Sage Journals: Discover world-class research

Abstract

Intergroup relation perspectives stem from research in Western contexts with clear distinctions between the dominant and nondominant groups. In South Africa, with at least 13 different cultural groups and 11 official languages, no group is dominant in all life spheres. We examine the relationship between identity and in-/out-group orientation across Black-Zulu, Coloured (mixed racial ancestry), Indian, and White-Afrikaans emerging adults (N = 390; 75% females, M _age = 19.97 years, SD = 2.44). Results indicate that personal identity for all groups and ethnic identity for Black-Zulu, Indian, and White-Afrikaans emerging adults were important for intergroup relations. Black-Zulu, Coloured, and Indian emerging adults distinguish themselves less from others, whereas White-Afrikaans emerging adults are less open to others. Ultimately, the complexity of intergroup relations in South Africa has implications for the effective transformation interventions needed to counter experiences of threat and make group boundaries more flexible for emerging adults.

Keywords

in-group out-group integrated threat theory minorities social identity theory South Africa

South Africa, with its host of cultural groups and languages, is known for its diversity. Race, ethnicity, tradition, and religion are prominent in most spheres of daily life (Adams, van de Vijver, & De Bruin, 2012). Although it is certainly not the most diverse country in the world (compared to countries such as Kenya and the Cameroon; see Adams et al., 2016), identity and intergroup relations in South Africa are nonetheless particularly complex—with cultural traditions that long preceded apartheid, that were reinforced by the legalized segregation of the apartheid system, and that still continue (Seekings, 2008). Contemporary South Africa is still known for the large economic, racial, and cultural differences between groups, with each ethnocultural group being confronted with real, yet very different forms of legal, political, and economic oppression (Adams, van de Vijver, De Bruin, & Torres, 2014; Ferguson & Adams, 2015). South Africa’s situation is atypical from the Western world in that there is no clearly dominant group (Adams et al., 2016). These complex intergroup relations in South Africa have important implications for in- and out-group categorization, intergroup relations, and the development of identity. The objective of this study is to examine how personal, ethnic, and religious identity aspects, important during all developmental stages of life, particularly prioritized during adolescence and emerging adulthood would inform us about emerging adults orientations toward both their own (in-) groups and other (out-) groups.

South Africa’s diversity is reflected in its higher education institutions (Nel, Nel, Adams, & De Beer, 2015). Serving as a microcosm of society, South African universities provide emerging adults with a dynamic and diverse environment in which to search for and form their identities. Emerging adults in the South Africa of today are living in a country remarkably different from the one in which previous generations did. The transition into a democratic dispensation during the 1990s created many more opportunities for interracial and intercultural contact for youth than their parents ever experienced (Bornman, 2011; Ferguson & Adams, 2015).

Intergroup Perspectives on Identity

We consider two theoretical perspectives on intergroup relations and their associations with identity. These are social identity approaches (SIAs; Haslam, 2001; Postmes, Tanis, & de Wit, 2001) and integrated threat theory (ITT; Stephan, Diaz-Loving, & Duram, 2000). SIA, which is based on social identity theory (SIT; Tajfel & Turner, 1986) and self-categorization theory (Turner, 1999), accounts for the manner in which individuals use social categories to define themselves and derive meaning from their involvement in group processes (Niens, Cairns, Finchelescu, Foster, & Tredoux, 2003). These theoretical perspectives consider the importance of identity and address how people construct meaning about themselves as either similar to (social identity) or distinct from (personal identity) others.

Social identity accounts for those components of an individual’s self-concept that stem from membership of a social group and the value, as well as the significance, attached to this group membership (Tajfel & Turner, 1986; Turner, 1999). It is formed through connections with and commitment to various social units or groups, including the beliefs, values, and goals that are aligned with group expectations (Adams, 2014). Two forms of social identity, namely, ethnic and religious identity, are of interest in this study, as they are regarded as important for the social identity development of the youth of South Africa (Adams et al., 2016; Sibusiso & Mdikana, 2013). These two social identity aspects are also generally regarded to be strongly related (Safran, 2006).

Ethnic identity is a dynamic multifaceted construct defined as the part of the self that stems from individuals’ knowledge of and commitment to their self-identified ethnic group (Cislo, 2008; Phinney & Ong, 2007). Ethnic identity includes components such as self-categorization; in-group attitudes, values, and beliefs; the importance of group membership; and behavioral involvement, as negotiated with a particular ethnic group. In the same vein, the process in which individuals experience a sense of belonging to a religious group and ascribe to its beliefs and/or practices defines their religious identity (Balkin, Schlosser, & Levitt, 2009). As with other forms of identity, religious identity is constructed by tradition and social discourse. Religious identity connects people to their religious beliefs or affiliations (Landman, 2013). Arnett and Jensen (2002) posited that exploration, which is a salient aspect of development during emerging adulthood, also includes the formation of distinct sets of beliefs and values regarding religious issues. In the South African context, where the majority of the population is religious, Sibusiso and Mdikana (2013) and Nel, Nel, Adams, and De Beer (2015) argue that religion provides youth with an ideological, social, and spiritual context for identity exploration; religion influences both their decision-making processes and how they interact with people who are culturally different from themselves.

Personal identity, on the other hand, is defined by various characteristics including values, beliefs, ideas, and emotions that individuals deem important to their sense of self (Schwartz, Montgomery, & Briones, 2006). The basis of personal identity resides more in idiosyncratic characteristics (differences in perceptions and actions) and tends to be more consistent across social settings compared to social identity which is more characteristically normative (the uniformity of perception and action among group members) and often varies based on the salience of group status in a specific setting (Adams, 2014; Stets & Burke, 2000; Tajfel & Turner, 1986). Thus, with a need for a more integrated understanding of the identity–intergroup relations and very little research on how personal identity relates to intergroup relations, we consider the inclusion of personal identity important for this study.

