a value and entails the public recognition and respect of the many cultures and cultural identities that populate a society. As such, multiculturalism “refers to the attitude in which groups value and actively support mutual cultural differences and equal chances and opportunities. This means that cultural diversity is … [appreciated] as important for the functioning of the society as a whole” (Arends-T´ oth & Van De Vijver, 2003). The current debate surrounding multiculturalism is quite heated. The significant variety of interpretations, conceptual definitions and meanings of multiculturalism (Colombo, 2015; Kymlicka, 2001; Reitz et al., 2009) reflects multiple concerns and claims about its potential benefits and disadvantages, also emphasizing important differences for dominant and subaltern ethnic groups. In fact, while some scholars emphasize that multiculturalism is a profitable solution for equitably managing cultural and ethnic diversity, others claim that it may provoke conflicts (Verkuyten, 2007) as it promotes ‘essentialism’ and impermeable group distinctions (Verkuyten, 2005). Within social and community psychology, multiculturalism is generally viewed as a desirable way of dealing with ethnic and cultural diversity (Trickett et al., 1994; Trickett, 1996). In particular, the endorsement of multiculturalism is expected to promote positive intergroup relations as it fosters mutual confidence and interpersonal trust (Berry et al., 2006), also decreasing racial biases (Plaut et al., 2009) and furthering more positive views of the ethnic outgroup members among the dominant ethnic group (Verkuyten, 2005; Wolsko et al., 2006). To the best of our knowledge, a number of studies have examined the differences in the endorsement of multiculturalism between dominant and subaltern ethnic groups (Davidov et al., 2008;Verkuyten, 2007), whereas relatively less is known about the potential differences within each group. Based on these premises, this study aimed at identifying distinct patterns of variables underlying individual differences in dominant ethnic members’ support of multiculturalism. In particular, using a person-centered approach (Coarsworrh et al., 2005; S. J. Schwartz & Zamboanga, 2008) and examining patterns of individual and contextual variables, our multinational research team sought to extract diverse groupings from a heterogeneous population of dominant ethnic individuals across three Southern European countries – i.e., Italy, Spain and Portugal. By doing so we intended to elucidate potential differences among the emerging profiles not only in terms of stronger or weaker support for multiculturalism but also in terms of co-occurring patterns of multiple factors (i.e., obstacles and resources) associated with this support. Theoretical background Within the psychosocial literature, the acculturation framework (Berry et al., 2006) has investigated multiculturalism as an acculturation preference held by the dominant ethnic group members involving the acceptance that the subaltern ethnic groups maintain their culture of origin while also having contact and participation in the mainstream culture (Arends-T´ oth & Van De Vijver, 2003). Yet, several studies have highlighted individual and contextual factors which are key to understanding the acceptance/opposition to multiculturalism. Group identification, and particularly national identification, has been proved to be an important correlate of inter-ethnic attitudes. In general, such a relationship is shaped by how individuals perceive and mean national identity (Badea et al., 2018). Particularly, inclusive forms of national identity, encompassing the values of diversity and common citizenship (i.e., civic nationalism) are often associated with positive inter-ethnic attitudes. On the contrary, exclusive forms of national identity, conceiving national groupings based on their ethnic ancestry (i.e., ethnic nationalism), are akin to generate negative responses towards newcomers and reduce the support for multiculturalism of dominant ethnic members (see Miglietta et al., 2018). In fact, ethnic forms of national identity were proved to be negatively linked to multiculturalism as members of subaltern ethnic groups are perceived as ‘alien’ to the national community (Sumino, 2017). This pattern has been mainly explained by the mediating role of the perceived out-group threat (Tip et al., A. Rochira et al. International Journal of Intercultural Relations 103 (2024) 102051 2 2012; Verkuyten, 2009). The perception of intergroup threat is also a key explanatory factor for the contact-multiculturalism relationship. Based on the wellestablished literature on inter-group contact (Pettigrew, 1998; Pettigrew & Tropp, 2006), research has stressed that positive inter-ethnic encounters are likely to enliven the endorsement of multiculturalism among dominant ethnic members via the reduced perception of outgroup threat. Moreover, existing studies show that regular contact with culturally and linguistically diverse persons can enhance positive intergroup attitudes (Callens et al., 2019; Williams & Johnson, 2011); in particular, the development of intimate intergroup relationships, such as