Salience of nationality in students’ spontaneous self-concept: A comparative study of a nationally homogeneous and a heterogeneous school context

Abstract

The study on which this article is based investigated the salience of nationality in adolescents’ self-concept as a function of the diversity of the school context. According to the distinctiveness postulate, people selectively attend to and encode aspects that are most distinctive. We therefore predicted that students in the nationally heterogeneous school environment would mention nationality more frequently in their spontaneous self-concept when questioned than would those in the more homogeneous school setting. Furthermore, we expected minority members to mention nationality more frequently. These predictions were confirmed and implications will be discussed.

Keywords

Distinctiveness postulate diversity nationality national identity self-concept youth

Introduction

‘International mobility, trade and employment is “becoming the norm”’ (Karim, 2012: 138). As a consequence of this increased mobility, the ethnic, linguistic and cultural compositions of communities and schools around the world are being re-defined. Cultural diversity and multicultural societies have become an established fact of life in many countries across the world (Barrett et al., 2011). This is a time when intermixing and integration are increasingly common. As early as 1978, McGuire et al. conducted a research project among 560 fourth-grade students in America, assessing the salience of ethnic identity in children’s spontaneous self-concept as a function of the ethnic mix at the school. Based on the distinctiveness postulate, they hypothesised that intermixing would heighten rather than erase consciousness of ethnicity. Their findings supported this hypothesis and the authors noted the social policy implications of the distinctiveness postulate, ‘some going in counterintuitive directions’ (McGuire et al., 1978: 512). Bringing children of different ethnic backgrounds together heightens rather than lessens awareness of the differences, especially for minority groups. Experienced educators within heterogeneous schools may be aware of this dynamic, but less experienced policy makers may formulate well-intentioned policies assuming that integration will be achieved simply by bringing people together. Given the increased mobility alluded to above, we reasoned that the implications of the distinctiveness paradigm are noteworthy and we set out to replicate McGuire et al.’s (1978) study, with some modifications, in a European context.

We compared adolescent school children at two secondary schools which differed with respect to their national mix of pupils, one being nationally homogeneous and the other being heterogeneous. The key question was whether this difference in school context would influence the salience of nationality in students’ spontaneous self-concept. In replication of the McGuire et al. study, the students were asked the non-directive questions ‘Tell us about yourself’ and ‘Tell us what you are not’ – eliciting, respectively, the students’ spontaneous affirmative and negated self-concept.

In line with the distinctiveness postulate, we hypothesised that students at the nationally heterogeneous school would mention nationality more frequently than students in the fairly mono-national school. In other words, we predicted that nationality would be more salient in the heterogeneous school students’ spontaneous self-concept. Before presenting the study in depth, we will explain the distinctiveness postulate in more detail, given its central importance to the research question. We will also outline our understanding of nations and national identity, and then proceed to a detailed description of the study. A comprehensive overview of the results will follow and conclusions will be drawn and implications discussed.

The distinctiveness postulate

As McGuire et al. (1978: 512) explain, the distinctiveness postulate was formulated to deal with the more general question of how people manage the problem of their senses being able to take in more information than they can meaningfully deal with in perception and memory. As was first suggested by Broadbent (1958), people essentially act like information-processing machines, applying selective filters. The distinctiveness postulate proposes that people selectively attend to and encode aspects that are most distinctive, which is an efficient information processing strategy. For example, a red triangle is noticed in a field of red circles (triangularity being the distinctive feature) and a green triangle would be noticed in a field of red triangles. These are very simple examples to illustrate the basic principle. Yet, the distinctiveness postulate can also be applied to self-perception: we notice in ourselves those aspects that are peculiar in our customary social environment. Distinctiveness affects the self-concept directly, as one notices one’s distinctive features, and indirectly by others responding to our peculiarities. Transferred to the school setting, the distinctiveness postulate implies, first, that nationality is more salient in people’s self-concepts in a nationally mixed environment than in a mono-cultural one. Second, nationality is more salient in the self-concept of minority group members than in the majority group.

The distinctiveness postulate therefore has important social policy implications. It implies that as schools integrate, children become more conscious of their nationality/ethnicity and more likely to define themselves in terms of it. This occurs as a basic perceptual phenomenon, regardless of the emotional climate as McGuire et al. (1978: 512) point out. McGuire et al. conducted their study at elementary and high schools in Chicago, United States. The ethnic background of the pupils was assessed by the researchers. Given the European setting, the focus of this study was on national rather than ethnic identity construal (Arts and Halman, 2006; Billiet, 2002).

Nation and nationality

Article 15 of the Universal Declaration of Human Rights (UDHR) asserts that (1) everyone has the right to a nationality and (2) no one shall be arbitrarily deprived of his nationality nor denied the right to change his nationality (United Nations, n.d.). Thus, nationality or the right to a nationality has been declared to be a universal human right. Yet, what is a nation? What characterises a nation?

