National Identity as Predicted by Ethnic Identity and Social Distance with Multiculturalism as Mediator: A Study Involving Chinese Indonesian Students in Jakarta

As a large multiethnic country, Indonesia has limited studies regarding factors that may influence the national identity of Indonesians of Chinese descent. This study examined the relations between four variables—ethnic identity, multiculturalism, social distance, and Indonesian national identity—between Chinese (n = 159) and non-Chinese Indonesian university students (n = 158) in Jakarta. A multiple-group path analysis was conducted to analyze data from Chinese and non-Chinese samples. The results revealed that structural covariance invariance was the best fit, describing ethnic identity’s direct prediction of national identity and social distance’s indirect prediction of national identity, fully mediated by multiculturalism. The mean score comparisons showed that both groups exhibited higher national identity than ethnic identity. A context of living in a global urban-metropolitan city may influence identity formation. Identitas Nasional Diprediksi oleh Identitas Etnik dan Jarak Sosial dengan Multikulturalisme Sebagai Mediator: Studi pada Mahasiswa Tionghoa di Jakarta


Introduction
Many studies have been conducted on ethnic and national identity or minority and majority groups (Verkuyten, 2018;Hindriks, Verkuyten, & Coenders, 2014). The majority of these studies were done by researchers in immigrant-receiving countries such as the Netherlands, Germany, Canada, and the United States. While many earlier studies used a multicultural approach, lately, some government leaders in these countries have declared that policies supporting multiculturalism have failed because many immigrants have divided loyalties, which undermines the national identity of their new countries (Verkuyten & Martinovic, December 2019½Vol. 23 ½ No. 2 2012. Recently, more and more studies have focused on relations among ethnic minorities, between ethnic minorities and the host (national) identity, and dual identity (Verkuyten, 2018).
Indonesia is unique in that it has more than 250 ethnic groups, but none of them can claim that they are really indigenous to the country because some of them immigrated to Indonesia many centuries ago. No single culture can declare itself as dominant or the majority, even the Javanese who constitute the majority in terms of number. Most studies in immigrant-receiving countries have dismissed multiculturalism because it does not support a cohesive national identity (Verkuyten & Martinovic, 2012). Instead, immigrants are expected to adopt the traditional values of the host country (Vasta, 2007). These immigrants already have a fixed sociocultural identity of their own, which may widely differ from that of the host country. Immigrants move to a new country to avoid several social circumstances that undermine their existence; thus, their social identity is deemed problematic. Going to and living in a new cultural environment does not mean abandoning their cultural heritage. Leaving their ethnic identity behind would probably be the best choice for a comfortable existence in the host country. Indonesia is different from immigrant-receiving countries because as a young country, Indonesia has been multicultural since its origins and continues to struggle in managing its diversity in the process of nation-building. We argue that most of the studies about ethnic and identity from immigrant-receiving countries cannot be directly applied to the Indonesian context because there seems to be no contradictions between its ethnic groups' and national identity. "Indonesia," the country's name, has been formally used since 1945. Previously, it consisted of several large and small kingdoms and sultanates interacting through trade. It was colonized by the Dutch, who used the divide-and-conquer strategy to keep their power over the locals' heirlooms. Through its proclamation of independence, Indonesia was primarily formed as a state-nation and not as a nation-state (Latif, 2011). Thus, the process of achieving a national identity resulting from acculturation and assimilation among its various ethnic groups became the ultimate focus of the founding fathers and the natural process of becoming Indonesian. This is consistent with what Rooseeno (2015) described as the transformation of a mosaic of ethnicities to become a multicultural Indonesian.
Considering the diversity in Indonesia and one's attempt to become Indonesian, multiculturalism is a very important aspect for intergroup relations in the country. According to Berry and Katlin (1995) and van de Vijver, Breugelmans, and Soekar (2008), multiculturalism is an attitude related to the acceptance of and support for the culturally heterogeneous composition of a society's population. We propose that multiculturalism is a better approach to studying the relations between ethnic and national identity in Indonesia.
Purpose of the study. Social identity theory (SIT) (Tajfel & Turner, 1986;Turner, 1999) provides a psychological basis for understanding multicultural dynamics for intergroup relations (Brown, 2000). Social identity refers to an individual's sense of internalized group membership. It is a person's sense of belonging to a certain social group that is important to him or her (Tajfel & Turner, 1979;Turner et al., 1987;Turner, 1999;Haslam, Reicher, & Platow, 2011). Heath and Demireva (2014) found that in a multicultural country such as Britain, an individual may have the same level of dual identity: that of their ethnicity and that of their host country. This dual identity is reported to support integration more than separation as what has been found in other immigrant-receiving countries in Europe. Considering the possibility of dual identity at the same level, in this study we are assuming that when an individual in Indonesia accepts that the country is multicultural, the following are possible: Hypothesis 1: Ethnic identity positively predicts national identity. Hypothesis 2a: Ethnic identity positively predicts multiculturalism. Hypothesis 2b: Multiculturalism mediates the relation between ethnic identity and national identity.
