Understanding minority feeling among people without a migration background: evidence from five majority-minority European cities

ABSTRACT

In many cities in Western Europe inhabitants without a migration background have become a local minority group among many others in so-called majority-minority neighbourhoods. These inhabitants remain the numerical and cultural majority in their respective countries as a whole. Being a (local) numerical minority does not automatically mean experiencing one’s position as that of a minority, a position related not only to the numerical representation of a group, but also to the group’s experience of (lower) status within society. Using a Social Identity Theory framework, I investigate the characteristics of people without a migration background that feel like a local minority while they remain a national majority. The results show only a minority of respondents who live in majority-minority neighbourhoods actually feel like a minority. Those who do tend to perceive a larger outgroup size in their surroundings, feel socially marginalised, and have a more exclusive understanding of national identity. Thus, people without a migration background who feel like a minority see the ingroup as a truer reflection of the national community and at the same time see the ingroup’s position as disadvantaged.

KEYWORDS:

• Majority-minority cities
• mixed neighbourhoods
• social identity
• feeling a minority
• marginalization

1. Introduction

The increasing diversity in Western European countries, a result of increasing international migration, is concentrated in urban areas (Edmonston and Lee 2018). These urban areas have turned into hubs of diversity, so much so that in many situations we see a demographic shift in which the original majority group, the inhabitants without a migration background, have become a numerical minority group among other numerical minority groups (Crul et al 2023). Therefore, while people without a migration background remain the culturally and numerically dominant group in these countries as a whole, in certain cities they have become a local numerical minority. Faced with rapidly changing cities and neighbourhoods, this new local minority is also faced with changes to the organisation of daily life and what constitutes the new normal (Verkuyten 2005; Vertovec 2019).

Despite the changing context in cities and neighbourhoods across Western Europe, large-scale empirical research on how people without a migration background experience becoming a minority is limited. Most research that addresses specifically how people experience the shift to a majority-minority environment is experimental work from the United States which looks into a fictive future majority-minority context (e.g. Craig and Richeson 2014; Outten et al. 2012). Some research looks into how being a minority changes the ways in which national-majority members relate to diversity (e.g. Jimenez 2017; Wessendorf, 2014). In Europe, research on the minority condition has focused largely on how this new context shapes attitudes and behaviours (e.g. Crul and Lelie 2017; Kraus 2023). However, previous research has not yet investigated how people experience emerging as a local minority, especially when they remain a national majority (and the nationally dominant group). In this paper, I examine how people without a migration background experience becoming a local numerical minority: do they perceive themselves as part of a minority altogether? What are the characteristics of those that do feel like a minority? This research is conducted in five so-called majority-minority cities (cities where the former ethnic majority, in this case the national majority, has become a local ethnic minority among many others). These cities, Amsterdam, Rotterdam, Antwerp, Malmö and Vienna, have seen their social and demographic landscape change in the past 50 years, largely due to migration. In all these cities there are well-established migrant communities.

Therefore the research question this paper poses is: ‘What characterizes the people without a migration background that feel like a minority in majority-minority neighbourhoods?’

To answer this question, I turn to literature on minority feeling and minority identity among traditional ethnic and cultural minorities. The literature indicates that minority feeling represents first and foremost a feeling of disadvantage and marginalisation for minority group members (Mulia et al. 2008; Wirth 1945). Minorities take in the position of a subordinate group in a social hierarchy because of the negative social value minority status is laden with (Sidanius and Veniegas 2000) which is intrinsically connected to disadvantage. Due to this disadvantageous positioning of the ingroup, minority status and minority identity is also connected with lower self-esteem and a more negative self-image (Verkuyten 1998). Nonetheless, people without a migration background generally do not poses objective markers of ethnic disadvantage, given that they remain dominant at the national level. By virtue of being part of the dominant group, they have access to a privileged social position (Bhopal 2018). However, previous research has shown that when faced with being a minority, they too can experience a sense of status and power disadvantage (Craig and Richeson 2014; Outten et al. 2012).

As Tripathy and Padmanabhan (2020) argue, minority condition is a community condition, and refers to the individuals seeing themselves as part of a ‘[…] larger narrative of community processes’ (11). Feeling like a minority can therefore be thought of as a reflection of group processes. In this paper, I argue that feeling like a minority refers to the positionality of the individual within the ingroup as compared to outgroups. In order to address the various groups processes at play, I turn to Social Identity Theory, which analyses a person’s sense of self through their group membership (Tajfel and Turner 2004).

In this paper, I analyse two group processes and investigate how they may characterise feeling like a minority while remaining the national majority: self-esteem and self-categorisation via the ingroup. I address self-esteem by looking at the role outgroup size plays, since becoming a minority refers primarily to the demographic relative group size, as well as by investigating the marginalisation aspects of feeling like a minority by looking at experiences of social marginalisation among people without a migration background. As these people without a migration background remain a national majority, I also look at self-categorisation by turning to how perceptions of an ethno-national identity are related to this new local minority status.