friendship, can lead dominant members to cultivate appreciation and concern for the situation of the subaltern ethnic groups hence increasing the endorsement of multiculturalism among them (Verkuyten & Martinovic, 2006). Tolerance is an important tenet in universalism, one of the ten basic human values according to Schwartz (S. H. Schwartz, 2017). In this regard, Verkuyten (2018) argues that divergence is unavoidable in pluralistic societies and tolerance towards diverse ways of life, practice and norms is pivotal to achieving peaceful intergroup coexistence (Verkuyten & Yogeeswaran, 2017). In fact, there is evidence that tolerance is positively associated with Multicultural Ideology among dominant ethnic population (Inguglia et al., 2017). In particular, Inguglia and colleagues proved that tolerance was likely to bolster the positive effect of Multicultural Ideology on citizens’ attitudes towards immigrants. Universalism is defined by tolerance, appreciation, concern for the wellbeing of others, both humankind and the more-than-human world, and understanding (Schwartz, 2012). Values have been conceived as trans-situational goals that serve as guiding principles in people’ lives (Schwartz, 1992) and their role in shaping attitudes and social behaviors has been widely investigated (Mannarini et al., 2021). Prior studies indicated that universalism is positively related to dominant members’ willingness to welcome migrant people and underpins a more positive view of the consequences of immigration (Davidov et al., 2008; Schwartz, 2011; Vecchione et al., 2012). Similarly, Sapienza and colleagues (Sapienza et al., 2010) found that self-transcendence values – e.g., universalism and benevolence – motivate individuals from dominant groups to opt for the acculturative strategy of integrationism-transformation (Bourhis et al., 1997); those with a high level of self-transcendence values appreciate inter-ethnic contact and culture maintenance and also agree to modify some aspects of their own culture of origin to promote better migrant social inclusion. Multiculturalism challenges dominant members to find ways to cope with a new social, cultural and political order (Verkuyten, 2009) that implies the acknowledgement and respect of subaltern ethnic group identities, cultures, and rights (Visintin et al., 2019). Some studies suggested that psychological (Zautra & Murray, 2004) and community resilience (Norris et al., 2008) might function, respectively, as individual and contextual resources enabling dominant members to pursue positive adaptation towards the cultural diversity of their territories, eventually fostering their support for multiculturalism (Verbena et al., 2021). However, the research findings were inconclusive. In fact, contrary to the authors’ expectations, the perception that the living context was endowed with social and material resources to deal successfully with the challenges of cultural diversity (i.e., community resilience) did not foster support of multiculturalism among dominant ethnic member participants. Diversely, the inner strengths and abilities enabling individuals to cope adaptively with stress and adversities (i.e., psychological resilience) were associated with the acceptance of cultural diversity and multiculturalism. As outlined above, several researchers have examined multiple individual and contextual variables associated with the support of multiculturalism among dominant members but a relatively smaller number of studies have considered the relation between participants’ responses to these different variables and the profiles emerging from them. To fill this gap, this study focuses on understanding the different profiles of obstacles and resources emerging among dominant members, beyond the distinction between supporters versus non-supporters of multiculturalism (Fox et al., 2013). The current study Privileging a person-centered approach, this study sought to identify subgroups of dominant members that reflected the interrelationships between the different variables examined in the literature. Specifically, national identity and intolerance were taken as obstacles whereas positive intergroup contact, psychological resilience and universalism were taken as resources for the endorsement of multiculturalism. As regards community resilience, given that prior research findings were inconclusive (McNeil-Willson et al., 2019; Verbena et al., 2021), we did not assume any particular classification. By employing Latent Profile Analysis [LPA], the study intended to describe the heterogeneity of respondents’ profiles which will potentially allow either higher or lower support for multiculturalism. Some previous investigations have employed LPA to examine profiles of individuals with similar patterns of acculturation styles, providing partial support for Berry’s model (Fox et al., 2013; Schwartz & Zamboanga, 2008). Other studies, instead, have used LPA to investigate attitudes (e.g., intolerance) towards subaltern groups differentiating subgroups of individuals based on distinct configurations, clusters, of significant factors (Dangubi´ c et al., 2021). All these investigations pointed out that LPA has the merit of elucidating