In a multicultural society, the perception of social distance among groups plays an important role. Social distance is a measure of social separation between groups caused by perceived or real differences between them as defined by well-known social categories such as ethnicity, religion, and nationality. The wider the social distance, the less likely a person feels close to members of other ethnic groups. Heath and Demireva (2014) found that in Britain, social distance between South Asian immigrants and Caucasians is wider than that between black people and Caucasians. Black people feel closer to the native group, so much so that they are open to intermarriages with white people; meanwhile, this idea is not well accepted by South Asians. This social distance was found to be associated with the perceived discrimination experienced by these groups. Following Heath and Demireva's findings, we propose the following: Hypothesis 3: Social distance negatively predicts national identity. Hypothesis 4a: Social distance negatively predicts multiculturalism. Hypothesis 4b: Multiculturalism mediates the relation between social distance and national identity. December 2019 ½Vol. 23 ½ No. 2 Contact hypothesis has been suggested as an instrument to improve intergroup relations by reducing prejudice (Dovidio, Glick, & Rudman, 2005;Pettigrew, Tropp, Wagner, & Christ, 2011). Several conditions need to be fulfilled for the contact hypothesis to work: the interacting groups have equal status, both groups have shared goals, contact is supported by law or local customs, the groups experience mutual interdependence, and contacts occur in friendly and informal settings (Allport, 1954;Amir, 1969Amir, , 1976Wilder, 1986;Dovidio & Gaertner, 2010;). A meta-analysis of studies on the intergroup contact theory found that intergroup contact reduces prejudice. The effect size is higher when the contact is formulated in a structured/institutionalized program (Pettigrew & Tropp, 2006). However, while intergroup contact is important toward building a positive effect of intergroup interaction, it is inadequate when dealing with intergroup conflict (Pettigrew & Tropp, 2006;Pettigrew, 2008;Pettigrew, Tropp, Wagner, & Christ, 2011).
Furthermore, Kende, Phalet, van den Noortgate, Kara, and Fischer (2017) observed cultural variance in the relation between contact and prejudice. The negative correlation between the two was found stronger in egalitarian cultures than in more hierarchical cultures. Since Indonesia is the latter (Hofstede, 2001), the negative relation between contact and prejudice is considered weak. To reduce prejudice among ethnic groups, a positive attitude toward diversity-multiculturalism-is necessary. When an individual believes that diverse groups experience the same situation and treatment, support for multiculturalism is achieved, and subsequently, national identity is attained.
Chinese Indonesians. The present study is part of a bigger study under the Psycho-cultural Consortium of Indonesia (Konsorsium Psikokultural Indonesia, or KPI) that aims to study the psychological aspects of five ethnic groups in Indonesia: Batakese, Minangkabaunese, Chinese, Javanese, and Balinese. This study focused on Chinese Indonesians in Jakarta. The 2010 national census shows that there are 2,832,510 Chinese Indonesians in the country, ranking 18th among Indonesia's ethnic groups, with the largest number (632,372 or 22.3%) living in Jakarta. Nevertheless, among the other ethnic groups, the Chinese are in a distinct position. Even though the mass immigration of the Chinese to Indonesia took place at the end of the 19th and the beginning of the 20th century, long before the country's independence, other ethnic groups in Indonesia have not yet fully accepted them as Indonesian (Lembong, 2008, Suryadinata, 2017. Many studies showed that Chinese Indonesians have experienced various forms of discrimination, especially during President Soeharto's reign. Examples of discriminative treatment are government regulations in the late 1960s that prohibited Chinese organizations and Chinese-speaking schools, the law that enforced changing Chinese names to Indonesian ones (Setijadi, 2016;Turner & Allen, 2007;Hudayah & Winarni, 2014), unequal education and job opportunities (Da Silva, 2010), and persecution or victimization in national or local conflicts (Purdey, 2006;Turner & Allen, 2007). One of the worst incidents happened in 1998, where many Chinese were killed and lost their homes and business, and many Chinese women were raped (Turner & Allen, 2007). In Indonesia, Chinese Indonesians have two group memberships, that is, as Indonesian and as Chinese. From an SIT perspective, the discrimination experienced by Chinese Indonesians may not make them feel positive about being Indonesians. In other words, their identification as Indonesian may be weaker than those of other ethnic groups in Indonesia. In turn, this will make them lean stronger toward being Chinese rather than being Indonesian. However, the more they hold on to their ethnic identity, the more difficult it is for them to mingle with other ethnics (Hoon, 2006). We propose that Chinese Indonesians' ethnic identity will be negatively correlated with their national identity as Indonesians.
Hypothesis 5: Multiculturalism correlates positively with national identity Most studies on Chinese Indonesians focused on them as targets of discrimination (Setijadi, 2016;Turner & Allen, 2007;Hudayah & Winarni, 2014), and only a few examined them from a different angle. Harjatanaya and Hoon (2018), who compared multicultural education in three schools, shows that teachers see nationalism as a means to protect the unity of the plural Indonesia. To date, there has been no data that shows the position of Chinese Indonesians regarding nationalism. In other words, some Chinese Indonesians, like other ethnic groups, may see themselves as part of Indonesia. In this case, Chinese Indonesians' ethnic identity will have a positive correlation with their national identity.
In the context of Chinese Indonesians, one would ask whether the discrimination experienced by Chinese Indonesians is due to their interactions with other ethnic groups (non-Chinese Indonesians). According to the social contact hypothesis, even though greater contact reduces prejudice, certain relationships are more complex than just contact. For example, factors such as intergroup threat and anxiety, cross-group friendships, intimacy, and high group salience also influence relations between contact and prejudice. Hence, we propose that for Chinese Indonesians, social distance correlates negatively with multiculturalism and national identity.
Even though Chinese Indonesians are citizens of Indonesia, they continue to be perceived as a foreign group, similar to other ethnics who come from other countries, such as Indians and Arabs. However, compared with these two groups, Chinese Indonesians have endured more discrimination from non-Chinese Indonesians (the "indigenous" group). We believe that this will provide a better understanding of Chinese Indonesians and non-Chinese Indonesians (indigenous ethnicities excluding other ethnicities such as Indian and Arabian) if we compare all the variables and the relationship pattern among the variables between the two groups.