This study takes place within a relatively new body of research on how the established ethnic majority deals with becoming a minority (e.g. Craig and Richeson 2014; Crul 2020; Jimenez 2017; Outten et al. 2012; Wessendorf 2014). While the phenomenon has been studied more extensively in the United States, contextual European responses to becoming a minority have not yet been examined in detail. Moreover, this study focuses specifically on being a minority within a context where this group remains a national majority. Within this context being and feeling like a minority is not a straightforward condition, since the national majority group is not structurally disadvantaged at the national level. This study therefore examines what characterises those who feel like a minority at the local level.

This study takes place at a time when national and ethnic identities are hotly contested in Western Europe. More and more people are living and working in diverse environments, thus diversity is present in many people’s lives, especially those living in big cities. Being exposed to diversity often leads to familiarity and improved interethnic relations (Kouvo and Lockmer 2013). However, not everyone reacts positively to living in diversity (Schlueter and Scheepers 2010). People can develop both negative attitudes and exclusionist behaviours toward people with a different ethnic background than themselves (e.g. Hooghe and De Vroome 2015). Furthermore, this demographic shift which places people without a migration background as a numerical minority is politicised by the populist right into presenting this group as disadvantaged (Kaufmann 2018). By understanding who feels like a minority among the nationally dominant group, and whether this feeling reflects similar experiences of disenfranchisement as for traditional minorities, we can address processes of further polarisation in society.

2. Theoretical overview

2.1. Understanding minority feeling among members of the national majority group

Before I discuss group processes in detail, I turn to the concept of being a minority in the literature, often explored in studies on majority-minority relations. A lot of this work points out that the ‘minority condition’ is not only a question of numerical size, but also a question of (lack of) privilege, power, and advantage (Outten et al. 2012; Strauss 2007; Van Amersfoort 1978; Wirth 1945). A pertinent real-life example of this process is the position that black inhabitants of South Africa had in the country during the Apartheid regime. White South Africans were strictly speaking a numerical minority and yet possessed the power and position of status and privilege over black South Africans.

Most of the literature on minority condition indicates that the state of being a minority, as well as the feelings and experiences associated with it, are partially consequences of objective markers of disadvantage – the minority group occupies a societal position in which it is faced directly with objective marginalisation, like structural discrimination in the labour market (Van Amersfoort 1978; Verkuyten 2008). The causes and effects of this minority status are well documented under national-level ethnic minority groups and migrants (e.g. David 2013; De Vroome and Hooghe 2015; Holden and Xanthos 2009; Lynam and Cowley 2007; Verkuyten and Thijs 2002). This societal position is often a result of inward and outward characteristics like skin colour or ethnicity which also sets minorities apart from the majority group and is used as motivation for discrimination (Sidanius and Veiegas 2000). People with a different skin colour and/or ethnicity than the national majority group are positioned as lower status in society (Bhopal 2018).

However, next to objective markers of disadvantage there is also an element of experience and self-image to being a minority, as also recognised in Wirth’s original work on minorities (1945), where feeling like a minority refers to being treated unequally in society as well as to feeling disadvantaged. Strauss (2007) indicates that seeing oneself as having minority status is a very important element of being part of a minority group. A minority status is also associated with lower ingroup esteem, and minority group members tend to feel alienated from mainstream society (e.g. Porter and Washington 1993).

As they are the national majority and the dominant group, people without a migration background who turn into a numerical minority locally don’t necessarily carry the negative ethnic markers of conventional minority groups, like being treated unequally or being structurally discriminated (on the basis of their skin colour or their ethnicity), especially at the national level. Nonetheless, some members of this national majority group may experience becoming a local numerical minority in terms of becoming a marginalised and threatened group, and therefore feeling like a minority, i.e. a non-dominant and lower status group, at least locally.

Experimental research in the United States shows that when white Americans are presented with the possibility of a majority-minority racial shift, they experience a threat to their group status and are more likely to express greater political conservatism and more fear and anger toward ethnic minorities (Craig and Richeson 2014; Outten et al. 2012). This research indicates that the idea of becoming a minority can lead to an experienced loss of power and status for people without a migration background.

Therefore, while the nationally dominant group might not possess the objective markers that indicate an ethnic minority social positioning, it is possible that when they are a local numerical minority, they can still perceive their own group’s positioning in society in relation to other ethnic groups as disadvantaged and marginalised. As the literature above shows, being a minority is relational and depends on majority-minority dynamics. To further elaborate how this group process may play out for the local minority that remains a national majority, I turn to Social Identity Theory.