relevant qualitative differences among subgroups of participants, going beyond a mere “positive-negative continuum” (Dangubi´ c et al., 2021). Inspired by this body of scholarship, this study went beyond a ‘pros vs. cons’ logic to pursue two main research goals. First, it aimed at unpacking the endorsement of multiculturalism (Berry, 2011) by examining the way in which distinct subgroups of dominant members combine the resources and obstacles identified in the literature in a different way, hence elucidating differences among the emerging profiles. Second, the various profiles emerging from the analysis were appraised in terms of their relationships with multiculturalism ultimately ascertaining how the subgroups presenting a different configuration of resources and obstacles were associated with either higher or weaker support of multiculturalism. Because, to the best of our knowledge, there is no research using LPA on a mixed pattern of resources and obstacles underlying the endorsement of multiculturalism, we did not formulate any specific hypotheses about the latent profiles of our participants but opted A. Rochira et al. International Journal of Intercultural Relations 103 (2024) 102051 3 to focus on categories based on patterns emerging from the data (bottom-up approach). The context of the study We collected our data in three Southern European countries, i.e., Italy, Spain and Portugal. In general, these three countries, together with Greece, are grouped together as "Southern Europe". "Southern Europe as a region did not emerge as a planned intellectual product or philosophical debate. It emerged from a chain of events over the last two centuries, passing through many formative stages" (Pedaliu, 2017, p. 14). Interestingly, the author notes that the severe economic and political crisis that began at the end of 2000 (Zamora-Kapoor1 & Coller, 2014) contributed to the formation of a kind of new regional identity that affected the social and cultural milieu of Italy, Portugal and Spain, which already shared some common background (e.g. the common Latin and Catholic tradition). In particular, the term "Southern European Syndrome" or "Southern European Exceptionalism" has been used to describe the decline in citizens’ political trust following the economic crisis. As for migration issues, since 2000, these three countries have been experiencing a substantial growth in the number of people arriving from non-EU countries, above all from the African continent. Commonly, from countries of emigration, particularly in the post-war period, they all turned into countries of immigration but also in countries of transit to other destinations, particularly in the Schengen area (Colucci & Gallo, 2018). Italy, Spain and Portugal were also included in the so-called Southern model of immigration because of the similarities in their immigration patterns (Baldwin-Edwards, 1999; Carta et al., 2005; King, 2000). Also, although immigration is a relatively recent phenomenon for Italy but not for Spain and Portugal, as these countries have a long history of (often forced) migration flows connected to colonial endeavors and imperial projects, 11 their experience implementing policies of social inclusion is relatively scarce. As a result, policies of social inclusion have often been inadequate and most migrants, particularly the illegal ones, have ended up working in informal and shadow economies (Ambrosetti & Paparusso, 2021). Despite these similarities, there are authors who claim that there are important differences among the countries of the Euro- Mediterranean region, which also includes the three countries examined in this study, even questioning the actual validity of the Southern model of immigration (Baldwin-Edwards, 2012). Important differences between Spain, Italy and Portugal concern, for example, their national social inclusion programs and the way immigration is regulated and managed at the national level. (La Spina, 2017). Social inclusion strategies pursued by Spanish policymakers revolve around three main pillars: the inclusion of migrant people in the workforce; the fight against discrimination, racism and xenophobia; and the appraisal of cultural diversity. More specifically, social inclusion is pursued through multi-level governance involving the central administration, local institutions and civil society organizations to address the needs of migrant populations across several life domains (i.e., education, employment, social services, health and housing) (https://ec.europa.eu/migrant-integration/governance/spain). However, according to the Migrant Integration Policy Index [MIPEX] – which evaluates social inclusion policies based on a set of formal indicators applied to definite policy areas (i.e., Labour market mobility, Family reunion, Education, Political participation, Long-term residence, Access to nationality, Anti-discrimination, and Health) – in Spain the procedure to acquire citizenship remains a major issue to migrants’ full access to rights and opportunities. More in general, Grad (2017) highlights a discrepancy between favorable attitudes towards ethnocultural diversity and the broader demand for assimilation within