Research questions.
This study aims to answer the following questions: Q1: Is the relationship model between multiculturalism, social distance, ethnic, and national identity described in Figure 1 the same between Chinese Indonesians and non-Chinese Indonesians? Q2: Are there mean differences in the level of multiculturalism, social distance, and ethnic and national identity between Chinese Indonesians and non-Chinese Indonesians?

Methods
Participants and data collection. This research recruited 317 university students (17-28 years old, M = 19.56, SD = 1.36; male = 29%) with non-Chinese (N = 158) and Chinese (N = 159) ethnic backgrounds from the departments of Asian literature and faculties of psychology in two universities in Jakarta. The participants self-identified their ethnic backgrounds. They indicated their ethnicity in the questionnaires. They were recruited via convenience sampling, which was tightly controlled. Classes were chosen based on the researchers' predetermined schedules. Lecturers of these classes gave their permission a week before the students were recruited. The participants filled out the questionnaires at the end of their classes. The questionnaires were also distributed to students online. The announcement and the link to the questionnaires were prepared by the student councils. This method served as a guarantee that we recruited the target population sample. The informed consent form was placed at the first page of the questionnaire, and the participants indicated their consent by their signatures (paper-and-pen questionnaires) or initials (online questionnaires). Ethical clearance was provided by the Atma Jaya Catholic University ethics committee.