2.2. Social identity theory

Social Identity Theory (SIT) outlines how individuals’ sense of self is (partially) driven by their group belonging (Islam 2014). According to SIT, individuals derive positive self-esteem (and maintain a positive self-image) partially through a positive social identity, which is derived through evaluating one’s ingroup more positively in comparison to outgroups (Tajfel and Turner 2004). One’s sense of the ingroup is in turn based on self-categorisation in social categories (Tajfel and Turner 2004). Social categorisations are ‘cognitive tools that segment, classify, and order the social environment, and thus enable the individual to undertake many forms of social action’ (283) and play an important role in social identity foremost through providing a system of reference for self-categorisation (Tajfel and Turner 2004). Due to the need to maintain or enhance individual self-esteem, insofar as one’s individual self-image derives from the social categories to which she perceives herself as belonging, one will strive for a positive distinction between the own ingroup and other outgroups (Tajfel and Turner 2004). If the positive distinction is threatened by the outgroup, individuals belonging to the ingroup can react in various ways (for instance, by disidentifying with the ingroup, or trying to leave the ingroup) (Riek, Mania, and Gaertner 2006). In and of itself this positive distinction of the ingroup need not lead to exclusion and negative reactions toward the outgroup (Hornsey 2008), but previous research shows that people can react with hostility toward an outgroup population as a defensive reaction to the perceived threat that the outgroup poses to the privilege or economic positioning of members of the in-group (e.g. Gorodzeisky and Semyonov 2020).

2.2.1 Self-esteem and the ingroup

2.2.1.1 Perceived outgroup size

One important aspect of how people define the ingroup is the way in which they perceive the outgroup. Outgroup size, both actual and perceived, has been linked to increased perceived ethnic threat as well as increased anti-immigrant attitudes (Schlueter and Scheepers 2010; Strabac 2011). Some researchers connect the negative outcomes related to outgroup size with the perceived (power) threat that members of the national majority can experience when faced with increased diversity (Pottie-Sherman and Wilkes 2017; Schlueter and Scheepers 2010; Strabac 2011). Research indicates that perceived outgroup size seems to be more important than actual outgroup size in predicting these outcomes (Pottie-Sherman and Wilkes 2017), possibly due to the visibility of the outgroup. Research points toward this visibility as a primary drive of negative reactions toward migrants in the neighbourhood (Van Heerden and Ruedin 2019), rather than the statistical presence of the outgroup per se, as certain subgroups might not be visibly identifiable as outgroups.

Alongside perceived outgroup size, a perceived change in size also affects negative outgroup attitudes (Van Heerden and Ruedin 2019). Research indicates that an increase in the relative size of an outgroup population is associated with an increase in perceived competition over economic resources, but also, very importantly, perceived competition over privilege (Gorodzeisky and Semyonov 2020).

Perceived outgroup size can also be related to positive interethnic attitudes because outgroup size, through structural meeting opportunities and familiarity with ethnic outgroups, can potentially lead to more contacts, more tolerant attitudes, and more friendship ties with said outgroup (e.g. Kouvo and Lockmer 2013; Schlueter and Scheepers 2010; Strabac 2011). However, the evidence above indicates that outgroup size can unsettle the self-image of the ingroup. Within a SIT framework, perceived outgroup size arguably poses a challenge to the positive distinction one makes between the ingroup and the outgroup. When confronted with increased ethnic diversity, people without a migration background are confronted with their own difference from the other and the value of what they consider normal (Verkuyten 2005). Therefore, the more ‘others’ around them, the more unsettled the image of the self, and hence, they would be more likely to see themselves as a minority.

Based on this work I formulate the first two complementary hypotheses.

Hypothesis 1a: The higher the perceived outgroup size in the neighbourhood is, the more people without a migration background are likely to feel like a minority.

Hypothesis 1b: The larger the perceived increase of outgroup size in a neighbourhood is, the more people without a migration background are likely to feel like a minority.

2.2.1.2. Social marginalisation

As identified above, marginalisation is a crucial element of feeling like a minority (David 2013; Outten et al. 2012). While most of this work on the concept of ‘minority’ and ‘feeling like a minority’ has usually been conducted within the framework of national-level ethnic minorities, research on majority-minority relations indicates that we may expect these feelings of disadvantage to be present in the national majority that finds itself a local numerical minority as well (e.g. Craig and Richeson 2014; Outten et al. 2012). Previous research that has looked at how various social groups without a migration background deal with diversity, indicates that certain groups in society are more likely to be affected by (increasing) ethnic diversity. For instance, research indicates that people in a weaker socioeconomic position are more likely to experience (economic) threat from various ethnic minority outgroups, and in turn exhibit more negative attitudes toward these outgroups (e.g. Coenders and Scheepers 2003; Scheepers, Gijsbert, and Coenders 2002; Gijsberts, Hagendoorn, and Scheepers 2017; Hjerm and Nagayoshi 2011). They are also more likely to perceive cultural threat from ethnic outgroups, likely because they have a harder time adapting to cultural diversity (Manevska and Achterberg 2013). A potential explanation for this could be that lower social status is associated with more difficulty in navigating social complexity (Van der Waal, de Koster, and van Noord 2017). Increased ethnic diversity, as a prominent part of the increased social complexity, may be perceived as threatening.