Within the remit of this article, only the surface of these big questions can be scratched and a few key aspects pointed out. The term nation can be traced back to the Latin term ‘natio’ which referred to birth, origin or people. In Roman times, the term described ‘others’ or ‘heathens’ (Rumpel, 1991). In the Middle Ages, we come across the territorial understanding of the term for the first time. The context was university registration – students had to state where they were from (Helmchen, 2005). Usually, two main interpretations of nation building can be found: one being based on ancestry or lineage (ius sanguinis) and the other based on affiliation or belonging (ius soli) (Schubert and Klein, 2006). The anthropologist Clifford Geertz (1973) describes a nation as a group based on primordial affiliations, reflecting an understanding based on ancestry. Today, most historians would agree that nations themselves can be considered a construct. An explanation of the essentialist versus constructivist understanding of nation using Luxembourg as an example is provided by Péporté (2011). Billig (1995: 24) explains that objective criteria such as language, religion or geography cannot be used to predict where state boundaries are: psychological considerations are decisive. Nations are in fact ‘imagined communities’ (Anderson, 2006).

National identity

The brief section above draws attention to the complexity of the concept ‘nation’ and the process of nation building. The ensuing question concerns the understanding of national identity, as construed by the individual. Arts and Halman (2006) explain that national identity refers to ‘perceived distinctiveness, a possibility to distinguish oneself or a group from others’ (p. 73). Condor (2011) explains national identity as the ways in which members of a national group reflexively understand themselves.

Nation and nationality have very different meanings from one country to another. In France, for example, nation has always had a very strong political connotation. In Central and Eastern Europe, as explained by Arts and Halman (2006), ‘the idea of nation came to have a particularly strong ethnic connotation’ (p. 73). Weinreich and Saunderson (2003) suggest that the terms national and ethnic can be used interchangeably if the context of growing up is taking into consideration:

The greater proportion of people, who are born into and experience their childhood within a well-defined peoplehood with a shared socio-cultural history, has the ethnic identity of that peoplehood. Often the term ‘nation’ or ‘ethnic group’ is used to refer to a peoplehood of this kind. (pp. 27–28)

Identity formation or construal is a topic in itself and only few key aspects can be looked at here. William James warned over a century ago that selfhood (including identity) is ‘the most puzzling puzzle with which psychology has to deal’ (James [1890] as cited by Owens [2003: 330]).

Individuals are faced with a lot more stimuli on a daily basis than they can attend to or process. The distinctiveness paradigm was offered as one cognitive strategy to deal with this information overload. The influence of cognitive structures on the selection and organisation of self-relevant information is also noted by Markus (1977). She states that people do not process information at random but develop what she has termed self-schemata. These self-schemata are derived from experience, and ‘Once established, these schemata function as selective mechanisms which determine whether information is attended to, how it is structured and how much importance is attached to it’ (Markus, 1977: 64). She explains that self-schemata are derived from specific experiences as well as general representations derived from repeated categorisations and the subsequent evaluation by the individual and others reacting to the individual. Information about the self, organised in self-schemata, is more easily accessed by the individual and is more salient to the individual. The organising function of the self-concept is also emphasised by Epstein (1973, 1979). He describes the self-concept as a self-theory the individual construes about himself and suggests that the self-theory is composed of a series of hierarchically organised postulates. Higher order postulates are more general and stable, and lower order postulates, which are derived from specific experiences, are more malleable. This structure allows continuity and stability as well as the assimilation of new experiences. Epstein notes that individuals are not necessarily conscious of their self-theories, and stresses the role of emotions in inferring postulates.

Self-theories are construed in interaction with the environment. Individuals’ self-conceptions of who they are and how they relate to others are greatly influenced by the interpersonal and group context in which they evolve and in which social comparisons are made. The importance of group membership is stressed in Tajfel and Turner’s (1979) social identity theory (SIT). SIT proposes that individuals seek a positive group identity; a positive self-concept based on their group membership through social comparison between their own and other groups.

The difficulties of adequately capturing an individual’s national or ethnic identity in quantitative surveys are specified by Billiet (2002). As he points out, it is a question of degree of identification and (simultaneous) multiple group membership. This is precisely why we opted for a non-directive qualitative measure and did not present our student samples with predefined categories. Concluding this section, we would like to emphasise that development of the self is a lifelong process in which adolescence is seen as a particularly important period. As Pinquart and Silbereisen (2000) indicate, the period of adolescence is special as young people change and adapt their self-definition through cognitive abilities acquired for self-reflection (p. 75). This is why we focused on adolescents. Looking at this age group also meant that the students in our sample could indicate their country of birth and state their nationality or nationalities themselves.

As explained above, the self-concept acts as a filter and organising principle for self-relevant experiences. Domains important to our self-concept are expected to be more salient and thus easily accessible in spontaneous retrieval. The distinctiveness postulate implies that we notice in our self-concept what makes us different from others in our social environment. We thus derived the following predictions for our study. First, nationality should be mentioned more frequently in the nationally mixed school context than in the more homogeneous school context (Hypothesis 1). Second, nationality should be mentioned more frequently by students belonging to minority groups (Hypothesis 2).

We identified two secondary schools which differed regarding their national mix of students (homogeneous vs heterogeneous). The two schools and their philosophies will be described in more detail in the following section.

Description of the schools

The homogeneous national school context was provided by a German, private, state-approved secondary school about 40 km to the west of Munich, Bavaria. Around 700 mostly baptised, mainly Catholic students attend this secondary school. Students could choose between a humanistic orientation and an emphasis on modern languages. All students start with Latin as first foreign language, followed by English. Later, ancient Greek or French and optionally Italian or Chinese can be chosen. It is the school’s philosophy to motivate students to actively engage in sports, music and other cultural activities. School fees are around 300 euros per year.