Group orientation and the role of threat

How groups relate to each other could be considered under the guise of group orientation, which is the extent to which individuals identify, make contact, and participate with in-group or out-group members (Ferguson & Adams, 2015). This implies that certain constructs and qualities, such as race, culture, or religion, inform group boundaries. Individuals are categorized by themselves or others according to the primary group to which they belong (in-group) and they use this categorization to compare themselves with other groups (out-group; Tajfel & Turner, 1986). Depending on the salience of self-categorization to a specific in-group, as well as its relative status within the group, boundaries can be more permeable (i.e., easier to cross for members from another group) or more rigid and clear-cut. As identification would inform in-group orientation, social and personal identity aspects would, therefore, help to define boundaries, which are then used in establishing a distinction between the in-group and out-groups; an in-group orientation can be expected to be a prerequisite for and to develop earlier than an out-group orientation. In an attempt to enhance their own in-group identity, individuals/groups seek to promote clear group boundaries and distinguish themselves positively from others (positive distinctiveness). A strong in-group orientation is thus associated with more clearly defined boundaries.

With respect to ITT (Stephan et al., 2000), realistic threat (the fear of out-group members threatening economic or physical well-being), symbolic threat (the fear of out-group members threatening cultural values and norms), intergroup anxiety (the fear of making contact with out-group members), and stereotypes (the beliefs—often negative—held about a group in its entirety) are related to the importance, amount, and quality of intergroup contact. Realistic threat and symbolic threat, in particular, are considered crucial for promoting the need for distinctiveness/separation. This need for distinctiveness/separation is particularly salient when groups differ in terms of social status, as groups with higher status often seek to distinguish themselves more (Anderson & Kilduff, 2009).

Emerging Adulthood in the South African Context

South African emerging adults are generally categorized into one of four major ethnocultural groups: Black people (79.4% of the South African population), Mixed Racial Heritage people (“Coloured” in the South African nomenclature, and this term and spelling will henceforth be used; 8.8%), Indian people (2.6%), and White people (9.2%; Statistics South Africa [StatsSA], 2012). These four major groups are further divided into various distinct cultural groups speaking 1 of the 11 official languages (Ndebele, Pedi, Sotho, Swati, Tsonga, Tswana, Venda, Xhosa, Zulu, Afrikaans, and English). Due to previous and current political, social, and economic segregation, relationships between ethnocultural groups have been a challenge in nation-building. Since the country’s transformation into a democratic dispensation in the 1990s, there have been various initiatives aimed at social and economic redress, as well as attempts to generate economic growth and facilitate the development of a stable Black, Coloured, and Indian middle class. According to Adams, van de Vijver, and De Bruin (2012), the Black and White groups are culturally and linguistically probably most distant from each other. Coloured and Indian groups are often grouped with the Black group. However, culturally, they are in many ways distinct from both Black and White groups.

There are emerging adults from four particular groups in this study: Black-Zulu, White-Afrikaans, Coloured, and Indian. In the current South African socioeconomic context, emerging adults from the Black-Zulu group (approximately 28.1% of the population) are part of the numerical majority and politically dominant Black group, which is more prominently represented in the South African government and the political sphere. Emerging adults from the White-Afrikaans group (5.3%) are part of the economically dominant White group, which has a proportionally higher socioeconomic status and is more affluent than the other groups (Adams et al., 2016). As a minority group, the dominant economic position of emerging adults from the White-Afrikaans group is currently threatened by the politically and numerically dominant Black group. Emerging adults from the Coloured and Indian groups are economically positioned between the Black and White groups, with the Indian group holding an economically stronger position than the Coloured group (StatsSA, 2012). However, neither during nor after the abolishment of apartheid have either the Coloured or Indian groups been in dominant positions, whether politically, economically, or socially.

In pluralistic societies such as South Africa, where ethnic identity might act as a buffer in situations of group conflict, threat, and competition for limited resources, in-group identification is of particular importance. Adams et al. (2016) found in their research that both Black and White South African emerging adults have salient ethnic identities. They argue that ethnic identity and in-group affiliation remain important in the history of ethnocultural divide and the continued polarization in postapartheid South Africa. Nel et al. (2015) are of the opinion that White-Afrikaans-speaking South Africans are confronted with various barriers related to stereotyping, social categorization, and symbolic threat. These authors found that both ethnic and religious identity are salient in the Afrikaans youth and influence how they behave in multicultural settings.

The Present Study

Much of the existing literature on identity and intergroup relations stems from research in Western contexts where there are clear dominant/majority and nondominant/minority groups (Smith, Fischer, Vignoles, & Bond, 2013). There is often a focus on ethnic identity and national identity and how these relate to the way in which these majority and minority groups relate to one another (Esses, Dovidio, Jackson, & Armstrong, 2001; Tsai, Chentsova-Dutton, & Wong, 2002). This is, however, not the case in many non-Western contexts, such as South Africa, where multiculturalism has evolved differently and where there are often different psychological dynamics at play. We set out to examine the associations between personal, religious, and ethnic identity and in-group and out-group orientation across South African emerging adults in four groups, namely, Black-Zulu, Coloured, Indian, and White-Afrikaans.

We investigated the relationships between identity and intergroup relations for emerging adults in the South African context. We wanted to examine which aspects of identity would be most informative of in-group orientation and how would this, in turn, inform orientation toward out-groups. A multicultural context facilitates the development of more salient social identities (Adams et al., 2016). While this fact is widely acknowledged, it is understudied in contexts where there is no clear distinction along dominant/nondominant and majority/minority lines. We examined how two presumably relevant social identity aspects, namely, ethnic and religious identity, inform group orientation. In order to build a more comprehensive understanding of the identity–intergroup relations model, we also included personal identity. We would argue that both personal and social identities can inform in-group and out-group orientation. More specifically, it can be expected that identities that emphasize group uniqueness will inform stronger in-group and weaker out-group orientations and that identities that emphasize inclusiveness will do the opposite. Ethnic identity is always in-group oriented, whereas religious identity can do the same but could also be inclusive (e.g., a strong Muslim identity could include Muslims from multiple ethnic groups and exclude Christians). Moreover, we expect that in-group orientations inform out-group orientations, as the latter develop later in life. As we do not expect that the associations will be invariant across groups, we refrain from specifying specific hypotheses. Rather, we have derived three research questions, which are presented in the model of Figure 1.

Figure 1.
Identity and group orientations model.