However, social marginalisation as such can be seen as more of a groups’ process than just a sentiment connected to economic disadvantage. Social marginalisation can be thought of as the experience of being unheard, unrecognized and undervalued (Vasas 2005). According to Honneth (1992) feeling unrecognized in society is not the same as experiencing economic distress or social repression, but also captures so-called ‘wounded dignity’ and experiences of disrespect (Juul 2010). In her work on marginalisation in the United States, Hochschild (2016) notes that her respondents felt largely culturally marginalized because they felt their values and the values of others like them were not being taken seriously by society at large, and were in fact often ridiculed. Hence, social marginalisation refers to feeling neglected and disrespected by society at large. Social marginalisation could therefore be seen as the experience of feeing unrecognized because of one’s group membership. Following a SIT logic in which the self is defined through belonging to the group, and through the image of the ingroup that one holds (Islam 2014), if one experiences their group as being comparatively disadvantaged, one’s image of the ingroup is also likely to be affected by this perception.

As a result my second hypothesis reads:

Hypothesis 2: The more socially marginalized people feel, the more they are likely to feel like a minority.

2.2.2. SIT and social categorisation: ethno-national identity

According to self-categorisation theory, a subset of Social Identity Theory, after a social identity becomes salient, it can serve as a lens through which those that self-categorize as this identity can perceive social reality. This so-called ‘group identity lens’ model posits that these social identities frame various reactions to social reality (Verkuyten 2009). One such example is the role that national identification plays in how people relate to ethnic diversity. Previous research has linked national identification, especially more exclusive and ethnic forms of national identity, to negative reactions toward the ethnic outgroup: nationalism and ethno-national identity are associated with negative attitudes toward immigrants (Coenders and Scheepers 2004; Esses et al. 2005; Pehrson and Green 2010). Nationalism is also strongly related to radical right-wing voting (Lubbers and Coenders 2017), which is in turn associated with anti-immigrant sentiment.

Often also called nativist identity, ethno-national identity sees the national self as based on a common ethnic heritage (Esses et al. 2005, 320; Pehrson and Green 2010). People position themselves in terms of the national ingroup, revolving around a homogenous ethnic group, and an outgroup, which is seen as fundamentally (and ethnically) different from the ingroup. Ethno-national identity can be seen as both essentialist and ascribed. One can only belong to the national ingroup if one was born from the ‘right’ parents and their parents before them and so on.

Ehtno-national identity represents a particular and exclusionist narrative on who is a true national. Based on the group identity lens model one can argue that for people who subscribe to an ethno-national conception of identity, boundaries between the nationally dominant group and migrant and other minority groups are more salient and impermeable. In this exclusive understanding of national identity, there is little space for ethnic outgroups. People who subscribe to such a view could be more likely to see a difference between themselves, perceived at the legitimate country nationals, and the others, and therefore feel like a minority because they feel like they legitimately should be a majority.

Therefore, I formulate hypothesis 3.

Hypothesis 3: The more people without a migration background subscribe to an ethno-national identity, the more they are likely to feel like a minority.

2.3. Confounding factors

Individual socio-economic background remains relevant for understanding marginalisation as a whole and its consequences. According to Robert Putnam: ‘In virtually all societies “have-nots” are less trusting than “haves”, probably because haves are treated by others with more honesty and respect – social trust tends to be expressed by the winners in society’ (2000, 138). Van der Waal and others (2017) also found that disadvantaged groups like the unemployed, people with low job security, lower income or lower capital, display higher rates of societal discontent. For this reason, I will include various measures of socio-economic position as control variables, such as education and income. In addition, political values are also relevant for social marginalisation, as evidenced by the tendency of those that experience social marginalisation to turn to voting for the radical right (Gidron and Hall 2017). Furthermore, both the concept of white minorities as well as ethno-national identity are highly politicized in the current Western political scene (Kaufmann 2018). For these reasons, political orientation will also be included as a control variable.

3. Method

3.1. Data collection

The data has been collected within the framework of the Becoming a Minority (BaM) project in five European cities where more than 50% of the inhabitants have a migration background: Amsterdam, Rotterdam, Antwerp, Malmö and Vienna. There are a total of 2440 respondents across the five cities (see Table 1). Responses have been collected in late spring, summer, and early fall of 2019. In all cities, sampling was conducted in neighbourhoods where more than 50% of the inhabitants have a migration background, or neighbourhoods approaching this percentage. The survey asks questions about living in diversity, from attitudes toward migrants and diversity to contacts with ethnically diverse groups to experienced discrimination, nationalism and so on. We pursued comparable sampling strategies in all cities, sometimes adjusting for strategies most appropriate for the context. See the appendix to the introduction of the special issue for a detailed description of the sampling (Crul et al. 2023). In all cities we drew a random sample from available databases, and in the case of Vienna we approached people via random walks, also resulting in random sampling.