The heterogeneous national school context was provided by the European School of Luxembourg, which has since been split into two separate schools. The Grand Duchy of Luxembourg is host to several European institutions, which led to the establishment of the very first European School in 1953. European Schools are divided into different language sections, representing the member states of the European Union (EU). Basic instruction is given in the official languages of the EU. The study of a first foreign language (English, French or German), known as L2, is compulsory throughout the school, from the first primary class. The aim of the European School system is stated as follows:

Educated side by side, untroubled from infancy by divisive prejudices, acquainted with all that is great and good in the different cultures, it will be borne in upon them as they mature that they belong together. Without ceasing to look to their own lands with love and pride, they will become in mind Europeans, schooled and ready to complete and consolidate the work of their fathers before them, to bring into being a united and thriving Europe. (Schola Europaea, n.d.)

European Schools are known for high standards in languages. Language classes are composed of mixed nationalities and are taught by native speakers. In the secondary school, classes in art, music and physical education are always composed of mixed nationalities. From the third class of secondary school, history and geography are studied in the pupil’s first foreign language (L2) and taught to groups of mixed nationalities. Thus, by definition, students from different, albeit mainly European, countries attend the school and intermixing is reinforced by bringing together different nationalities in different subjects. At the time of the study, 2265 students attended the secondary school. For European Institution employees, school fees are paid for by the institutions; for other students, fees of around 10,700 euros per year are charged. Given their locations, the German school attracts foremost students of German nationality whereas the European School of Luxembourg attracts a wide European student base. The different composition of the student body in terms of nationalities is conceptualised as the main independent variable within the study.

Theoretical and practical considerations guided the decision regarding the measurement of the dependent variable: the salience of nationality in youths’ spontaneous self-concept. A key criterion for the dependent variable measure was non-directedness of the measure. In the original McGuire et al. (1978) study, the affirmative and negated spontaneous self-concepts were measured. We wanted to retain this split for this study as we believe, in line with identity-construal theories (Weinreich and Saunderson, 2003), that the negated self-concept is highly informative, given the potential positivity bias of the affirmative self-concept. In the original study, the test modus was to hold separate interviews with each child, with a time limit of 5 minutes for the affirmative interview and 3 minutes for the negated self-concept. Resource and time constraints would not allow us to replicate this mode of data gathering and we opted for survey methodology instead. As noted above, we decided to focus on adolescents who engage in reflections about themselves; this age group of students is able to complete questionnaires.

In the style of the non-directive Twenty Statement Test (TST; Kuhn and McPartland, 1954), we asked the students Who am I? and Who am I not? As the test title implies, the original version of the TST uses 20 statements but only focusing on affirmative self (I am). We decided to ask the students for five instances of the positive and five of the negated self-concept, resulting in 10 statements per student. Several reasons led us to use a shortened list. First, we anticipated attention problems in school-age children. Second, we wanted to reduce the stressors of writing and spelling. Third, we expected a more consistent pattern of actually obtaining five answers per self-concept per student, rather than a high variability when 20 answers have to be elicited. Most crucially, we assumed – in line with the self-theories outlined above – that the most salient aspects would more easily accessible and be mentioned in either of the top five statements.

Changes in the number of statements have been made before (Halkoaho, 2012, or Niebergall, 2010), and this shortening has proven to be useful, depending on the aim of the study. Halkoaho reduced the number of statements to three. Cousins (1989) used all 20 statements in his comparative study of self-perception in Japan and the United States, and asked the students to circle their top five statements. The analysis of the TST statements showed that the top five responses were similar to the full-form responses. Even though we did develop a coding system for our study and analysed all answers, we were primarily interested in the question of whether nationality markers would be mentioned or not. Contrary to other studies using TST (e.g. Bond and Cheung, 1983), there was no explicit time limit given to the students.

The materials were first piloted with students at a Luxembourgish secondary school, then tested with students attending the German school and the European School. The pilot study confirmed our decision to concentrate on adolescents as they have a greater level of self-reflection than younger children. Crucial feedback during this pilot phase was also that the instruction to write answers to the simple question ‘Who am I?’ was too abstract for the students. We therefore decided to add clarifying text, which resulted in the final measure described below.

Method

Measure

The dependent variable, salience of nationality in youths’ spontaneous self-concept, was measured using the following statements: ‘Who am I? Just imagine you have to describe yourself to someone. Could you please list five things which come to your mind?’ For the negated self-concept we added ‘Tell us five things you are not. Again, just list things that come to your mind (but please be serious)’.

This qualitative measure was complemented by a quantitative part in the form of a short demographic questionnaire.

Procedure

We presented our project outline to the respective headmasters of each school and were pleased to obtain permission to conduct our study. As the subjects were minors, parent consent forms were distributed at each school well ahead of the study. No parent refused participation of their child. The test material (in German) was the same for each school. For the European School, all materials were also available in French and English. For some European School language sections (Finnish and Danish), material was therefore provided only in their second language (L2) and students had the choice between the three languages. As we addressed secondary students in Years 3 and 4, an excellent command of the second language could be expected. The students completed the survey at school in their classrooms, observed by their teacher. The second author was also present at the German school. Students completing the questionnaire in the classroom, surrounded by classmates, was also assumed to reinforce the school context.

