My Nationality or My Ethnicity? The Pride and Attitudes of Balinese Youth as Citizen of Indonesia

Prior research on national identity focused on immigrants and racial discrimination. The current study aims to analyze national identity in Indonesia, a country with various ethnicities, especially on Bali. We hypothesized that multicultural attitudes, ethnic identity, and multicultural communication predict national identity. Four hundred undergraduate students completed the questionnaires. Data was analyzed with regression analysis entry method. Our hypothesis was partially supported. The contextual concept and understanding were discussed in explaining the research findings along with its implications. Kebangsaanku atau Etnisku? Kebanggaan dan Perilaku Pemuda Bali sebagai Warga Negara


Introduction
Social identity theory (SIT) and self-categorization theory (SCT) were originally developed by Tajfel and Turner (1986) and were further refined (Turner, 1999). Brown (2000) argued that SIT and SCT provide a psychological foundation for understanding intergroup relations' dynamics. Social identity consists of ethnic, cultural, national, and religious aspects. It affects how individuals and groups define membership or boundaries (Verkuyten, 2011). SIT holds that individuals shape their own identity based on external or social criteria such as their membership, for example, in a music fan club, students' council, or political party. This identity then involves a categorization process of defining who comprise the in-group and out-group.
National identity as part of SIT was examined thoroughly through many previous studies (Arts & Halman, 2006;Polzin, 2015). These studies were conducted in a European context among different countries/nations within the union. Applying that focus to Indonesia would be an interesting approach. Indonesia as a country has diverse ethnicities (such as Java, Sunda, Sumatera, Kalimantan, and Sasak), which make up the nation. This complexity makes Indonesia This research focuses on identity as primarily Indonesian. Previous research regarding Indonesian as a national identity had been conducted (Aspinall, 2015;Basyari, 2013;Susanto, 2016). These studies talked about the shaping of national identity through building character or narratively analyzing national identity in Indonesia. Previous research did not discuss how national identity related to different variables but displayed its depth and intricacy in Indonesia. İnaç and Ünal (2013) stated that national identity is a complex concept that has several perspectives. National identity can be viewed as the transcendence of various ethnic identities. National identity can also be seen as a melting pot of diverse ethnicities and cultures within a nation. National identity is also understood as the primary characteristic of a nation.
Rusciano (2003distinguishes two perspectives on national identity, "Selbstbild" and "Fremdbild." The concept of Selbstbild is the citizen's image or perception of their own country/nation. The concept of Fremdbild is a nation's own image with respect to its international position. Through this framework, national identity is determined by each citizen's pride and the nation's orientation/position internationally. Based on Rusciano's idea, national identity in this research is defined as individuals' thoughts, feelings, and behaviors as Indonesians. National identity is closely related to ethnic identity (Fenton, 2011). National identity is heavily influenced by a person's membership of a primary ethnic group (Green, 2017). Primary ethnic group membership is how people identify with a specific ethnic group. For example, someone identifies himself/herself as Javanese or Minangnese. Ethnic identity affects national identity formation. Ethnicity (and a respective ethnic group), as part of a nation, may contribute to national identity. On the other hand, Fuller-Rowell, Ong, and Phinney (2013) stated that national identity influences ethnic identity. Perceived discrimination and weaker national identity increase ethnic identity commitment. Conflicting ideas arise from those discoveries leading to this research to examine whether ethnic identity predicts national identity.
Within the context of ethnic identity as a scientific subject, Phinney is one of the most prominent figures. Phinney (1991) argued that a lot of research based ethnic identity conceptualization on intuition, expert consultation, and cultural background characteristics. Phinney (1991) found that there are four components of ethnic identity: 1) self-identification, 2) belongingness, 3) attitudes, and 4) participation in cultural activities. For the purposes of this research, ethnic identity is defined as the individual's subjective evaluation (thoughts, feelings, and behaviors) regarding their membership of a particular ethnic group.
People who score higher on national identity are more likely to hold positive attitudes toward other people from different cultural backgrounds (Breton, 2015). However, other research also pointed out that national identity negatively correlated with collective threat and has an indirect effect on multicultural attitudes (Sumino, 2017). Hamamura (2016) argued that there is a disharmonious relationship between social identity and attitudes toward cultural diversity. "Multicultural attitudes" refers to an attitude related to the political ideology expressed through acceptance and support of a heterogeneous society (Berry & Kalin, 1995). Attitudes toward multiculturalism or cultural diversity may influence one's national identity. A positive attitude to other people from a different cultural background (which can be found often in Indonesia) is hypothetically linked to a higher sense of pride in being Indonesian. However, another perspective argued that positive multicultural attitudes are related to a lower level of national identity (Yitmen & Verkuyten, 2018). Diverging opinions encourage the authors to investigate relationship between the two.
To this extent, descriptions about identity is always associated with location and place, especially within the context of nationality and ethnicity. Ellis (2005) said that location plays an important role in shaping individual identity. Muttaqin and Ekowarni (2016) also argued that identity is inseparable from its contextual aspect. Some scholars believe that the contextual aspect is an integral part of identity formation, and there is mutual connection between the person and the place itself (Arbab, Azizi, & Zedarbast, 2018;Schachter, 2005). Therefore, this research also takes into consideration how external experience affects national identity. Research has shown that socialization about cultures and ethnicities may influence social identity (Mohanty, Keokse, & Sales, 2006;Ochs, 1993). In this research, this variable is represented by multicultural communication within families and on campus. Family and campus are chosen because of their proximity to the participants' age group (late adolescent and young adult).
As stated before, this research will focus solely on late adolescent and young adult identity. This notion is based on age group and its related developmental stage, which relates to identity crisis and identity formation (see Erikson, 1956). Prior research on social identity also showed these age groups as participants (Bakken, Sandøy, & Sandberg, 2017;Benson & Elder, 2011;Crocetti & Rubini, 2017;Crocetti, Sica, Schwartz, Serafini, & Meeus, 2013). These studies centered on external variables such as psychosocial maturity, age perception, or delinquency. Thus, the authors believe current research can offer a fresh focal point with regard to national identity.
The decision of the context where the research would be done also needs to be made. As noted before, Indonesia has rich ethnicities and cultures. This research opted to focus on one ethnic group in Indonesia at this stage, that is, Bali. Choosing Bali was based on its uniqueness and affluence in tradition and cultural background. Bali, historically, has been drawing attention as a tourism spot and also as research subject (Clifford Geertz famously conducted ethnographic research in Bali). Balinese culture has been under the spotlight for research over the years (MacRae, 2017;Nordholt, 1994). Balinese ethnic identity itself also serves as tourist attraction in itself (Picard, 2008). Hence, this research selected Bali as the primary location for research.
From major perspectives, studying national identity may contribute to a better understanding of intergroup relations and provide input for policy to manage contact between ethnicities in Indonesia. On an individual level, identity was linked to self-esteem (Luyckx et al., 2013;Phinney, 1991). It shows an all-round benefit on observing identity as Indonesians. Previous research dealt with how to construct and/or strengthen Indonesian national identity (Grzywacz, 2013;Siebers, 2019). Little has been said about people's self-identity as Indonesian, or membership within an ethnic group (Bali) and the multiculturalism context. Thus, the present authors hypothesize ethnic identity, Indonesian multicultural attitudes, and multicultural communication on campus and in the family can predict national identity (Indonesian identity) in emerging adulthood (late adolescent to young adult) in Bali (H1). We further hypothesize that there is a positive correlation between national identity and ethnic identity (H2). Individuals with a positive attitude toward multiculturalism will also be more proud of their own identity as Indonesian (H3). Lastly, we predicted that the higher rate of multiculturalism communication (within the family and on campus) may be related to higher level national identity (H4 & H5).