Instruments.
In this study, we adapted the multiculturalism scale developed by van de Vijver, Breugelmans, and Soekar (2008), which measures acceptance of and support for a society's plural nature.
Their report found that the scale was applicable across cultures (across nations). We used this scale to measure how cultural diversity is accepted and supported in Indonesia (for example, "I think the existence of diverse cultural groups in Indonesia will weaken Indonesia's unity" (reversed score)). The tool consisted of six items with a five-point Likert scale from "strongly disagree" (1) to "strongly agree" (5).
Ethnic identity was measured using the Multigroup Ethnic Identity Measure (MEIM) developed by Phinney (1992). The scale measures the strength of association between an individual and their ethnicity. It measures a person's pride, affection, and sense of belongingness to their ethnic background and their efforts to understand their cultural background. This scale has 12 items with a five-point Likert scale ranging from "strongly not correspond to me" (1) to "strongly correspond to me" (5). Examples of the items are "I am happy that I am a member of the group I belong to," and "I have a strong sense of belonging to my own ethnic group." To measure social distance, we used the Inclusion of the Others in the Self Scale developed by Aron, Aron, and Smollan (1992 (2014) studying inclusive identity among adolescent and emerging adults. The modification was made by adding eight items to the original regional identity scale (Asian/African/European (four items)) pertaining to one's national identity as an Indonesian. An example of the adapted items is "I am proud of being Indonesian" (original: "I am proud of being Asian") whereas an example of the modified items is "Being Indonesian is meaningful for me." This scale has a total of 12 items.
Instrument adaptation. The instruments were adapted by following the process described by Abubakar, Dimitrova, Adams, Jordanov, and Stefenel (2013). This covered five steps: forward translation, back translation, harmonization of items with cultural aspects, cognitive check (concept and semantic equivalence), and psychometric analysis. Forward translation was achieved by translating the English version to Bahasa Indonesia in a group. The translators were the first author with eight AJCU students at their last year (six from psychology and two from the English department). Each group member translated each item by themselves and then discussed their translations with the group. In the discussion, the translation also involved the cultural aspect of the items. Subsequently, the translated version was back-translated from Bahasa Indonesia to English by another psychologist. The discussion on the harmonization of items with cultural aspects included colleagues who are trained social psychologists. The result of the discussion included changing the words "race" to "ethnic" in the sentences. Equivalence of concepts and semantics was also achieved by two other colleagues who confirmed the same psychological meaning of the original and back-translated items. The results showed that the meaning of the two versions were 80% similar. The last step, psychometric analysis, showed adequate measurements (see Table 1).

Statistical analysis.
We tested the hypotheses by applying a multiple-group path analysis. Through this method, models of variable relationships between Chinese and non-Chinese groups were compared. The results determined whether the relationship model (configural invariance) is the same for the two groups. It also analyzed whether the weight of the predictions (structural invariance), the covariance (structural covariance), or the error of measurement (structural residual) are similar (Milfont & Fischer, 2010).

Results
The multiple-group path analysis revealed that structural covariance was the most restrictive with an acceptable fit (Table 2) The path in Figure 2 showed that for both groups, national identity was directly associated with ethnic identity (βEI-NI Chinese = 0.22; βEI-NI Non-Chinese = 0.27; p < 0.001) and multiculturalism (βM-NI Chinese = 0.13; βM-NI Non-Chinese = 0.14; p < 0.05). Social distance correlated with multiculturalism (βSC-M Chinese = −0.20; βSC-M Non-Chinese = −0.22; p < 0.001) but not with national identity. This means social distance indirectly predicts national identity, fully mediated by multiculturalism (βindirect SC-NI Chinese = 0.03; βindirect SC-NI Chinese = 0.03; p < 0.05). Also, when ethnic identity is stronger, national identity will be stronger too. The closer a person is to others from different cultural background, the higher their multiculturalism, and this will subsequently increase national identity. Further, the independent t-test analysis showed that the level of multiculturalism, social distance, and national identity of the Chinese group is significantly lower than that of the non-Chinese group ( Table 3). The descriptive statistics also revealed that national identity is higher than ethnic identity in both groups. The difference between the two variables is −4.89 (p < 0.001) in the Chinese group and −8.58 (p < 0.001) in the non-Chinese group.  Note: Bold = Chinese group, Italics = Non-Chinese group, * = p < 0.05, *** = p < 0.001