In the next section, we will first provide a general description of the participant sample, followed by a more detailed description of the national makeup. Subsequently, the quantitative and qualitative results concerning salience of nationality will be presented.

Samples

Descriptive

The German school sample consisted of 117 students and the European School sample of 106 students, but two subjects had to be excluded, as the questionnaires had not been completed properly. This makes for a total sample of 221 students. The students were aged between 12 and 16 years, and the average age was 13.64 years (SD = 1.72). Gender was equally distributed in both groups (in total: 49% female; 51% male). Participants at the German school attended Years 7 and 8 (7. and 8. Klasse), corresponding to secondary Years 3 and 4 (S3 and S4) at the European School, where the following language sections (LS) participated: English (EN; n = 45), German (DE; n = 25), French (FR; n = 14), Finnish and Danish (FI; DK; n = 10 each). As anticipated, the nationality makeup of the student samples at the two schools was quite different.

National makeup of the school samples

Uniformly, all students at the German school defined their first nationality as German. Only four students (3%) mentioned a second nationality (two Italian, one Austrian and one Bavarian). Only one student was born in another country than Germany, namely England. Regarding mother tongue, all defined their first mother tongue as German. As second mother tongue, three named Italian, and one named English. It is also noteworthy that three specified the regional dialect Bavarian and one even specified the more specific Lower Bavarian as second language. A third nationality or mother tongue was not mentioned. Nearly all mothers (97.4%) were born in Germany. The three exceptions were born in Ireland, Spain and Austria. The fathers were also mostly born in Germany (95.7%), with five exceptions: two were born in Italy, one in Austria, one in Cameroon and one in Congo.

At the European School, 20 different first nationalities were mentioned, not counting regional variations within the United Kingdom. The nationalities are listed in Table 1.

Table 1.

Frequency of first-named nationalities at the European School.

First-named nationality	Frequency of mentions
British, including separate mentions of English, Welsh, Scottish and Northern-Irish	17
German	12
Irish	10
Danish	10
Finnish	9
French	8
Luxembourgish	7
Belgian	6
Lithuanian	5
Italian	4
Austrian	3
Hungarian	2
Maltese	2
Slovakian	2
American	1
Bulgarian	1
Canadian	1
Croatian	1
Latvian	1
Polish	1

Forty-five (43%) of the 104 subjects named a second nationality and 11% also mentioned a third nationality. Concerning first-named mother tongue, English was most frequent (n = 29), followed by German (n = 18), French (n = 16), Finnish (n = 10) and Danish (n = 9). Five students also named Lithuanian as their first mother tongue, two Maltese, two Luxembourgish, two Italian, two Hungarian and two Slovakian. Seven more languages were named once. Just over a third of the students also named a second mother tongue and five students even a third. Regarding students’ country of birth, Luxembourg (n = 41) was most dominant. Parents’ country of birth was also much more diversified than in the German sample, as can be implied by the first-named nationalities of the children.

These comparative breakdowns of nationalities, country of birth of students and parents as well as languages spoken illustrate that we did indeed manage to identify two schools which differed in the composition of the national mix of children.

Results

Distribution of self-concept-related statements

Our student sample of 221 produced a total of 2311 responses to the questions of ‘Who am I’/’Who am I not’. We coded all responses and developed a specific taxonomy. The lion’s share of responses (22%), which included attributes such as friendly, nice or helpful, were classed as ‘agreeable’. These morally good attributes were the most frequently named category in students’ spontaneous self-concept. This finding matches the Shell Youth Study which showed that mandatory social and moral rules are important to youths (Albert et al., 2010). The next most frequently mentioned categories were leisure activities and interests (14%) followed by physical descriptions (11%; e.g. height, weight, hair/eye colour, etc.). Apart from differences in points of emphasis, for example, stronger language focus at the European School or more mentions of leisure activities (sports and music) at the German school, there were few significant differences between the two school contexts. Youth-specific topics dominated the responses.

Nationality markers were defined as any mentions of a specific nationality, native country or country of birth. Overall, nationality markers were not mentioned so frequently (0.87% of the total responses).This result is, however, in line with McGuire et al.’s study. In their sample of n = 560 respondents, only 3% of the children made any mention of ethnicity. By design, only half of our sample was nationally mixed and, following Hypothesis 1, we expected mentions of nationality markers to come only from half or our total sample of 221. Thus, proportionally, the absolute number of mentions is comparable. In the following section, we will test whether or not the school context influenced the rate of mentions.

Testing of Hypothesis 1: salience of nationality in the homogeneous versus heterogeneous context

We did indeed find that context matters: there were significantly more mentions of nationality at the European School compared to the German school. The Chi-square test (with Yates continuity correction) indicates a highly significant association between school context and the occurrence of nationality markers, χ² (1, 221) = 13.35, p = .00, phi = .26. Thus, these data support the hypothesis that a mixed national context increases students’ awareness of nationality. Within the European School, 16.3% of students mentioned nationality within their spontaneous self-concept, compared to 1.7% at the German school. In line with McGuire et al.’s (1978) findings, there was no gender effect within our student sample, χ² (1, 221) = .42, p = .26, phi = .06.

