In- and out-groups across cultures: Identities and perceived group values
Introduction
Human beings exist within groups, and conflict between those groups stems at least in part from perceptions that we have about our own groups' moral beliefs as compared with others'. Such beliefs contribute to and are the result of intergroup prejudices. Intergroup conflict is conceptualized as either based on in-group favoritism (Brewer 1999) or derogation of out-groups, with group members holding prima facie dislike of out-groups rooted in feelings of prejudice derived from perceived attributes of that out-group (Fiske et al. 2007). Evidence for a single mechanism is scarce. While existing literature converges on the idea that humans evolved in contexts of intra and inter-tribal social interactions leading to the development of categorically selective in-group empathy (Awad et al., 2018; Buchanan and Powell 2018; Firat and Boyer 2015; Tomasello 2016), there are steps missing toward developing a fuller understanding of these procesess. We have little cross-cultural exploration about those groups that typically are considered in-groups. Additionally, while we know a lot about moral values people report at the indvidual and societal levels, we know less about how third-order values – beliefs about the values of in- and out-groups – are distributed.
Morality is central to how group members perceive themselves and evaluate others (Leach et al. 2007; Moscatelli et al., 2019); perceptions of moral divisions (rather than objective moral opinion differences) are salient in forming in- and out-groups (Waters 1990). These purported divisions are often used as post-hoc moral justifications (Vaisey 2009) for drawing social boundaries (Lamont 1992), and varying categories can be attached to a notion of a dangerous ‘other’ (Flores and Schachter 2018) that has important social, economic and political ramifications. Much of the research in this arena has been done with artificially constructed groups or with researchers presupposing the targeted out-groups, even though people hold a host of social identities – shared sets of meanings with other members of social collectives -- as well as prejudices toward different types of out-groups. Little work has explored the range of values attached to the social groups that people find important – and distance themselves from – outside of the laboratory and across cultures.
We present a demography of potential in- and out-groups that people from four distinct cultural quadrants (Inglehart and Baker 2000) report as meaningful, alongside a novel measurement of perceived group ‘third-order’ values (Firat et al. 2018). These ‘beliefs about a group's beliefs’ capture the putative distinctions people draw towards their out-groups across cultures. We model beliefs people hold about the moral values of self-identified in- and out-groups using original, cross-cultural data.
People define themselves and others in ways that have consequences for the allotment of economic, social and political resources. Carving the world into ‘us’ and ‘them’ is a fundamental aspect of social cognition (Berreby 2005). People are members of many social collectives and as such identify with various ethnic, racial, gender, religious, and other sorts of categories. Social identity theory suggests that people prioritize in-group members (Dovidio et al. 2002) and demonstrates how we differentiate in-groups from out-groups.
In-group identification, even based on arbitrary characteristics, reveals strong group biases and favoritism (Tajfel 1982) and creates discriminatory perceptions and behaviors even against arbitrary out-groups. Modern social groupings are more depersonalized, contributing to a sense that in-group norms and customs take on moral authority (Brewer 1999).
Social identity theory (Tajfel and Turner 1979) suggests not only that people prioritize in-group members, even artificially created in-groups, but more strongly suggests that differentiation from out-groups relies on finding and developing a sense of ‘positive distinctiveness’ (Brewer 1999). Many phenomena flow from this basic insight, ranging from a sense that people expect more cooperation from in-group members (Romano et al., 2017) to outright hostility toward out-groups. These differential evaluations and feelings can be generated based on manufactured in-groups which, while short-lived and artificial, still reveal strong in-group biases and favoritism based on an arbitrary social categorization (e.g., Tajfel et al., 1971; Perreault and Bourhis, 1999). Such groupings, once defined for participants, creates discriminatory perceptions and behaviors against arbitrary out-groups (Rabbie et al., 1969, Tajfel et al., 1971).
These biases are even stronger when applied to meaningful, real-world in- and out-groups (Brewer, 1991; Turner, 1987). This suggests that identification and even temporary identities lead in-group favoritism, or ethnocentrism for more established groups. This strong favoritism toward the in-group possibly can be decoupled from out-group antipathy, a matter of debate in the psychological literature. People favor even meaningless in-groups, and some suggest that evidence supports the idea that out-groups lead to hostility and discrimination (e.g., Perry et al., 2018).