Research Question 1: How are personal, religious, and ethnic identity aspects associated with in-group orientation across Black-Zulu, Coloured, Indian, and White-Afrikaans emerging adults?

Research Question 2: How is in-group orientation associated with out-group orientation across Black-Zulu, Coloured, Indian, and White-Afrikaans emerging adults?

Research Question 3: Does in-group orientation mediate the relationship between personal, religious, and ethnic identity aspects and out-group orientation across Black-Zulu, Coloured, Indian, and White-Afrikaans emerging adults?

Method

Participants and Procedure

Data were collected as part of a larger study on identity, personality, culture, and well-being among undergraduate students from several universities in South Africa. The sample consisted of 390 participants (75% females, M _age = 19.97 years, SD = 2.44) and comprised participants from South Africans of White-Afrikaans¹ (n = 120; 67% females, M _age = 19.81 years, SD = 2.40), Coloured (n = 115; 84% females, M _age = 20.07 years, SD = 2.84), Indian (n = 75; 81% females, M _age = 19.61 years, SD = 1.64), and Black-Zulu (n = 80; 68% females, M _age = 20.43 years, SD = 2.46) descent.

Measures

Sociodemographic questionnaire

Participants provided their age, gender, and the highest level of education of both parents (averaged to ascertain Social Economic Status [SES]). Chi-square analysis indicated significant differences across ethnocultural groups in terms of gender, χ²(3, N = 389) = 12.90, p = .005, with more females in the White-Afrikaans and Coloured groups. Analyses of Variance indicated no group differences in terms of age, F(3, 386) = 1.71, p = .164, ηp² = .01, and that White-Afrikaans participants had higher SES, F(3, 386) = 18.96, p < .001, ηp² = .13.

Identity

Participants completed three identity measures. The first was an adapted version of the identity subscale from the Erikson Psychosocial Stage Inventory (Rosenthal, Gurney, & Moore, 1981), which measures personal identity. It comprises two subscales, personal identity synthesis and personal identity confusion with 6 items each, rated on a 5-point Likert-type scale ranging from 1 (not applicable to me) to 5 (always applicable to me). Items include “I change my opinion of myself a lot” and “I know what kind of person I am.” All 12 items are used to compute a general score for personal identity with Items 1, 2, 7, 10, 11, and 12 reverse scored.

Two scales that measure social identity were administered. The first was the Religious Identity Scale Short version (Adams, 2014), which measures how individuals feel about their religious views, as a measure of the centrality of a person’s religious beliefs and groups. It is a unidimensional scale with 6 items rated on a 5-point Likert-type scale ranging from 1 (not applicable to me) to 5 (always applicable to me). Items include “I perceive myself as a member of my religious community” and “My religious beliefs will remain stable.” We calculated a mean score for the 6 items.

The second was the Belonging subscale of the multiethnic identity measure (Phinney, 1992), which measures ethnic identity. The full measure (comprising 12 items) has two subscales, namely, Ethnic Identity Exploration and Ethnic Identity Belonging. In this study, we used the latter subscale with items such as “I have a strong sense of belonging to my own ethnic group.” The scale has 7 items, rated on a 4-point Likert-type scale ranging from 1 (strongly disagree) to 4 (strongly agree).

Group orientations

Participants completed the 48-item South African Cultural Orientation subscale from the Acculturation Rating Scale for South Africans (ARSSA; Ferguson & Adams, 2015). The scale measures a participant’s orientation toward the four major South African ethnocultural groups (four subscales) and general language usage (fifth subscale) on a 5-point Likert-type scale ranging from 1 (none or not at all) to 5 (very much or always). Each of the first four subscales has 11 items that measure contact, participation, associations, and identification with each of the four major ethnocultural groups in South Africa. Items in each subscale are similar but refer to different ethnocultural groups: (a) the Black South African Orientation subscale (BSAOS; e.g., “I associate with Black/African South Africans”), (b) the Coloured South African Orientation subscale (CSAOS; e.g., “I associate with Coloured South Africans”), (c) the Indian South African Orientation subscale (ISAOS; e.g., “I associate with Indian South Africans”), and (d) the White South African Orientation subscale (WSAOS; e.g., “I associate with White South Africans”). The final subscale, (e) General Language Usage subscale (GLUS), was added to the ARSSA for the purposes of this study. This subscale has 4 items that measure preference of language usage in verbal and written communication (e.g., “I write (letters, e-mails, and other correspondence) in my home language”). We used principal component analysis to confirm the unidimensional structure of the five subscales that measure intergroup relations.

In-group and out-group orientations

After computing mean scores for each subscale, we computed in-group and out-group orientations for each participant. The score for the in-group orientation was the average of the mean scores of BSAOS and GLUS for Black-Zulu emerging adults and WSAOS and GLUS for White-Afrikaans emerging adults, with CSAOS and GLUS for Coloured emerging adults and ISAOS and GLUS for Indian emerging adults. The out-group orientation score comprised the mean scores of the remaining orientations for each group. For example, for Black-Zulu emerging adults, the mean score of CSAOS, ISAOS, and WSAOS was calculated to indicate out-group orientation. Measurement invariance was not calculated, as different target groups are presented in the items for each of the in-group and out-group orientations, which means that these items are not directly comparable (see Ferguson & Adams, 2015, for a similar argument).

Statistical Analyses

Our analytic process entailed three aspects, namely, preliminary analyses, descriptive statistics, and model testing. For the preliminary analyses, we assessed the internal consistencies of all the measures using Cronbach’s α, which is considered acceptable above .70 (Tabachnick & Fidell, 2001), using IBM SPSS version 22. We evaluated the measurement invariance as three nested models at configural, metric, and scalar level for the identity measures in AMOS version 22. In terms of general model fit, we first considered the normed Chi-square (χ² /df) for reasonable (χ² /df < 5) and good (χ² /df < 2) fit (Bollen, 1989). Next, the Tucker–Lewis index and the comparative fit index (CFI) were assessed for reasonable (>.90) and good (>.95) fit, as well the root mean square of approximation (RMSEA <.08 indicates a reasonable fit and RMSEA <.06 indicates a good fit). In terms of nested models, the change in the χ² from less restrictive to more restrictive models should not be significant. However, considering that the χ² is sensitive to sample size, the change in CFI is often considered more informative for comparing more restrictive with less restrictive models, with a value of less than or equal to .01 between two models pointing to a good fit of the more restrictive model (Milfont & Fischer, 2010).