Table 1. Descriptive statistics.

	nOBS	Min	Max	Mean/%	SD
Social marginalisation (scale)	2321	1	5	2.77	.81
Disinterested politicians	2287	1	5	3.17	1.06
Political cynicism	2304	1	5	3.38	1.05
Disrespected values	2300	1	5	2.69	1.13
Systematic neglect	2297	1	5	2.29	1.16
No media representation	2240	1	5	2.78	1.06
No respect in society	2318	1	5	2.32	.94
Ethno-national identity (scale)	2384	1	5	2.79	.99
Birth in country	2384	1	5	2.6	1.19
Country ancestry	2377	1	5	2.31	1.16
Having citizenship	2362	1	5	3.45	1.19
Perceived outgroup size	2371	1	5	3.26	.92
Perceived increase outgroup size	1836	1	5	3.42	.92
Feeling like a minority
Yes	648			28%
No	1665			72%
Gender
Female	1314			54.2%
Male	1110			45.8%
Age	2440	25	46	34.51	6.06
Education	2329	1	7	5.69	1.54
Income	2209	1	5	3.15	1.2
Income deprivation	2369	1	4	1.94	.84
Subjective social status	2345	1	10	6.74	1.66
Right-wing orientation	2217	1	7	3.38	1.54
Sexual orientation
Heterosexual	2072			89.4%
LGBTQ+	224			9.7%
Other	21			.9%
City
Amsterdam	422			17.3
Rotterdam	427			17.5
Malmö	561			23
Vienna	400			16.4
Antwerp	630			25.8

The sample includes only people without a migration background (i.e. those who were born in their country of residence, with both parents born in said country). 54.2% of the sample is female. The sampling focused only on the age range between 25 and 45 years old. This choice is intended to sample only those people that are actively making major life choices and therefore are actively engaged with considering diversity, whether positively or negatively, in their life choices. The average age is 34.5 years old.

3.2. Variables

3.2.1. Feeling like a minority

Feeling like a minority is measured by a dichotomous variable asking respondents whether they feel like a minority in their neighbourhood. The responses have been coded so that one means yes and zero means no. 28% of the sample feels like a minority.

3.2.2. Independent variables

The main independent variables used to understand feeling like a minority are perceived outgroup size, increase in perceived outgroup size, social marginalisation and ethnonational identity. Perceived outgroup size in the neighbourhood is measured by asking respondents to estimate how many people with a migration background live in their neighbourhood. The answer categories are on a scale with five categories, from ‘a neighbourhood where almost no one has a migration background’ to ‘a neighbourhood where almost everyone has a migration background’. Increase in perceived outgroup size in the neighbourhood is measured by asking respondents whether they think that the number of residents with a migration background in their neighbourhood has changed in the past five years. They answer on a five-point scale, ranging from ‘strongly decreased’ to ‘strongly increased’. Social marginalisation is measured via six items on issues like feeling respected in society, feeling neglected in society, feeling respected by politicians and so on. This scale is a reflection of adapted measures from Swyngedouw et al. (2016) as well as questions constructed based on the literature I discussed on social marginalisation.¹ The answer categories have been recoded to range from 1 – ‘not socially marginalized at all’ to 5 – ‘very socially marginalized’. Cronbach’s alfa is .85, indicating that this is a reliable scale, This is confirmed by an exploratory factor analysis which identifies one factor. The explained variance of this factor is 57.6%. For the analysis I composed a scale based on the mean of the six items when at least four of the items have non-missing values.

Ethno-national identity has been measured via three items which refers to different beliefs about who belongs to the national group (EVS 2022). These questions refer to ethnic criteria for being truly [nationality], for instance ancestry.² The items have been coded so that 1 reflects thinking that this specific aspect of national identity is irrelevant, and 5 reflects thinking this is very relevant to becoming [nationality]. The Cronbach’s alpha for these six items is .79. For the analysis I composed a scale based on the mean of the three items when at least two of the items have non-missing values.

3.2.3. Control variables

In all analyses, seven control variables have been included. These include demographic variables that may have an effect on the outcome variables, such as gender, age and education, as well as variables that can be considered relevant when I take into account social marginalisation, like income. Education has been coded according to the ISCED 1997 education coding scheme (UNESCO 2012) in which 1 represents no education and 7 represents the second stage of tertiary education. Income has been coded according to the European Social Survey coding scheme (ESS 2018), in which it is ensured that categories are comparable across countries. The lowest income possible is represented by a 1, while the highest is represented by a 5. Income deprivation has been measured by asking respondents to assess whether they can make do with their current household income. They could answer from 1 – ‘I/we have more than enough, we can easily save money’ to 4 – ‘I/we do not have enough and often have difficulties getting along’. Subjective social status is measured by asking respondents to place themselves in a ladder representing their social position in society, where 1 represents a very low position and 10 represents a very high position (Adler and Stewart 2007). I also include political orientation. This variable is measured via a seven point scale asking respondents to self-identify from ‘very left’ to ‘very right’. When looking at the predictors of feeling like a minority, I have also included a dichotomous variable which refers to sexual orientation (heterosexual or LGBTQ+). This is done in order to see whether feeling like a minority reflects feeling like a sexual minority.

3.3. Model and analytical strategy

The individual responses in our survey are nested within neighbourhoods and, in turn, cities. For this reason, I use multilevel binomial regression analysis as an analytical strategy. This enables us to investigate individual-level relationships nested in neighbourhood and city clusters. I conduct a multilevel binomial logistic regression to test all the hypotheses, which have as a dependent variable the dichotomous variable ‘feeling like a minority’. I use listwise deletion in all analyses, i.e. all respondents with at least one missing value in any of the variables used are excluded from the analysis. I introduce the variables step-wise: I begin with an empty model to determine whether there is a multilevel structure, I subsequently include the control variables, and after that I include all the main predictors.