the Spanish population. With regard to the Italian context, immigration is a highly politicized issue (Geddes & Pettrachin, 2020) with a great emphasis on securitizing and restricting migrant flows (e.g., Esposito et al., 2021; Novara et al., 2021). In Italy, regional governments have the competence to plan and implement social inclusion policies with considerable differences and disparities across territories in terms of programs and activities (e.g., https://ec.europa.eu/migrant-integration/country-governance/governance-migrant-integration-italy_ en). As for the legislation, Italy embraced a restrictive approach to the naturalization of migrants, rather promoting their temporary and precarious integration (Paparusso, 2019). According to the MIPEX evaluation, political participation and access to citizenship are the two major policy obstacles to full migrant inclusion. As Neto and Neto (2017) suggest, Portugal represents a highly interesting case study for examining multiculturalism within the dominant ethnocultural population. In Portugal, the Plan for Immigrants’ Integration (Plano de Integraç˜ ao de Imigrantes-PII) is pursued at various levels (e.g., national and local) and includes multiples programs such as language courses, civic education, work orientation and social and intercultural activities (https://ec.europa.eu/migrant-integration/country-governance/governancemigrant-integration-portugal_en). Following the MIPEX, Portugal is one of the leading European countries where social inclusion takes place as a two-way process and allegedly is “far ahead of countries like Italy and Spain”. However, and despite the good intentions, many have highlighted the operational limitations of such responses and the persisting barriers for migrants in accessing their basic rights, such as health, housing, employment and education (Casquilho-Martins & Ferreira, 2022; Fonseca et al., 2021; Pires et al., 2020; Raposo & Violante, 2021). In conclusion, besides the assessment of the social inclusion policies ratified in the three European countries in question, it is worth noticing that an important gap emerges between what is written on paper (e.g., policy pronouncements) and what occurs in the field (i. e., migrants’ effective access to rights and opportunities). 11 Italy, too, had a colonial empire between 1882 and the end of the World War II. The difference with Portugal and Spain is that they have a much longer colonial history. A. Rochira et al. International Journal of Intercultural Relations 103 (2024) 102051 4 Method To pursue the research objectives, a multi-lingual questionnaire (Italian, Spanish, and Portuguese) was formulated by our transnational research team. Data were collected from the beginning of May until the end of September 2021 in an online survey administered through a Google form. A snowball sampling strategy was used; an invitation to participate and the link to the questionnaire were disseminated by the research team as well as by trained psychology students to their personal, professional, and social networks via email, social media, and word of mouth. After completing the survey, respondents were asked to forward the link to others in their own networks. Participation was voluntary and anonymous. Before starting the questionnaire, participants were provided with informed consent detailing the purpose of the research, the survey content and procedure as well as their right to withdraw from the study at any time. The questionnaire took about 20 min and no incentives were offered to participants. All procedures implemented in the study are consistent with the APA’s ethical standards and the 1964 Helsinki Declaration. The project was reviewed and approved by the Human Research Ethics Committee of the University of [BLINDED for PEER REVIEW]. Participants This study is based on an online survey applied to a non-probability convenience sample reached by means of a snowball procedure. Participants were invited to take part in a study on migration and citizens’ relationships with migrants. A total number of 1837 respondents from the three European countries completed the survey. To be specific, 725 respondents were from Spain (F =64.14 %), 709 were from Italy (F =51.19 %), and 403 were from Portugal (F =73.7 %). Given the non-probability nature of the sample and its unbalanced distribution by sex assigned at birth and country of residence, we extracted a second sample (Sample 1) from the first one (Sample 0) to improve the data balance (for a similar procedure see Salvatore et al., 2018). Sample 1 was a stratified, non-proportional quota sample randomly extracted from Sample 0. Specifically, we divided Sample 0 by country of residence and then randomly selected a predefined number of subjects by gender (F=106, M=106) for each country, thus ultimately obtaining the same number of participants for the three European contexts involved in the study. Having 212 subjects per country was considered suitable for optimizing the size and gender balance of the three national samples. Female respondents had a mean age of almost 36 years (mean [M]=35.9, standard deviation [SD]=13.44), while men had a mean age of 37 years (mean [M]=37.19, standard deviation [SD]=14.16). As far as education is concerned, the majority of the participants (56.23 %) had an upper secondary education (ISCED 3), 37.07 % lower secondary education (ISCED 2), 34.73 % a primary education (ISCED 1), while only 8.11 % held a Bachelor or Masters level (ISCED 6 and 7 according to the International Standard Classification of Education 2011 – ISCED 2011). 