When analysing the data, we also noticed that 11 students, 10 of whom attended the European School, mentioned in the negated self-concept ‘I am not racist’. Even though this statement is not a nationality marker in the above-defined sense, this statement suggests an awareness of nationalities.

Figure 1 illustrates the results, if these anti-racism statements are added to the nationality markers.

Figure 1.

Percentage of nationality and anti-racism markers as a function of school context.

The Chi-square test is also highly significant, χ² (1, 221) = 23.37, p = .00 and the effect size of phi = .34 suggests a medium effect using Cohen’s (1988) criteria. We broke down occurrence by language section and the results are presented in Figure 2. The German school is also included for comparison purposes.

Figure 2.

Percentage of nationality marker/anti-racism mentions per language section.

Salience of nationality was highest within the French section (28.6%). The lowest mentions of nationality occurred in the German section, but five (half) of all the anti-racism mentions originated in this group. No language section within the European School reached lower scores for salience of nationality or anti-racism than the German school sample, supporting Hypothesis 1.

Testing of Hypothesis 2: salience of nationality in the majority versus minority group

Within the German school sample, all students mentioned German as their first nationality, and all but one student were born in Germany. For a differentiator between minority and majority, we chose the ‘second-named nationality’ and defined those with more than one nationality as ‘minority’. Overall, there were only two mentions of nationality, and indeed one of these mentions was from a minority child in the above-defined sense (‘I am half German, half Cameroonian’), which is commensurate with Hypothesis 2. The other mention was by a (majority) German student (‘I am not Italian’), but without further context information this statement is difficult to interpret. A within-group comparison between minority and majority is thus not applicable for the German school.

While we did not observe sufficient minority members within the homogeneous school context, we had to determine a meaningful definition of minority versus majority in terms of nationality that reliably covered the huge variability observed within the heterogeneous context. By default, the school is composed of different language sections, and we decided to look at each language section separately as a reference group. However, the language sections themselves are composed of different nationalities, such as French and French-speaking Belgians for the French section, or the different nations of the United Kingdom, Ireland and Malta in the English section. The language L1 is actually the lowest common denominator between the students and does not show the complex mix of nationalities at the individual student level. This point will be illustrated using the French section as an example. In Table 2, nationality attributes in the form of country of birth (COB) of the student, parents, first nationality and first-named mother tongue are listed. Second or third nationality or mother tongues are not included as these do not apply to all students.

Table 2.

National background analysis: Nationality marker grid French-language section.

Student	Nationality markers
	COB student	COB mother	COB father	First nationality	First mother tongue
1	DK	DK	PT	LUX	FR
2	FR	FR	FR	FR	FR
3	FR	FR	FR	FR	FR
4	BE	BE	IT	BE	FR
5	LUX	FR	UK	IR	EN
6	IT	IT	IT	IT	IT
7	BE	BE	BE	BE	FR
8	FR	FR	FR	FR	FR
9	LUX	UK	BE	BE	FR
10	LUX	BE	PT	BE	FR
11	FR	FR	GK	FR	FR
12	FR	FR	FR	FR	FR
13	LUX	LUX	LUX	LUX	FR
14	FR	FR	FR	FR	FR

COB: country of birth; BE: Belgium/Belgian; DK: Denmark/Danish; EN: England/English; FR: France/French; GK: Greece/Greek; IT: Italy/Italian; LUX: Luxembourg/Luxembourgish; PT: Portugal/Portuguese.

Highlighted are the francophone ‘mono-national’ students (n = 6). Who should be defined as minority within this class? Those whose first nationality is not French or Belgian? Those whose parents are of mixed nationality, or even the highlighted francophone students who are in fact, numerically, in the minority?

To define ‘majority’ within the language sections, we created a ‘homogeneity’ index, which will be illustrated using the French class as an example: a total of 70 attributes defining the composition of the national background are possible (5 Nationality Markers ×14 Students = 70). We considered the French-speaking countries (France and Belgium) as the default group, and for this class a total of 47 French or Belgian entries can be counted. We then calculated the proportion of main group entries to the total (47/70 × 100), which shows that 67% of attributes belong to the two ‘default’ countries (majority). The various language sections differ regarding nationality mix, which is shown in Figure 3.

Figure 3.

Percentage of main country attributes in each language section.

As can be seen, the Scandinavian sections are much more homogeneous than the other language sections. Homogeneity is overstated, especially for the English section, as the pool of ‘default’ nationalities is already quite wide. For the Finnish section, ‘majority’ will be easier to define than for the heterogeneous English sections.

We classified students’ responses into ‘majority’ and ‘minority’ according to the nationality marker grid (Table 2) and noted the mentions of nationality markers. The findings are summarised in Table 3.

Table 3.

Number of nationality markers mentioned by language section.