4. Results

Table 1 contains descriptive statistics of the variables. The most important outcome concerns the frequency of the variable feeling like a minority. Among our respondents from the national majority group, almost one out of four (28%) feel like a minority in the neighbourhood.

Regarding the control variables, we note that the sample leans toward left of centre regarding political orientation. The sample is also rather highly educated: the education mean of 5.69 (SD = 1.54) indicates an education level closest to first stage of tertiary education between on the ISCED scale. In terms of income, the sample scores around the midpoint of the scale, indicating an average mean income across respondents in all cities. On average, the respondents do not experience income deprivation, in fact they score around ‘I/we can get by without any problems’. People score rather high on subjective social status as well (mean = 6.74, SD = 1.66), indicating that they see themselves as highly placed on the societal ladder based on their income, education level, and job. Therefore, all in all the sample does not score high on economic marginalisation.

Turning to social marginalisation, we see a similar pattern. Feelings of social marginalisation are somewhat present and yet not very high: they remain under the midpoint of the scale (mean = 2.77, SD = .81), among which feelings of political marginalisation are particularly widespread. In fact, only for the items denoting political marginalisation do respondents score on average above the mid-point of the scale. For the items focusing exclusively on other forms of social marginalisation respondents score under the mid-point of the scale. On average, respondents score somewhat above the mid-point of our scale on ethno-national identity. Note however that people consider the criterium of citizenship more important than the criteria of ancestry and country of birth.

4.1. Minority feeling

At first, I analyse the predictors of feeling like a minority through a multilevel binomial logistic regression. The data are hierarchically structured and respondents are nested within neighbourhoods and cities therefore I start by allowing intercepts to vary at the city level. The results indicate that intercepts do not vary at the city level: b = .04, p = .27. I proceed to allow intercepts to vary at the neighbourhood level. As can be seen in Table 2, this test is significant, therefore it can be ascertained that the individual responses to the dependent variable indeed vary significantly across neighbourhoods. In other words, the neighbourhood level has a significant effect on individuals’ likelihood of feeling like a minority. I continue to test the hypotheses using a multilevel design with individuals nested within neighbourhoods, but not within cities.

Table 2. Multilevel binomial logistic regression feeling like a minority.

	Model 1: Intercepts only	Model 2: Controls	Model 3: Predictors
Effects on minority feeling	b (SE)^+	b (SE)^++	b (SE)^+++
Individual-level fixed effects
Intercept	-.92(.09)***	−1.38(.64)*	−10.25(1.02)***
Female	n/a	.09(.13)	.04(.16)
Age	n/a	−02(01)	-.02(.01)
Income	n/a	-.00(.07)	.15(.09)
Income deprivation	n/a	.23(.09)*	.09(.12)
Education	n/a	-.05(.05)	.02(.06)
Subjective social status	n/a	-.03(.05)	-.05(.07)
Heterosexual	n/a	-.49(.2)*	-.42(.26)
Right wing political orientation	n/a	.38(.04)***	.22(.06)***
Perceived ethnic composition neighbourhood	n/a	n/a	1.64(.12)***
Perceived change in ethnic composition	n/a	n/a	.39(.1)***
Social marginalisation	n/a	n/a	.58(.12)***
Ethno-national identification	n/a	n/a	.36(.09)***
Random effects
Variance neighbourhood level	.85(.17)***	.88(.19)***	.25(.13)
Akaike Corrected Information Criterion	10084.89	8318.38	7193.071

Note: *** two-tailed p < .001, ** p < 0.01, * p < .05

⁺ valid n = 2217

⁺⁺ valid n = 1778

⁺⁺⁺ valid n = 1350.

As can be seen in the third column of Table 2, the model including the control variables, income deprivation is only weakly significantly related to a stronger likelihood of feeling like a minority. Noteworthy is that other indicators of individual socio-economic position are not related to feeling like a minority. Furthermore, a heterosexual orientation is also related to feeling less like a minority. Finally, right-wing political orientation is strongly related to the likelihood of feeling like a minority.³

In the fourth column of Table 2, the results of the model including all predictors can be seen. These results confirm hypotheses 1a and 1b, 2, and 3. The larger the perceived outgroup size, the higher the likelihood of feeling like a minority, b = 1.64, t = 13.87, p < .001 (hypothesis 1a). When respondents perceive an increase in outgroup size in the past few years, they are also more likely to feel like a minority, b = .39, t = 3.98, p < .001 (hypothesis 1b). The more socially marginalized respondents feel, the more likely they are to feel like a minority as well, b = .58, t = 4.9, p < .001 (hypothesis 2). Ethno-national identity has a positive and significant result as well, b = .36, t = 3.86, p < .001 (hypothesis 3). With regards to controls, we see that in the main model which includes all the predictors, only right-wing political orientation remains relevant and significantly increases the likelihood of feeling like a minority, a result in line with the expectation that the discourse on who becomes a white minority is politicized from the right. All other control variables become insignificant, which indicates that the effects of income deprivation are not relevant anymore when considering perceived (ethnic) outgroup size.⁴ Interestingly, the effect of sexual orientation also disappears, indicating that feeling like a minority does not capture feeling like a sexual minority when considering perceived (ethnic) outgroup size and change in outgroup size.