12 A small number of participants had a different type of education (e.g., private courses) (0.76 %), while a total of 0.163 % of respondents declared no educational background. The sociodemographic characteristics of the overall sample segmented by country are reported in Table 1. Instruments The questionnaire consisted of the following measures. National Identity. A single ad hoc item was adopted to rate participants’ self-identification with their nation-state on a 5-point Likert scale ranging from 1 (not at all important) to 5 (extremely important) (“How important is it for you to be [Italian/Spanish/ Portuguese]”?). Intolerance ( α =0.843). Eight items were drawn from the Extremism Scale developed by Ozer and Bertelsen (2018) to capture participants’ intolerance toward culturally diverse groups on a 7-point Likert scale ranging from 1 (strongly disagree) to 7 (strongly agree). Examples of items are “The groups in the society that don’t support the good and correct life should be deprived of their rights”; “It is wrong to make compromises with what one stands for”. Interethnic Contact ( α =0.708). Three ad hoc items were created to assess the frequency of participants’ interactions with migrant people in three different relational contexts, i.e., work/school, family, and friends/peers, on a 5-point Likert scale ranging from 1 (Never) to 5 (Everyday). Psychological Resilience ( α =0.844). The Connor-Davidson Resilience scale (CD-RISC), developed by Campbell-Sills and Stein (2007), was employed. The scale comprises ten items rated on a 5-point Likert scale ranging from 0 (not true at all) to 4 (true nearly all the time). Examples of items are “I’m able to adapt to change”; “I can deal with whatever comes”; “I can achieve goals despite obstacles”. Community Resilience ( α =0.81). The 5-item Resources subscale of the “Communities Advancing Resilience Toolkit” (CART) developed and validated by Pfefferbaum et al. (2013) was used. Items were rated on a 5-point Likert-type scale ranging from 1 (strongly disagree) to 5 (strongly agree). The scale assesses participant’s perception of the availability of material resources (e.g., natural, physical, information, human, social, and financial capital) within their community. Examples of items are “My community supports programs for children and families”; “My community has the resources it needs to take care of community problems (resources include, for example, money, information, technology, tools, raw materials, and services)”. 12 UNESCO, Institute for Statistics, 2012. International Standard Classification of Education. ISCED 2011. Available at: http://www.uis.unesco. org/Education/Documents/isced-2011-en.pdf A. Rochira et al. International Journal of Intercultural Relations 103 (2024) 102051 5 Universalism ( α =0.809). Three items were drawn from the Portrait Value Questionnaire (PVQ, Schwartz, 2007). The items describe a person’s goals that underpin Universalism value type. More specifically, the respondents were asked to assess the extent to which the person described in the statement was similar to them on a six-point scale ranging from 1 (very similar) to 6 (nothing like me). The original items were: “They think it is important that every person in the world be treated equally. They believe everyone should have equal opportunities in life”; “It is important to them to listen to people who are different from them. Even when they disagree with them‚ they still want to understand them”; “They strongly believe that people should care for nature. Looking after the environment is important to them”. Multiculturalism ( α =0.671; ω =0.726). The 4-item Multiculturalism subscale of the Mutual Intercultural Relations in Plural Societies (MIRIPS) scale was used to measure participants’ acculturation expectations (Berry, 2017). Participants were asked to rate their level of support for immigrants’ social inclusion and respect for cultural diversity on a 5-point Likert scale ranging from 1 (totally disagree) to 5 (totally agree). Examples of items are “Immigrants should be fluent both in [Italian/Spanish/Portuguese] and in their own language”, “I feel that immigrants should maintain their cultural traditions but also adopt those of Italians/Spanish/Portuguese”. Sociodemographics. Participants were asked to report their age, gender, level of education, occupation, and place of residence. Data analysis Statistical analyses were performed using SPSS 26, R (R Core Team, 2020, https://www.R-project.org/), and Jamovi (the Jamovi Project, 2022, Version 2.3, https://www.jamovi.org). Missing data analysis found that less than 5 % of the data were missing and Little’s MCAR Test (Little, 1988) indicated that data were missing completely at random. Missing data were imputed at item level and were replaced with the