We were also interested in mean group differences. Mean differences were compared using one-way ANOVAs with the ethnocultural group as the independent variable and all identity and group orientation variables as dependent variables. Finally, to answer our research questions, we tested a multigroup path model in which in-group orientation mediated personal, religious, and ethnic identity aspects and out-group orientation (as presented in Figure 1). Similar criteria were used to assess the fit of these nested models as with the multigroup CFA models, which were used to assess measurement invariance.

Results

Preliminary Analyses

The internal consistencies, presented in Table 1, were acceptable for all measures. In Table 2, we present the measurement invariance results. We obtained configural and metric invariance for all three identity measures and partial scalar invariance for personal (intercepts constraints for Items 7 and 9 were released) and ethnic identity (intercepts constraints for Item 2 was released). Full scalar invariance was obtained for religious identity (van de Vijver & Leung, 1997). Our analyses suggested that a direct comparison of mean scores across cultural groups was not warranted. However, given that we obtained partial scalar invariance for most scales and the fit values of the scalar invariance tests were not poor, we decided to test group differences in means; these outcomes should be interpreted with caution, given the incomplete evidence for scalar invariance.

Table 1.
Scale Reliabilities Across Ethnocultural Groups.

Measures Items Black- Zulu EA Coloured EA Indian EA White- Afrikaans EA

Identity

Personal identity 12 .78 .86 .83 .81

Religious identity 6 .88 .92 .95 .90

Ethnic identity (belonging) 7 .84 .89 .94 .91

Intergroup relations

BSA Orientation Scale 11 .70 .78 .80 .88

CSA Orientation Scale 11 .78 .86 .85 .90

ISA Orientation Scale 11 .79 .79 .76 .89

WSA Orientation Scale 11 .77 .79 .69 .80

General Language Usage Scale 4 .74 .81 .91 .77

Note. Measurement invariance not calculated for Intergroup Relation subscales as different target groups was used to measure in-group and out-group orientations. BSA = Black South African; CSA = Coloured South African; EA = Emerging Adults; ISA = Indian South African; WSA = White South African.

Table 2.
Measurement Invariance for Identity Measures.

Models χ²/df TLI CFI RMSEA AIC BCC

Personal identity

Configural invariance 1.66* .88 .92 .04 656.99 715.42

Metric invariance 1.61* .89 .91 .04 635.70 684.17

Scalar invariance 1.66* .88 .89 .04 635.59 672.10

Partial scalar invariance 1.58* .90 .90 .04 616.57 655.08

Religious identity

Configural invariance 4.48* .87 .93 .10 305.25 317.20

Metric invariance 3.42* .91 .93 .08 288.78 298.23

Scalar invariance 2.93* .93 .92 .07 280.20 286.67

Ethnic identity (belonging)

Configural invariance 3.42* .87 .92 .08 359.40 375.53

Metric invariance 2.99* .90 .91 .07 353.07 365.73

Scalar invariance 2.79* .91 .90 .07 355.09 363.72

Partial scalar invariance 2.78*** .91 .90 .07 351.83 361.04

Note. When reading comparing models, the partial scalar invariance model needs to be compared with the metric invariant model. CFI = comparative fit index; TLI = Tucker–Lewis index; RMSEA = root mean square error of approximation; AIC = Akaike information criterion; BCC = Browne-Cudeck Criterion.

***p < .001.

Mean Score Comparisons

Separate one-way ANOVAs indicated that there were significant differences across groups for religious identity, in-group orientation, and out-group orientation. As can be seen by the means, standard deviations, F statistics, and eta-squares (η²) presented in Table 3, religious identity was significantly more important for White-Afrikaans emerging adults compared to that of the other groups. In-group orientation increases across groups in the following order: Coloured, Black-Zulu, Indian, and White-Afrikaans. Out-group orientation is highest in the Black-Zulu, Indian, and Coloured groups and significantly lower in the White-Afrikaans group.

Table 3.
Ethnocultural Group Mean Differences for Identity and Intergroup Relations.

Black-Zulu EA Coloured EA Indian EA White-Afrikaans EA Ethnocultural Group

Measures M (SD) M (SD) M (SD) M (SD) F(3, 386) η²

Identity

Personal identity 3.89 (0.56) 3.80 (0.67) 3.78 (0.58) 3.76 (0.59) 0.80 .01

Religious identity 3.69 (0.83)_a 3.66 (1.01)_a 3.51 (1.10)_a 3.94 (0.75)_b 3.85* .03

Ethnic identity (belonging) 3.13 (0.55) 3.10 (0.58) 3.16 (0.63) 3.27 (0.58) 0.31 .00

Intergroup relations

In-group orientation 3.78 (0.56)_a 3.58 (0.46)_a 4.05 (0.73)_b,c 4.41 (0.54)_b,d 45.88* .26

Out-group orientation 3.01 (0.45)_d 2.60 (0.51)_b,e 2.75 (0.57)_c,e 2.02 (0.73)_a 52.36* .30

Note. Means with a different subscript are significantly different in a post hoc test. EA = Emerging Adults. Black-Zulu EA: n = 80; Coloured EA: n = 115; Indian EA: n = 75; White-Afrikaans EA: n = 120.

p < .05. p < .01. p < .001.

Model Assessment: Identity and In-Group and Out-Group Orientations

In the multigroup path model, the structural weights model assesses similarities of regression weights across groups. When all parameters were constrained to be equal across groups (structural weights model), the fit was poor in relation to the unconstrained model (see Original Model tested in Table 4). Firstly, the modification indices suggested that two paths seemed to differ in at least one group. These paths were between (in the order in which they were released) (a) in-group orientation and out-group orientation for the White group and (b) ethnic identity belonging and in-group orientation for the Coloured group. As can be seen in Table 4 (Modified Model), releasing these constraints resulted in an acceptable structural weights model, χ²(22, N = 390) = 40.41, p = .010, χ²/df = 1.84, CFI = .93, RMSEA = .05 in comparison with the unconstrained model (ΔCFI = .010).