Social psychologists disagree whether in-group favoritism necessarily leads to the derogation of out-groups, or if only specific out-groups garner antipathy. For example, racially biased behavior and attitudes can occur without an explicit racial hatred by manifesting as an automatic avoidance of the racial out-group (Dovidio et al. 2002). Brewer (2007) argues that in-group biases are less about antipathy toward out-groups and more based on evolutionarily shaped aspects of trust and cooperation, a view with growing support (Greenwald and Pettigrew, 2014). Out-groups garner antipathy if they are perceived as violating in-group moral standards (Allport 1954; Weisel and Böhm 2015).
In contrast, other scholars (for a review, see Kinder and Kam 2009) suggest that a fundamental dislike for out-groups is at the core of observed discrimination, as well as prejudiced attitudes and behaviors at both individual and structural levels. People differ in the intensity of out-group dislike (Stets and Burke, 2000), and a complication in adjudicating this issue is that most of the empirical work on perception, categorization, and behavior relies on specific, mutually antagonistic identities (e.g., Israelis vs. Palestinians) struggling over tangible resources, like land (Bobo 1983). At the extreme, this evidences dehumanization (Bandura et al., 1996; Haslam and Loughnan 2014), the denial of the humanity of a particular out-group, which often dovetails with extremely bad behavior toward groups (e.g., Luft 2015; Osofsky et al. 2005; Staub 2004) considered outside a perceived sphere of moral acceptance. We know less about how people concurrently perceive a range of potential outgroups.
Factors contributing to in-group valuation and/or out-group derogation may not be universal across identities, contexts, and societies; evidence suggests they are distinct constructs and depend on cultural context (Hamley et al., 2020). For example, men discriminate more in favor of in-groups than women (Romano et al., 2017). Strength of identification with one's identity contributes to in-group preferences (Leszczensky and Pink 2019), and out-group discrimination may not be based on antipathy as much as a lack of positive feeling (Firat et al., 2017). Motivations may change over time: Iyengar and Krupenkin (2018, p. 212) find that, over time, what were once positive in-group political feelings (in the U.S. context) has shifted toward “negativity toward the opposition that is the stronger electoral motive.”
For purposes of discrimination and inequality research, some out-groups are self-evidently relevant and become the object of inquiry, and within social psychology putative out-groups are either manufactured or based on obvious real-world situations. However, in a pluralistic society, the relationship between any important identity for a group member and those not in the group is likely contextual, and current approaches downplay ecological validity in the spirit of identifying important mechanisms. Studying out-groups implicates work on stereotyping (see Fiske, 2018 for a brief overview), a process that folds into the nature of what we perceive out-groups as believing. Qualitative work suggests that we draw moral boundaries between the perceptions of what our in-groups and out-groups value (e.g., Lamont 2000). However, such groups in experimental and qualitative traditions are often predetermined by the researcher, when in practice people hold a variety of in-groups and out-groups as meaningful social units.
We advocate for greater ecological validity about processes underlying the perception of in- and out-groups by exploring respondent-prioritized groups alongside a new measure of perceived group morality. Moral concerns allow us to bridge the study of people's valued in-groups and the associations they make with those they consider out-groups alongside the perceived moral priorities of those groups; this can point toward global patterns and cultural distinctions.
Humans perceive groups, theirs and others', as sharing characteristics about group values, attitudes, and orientations. People's self- and other-ratings conform more to popular stereotypes than they are objective indicators of the actual ratings from members of those groups (e.g., Scheffer et al., 2020). Perceptions, as a century of social psychology has taught, shape behavior and influence social, political and economic decisions. Moral values are chief among the ways we define our groups (Waters, 1990), and we use these beliefs to delimit purported moral boundaries as symbolic treatments that are experienced as fundamental to in-groups (Lamont 2000). Such moral criteria serve to maintain and create institutionalized social differences (Lamont and Molnár, 2002). Some (Beierlein et al. 2016) suggest moral boundaries express majority enmity toward societal out-groups, while others (Scott 1976) suggest that disadvantaged groups draw moral boundaries to maintain collective self-worth in the face of economic deprivation. Moral boundaries are employed to justify exclusionary government policies (see Pachucki et al. 2007), and are related to racial/ethnic discrimination (Wimmer 2008, 2013). In the American case, lines seem to be drawn whereby non-religious people are perceived as immoral (Edgell et al. 2006).
Criteria for streamlining the range of potential moral boundaries can be found in measures of human values (Eisenberg et al. 1989; Hitlin 2007; Miles and Vaisey 2015). Values are “broad, desirable goals that serve as standards for evaluating whether actions, events, and people are good or bad” (Schwartz 2011, p. 464). The basic structure of values exists across societies (Schwartz et al., 2012), and is reflected at both the individual and societal level (Borg et al. 2017). This well-established model finds evidence for two core dimensions along which lie an array of 10 values found in all literate cultures (see Fig. 1).