Multiple Imputation procedure on SPSS. All the variables inserted in the analysis were standardized. To identify meaningful Multiculturalist Profiles among our participants, a latent profile analysis with maximum-likelihood and fullinformation maximum likelihood was performed (Williams & Kibowski, 2016). Using the tidyLPA package in R (Rosenberg et al., 2018), an ascending number of profile solutions with one to six profiles was specified and the fit of each profile solution was assessed based on the following fit statistics, Akaike Information Criterion (AIC), Bayesian Information Criterion (BIC), Entropy, Parametric Bootstrapped Likelihood Ratio Test (BLRT), and (BLRT_p), Consistent Aikake information criterion (CAIC), log-likelihood (LogLik), Approximate weight of evidence (AWE), Classification Likelihood Criterion (CLC), Kullback information criterion (KIC), and Sample size-adjusted Bayesian information criterion (SABIC). In addition to statistical parameters, the interpretability, theoretical relevance and model parsimony were also taken as criteria for identifying the best profile solution among the six assessed. After completing the LPA procedure, an ANOVA analysis was performed using Jamovi to test whether the identified profiles differ significantly from each other when it comes to their association with multiculturalism. Table 1 Level of Education of the sample by country. Level of Education Country Spain Italy Portugal n % n % n % Primary education 0 0 31 14.62 3 1.41 First stage secondary education 17 8.02 105 49.53 41 19.34 Second stage secondary education 118 55.66 28 13.21 104 49.06 First level tertiary education 48 22.64 31 14.62 37 17.45 Second level tertiary education 27 12.74 17 8.02 27 12.74 Post tertiary education 2 0.94 0 0 0 0 Table 2 Inter-correlations between resource and obstacle variables. Variable M SD 1 2 3 4 5 6 7 1. National Identity 3.88 0.99 —      2. Intolerance 17.5 8.39 .10*—     3. Interethnic Contact 7.65 2.86 −.05 −.17*** —    4. Psichological Resilience 41.3 5.91 .16*** .07 .06 —   5. Community Resilience 16 4.13 .17*** −.11** .09*.06 —  6. Universalism 16.4 2.32 −.09*−.41*** .14*** .10** .028 — 7. Multiculturalism 16.1 3.06 −.01 −.22*** .11** .09*−.034 .25*** — Note. * p <.05, ** p <.01, *** p <.001 A. Rochira et al. International Journal of Intercultural Relations 103 (2024) 102051 6 Results In this section we report the findings emerging from the Descriptive Analytics and the Latent Profile Analysis. For greater clarity, in the results description, National Identity and Intolerance are referred to as personal obstacles, Universalism and Psychological Resilience are referred to as personal resources, Interethnic Contact and Community Resilience are referred to, respectively, as social and community resources. Descriptive statistics Prior to implementing LPA, inter-correlations between resource and obstacle variables were calculated. The results are shown in Table 2. National Identity and Intolerance, which were seen as obstacles to the endorsement of multiculturalism, were positively though weakly correlated and the latter was negatively and mildly correlated with multiculturalism. Diversely, National Identity was positively associated with both Psychological Resilience and Community Resilience whereas a negative but feeble association emerged between National Identity and Universalism. Intolerance was negatively and strongly associated with personal resources (i.e., Universalism) and negatively and feeble associated with social (i.e., Interethnic Contact) resources; further it was weakly and negatively related to Community Resilience. Psychological Resilience was positively but weakly associated with Universalism and Multiculturalism. After conducting correlation analysis, we performed the Latent Profile Analysis and the fit indices for the LPA are presented in Table 3 (i.e., reports AIC, BIC, Entropy, BLRT_val, BLRT_p, CAIC, LogLik, CLC, KIC, and SABIC statistics for all the solutions). Table 3 shows the fit statistics for models with different profile solutions. To determine the optimal number of profiles, we used several fit statistics, including Akaike’s information criterion (AIC), sample size adjusted Bayesian information criterion (BIC), entropy, while also considering LRT_val, BLRT_p, CAIC, LogLik, CLC, KIC and SABIC values (Magidson & Vermunt, 2002; Nylund, Asparouhov, & Muth´ en, 2007). These criteria helped to identify the best fitting models among different alternatives, with lower AIC and BIC values indicating a better fitting model (Muth´ en, 2004). In addition, we examined the entropy of each profile to determine the quality of the classification, with entropy values above 0.7 indicating good classification of individuals into profiles. Furthermore, we examined the results of the BLRT to obtain p-values below 0.05, which indicate whether the addition of a profile (k-1 profile model compared to k profile model) leads to a statistically significant improvement in model fit. BLRT was significant across all models. Based on these criteria, as shown in Table 3, the five- and six-profile solutions had the lowest fit indices. In particular, the five-profile solution had the lowest