Language section	Number of students belonging to			Nationality marker mentioned by student belonging to
Language section	Minority	Majority	Total	Minority	Majority	Total
French	8	6	14	3	2	5
German	21	4	25	2	1	3
Finnish	4	6	10	0	2	2
Danish	5	4	9	2	0	2
English	31	14	45	5	0	5
Total	69	34	103	13	4	17

Table 3 illustrates two points. The cell size for some entries makes statistical analysis difficult and, within the heterogeneous environment, minority children are in fact in the majority. Against this background, qualitative findings regarding mentions of nationality markers will be presented, starting with the now familiar French class. Out of a class of 14, there were five mentions of nationality markers in total. Two ‘minority’ children referred to their mixed parentage status (‘I am Portuguese and Danish’, ‘I am Belgian and English’). Two students addressed their relationship with Luxembourg (their host country), distancing themselves from Luxembourg (‘I am not Luxembourgish’), asserting their nationality (one each from the majority/minority). As pointed out above, the mono-national students are in the minority within this particular French class and one student actually commented ‘Je suis d’une seule nationalité’ – I am of one nationality (all nationality attributes were French). Regarding our definition, this student belongs to the ‘majority’, but his status within this group is actually ‘minority’ so this comment can also be explained using the distinctiveness postulate and is in line with Hypothesis 2.

Within the English section, all five responses came from what we defined as ‘minority’ students. Three times origins were mentioned (two times ‘I am Lithuanian’ and ‘I am Slovakian’) and one student stated mixed parentage: ‘I am half Dutch and half Scottish’. A further student commented on the relationship with Luxembourg and asserted his nationality: ‘I am English. I live in Luxembourg, speak Luxembourgish, but I am English’.

The German section was the most heterogeneous in this study. One student commented on his mixed nationality (half French, half German) and another student asserted his nationality: ‘I am Luxembourgish’ (this student also mentioned being ‘of colour’). One girl mentioned in first place within the negated self-concept: ‘Ich bin keine Deutsche. Manche meinen das, da ich so spreche, aber ich hasse das’ (I am not German. Some think that I am because I speak German, but I hate that). This student belonged to the predefined majority group, but belonged to a minority within the majority. A fifth of the class mentioned anti-racism statements.

In the fairly homogeneous Danish section, two children mentioned being of mixed nationality (I am Danish and French). Two majority Danish children mentioned ‘not being racist’. Finally, in the Finnish section two (majority) students asserted ‘being from Finland’ in their spontaneous self-concept.

As indicated above, the definition of majority versus minority is quite complex within the European School context. However, a closer look at the responses within the context of each language section shows that if nationality markers are mentioned, they are mentioned by minorities. Yet, the definition of minority varies. Minority may constitute having a different nationality from the ‘language section default’, as is the case with the Lithuanians/Slovaks in the English section, or the ‘minority default’ as exemplified by Austrians within the German section. Alternatively, minority status may refer to being of mixed nationalities or even being of one nationality. Minority status may also be defined with reference to the wider school context. The Finnish section, for example, is quite small compared to other language sections, which may explain the number of mentions of Finnish origins. Some students also raised the issue of living in Luxembourg, their current country of residence, and asserted their nationality. In conclusion, our qualitative data tend to support Hypothesis 2.

Discussion

Hypothesis 1 could be clearly accepted – context does matter: commensurate with the distinctiveness postulate, salience of nationality was significantly higher in the heterogeneous school context than in the homogeneous school context. A qualitative and quantitative analysis of responses also points to the acceptance of Hypothesis 2, salience of nationality being higher in minority members. It should, however, be mentioned that, overall, mentions of nationality markers were relatively small. We would like to point out several specific observations.

The European School’s credo ‘Without ceasing to look to their own lands with love and pride, they will become in mind Europeans’ was quoted above. The results from this study suggest that the first part of the school’s credo is achieved – the students do not forget about where they are from; awareness of nationality is, in fact, enhanced. Yet, not a single child mentioned ‘European’ as an attribute. This supra-national category did not feature in students’ spontaneous self-concept. This finding may be explained by the fact that different domains are evoked when evaluating group membership regarding nationality and supra-national entities. Social identity theory differentiates three components, namely cognitive, affective and evaluative structures of identification. Within the European School credo, the home country is couched in the terms ‘love and pride’, which evoke affective and evaluative connotations, and emotions are important in inferring postulates as was noted above. On the other hand, one should become ‘in mind’ European – a cognitive aspect. An attribute mainly anchored in the cognitive sphere may not be triggered in the spontaneous self-concept. The survey of orientations of young men and women to citizenship and European identity of several regions in the EU (Jamieson, 2005) also showed that high exposure to Europe does not ensure high personal identification with Europe.

The European School produced the highest number of nationality markers and the vast majority of ‘I am not racist’ statements in the negated self-concept. This suggests that there is an awareness of other nationalities and tolerance. Five of the 10 mentions came from students attending the German section. In his recent comparative study of personal and social identity construal across Europe, Schmidt-Denter (2011) found that Europeans on the whole are becoming more homogeneous regarding their description of their personal identity. However, there is diversity in the sphere of social identity as this construal process is more influenced by cultural and socio-historic factors. Schmidt-Denter found that Germans are, in general, highly self-critical and aware of the aftermaths of the Second World War. It comes, therefore, as no surprise that half of the anti-racism mentions were in the German section.

Across language sections, the issue of third-country nationality was broached. Usually, the children distanced themselves from their (current) host country, asserting their nationality. Those students usually expressed discomfort with their current situation, which probably explains the salience of this issue.