Table 4.
Fit Statistics for Multigroup Model.

Models χ²/df AGFI TLI CFI RMSEA Δχ² Δdf

Original model

Unconstrained 2.32 .87 .79 .94 .06 — —

Structural weights 3.20* .82 .65 .79 .08 48.97* 12

Structural covariances 3.01* .84 .68 .67 .07 49.58* 18

Structural residuals 3.94* .79 .53 .44 .09 63.06* 6

Modified model

Unconstrained 2.32 .87 .79 .94 .06 — —

Structural weights 1.84* .90 .87 .93 .05 12.63 10

Structural covariances 2.25* .87 .80 .80 .06 49.58* 18

Structural residuals 3.23* .81 .65 .59 .08 58.44* 6

Note. Most restrictive model with a good fit is in italics. AGFI = adjusted goodness-of-fit index; TLI = Tucker–Lewis index; CFI = comparative fit index; RMSEA = root mean square error of approximation.

p < .05. p < .01. p < .001.

As illustrated in Figure 2, we found that personal identity was predictive and that religious identity was not predictive of in-group orientation for all groups of emerging adults and that ethnic identity was predictive of in-group orientation for emerging adults of all groups except for the Coloured group (Research Question 1). With respect to Research Question 2, in-group orientation was positively associated with out-group orientation for the Black-Zulu, Coloured, and Indian emerging adults, but was negatively associated for White-Afrikaans emerging adults. In terms of Research Question 3, we found that in-group orientation mediated the relationship between (a) personal identity and out-group orientation for all emerging adult groups, (b) ethnic identity and out-group orientation for Zulu-Black, Coloured, and Indian emerging adults, with personal identity negatively predicting out-group orientation; and (c) did not mediate the relationship between religious identity and out-group orientation.

Figure 2.
Integrated identity and group orientations model. A = White-Afrikaans White, C = Coloured, I = Indian, Z = Black-Zulu, ethnic identity (B) = ethnic identity belonging. p < .05. p < .01. p < .001.

Discussion

The objective of this study was to assess the relationship between identity and in-group/out-group orientations across four South African groups of emerging adults, namely, Black-Zulu, Coloured, Indian, and White-Afrikaans. Results indicated that identity aspects were reasonably strong (above the scale midpoint) for all emerging adults and that groups clearly favored their own group over others. More specifically, White-Afrikaans emerging adults present with the strongest in-group and weakest out-group orientation. In contrast, emerging adults from the Black-Zulu group seem more open to contact with others (a strong out-group orientation combined with a moderate in-group orientation). Coloured emerging adults seemed neither to emphasize their own group membership nor to seek contact with out-groups (lowest in-group and moderate out-group orientation); Indian emerging adults were quite open to contact with others, while distinguishing themselves more clearly from others (high in-group orientation and moderate out-group orientation).

Identity and Group Orientation Across Ethnocultural Groups

In terms of our expected model, we found that the global patterns were similar for Black-Zulu, Coloured, and Indian emerging adults, but the pattern was quite different for White-Afrikaans emerging adults. While we expected that a group’s ethnicity and religion usually walk hand-in-hand and that there is often no separation between the two (Safran, 2006), ethnic identity presents or restricts an avenue for contact with others, whereas religious identity seemed unrelated for all groups. We found that South African emerging adults seem to detach themselves somewhat from their religion when relating to their own group and other groups (cf. Arnett & Jensen, 2002).

As the group that stands most clearly apart from the others, White-Afrikaans emerging adults identify more strongly with their own group and reveal a stronger sense of an “us” versus “them” demarcation. Personal and social identity aspects relate to in-group orientation similarly in other groups; yet, mean scores are not the same across groups, a strong in-group orientation allows White-Afrikaans emerging adults to distinguish themselves from others at a group level. As the historically politically, economically, and socially dominant group, this clear distinction may be due to White-Afrikaans emerging adults experiencing the South African transition to a multicultural society as the most threatening, both symbolically and realistically (Stephan et al., 2000).

The pattern discussed above is in sharp contrast to the openness presented by other groups, in particular, Black-Zulu emerging adults, which may be due to their current status as part of the politically dominant group and the largest of the Black South African groups. According to Mtshelwane, Nel, and Brink (2016), Black-Zulu group members are traditional and proud of their heritage and cultural rituals. They may, therefore, experience less threat to their group identity and may be more open to other groups. Another reason for their openness could be that this group, being one of the groups most heavily discriminated against in the past, would have much to gain economically from contact with other groups, in particular, the White-Afrikaans group. Despite being a small minority (2.6% of the South African population), the Indian group has become the second most affluent group in South Africa in the last 25 years (StatsSA, 2012). Their openness to other groups also seems regulated by their strong ethnic identity (similar to what we found for the Black-Zulu group).

The results of this study portray interesting similarities and differences with a somewhat similar study conducted by Duckitt, Callaghan, and Wagner (2005). While the four ethnic groups that participated in their study (Africans/Blacks, Indians, White English, and White Afrikaans) differ somewhat from the current study, some comparison is possible. Like Duckitt et al., we found pronounced in-group favoritism. Furthermore, Duckitt et al. found marked reciprocal hostility between Blacks and Afrikaners (for both Blacks and Afrikaners, positive in-group attitudes were associated with negative evaluations of the other).

The result of our study replicates this finding for the White-Afrikaans group (still with strong in-group and weak out-group orientations). However, the Black-Zulu groups showed the opposite pattern (with much more openness). Duckitt et al. (2005) explained their findings by referring to how the experience of discrimination and prejudice relates to negative intergroup relations (especially in South Africa’s past). This argument might still be true for the White-Afrikaans group (the transitions in South Africa as threateningly challenging their historically dominant position) but not for Black-Zulu emerging adults (who moved from being discriminated against in the past to gaining political and potentially more economic power).