Schwartz's extensive body of work suggests a core subset of potential human values that exist across cultures based on the needs that social groups have to survive (Schwartz and Bilsky, 1987; Schwartz et al., 2012). The basic structure (see Fig. 1) is represented in hundreds of samples in all literate societies (Schwartz, 2004; Spini, 2003). They are indirectly linked with situated behavior, actions that have a variety of other pressures that influence the person (Eyal et al., 2009; Roccas and Sagiv, 2017).
Values offer windows into both individual and group experiences of culture and an important moral dimension for understanding perceptions of in- and out-groups; people draw boundaries magnifying in-group cohesion through differentiation from the perceived values of out-groups (Kinder and Kam 2009; Lamont 2000; Tajfel and Forgas 2000). Morality is central to how people evaluate in- and out-groups (Ellemers, van den Bos and Compass 2012). Values have been extensively employed at both the individual and collective levels (Hitlin and Piliavin 2004).
Researchers to date have typically estimated group-level values through aggregating individual survey responses. Instead, we incorporate a novel measure of third-order values (Firat et al., 2018) to measure moral perceptions toward a range of meaningful societal groups. We present direct responses of beliefs about groups' beliefs, rather than relying on simple first-order aggregation. Perceptions of beliefs are core for how people define in- and out-groups, more so than the actual values held by those groups. Our measurement follows recent treatments of collective beliefs that are exploring the relationship between levels of belief, like Melamed et al. (2019) who find that third-order beliefs about norms can override individual's own first-order beliefs in certain circumstances. O'Dwyer and Ҫoymak (2020) suggest ‘meta-representations’ as perceptions of what ‘most others’ value (see also Portelinha and Elcheroth 2016) that play an important role in shaping individual attitudes and behavioral intentions.
Building on recent sociological engagement with the cross-cultural properties of measuring values (Miles 2015; Longest et al., 2013), we incorporate a novel measure based on the well-established cross-cultural Schwartz research program to examine the nature of third-order group values for in-groups and ecologically meaningful out-groups. Our measure of ‘third-order’ beliefs (see also Correll et al., 2017) captures the perceived values held by ‘most people’ in a group. This illuminates how boundaries are drawn around the perception that one's in- and out-groups share values. Values measured in this fashion can sometimes be labeled as ‘culture’, but our approach assumes that third-order understandings of values can be applied to various social groupings within a society, as people make intra-cultural distinctions along moral lines that, the literature suggests, is used to delineate and justify prioritizing in-groups over out-groups. Drawing these lines can represent responses to perceived social threat (e.g., Quillian 1995; Abascal 2020) or part of a complicated interplay based on social role location (Koenig and Eagly 2019).
While scores of studies aggregate individual values to make claims about groups, organizations, and nations, our proposed measure asks people to select values that they believe are important to various in- and out-groups. That tells us something apart from aggregating beliefs; it tells us what values are understood to be important. Aggregation assumes that people share the values of their chosen categories, and little to no research that we are aware of asks people to select – in this sort of systematic way – the perceived beliefs of non-in-groups; many groups in society are not out-groups, per se, but do not fall into the umbrella of groupness. Our approach begins to examine these beliefs. People evidence psychological discomfort when an in-group, not an out-group, violates a personal value, something that mediates disidentification with their group through a feeling of dissonance (Glasford et al., 2008).
Research on values-as-aggregations falls in the same pattern as other work that predetermines the in- and out-groups under exploration, often measuring tolerance toward people who differ from societal norms in their socio-demographic characteristics (e.g., ethnicity, religion) or in their personal attitudes and preferences. Sagiv and Schwartz (1995) reasoned that universalism values (a self-transcendence value), which express concern for the welfare of all others including those whose lifestyle differs from one's own, fosters tolerance. Values shape reactions to members of minority groups by affecting how people react to diversity in society (Roccas and Amit, 2011). Attributing high importance to universalism and openness to change values, and low importance to conservation values, relates to tolerance toward various minority groups by dominant group members in several cultures. Network diversity was positively related to tolerance only among people who attributed low importance to conservation values (Bloom and Bagno-Moldavsky, 2015).