Another interesting finding in this context is the emphasis on regions, which particularly showed in the Anglophone sample, but also in the German school sample. In the English section, most students were quite specific regarding their country of origin – England, Scotland, Wales and Northern Ireland were mentioned, very few students said ‘British’ and no student mentioned ‘UK’. This matches Condor’s (1996, 2000) finding that the self-identification ‘UK’ was never used. The particularity of ‘regionalism’ was also found within the German school context, where one student mentioned Bavarian as second nationality and three named their mother tongue to be Bavarian, one even specifying Lower Bavarian. ‘Bavarian’ was not triggered in the spontaneous self-concept, but in the quantitative part these students volunteered this additional information. We can thus infer that this must have been important to them – that these students possibly wanted to differentiate themselves from the wider national context. This touches on one of the limitations of the study: we can analyse if nationality was mentioned, but we can only speculate on why. Distinctiveness does play a key role; we could show that minority status increases salience. However, not all students of minority status did mention nationality and we do not know what prompted those who did mention nationality markers. Does being of mixed nationality, for example, provide a sense of pride or a unique feature that serves the core motive of self-enhancement (Fiske, 2010)? From the data, we cannot tell which reference group (e.g. class, school, family, country and nation) the students used for social comparison. Furthermore, we cannot tell how nationality is evaluated. A follow-on study using a different methodology could elaborate on the inter-individual differences regarding salience and evaluation of nationality.

A further limitation of the study was the relatively small class sizes of some language sections, which did not allow statistical analysis across the language sections. In addition, not all language sections of the European School were represented in the study. Furthermore, the homogeneous sample was provided by a school based in Germany. As alluded to above, the German responses may be influenced by the Holocaust and may have caused less salience of nationality within this sample. It would be interesting to complement data from homogeneous contexts in different countries and different educational settings.

Conclusion

The purpose of this study was to investigate salience of nationality as a function of the school context in students’ spontaneous self-concept. We used a non-directive, qualitative measure, allowing students to answer spontaneously. Great care was taken not to prime students or to direct them in any way. The answers should thus be authentic, reflecting attributes that are important to the students and that characterise them. We could show that a nationally heterogeneous context raises awareness of nationalities, especially for minority students, even though the definition of ‘minority’ is quite complex in itself. The hypotheses were drawn from the distinctiveness postulate and our findings confirm McGuire et al.’s (1978) results.

We would like to point out several specific observations. As context does seem to matter regarding students’ self-identification with nationality, it is important for school practice to consider the results of this study. Just bringing students of different nationalities together does not mean that nationalities are ignored and a supra-national category such as ‘European’ or ‘international’ is adopted. On the contrary, we could show empirically that awareness of nationalities is enhanced in a nationally heterogeneous context.

The results from the Anglophone section also suggest that a sense of regional pride is reinforced. However, putting students together under a European ‘cloak’ does not seem to enhance identification with Europe to the extent that ‘European’ becomes salient in students’ spontaneous self-concept. We could also show that the definition of majority/minority was quite fluid within the heterogeneous context. This could also be seen as an enhanced opportunity to practise perspective taking – alternating between the majority and minority perspective, depending on context. Increased exposure to different nationalities raises awareness of differences but, as evidenced by the spontaneous mentions of anti-racist statements, also awareness of respectful living with each other. Concluding with the introductory statement, at a time when international mobility and trade are becoming the norm (Karim, 2012), it is important to appreciate these complex dynamics when bringing people of different nationalities or backgrounds together.

Footnotes

Acknowledgements

We wish to thank the Rhabanus-Maurus-Gymnasium St Ottilien and the European School of Luxembourg for their kind cooperation.

Funding

This research was funded in part by a doctoral grant to the first author by the Luxembourg National Research Fund.

Author biographies

Elke Murdock is a Researcher (Chercheur en Formation) at the University of Luxembourg. She has a degree in Psychology from the Saarland University, a Master’s degree from Western Washington University, Bellingham, WA, USA, which focused on cross-cultural psychology, and an MBA from Heriot-Watt University, Edinburgh. Since 2010 she has been working on her PhD project: ‘The construal of national identities within a multinational context’ at the University of Luxembourg.

Franziska S Hirt earned a trilingual Bachelor of Science in Psychology at the University of Luxembourg. Her Bachelor’s thesis, with the title ‘Salience of nationality in the spontaneous self-concept: A comparison between a homogeneous versus heterogeneous national context’, brought her into contact with cross-cultural research. Currently, she is completing a Master’s degree in work and organisational psychology at the University of Bern, Switzerland.

Dieter Ferring is Professor of Developmental Psychology and Geropsychology and Head of the Integrative Research Unit on Social and Individual Development at the University of Luxembourg, Luxembourg. His main research areas and publications cover life-span development and ageing, including intergenerational relations, self-regulative processes and identity under a life-span perspective, coping and regulation of subjective well-being in old age, as well as technology use and ageing.

References

Albert

Hurrelmann

Quenzel

(2010) 16. Shell Jugendstudie – Jugend 2010: Flyer. Available at: http://www.shell.de/home/content/deu/aboutshell/our_commitment/shell_youth_study/downloads/ (accessed 22 April 2013).

Anderson

(2006) Imagined Communities: Reflections on the Origin and Spread of Nationalism. London: Verso Books.