While it may be a possibility that the nonsignificant relationship between ethnic identity and in-group orientation (β = .16) may be due to sufficient statistical power in our Coloured sample. This lack of association, compared with the other groups, may also be explained by the fact that Coloured individuals have struggled to find their place in South African society (Dannhauser, 2006). The Coloured group showed a pattern of average scores of personal and religious identity and low scores on ethnic identity and low scores on in-group orientation. This pattern confirms the lack of a clear, positive distinctiveness of the own ethnocultural group. While considered homogenous, this group is historically quite heterogeneous. The definition of what it means to be Coloured is complex. Within this group, there is no clear shared culture, religion, language, or ethnic descent. Sakamoto (2008) states that the categorization of Coloured individuals as “mixed-race” or “half-breed” makes it controversial for the community itself and develops a sense of distancing themselves from being cast as such. It would, therefore, make sense that emerging adults from this group may find it more difficult to navigate not only their own in-group-related identity processes but also how they relate to out-groups, as there is a lack of clarity as to where they are positioned in the South African cultural plethora.

Implications

This study has several implications for SIA (Postmes et al., 2001) and ITT (Stephan et al., 2000). Firstly, personal identity and associated intrapersonal aspects (that which differentiates us from others) seem important for studying intergroup relations, as many social psychologists focus primarily on social identity aspects such as cultural, ethnic, or religious identity. Also considered an important aspect of the SIT theoretical framework (Tajfel & Turner, 1986), the value of personal identity requires further scrutiny, notably as it relates to intergroup relations. In addition, social identity aspects in themselves seem to function rather separately across the different groups, requiring more integrated models to assess the relationship between identity and intergroup relations.

Secondly, it is clear from this study that in the absence of a clearly defined dominant group, all groups experience some form of threat (Stephan et al., 2000). Interestingly, groups deal with this threat very differently. Emerging adults from the different ethnocultural groups experience in-groups and out-groups very differently. The threat model holds only for White-Afrikaans emerging adults, as in-group and out-group orientations are stronger and more focused on maintaining the status quo. Threat theory seems to apply less to the other groups, in particular to the low in-group and moderate out-group orientation of the Coloured group, which do not seem to be based on perceived threats but may be based more on a historically less-developed sense of unity within the group.

Limitations and Recommendations

The convenience sampling and modest sample sizes, with an overrepresentation of females in each ethnocultural group, make it challenging to generalize our results. More equivalent numbers of participants reflecting the gender distribution and ethnocultural makeup of South Africa are needed. In addition, this study only included student samples of emerging adults, which do not represent the variety of emerging adults in the South African context. A more representative sample of all emerging adults is required before generalizations about the true manifestation of identity and group orientation in South Africa can be made. South Africa is a highly diverse context, with approximately 13 different ethnocultural groups. We have considered only four such groups in this study. A more comprehensive study is required to truly understand the intergroup dynamics at play within this country. Additional research is needed in order to further investigate the predictive value of personal and social identity aspects on group boundaries, especially with the emerging adults in non-Western multicultural contexts. In order to better understand the reasons for the results, a causal inquiry may be conducted by utilizing a longitudinal design.

Conclusion

In a society such as South Africa, where all groups experience some form of threat and deal differently with threats specific to their group, emerging adults favor their own groups and there is a general sense of “us” versus “them.” However, in-group and out-group orientation tend to be positively associated unless the group feels challenged, as with the White Afrikaans group. Emerging adults, who will soon take over the social, political, and economic sectors as the next generation workforce, leaders, and politicians, are clearly developing a sense of who they are and how they relate to others. Their identities, both personal and social, seem not only important for how they see their own groups but how they relate to others. It is important to know that, as many societies are becoming more multicultural and diverse, the challenge is to develop policies that can create an enabling environment for intergroup relationships and emphasize development toward a context which embraces difference. It is clear that out-group difference and threat are prominent themes in today’s society. In this study, our objective was to consider the importance of these issues during emerging adulthood, where it may still be possible to develop effective transformation interventions to counter experiences of threat and to make group boundaries more flexible.

Measures	Items	Black- Zulu EA	Coloured EA	Indian EA	White- Afrikaans EA
Identity
Personal identity	12	.78	.86	.83	.81
Religious identity	6	.88	.92	.95	.90
Ethnic identity (belonging)	7	.84	.89	.94	.91
Intergroup relations
BSA Orientation Scale	11	.70	.78	.80	.88
CSA Orientation Scale	11	.78	.86	.85	.90
ISA Orientation Scale	11	.79	.79	.76	.89
WSA Orientation Scale	11	.77	.79	.69	.80
General Language Usage Scale	4	.74	.81	.91	.77

Models	χ²/df	TLI	CFI	RMSEA	AIC	BCC
Personal identity
Configural invariance	1.66***	.88	.92	.04	656.99	715.42
Metric invariance	1.61***	.89	.91	.04	635.70	684.17
Scalar invariance	1.66***	.88	.89	.04	635.59	672.10
Partial scalar invariance	1.58***	.90	.90	.04	616.57	655.08
Religious identity
Configural invariance	4.48***	.87	.93	.10	305.25	317.20
Metric invariance	3.42***	.91	.93	.08	288.78	298.23
Scalar invariance	2.93***	.93	.92	.07	280.20	286.67
Ethnic identity (belonging)
Configural invariance	3.42***	.87	.92	.08	359.40	375.53
Metric invariance	2.99***	.90	.91	.07	353.07	365.73
Scalar invariance	2.79***	.91	.90	.07	355.09	363.72
Partial scalar invariance	2.78***	.91	.90	.07	351.83	361.04

	Black-Zulu EA	Coloured EA	Indian EA	White-Afrikaans EA	Ethnocultural Group
Identity
Personal identity	3.89 (0.56)	3.80 (0.67)	3.78 (0.58)	3.76 (0.59)	0.80	.01
Religious identity	3.69 (0.83)_a	3.66 (1.01)_a	3.51 (1.10)_a	3.94 (0.75)_b	3.85*	.03
Ethnic identity (belonging)	3.13 (0.55)	3.10 (0.58)	3.16 (0.63)	3.27 (0.58)	0.31	.00
Intergroup relations
In-group orientation	3.78 (0.56)_a	3.58 (0.46)_a	4.05 (0.73)_b,c	4.41 (0.54)_b,d	45.88***	.26
Out-group orientation	3.01 (0.45)_d	2.60 (0.51)_b,e	2.75 (0.57)_c,e	2.02 (0.73)_a	52.36***	.30