Discussion
The current study aimed to test the relation between multiculturalism, social distance, and ethnic identity and national identity and whether this model shows invariance for Chinese and non-Chinese groups. We also intended to compare the level of these four variables between these groups. The results showed that the relationship model of the variables is the same for the two groups. The prediction of national identity by ethnic identity (H1) was confirmed in both groups. However, the relation between ethnic identity and multiculturalism (H2a) and the function of multiculturalism as a mediator for the relation between ethnic identity and national identity (H2b) were not confirmed.
The relation between social distance and national identity (H3) was not proven, but its association with multiculturalism (H4a) was significant. Because of the observed correlation between multiculturalism and national identity (H5), the hypothesis that multiculturalism mediates the relation between social distance and national identity (H4b) is confirmed.
These results showed that for all groups, the structure, direction, and weight of the determinants of national identity formation are the same. No cultural variance was found between the Chinese and non-Chinese groups in the three variables that form Indonesian national identity. The positive association between ethnic and national identity permits the possibility of dual identities to coexist; in addition, the stronger the ethnic identity, the stronger the national identity. According to Heath and Demivera (2014), the willingness to adopt the host country's identity is determined when individuals perceive that they are not discriminated. It is likely that in Indonesia, these young adults feel safe and comfortable with the social environment related to their ethnic identity. This finding is supported by the existence of multiculturality and plurality in Indonesia.
The negative association between social distance and multiculturalism and the mediating function of multiculturalism between social distance and national identity suggest that for all participants, the stronger the perceived social distance among the diverse ethnic groups, the more difficult it is to achieve a strong national identity. On the other hand, following the contact hypothesis theory (Pettigrew & Tropp, 2006), the findings also showed the important role of multiculturalism as a mediator to reduce prejudice among ethnic groups, which can subsequently strengthen national identity. From this perspective, we argue that attempts to diminish multiculturalism in Indonesia will be counterproductive for its growth as a nation. Nevertheless, a random contact in itself is assumed insufficient to form national identity.
Contact should be formulated in a more structural and institutionalized program to encourage individuals to respect multiculturalism, equip them with the ability to adapt and manage diversity, and make them proud of Indonesia as a multicultural nation. The t-test analyses showed that the Chinese group has significantly lower mean scores than the non-Chinese group for multiculturalism, social distance, and national identity. The Chinese group still appears to hesitate with respect to being more open to other cultures. However, the mean ethnic identity scores are significantly lower compared to those for national identity in both groups. This indicates that all participants feel prouder as Indonesians rather than as their ethnic backgrounds. The effect size of this difference is medium for the Chinese group and large for the non-Chinese group. Considering the context of living in Jakarta, a global urban-metropolitan city, the participants may go beyond their ethnic identity and national identity and adopt a global citizen identity.
Limitation of the study. The sampling method in this study may misrepresent the Chinese population in Jakarta. For example, during recruitment, we did not ask information about participants' subethnicity and the number of ancestral generations who have lived in Indonesia. Ethnic background was also identified based on self-descriptions, which may contain personal biases on the respondent's identity. Future studies may alleviate this problem by using a list of Chinese and non-Chinese characteristics to identify the participants more objectively.

Conclusion
This study concludes that even though the measurement of all variables is different among the Chinese and non-Chinese groups, national identity formation (being Indonesian) as predicted by ethnic identity and social distance and mediated by multiculturalism, applied to both groups equally. These emerging adults are also showed higher national identity than ethnic identity.
Future study. The correlation between social distance and national identity was not significant (p = 0.56). The insufficient number of participants may have influenced this result. Hence, we suggest using a larger sample to test this relationship in future studies.

Intervention program.
The results showed that multiculturalism plays an important role as mediating variable in the relation between social distance and national identity. This finding can be used as a basis to promote multiculturalism through cultural exchange programs, multicultural curriculums, and diversity workgroups/projects. Such programs could build more positive attitudes toward diversity in Indonesia and make participants feel safe to be closer to other groups and, subsequently, make them feel proud as Indonesians.