Arts

Halman

(2006) National identity in Europe today. International Journal of Sociology 35(4): 69–93.

Barrett

Flood

Eade

(eds) (2011) Nationalism, Ethnicity, Citizenship: Multidisciplinary Perspectives. Newcastle: Cambridge Scholars Publishing.

Billiet

(2002) Questions about national, subnational and ethnic identity. European Social Survey: Core Questionnaire Development. Available at: http://www.europeansocialsurvey.org/docs/methodology/core_ess_questionnaire/ESS_core_questionnaire_national_ethnic_identity.pdf (accessed 22 April 2013).

Billig

(1995) Banal Nationalism. London: SAGE.

Bond

Cheung

(1983) College students’ spontaneous self-concept: the effect of culture among respondents in Hong Kong, Japan, and the United States. Journal of Cross-Cultural Psychology 14(2): 153–171.

Broadbent

(1958) Perception and Communication. London: Pergamon Press.

Cohen

(1988) Statistical Power Analysis for the Behavioral Sciences, 2nd edn. Hillsdale, NJ: Lawrence Erlbaum Associates.

10.

Condor

(1996) Unimagined community? Some social psychological issues concerning English National Identity. In: Breakwell

Speri

(eds) Changing European Identities: Social Psychological Analyses of Social Change. Oxford: Butterworth Heinemann, pp. 41–64.

11.

Condor

(2000) Pride and prejudice: identity management in English people’s talk about ‘this country’. Discourse & Society 11(2): 175–205.

12.

Condor

(2011) Towards a social psychology of citizenship? Introduction to the special issue. Journal of Community & Applied Social Psychology 21(3): 193–201.

13.

Cousins

(1989) Culture and self-perception in Japan and the United States. Journal of Personality and Social Psychology 56(1): 124–131.

14.

Epstein

(1973) The self-concept revisited or a theory of a theory. American Psychologist 28: 404–416.

15.

Epstein

(1979) Entwurf einer Integrativen Persönlichkeitstheorie. In: Filipp

(ed.) Selbstkonzept-Forschung: Probleme, Befunde, Perspektiven. Stuttgart: Klett-Cotta, pp. 15–45.

16.

Fiske

(2010) Social Beings: Core Motives in Social Psychology. Hoboken, NJ: Wiley.

17.

Geertz

(1973) The Interpretation of Cultures. New York: Basic Books.

18.

Halkoaho

(2012) Identity-related media consumption a focus on consumers’ relationships with their favorite TV programs. PhD Dissertation, University of Vaasa, Vaasa.

19.

Helmchen

(2005) Die Entstehung der Nationen im Europa der Frühen Neuzeit: ein integraler Ansatz aus humanistischer Sicht, vol. 10. Bern: Peter Lang.

20.

Jamieson

(2005) Orientations of young men and women to citizenship and European identity (Final Report). Available at: http://www.sociology.ed.ac.uk/youth/final_report.pdf (accessed 1 May 2014).

21.

Karim

(2012) The co-existence of globalism and tribalism: a review of the literature. Journal of Research in International Education 11(2): 137–151.

22.

Kuhn

McPartland

(1954) An empirical investigation of self-attitudes. American Sociological Review 19(1): 68–76.

23.

McGuire

Child

. (1978) Salience of ethnicity in the spontaneous self-concept as a function of one’s ethnic distinctiveness in the social environment. Journal of Personality and Social Psychology 36(5): 511–520.

24.

Markus

(1977) Self-schemata and processing information about the self. Journal of Personality and Social Psychology 35(2): 63–78.

25.

Niebergall

(2010) Promoting positive identity among children in a school curriculum. Master’s Thesis, The University of Kansas, Lawrence, KS.

26.

Owens

(2003) Self and identity. In: Delamater

(ed.) Handbook of Social Psychology. New York: Springer Science & Business Media, pp. 205–232.

27.

Péporté

(2011) Constructing the Middle Ages: Historiography, Collective Memory and Nation-Building in Luxembourg, vol. 3. Leiden: Brill.

28.

Pinquart

Silbereisen

(2000) Das Selbst im Jugendalter. In: Greve

(ed.) Psychologie des Selbst. Weinheim: Beltz, pp. 75–95.

29.

Rumpel

(1991) Kulturelle Identität. Eine Konzeptanalyse im interdisziplinären Raum. Unpublished Diplomarbeit, University of the Saarland, Saarbrücken.

30.

Schmidt-Denter

(2011) Die Deutschen und ihre Migranten: Ergebnisse der europäischen Identitätsstudie. Weinheim: Beltz Juventa.

31.

Schola Europaea (n.d.) School homepage. Available at: http://www.eursc.eu/index.php?id=132 (accessed 29 April 2013).

32.

Schubert

Klein

(2006) Das Politlexikon. Bonn: Dietz Verlag.

33.

Tajfel

Turner

(1979) An integrative theory of intergroup conflict. Social Psychology of Intergroup Relations 33: 47.

34.

United Nations (n.d.) The universal declaration of human rights. Available at: http://www.un.org/en/documents/udhr/ (accessed 29 April 2013).

35.

Weinreich

Saunderson

(2003) Analysing Identity: Cross-Cultural, Societal and Clinical Contexts. Hove: Routledge.