A similar pattern emerged in studies on attitudes toward immigration: Attributing high importance to universalism values related positively and attributing high importance to conformity and tradition values related negatively to favorable attitudes toward immigration (Davidov and Bart, 2012; Davidov et al., 2008, Davidov et al., 2014). This pattern holds not only for tolerance toward ethnic out-groups but also for tolerance toward other minority groups. Studies of attitudes toward homosexuality in Europe showed the same pattern. The more important openness to change and universalism values are to people, and the less important conservation and power values, the more positive their views of homosexuality (Kuntz et al., 2015; Donaldson et al., 2017; Licciardello et al., 2011).1 Surveys of attitudes toward a variety of minority groups (e.g., sexism, anti-Semitism, anti-foreigner attitudes, anti-Muslim attitudes) also reveal that universalism and conservation values relate directly to tolerance of those who differ from the dominant group in society (Beierlein et al. 2016).
We hypothesize that certain in-groups will be prioritized across societies, beginning with family membership and extending to religious groups. When respondents were asked to select the moral priority of various potential societal groups, “there was a relatively consistent order of moral priority: family, friends, and in-group members were seen as relatively central, whereas out-group members and nonhuman targets were seen as relatively distal” (Crimston et al., 2016, p. 650). Our data allow us to capture the two elements: a) looking at the prioritized in-groups and out-groups given an array of possibilities, and b) examining the perceived content of the moral beliefs attributed to both kinds of groups. This lets us explore, say, whether family membership is prioritized in very different cultures and whether third-order beliefs about moral boundaries attached to families differ. We can look at similar patterns for out-groups (i.e., are certain third-order beliefs attached to certain out-groups across cultures?), while also discerning which groups are most often identified as out-groups.
Section snippets
Data
Data were collected from France, South Korea, Turkey and the United States between 2015 and 2016 through online surveys administered by panel research companies. This country-level selection was guided by previous distinctions between traditional vs. secular-rational values dimension and survival vs. self-expression values dimension (Inglehart and Baker 2000) and individualistic vs. collectivistic societies (Hofstede 2001); the United States and France as individualistic, higher self-expression
Measures
Identification Measures: We asked people to select valued in-groups using an established item from the International Social Survey Program (ISSP) 2003 (occupation, race/ethnicity, gender, age group, religion, political orientation, nationality, family status, social class, and the part of the country they live in).
Out-group identification is measured by the widely-used social distance question taken from the World Values Survey (2005–2009) with a slight modification; we asked respondents to
Results
Despite presumed cultural differences across four countries, we find family status is the pan-culturally top-ranked in-group (see Table 1). Occupation is another important in-group across our four-country samples. On the other hand, we find ‘different political orientation’ is reported as one of the top three least preferred out-groups across our four national samples, with many cultural differences in the rest of the responses.6
Conclusion
We find evidence for cross-cultural patterns in the perceived moral content of in- and out-groups. Values assigned to gays and lesbians, political orientation, family status, and occupation cluster meaningfully across societies (with a few Korean exceptions). Even though social psychological theories of culture have long argued that societies can be divided as individualistic vs. collectivistic based on the ways the people perceive their social landscapes, we find that people with historically
Acknowledgments
The research reported here was funded in whole or in part by award W911NF-13-1-0342 from the U.S. Department of Defense and the U.S. Army Research Office/Army Research Laboratory under the Minerva Research Initiative. The views expressed are those of the author(s) and should not be attributed to the Department of Defense or the Army Research Office/Army Research Laboratory. Authors declare no competing interests. Data available upon request and will be deposited at ICPSR at a future date.
References (94)
- et al.
Universalism, conservation and attitudes toward minority groups
Soc. Sci. Res.
(2016) - et al.
When values matter: expressing values in behavioral intentions for the near vs. distant future
J. Exp. Soc. Psychol.
(2009) - et al.
Coalitional affiliation as a missing link between ethnic polarization and well-being: an empirical test from the European social survey
Soc. Sci. Res.
(2015) - et al.
Universal dimensions of social cognition: warmth and competence
Trends Cognit. Sci.
(2007) - et al.
Intragroup dissonance: Responses to ingroup violation of personal values
J. Exp. Soc. Psychol.
(2008) - et al.
Intergroup contact, value system and the representation of homosexuality
Proc. Soc. Behav. Sci.
(2011) - et al.
Morality and politics: comparing alternate theories
Soc. Sci. Res.
(2015) - et al.
Boundary processes: recent theoretical developments and new contributions
Poetics
(2007) - et al.
Group heterogeneity and tolerance: The moderating role of conservation values
J. Exp. Soc. Psychol.
(2011) Universals in the content and structure of values: theoretical advances and empirical tests in 20 countries
Adv. Exp. Soc. Psychol.
(1992)