Models	χ²/df	AGFI	TLI	CFI	RMSEA	Δχ²	Δdf
Original model
Unconstrained	2.32**	.87	.79	.94	.06	—	—
Structural weights	3.20***	.82	.65	.79	.08	48.97***	12
Structural covariances	3.01***	.84	.68	.67	.07	49.58***	18
Structural residuals	3.94***	.79	.53	.44	.09	63.06***	6
Modified model
Unconstrained	2.32**	.87	.79	.94	.06	—	—
Structural weights	1.84*	.90	.87	.93	.05	12.63	10
Structural covariances	2.25***	.87	.80	.80	.06	49.58***	18
Structural residuals	3.23***	.81	.65	.59	.08	58.44***	6

Footnotes

Authors’ Note

The opinions expressed and the conclusions reached are those of the authors and should not be attributed to National Research Foundation.

Author Contribution

B. G. Adams contributed to conception and design, acquisition, analysis, and interpretation; drafted the manuscript; critically revised the manuscript; gave final approval; and agrees to be accountable for all aspects of work ensuring integrity and accuracy. L. Naudé contributed to acquisition and interpretation, drafted the manuscript, gave final approval, and agrees to be accountable for all aspects of work ensuring integrity and accuracy. J. A. Nel contributed to acquisition and analysis, drafted the manuscript, gave final approval, and agrees to be accountable for all aspects of work ensuring integrity and accuracy. F. J. R. van de Vijver contributed to interpretation, drafted the manuscript, critically revised the manuscript, gave final approval, and agrees to be accountable for all aspects of work ensuring integrity and accuracy. S. Laher contributed to acquisition, gave final approval, and agrees to be accountable for all aspects of work ensuring integrity and accuracy. J. Louw contributed to acquisition, gave final approval, and agrees to be accountable for all aspects of work ensuring integrity and accuracy. F. Tadi contributed to acquisition, gave final approval, and agrees to be accountable for all aspects of work ensuring integrity and accuracy.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The financial assistance of the National Research Foundation (NRF—grant number: 74653) toward this research is hereby acknowledged.

Note

References

Adams

B. G.

(2014). I think therefore I am…I think? The complexity and diversity of identity (Published PhD thesis). Tilburg University, the Netherlands, Ridderprint. Retrieved from https://pure.uvt.nl/portal/files/9286709/Adams_I_think_24_09_2014_emb_tot_24_09_2015.pdf

Adams

B. G.

Abubakar

van de Vijver

F. J. R.

De Bruin

Arasa

Fomba

… Murugami

(2016). Ethnic identity in emerging adults in Sub-Saharan Africa and the US, and its associations with psychological well-being. Journal of Community & Applied Social Psychology, 26, 236–252. doi:10.1002/casp.2247

Adams

B. G.

van de Vijver

F. J. R.

De Bruin

G. P.

(2012). Identity in South Africa: Examining self-descriptions across ethnic groups. International Journal of Intercultural Relations, 36, 377–388. doi:10.1016/j.ijintrel.2011.11.008

Adams

B. G.

van de Vijver

F. J. R.

De Bruin

G. P.

Bueno Torres

(2014). Identity in descriptions of others across ethnic groups in South Africa. Journal of Cross-Cultural Psychology, 45, 1411–1433. doi:10.1177/0022022114542466

Anderson

Kilduff

G. J.

(2009). The pursuit of status in social groups. Current Directions in Psychological Science, 18, 295–298.

Arnett

J. J.

Jensen

L. A.

(2002). A congregation of one individualized religious beliefs among emerging adults. Journal of Adolescent Research, 17, 451–467. doi:10.1177/0743558402175002

Balkin

R. S.

Schlosser

L. Z.

Levitt

D. H.

(2009). Religious identity and cultural diversity: Exploring the relationships between religious identity, sexism, homophobia, and multicultural competence. Journal of Counseling & Development, 87, 420–427.

Bollen

K. A.

(1989). Structural equations with latent variables. New York, NY: Wiley.

Bornman

(2011). Patterns of intergroup attitudes in South Africa after 1994. International Journal of Intercultural Relations, 35, 729–748. doi:10.1016/j.ijintrel.2011.06.006

10.

Cislo

A. M.

(2008). Ethnic identity and self-esteem: Contrasting Cuban and Nicaraguan young adults. Hispanic Journal of Behavioral Sciences, 30, 230–250. doi:10.1177/0739986308315297

11.

Dannhauser

P. D.

(2006). Representation of coloured identity in selected visual texts in Westbury, Johannesburg (Unpublished master’s dissertation) University of Witwatersrand, Johannesburg, South Africa.

12.

Duckitt

Callaghan

Wagner

(2005). Group identification and outgroup attitudes in four South African ethnic groups: A multidimensional approach. Personality and Social Psychology Bulletin, 31, 633–646. doi:10.1177/0146167204271576

13.

Esses

V. M.

Dovidio

J. F.

Jackson

L. M.

Armstrong

T. L.

(2001). The immigration dilemma: The role of perceived group competition, ethnic prejudice, and national identity. Journal of Social Issues, 57, 389–412. doi:10.1111/0022-4537.00220

14.

Ferguson

G. M.

Adams

B. G.

(2015). Americanization in the rainbow nation: Remote acculturation and psychological well-being of South African emerging adults. Emerging Adulthood, 1–15. doi:10.1177/2167696815599300

15.

Haslam

S. A.

(2001). Psychology in organizations: The social identity approach. London, England: Sage.

16.

Landman

(2013). The (de)construction of religious identity in oral history research in South Africa. Studia Historiae Ecclesiasticae, 39, 1–14.

17.

Milfont

T. L.

Fischer

(2010). Testing measurement invariance across groups: Applications in cross-cultural research. International Journal of Psychological Research, 3, 111–130.

18.