BIC, CAIC, and AWE, while the six-profile solution presented the lowest AIC, LogLik, CLC, KIC, SABIC. Moreover, the five-profile solution had a satisfactory Entropy value (0.75), whereas the Entropy value of the six-profile solution was unacceptable (0.617), thus suggesting that the former indicated a more precise assignment of participants to the latent profiles extracted (Wang et al., 2017). Lastly, in determining the preferred solution, five- and six-profile solutions were also judged in terms of their interpretability and conceptual clarity. Also in this case, the five-profile solution appeared a better option. The five profiles that emerged are: Cosmopolitans, Glocals, Parochials, Resilient Intolerants and Disengaged. Cosmopolitans (n. 44; 6.9 % of the sample). Overall, this profile is characterized by the lowest presence of obstacles. In fact, compared to the other four profiles, this subgroup includes the participants with the lowest levels of both national identity and intolerance. At the same time, it consists of individuals with the highest levels of interethnic contact and universalism whose mean scores are both above the average level and rank highest among all the five profiles. As for the other resources, persons in this subgroup reported low levels of community and psychological resilience, with the former being below the average level and the latter almost reaching it. Glocals (n. 391; 61.5 % of the sample). This is the largest and most resourceful subgroup of participants, as those belonging to this profile show a high level of personal resources. In fact, it can be distinguished from the other profiles by significantly lower intolerance and higher national identity, universalism, and psychological resilience. Furthermore, individuals within this profile reported average Table 3 Fit indices for Latent Profile Analysis. Profile 123456 AIC 31,297.08 30,344.64 30,233.05 29,908−16 29,734.47 29,707.22 BIC 31,363.27 30,449.44 30,376.46 30,090−18 29,955.10 29,966.46 Entropy 1.00 0.95 0.77 0.76 0.75 0.62 BLRT-val 96,643.794 12,559.269 33,889.134 18,769.036 4.125.189 BLRT_p 0.01 0.01 0.01 0.01 0.01 CAIC 31,375.27 30,468.44 30,402.46 30,123.18 29,995.10 30,013.46 LogLik −15,636.54 −15,153.32 −15,090.52 −14,921.08 −14,827.23 −14,806.61 AWE 31,487.46 30,647.35 30,648.33 30,435.69 30,374.23 30,459.47 CLC 31,275.08 30,308.54 30,182.59 29,845.68 29,655.98 29,614.45 KIC 31,312.08 30,366.64 30,262.05 29,944.16 29,777.47 29,757.22 SABIC 31,325.15 30,389.08 30,293.86 29,985.34 29,828.02 29,817.14 A. Rochira et al. International Journal of Intercultural Relations 103 (2024) 102051 7 levels of social and community resources, i.e., interethnic contact and community resources. Parochials (n. 51; 8 % of the sample). Individuals in this subgroup have neither personal nor social resources at their disposal. In fact, profile 3 is characterized by below-average scores of psychological resilience, universalism, and interethnic contact; it is worth noticing that the score on universalism is the lowest among the five profiles. In addition, participants in this profile perceive that their community is not equipped with material resources, as indicated by the mean score of community resilience which is below the average level. As for obstacles, this subgroup includes participants with a relatively strong national identity and intolerance orientation. Resilient Intolerants (n. 39; 6.1 % of the sample). As with profile 3, profile 4 includes participants with above-average levels of obstacles. In particular, the value of intolerance ranks highest for members of this subgroup. However, unlike profile 3, these participants report above-average levels of psychological resilience. However, all the other resources are lacking: participants barely have contact with migrant people, perceive a scarcity of material resources within their community, and are not influenced by universalistic values in their actions. Disengaged (n.111; 17.5 % of the sample). Compared to the other profiles, profile 5 comprises individuals with neither obstacles nor resources. In fact, this subgroup comprises participants with the lowest level of psychological resilience as well as with a relatively low level of national identity, intolerance, and interethnic contact. Lastly, persons in this subgroup reported mean scores of universalism and community resilience that are around the average or slightly below. Fig. 1 shows an overall picture of the combination of resources and obstacles towards the endorsement of multiculturalism for each profile. Focusing on the countries involved in the study, Table 5 displays the means of the resource and obstacle variables in the three national samples. Further, as for the distribution of the five profiles across the three European countries, Table 6 displays a varied image; specifically, though the Glocal profile constitutes the largest subgroup in all the countries, it is quite numerous in Portugal. Diversely, the Parochial and the Resilient Intolerants are more prevalent in Italy and barely existing in Spain and Portugal. The Disengaged profile is the second largest subgroup in Spain, that has also the larger proportion of the Cosmopolitan