The multilevel effect of neighbourhood of residence disappears in the model including perceived outgroup size and perceived change in outgroup size (see footnote 4) which indicates that possible neighbourhood effects are captured by perceptions of outgroup size and perceived changes therein.

5. Conclusion and discussion

In this paper I have attempted to describe the characteristics of people without a migration background that feel like a minority in majority-minority neighbourhoods. I took previous research on ethnic minorities and national-level minorities as my departure point to argue that a minority feeling represents a sense of the ingroup as disadvantaged as compared to the outgroup. Based on this, I argued that feeling like a minority is a reflection of groups’ processes, as it reflects how individuals experience their positioning within the ingroup while comparing the ingroup to relevant outgroups.

First of all, regarding our descriptive findings, it must be highlighted that most people without a migration background living in majority-minority neighbourhoods do not feel like a minority (72%), a result consistent with their positioning as the dominant national group. As I pointed out in our introduction, people without a migration background do not possess markers of ethnic disadvantage that other ethnic groups do, and are in a privileged position as compared to traditional national-level minority groups (Bhopal 2018). Nonetheless, it is important to note that, despite their dominant position in the society, more than one out of four national majority residents feel like a minority in their neighbourhood. Hence, it is important to increase our understanding of such minority feelings, as they can have consequences for the further polarisation of society (Meuleman et al. 2020).

People who estimate higher levels and an increase in outgroup size in their immediate surroundings feel more often like a minority. Based on previous research, I argued that this represents concerns for the position of the ingroup with regards to outgroups. Because outgroups become more present and established, respondents may feel that the image of the ingroup is unsettled, and therefore feel like a minority.

Furthermore, the likelihood of feeling like a minority is stronger among those who experienced social marginalisation. The second result is in line with long-standing research about ethnic minorities which links feeling like a minority to feeling disempowered and disrespected (Outten et al., 2012; Van Amersfoort 1978; Wirth 1945; Keskiner and Waldring 2023). Feeling like a minority seems to be tied to a sense of social marginalisation. This indicates that when people see the position of the group they belong to as disadvantaged, people see themselves also as members of a minority group. This can also be seen in the strong and positive relationship between right wing political orientation and feeling like a minority. The discourse on people without a migration background becoming a minority has been used often by the populist right to highlight what they see as a problem of society being taken over by migrants (Kaufmann 2018).

In addition, I also looked at how the way in which people without a migration background relate to their ingroup relates to feeling like a minority. Based on social categorisation theory (Tajfel and Turner 2004) I expected people with a more exclusive understanding of national identity to feel more like a minority, which was indeed the case. These results are in line with the ‘group identity lens’ model, which posits that people understand social reality through the lens of their salient social identity (Verkuyten 2009). Seeing the nation state as exclusively belonging to a narrow ingroup, with no space for the ethnic Other, means that people without a migration background also see themselves as fundamentally different, more deserving of social status, and a minority.

This relationship could also be due to exclusive national identity being a reaction to the status loss associated with being a minority and being socially marginalized. Identifying with a more exclusive understanding of the nation state could be a way to compensate for this status loss. Highlighting a more restrictive understanding of the nation-state – which makes the latter accessible only to them and people like them – can make people without a migration background feel in a position of superiority with regards to the other ethnic outgroups in their local context, hence symbolically compensating for the status loss. Previous experimental evidence also indicates that (future) former majorities that perceive status and power loss are more likely to increase their identification with the ingroup (Outten et al. 2012), but further research should investigate this relationship more thoroughly.

The results of this paper support a stream of research which shows that minority feeling is a condition representing more than just the numerical size of the group – it captures experiences of disadvantage and disrespect. The more socially marginalized people feel (i.e. less respected, recognized and valued by society at large) the more likely they are to feel like a minority (i.e. disadvantaged and powerless). Therefore, even in the absence of national-level objective markers of disadvantage, people can feel like a minority. Moreover, this paper shows how feeling like a local minority when one is a national majority relates to the very way in which people give meaning to the national majority.

No significant differences in feeling like a minority between the various cities were noted, indicating that people across these ethnically diverse cities in four countries share a rather similar degree of feeling like a minority. There were differences at the neighbourhood level, but these differences disappeared once I accounted for perceived increases in outgroup size. Further research could look more in-depth into perceived outgroup size, taking into account the visibility of various ethnic minorities, as well as other characteristics of neighbourhoods that play a role in interethnic contact.