Mtshelwane

Nel

J. A.

Brink

(2016). Impression management within the Zulu culture: Exploring tactics in the work context. South African Journal of Industrial Psychology, 42, 13–25. doi:10.4102/sajip.v42i1.1325

19.

Nel

J. A.

Adams

B. G.

De Beer

L. T.

(2015). Assessing cultural intelligence, personality and identity amongst young white Afrikaans-speaking students: A preliminary study. South African Journal of Human Resource Management, 13, 1–12. doi:10.4102/sajhrm.v13i1.643

20.

Niens

Cairns

Finchilescu

Foster

Tredoux

(2003). Social identity theory and the authoritarian personality theory in South Africa. South African Journal of Psychology, 33, 109–117. doi:10.1177/008124630303300206

21.

Phinney

J. S.

(1992). The multigroup ethnic identity measure: A new scale for use with adolescents and young adults from diverse groups. Journal of Adolescent Research, 7, 156–176. doi:10.1177/074355489272003

22.

Phinney

J. S.

Ong

A. D.

(2007). Conceptualization and measurement of ethnic identity: Current status and future directions. Journal of Counselling Psychology, 54, 271–281. doi:10.1037/0022-0167.54.3.271

23.

Postmes

Tanis

de Wit

(2001). Communication and commitment in organizations: A social identity approach. Group Processes & Intergroup Relations, 4, 227–246. doi:10.1177/1368430201004003004

24.

Rosenthal

D. A.

Gurney

R. M.

Moore

S. M.

(1981). From trust on intimacy: A new inventory for examining Erikson’s stages of psychosocial development. Journal of Youth and Adolescence, 6, 525–537. doi:10.1007/bf02087944

25.

Safran

J. D.

(2006). The relational unconscious: The American enchanted interior and the return of the repressed. Contemporary Psychoanalysis, 42, 393–412. doi:10.1080/00107530.2006.10747112

26.

Sakamoto

(2008). Coloured identity and cultural transformation. In Gordimer

(Ed.), My son’s story (pp. 313–330). New York, NY: Bloomsburg Publishing PLC.

27.

Schwartz

S. J.

Montgomery

M. J.

Briones

(2006). The role of identity in acculturation among immigrant people: Theoretical propositions, empirical questions, and applied recommendations. Human Development, 49, 1–30. doi:10.1159/000090300

28.

Seekings

(2008). The continuing salience of race: Discrimination and diversity in South Africa. Journal of Contemporary African Studies, 26, 1–25. doi:10.1080/02589000701782612

29.

Sibusiso

Mdikana

A. A.

(2013). Religious identity and decision making among IsiXhosa-speaking youth. Journal of Psychology in Africa, 23, 623–624. doi:1080/14330237.2013.10820676

30.

Smith

P. B.

Fischer

Vignoles

V. L.

Bond

M. H.

(2013). Social psychology across cultures: Engaging with a changing world. London, England: Sage.

31.

Statistics South Africa. (2012). Census 2011: Statistical Release. Retrieved from http://www.statssa.gov.za/Publications/P03014/P030142011.pdf

32.

Stephan

W. G.

Diaz-Loving

Duran

(2000). Integrated threat theory and intercultural attitudes: Mexico and the United States. Journal of Cross-Cultural Psychology, 31, 240–249. doi:10.1177/0022022100031002006

33.

Stets

Burke

P. J.

(2000). Identity theory and social identity theory. Social Psychology Quarterly, 63, 224–237.

34.

Tabachnick

B. G.

Fidell

L. S.

(2001). Using multivariate statistics (4th ed.). Needham Heights, MA: Allyn and Bacon.

35.

Tajfel

Turner

J. C.

(1986). The social identity theory of intergroup behavior. In Worchel

Austin

W. G.

(Eds.), Psychology of intergroup relations (pp. 7–24). Chicago, IL: Nelson-Hall.

36.

Tsai

J. L.

Chentsova-Dutton

Wong

(2002). Why and how researchers should study ethnic identity, acculturation, and cultural orientation. In Nagayama Hall

G. C.

Okazaki

(Eds.), Asian American psychology: The science of lives in context (pp. 41–65). Washington, DC: American Psychological Association. doi:10.1037/10473-002

37.

Turner

J. C.

(1999). Some current issues in research on social identity and self-categorization theories. In Ellemers

Spears

Doosje

(Eds.), Social identity (pp. 6–34). Oxford, England: Blackwell.

38.

van de Vijver

F. J. R.

Leung

(1997). Methods and data analysis for cross-cultural research. Thousand Oaks, CA: Sage.

	Black-Zulu EA	Coloured EA	Indian EA	White-Afrikaans EA	Ethnocultural Group
Measures	M (SD)	M (SD)	M (SD)	M (SD)	F(3, 386)	η²
Identity
Personal identity	3.89 (0.56)	3.80 (0.67)	3.78 (0.58)	3.76 (0.59)	0.80	.01
Religious identity	3.69 (0.83)_a	3.66 (1.01)_a	3.51 (1.10)_a	3.94 (0.75)_b	3.85*	.03
Ethnic identity (belonging)	3.13 (0.55)	3.10 (0.58)	3.16 (0.63)	3.27 (0.58)	0.31	.00
Intergroup relations
In-group orientation	3.78 (0.56)_a	3.58 (0.46)_a	4.05 (0.73)_b,c	4.41 (0.54)_b,d	45.88***	.26
Out-group orientation	3.01 (0.45)_d	2.60 (0.51)_b,e	2.75 (0.57)_c,e	2.02 (0.73)_a	52.36***	.30

When There Are Only Minorities

Abstract

Keywords

Intergroup Perspectives on Identity

Group orientation and the role of threat

Emerging Adulthood in the South African Context

The Present Study

Method

Participants and Procedure

Measures

Sociodemographic questionnaire

Identity

Group orientations

In-group and out-group orientations

Statistical Analyses

Results

Preliminary Analyses

Mean Score Comparisons

Model Assessment: Identity and In-Group and Out-Group Orientations

Discussion

Identity and Group Orientation Across Ethnocultural Groups

Implications

Limitations and Recommendations

Conclusion

Footnotes

Authors’ Note

Author Contribution

Declaration of Conflicting Interests

Funding

Note

References