profile. Regarding the differences among the five profiles in terms of their relationship with multiculturalism, the results of the ANOVA yielded a significant overall main effect [F(4, 115) =10.2, p <.001] for multiculturalism. The post hoc test (Table 7) indicated that the Cosmopolitans (M=16.8; SD=2.76) and the Glocals (M=16.7; SD=2.58) subgroups had higher score in multiculturalism than the Parochials (M=13.8; SD=3.74) and Resilient Intolerants (M=14.5; SD=3.57) profiles which, in turn, scored lower than the Disengaged profile (M=16.4; SD=2.70). Finally, Fig. 2 shows a whole portrait of the five profiles as regards their association with multiculturalism, revealing that both the intolerants subgroups are quite unlikely to endorse multiculturalism whereas the other three subgroups exhibit a clear multiculturalist orientation. Discussion Overall, the research findings corroborated that there are different ways to either support or reject multiculturalism and that privileging a person-centered approach can offer a more careful account of such a variety beyond a mere ‘pros vs. cons. logic’. In particular, the results bolstered the categorization of universalism (Sapienza et al., 2010; Vecchione et al., 2012), interethnic contact (Verkuyten, 2005; Wolsko et al., 2006) and psychological resilience (Verbena et al., 2021) as resources, as well as of intolerance Fig. 1. Distribution of resources and obstacles towards the endorsement of the Multiculturalism among the five-class latent profiles. A. Rochira et al. International Journal of Intercultural Relations 103 (2024) 102051 8 (Dangubi´ c et al., 2021) as an obstacle towards the endorsement of multiculturalism (Badea et al., 2018; Visintin et al., 2019). National Identity was a notable exception (Sumino, 2017; Tip et al., 2012). In fact, the research outcome unveiled that national identity is significantly associated with multicultural attitudes, and is not exclusionary per se (Verkuyten, 2005; Verkuyten, 2007). Rather, the findings echoed previous evidence attesting that the relation between national identification and inter-ethnic attitudes is multifaceted and depends on how the former is meant and perceived by individuals. In particular, our results corroborated the distinction between ethnic and civic forms of national identity showing that it can function as either an obstacle or a resource towards the endorsement of multiculturalism depending on specific patterns of co-occurring variables. Considering the Cosmopolitan subgroup, the findings aligned with the multiple studies proving that national identity is a key predictor of negative attitudes towards subaltern ethnic groups and low support for multiculturalism (Badea et al., 2018; Verkuyten, 2009). On the other side, taking into account the Glocal subgroup, the research results suggest that dominant ethnic people who identify strongly with their nationality ingroup do not automatically engage in exclusionary attitudes towards subaltern ethnic outgroups. This is particularly true when their motivational goals entail understanding, appreciation, and full acceptance of diversity (i.e., universalism), also in case of differing opinions, values and beliefs (i.e., tolerance) (Verkuyten et al., 2021). We might argue that, in this case, national identity plays an inclusive role as the personal values associated with it are themselves inclusive (Breton, 2015) and, at the same time, the norms and beliefs of the subaltern outgroups are perceived as compatible with that of the dominant ethnic culture (Verkuyten & Yogeeswaran, 2017). Table 4 Mean Scores of obstacles and resources for each profile. Variable Profile Cosmopolitans Glocals Parochials Resilient Intolerants Disengaged National Identity −2.03 0.33 0.13 0.25 −0.16 Interethnic Contact 0.452 0.171 −0.26 −0.888 −0.30 Intolerance −0.48 −0.20 1.10 1.73 −0.21 Universalism 0.51 0.34 −2.65 −0.51 0.02 Psychological Resilience −0.07 0.35 −0.32 0.6 −1.03 Community Resilience −0.39 0.16 −0.10 −0.70 −0.01 Table 5 Total sample and Countries’ means and standard deviations. Country Multiculturalism National Identity Intolerance Universalism Interethnic Contact Psychological Resilience Community Resilience M SD M SD M SD M SD M SD M SD M SD Spain 16.40 2.52 3.59 0.94 13.7 1.77 16.8 1.77 7.64 2.68 39.3 5.85 17.60 3.49 Italy 15.90 3.64 3.91 0.97 21.7 2.65 15.70 2.65 6.55 2.60 43.3 5.43 13.60 4.04 Portugal 16.10 2.79 4.36 0.94 16.9 2.31 16.90 2.31 9.59 2.56 41.4 5.63 17.60 3.16 Total 16.10 3.06 3.88 0.99 17.5 8.39 16.40 2.32 7.65 2.86 41.3 5.91 16.00 4.13 Table 6 Profile distribution per Country. Profile Country Spain (%) Italy (%) Portugal (%) Cosmopolitans 10.40 5.20 5.20 Glocals 54.20 50.00 80.20 Parochials 2.80 17.45 3.80 Resilient Intolerants 0.90 16.50 0.90 Disengaged 31.60 10.80 9.90 Table 7 Post hoc Turkey Multiple Comparison among the 5 Profiles. Cosmopolitans Glocals Parochials Resilient Intolerants Disengaged Cosmopolitans —    Glocals 0.03 —   Parochials 1.02 *** 0.99 *** —  Resilient intolerants 0.77 ** 0.75 *** −0.246 — Disengaged 0.12 0.095 −0.90 *** −0.65 *** — Note. *p <.05, ** p<.01, *** p<.001 A. Rochira et al. International Journal of Intercultural Relations 103 (2024) 102051 9