It is noteworthy that the sample is rather non-marginalised economically. This could be an explanation for the non-significant results of the control variables used to measure the socio-economic positioning of respondents. Across all cities, respondents show high levels of education and tend to not experience income deprivation or low social status. Nonetheless, the conclusions should be interpreted bearing in mind that the BaM target group is between 25 and 45 years old. Within this group we conducted random sampling, therefore we believe the background characteristics in the BaM survey reflect the overall sampled population. While we have taken measures to reduce self-selection and non-response bias (see Crul et al. 2023), both forms of bias could still be present in this paper, as they are known to affect research on public opinions (Dahlgaard et al. 2019). As this might have an impact on the responses, future research could consider including secondary data analysis and further investigate whether economic marginalisation plays indeed a role as an indicator of minority feeling. Nonetheless the present research establishes the relationship between social marginalisation (i.e. experiences of marginalisation) and feeling like a minority among respondents without a migration background.

Limitations and further research

In this paper, I use a broadly defined variable referring to feeling like a minority. While I did my best to exclude other dimensions of minority experience, like sexual orientation, the concept remains broad. Feeling like a minority can refer to more complex ways in which people self-identify and self-categorise, therefore future research should examine the intricacies of being a minority in multiple dimensions in increasingly diverse societies, including identities and group membership. The intersecting identities that influence one’s life might affect feeling like a minority in different ways. For instance, social marginalization may capture various, intersecting, complementary, and competing group memberships and group identities, and people may feel like a minority in reaction to one or many the various identities.

When looking at outgroup size, I did not consider how visible markers of otherness (like skin colour or religious clothing) can affect how people react to diversity. In previous research the visibility of others has been shown to affect negative attitudes toward different ethnic groups (Van Heerden and Ruedin 2019). Therefore, future research should also examine the role played by race and other visible markers of otherness in feeling like a minority.

The relationship between social marginalisation and feeling like a minority is an issue that should be addressed more largely in society. Far from everyone feels like a minority, and yet those that do engage with more exclusive attitudes and, potentially, behaviours. When individuals see their group’s societal positioning as diminishing, they can react through discrimination and prejudice toward outgroups (Meuleman et al. 2020), thus despite their relatively small share in our research, people who feel like a minority can contribute to a further polarisation of these diverse environments and to negative interethnic relations. Understanding disempowerment and how it can perpetuate further disempowerment (of the ingroup or the outgroup) is key in talking about the diverse society. More exclusive national identity is associated with voting for the radical right, which contributes to more adversarial interethnic relations. To understand this, future research must also look into the concrete behavioural consequences of feeling like a minority (whether positive or negative) and how this affects living in diversity. Furthermore, with the current data I am unable to compare our respondents with people without a migration background who remain a local majority. It is possible that a more exclusive national identity is generally related to feeling like a minority, regardless of whether this is a demographic reality, and further research should examine this more closely.

Further research should also investigate more in detail under which conditions of diversity people are less likely to feel like a minority, including conditions under which (perceived) outgroup size would contribute to more familiarity with ethnic outgroups and could lead to the development of a superordinate common identity.

Contribution

This paper is a ‘thick description’ within the limits of statistical research in which I show what feeling like a minority represents for people without a migration background, namely how the respondent relates to the ingroup, and how they see the position of the ingroup as compared to outgroups. In general, those that feel like a minority are those that perceive a narrower and more exclusive ingroup, of which they themselves are the true legitimate members, and who at the same time perceive the position of this (narrow) ingroup as threatened by the presence of outgroups. By turning to Social Identity Theory for an understanding of how self-esteem and self-image is related to group membership, we can think of feeling like a minority as a self-image ‘proxy’. One feels like a minority when one perceives their own status as diminished, through the status loss of the ingroup, and also when one thinks one has more right to this status as a legitimate member of an exclusive national identity.

Acknowledgements

I would like to thank Marcel Coenders, Maurice Crul, Elif Keskiner and the reviewers for their valuable comments and suggestions on earlier versions of this manuscript.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by the European Research Council (ERC) Advanced Grant # 741532.

Notes

1 The items are: ‘Politicians in [country] have an interest in the problems of people like me’; ‘Political parties in [country] are only interested in my vote, not in my opinion’; In this country, the norms and values of people like me are not taken seriously anymore’; ‘People like me are being systematically neglected in this country’; ‘I feel that people like me are positively represented in the media’; ‘People like me are respected by [country] society’.

2 The items are: ‘To have been born in [country]’; ‘To have [country] ancestry’; ‘To have [country] citizenship’;

3 Given that political orientation is such a significant control variable, we conducted an additional analysis in which we tested the controls-only model excluding political orientation to see whether any of the control variables are significant in that case. We see that education is weakly significant, beta = -.11, t = -2.49, p<.05. This indicates that right-wing political orientation captures the effect of education level (therefore the effect of education is explained through the political orientation that these people have). However, this is not a highly significant result.

4 We enter the main predictors one by one to see which predictors capture the (weak) effects of the control variables income deprivation and heterosexual orientation. Income deprivation becomes insignificant when including perceived outgroup size, while LGBT+ becomes insignificant when including change in perceived outgroup size. Once the latter is included, the multilevel effect of the neighbourhood of residence also disappears.