Methods
Participants. Participants in this study were undergraduate students from a private university on Bali, aged 17-25 years old, who identify themselves as Balinese. We employed convenience sampling. We were given permission to come into classrooms and distribute the questionnaires to select participants. Before handing out the questionnaires, we asked students their ethnic group affiliations, and only gave the questionnaires to students who identified themselves as Balinese and were willing to participate. To cross-check, the questionnaire asked for their ethnic group affiliation (Bali) and also that of their parents. A total of 409 students completed the questionnaires. 400 participants were selected (128 males, 272 females). Nine participants/questionnaires were rejected for several reasons such as a missing field (participants did not identify their ethnic group) and where both parents did not come from Bali.
Instruments. This study used five scales to measure national identity, ethnic identity, multicultural communication in the family, multicultural communication on campus, and attitudes toward multiculturalism. These scales were adapted to the Indonesian context and were administered for tryout process on 800 students from four big cities in Indonesia (Jakarta, Medan, Bali, Padang). All of the scales used 5-point Likert scales (1 = strongly disagree, 5 = strongly agree). The result of the pilot stage is displayed in Table 1.
The national identity scale was adapted from Ashmore, Deaux, and McLaughlin-Volpe (2004) to measure more comprehensive identity within a national context. This scale measured participants' identity as Indonesian citizens. In the adapted version, all 12 items were found to be reliable.
Ethnic identity was measured by the Multigroup Ethnic Identity Measure (MEIM) developed by Phinney (1992) to measure the participants' identity as members of a local ethnic group. MEIM was adapted to Indonesian and all the items were found to be reliable. Phinney also added some open questions about the participants' ethnic affiliation that were used to define participants' ethnic group when they completed the questionnaires.
The Multicultural Communication Scale was adapted from Brown and Krishnakumar (2007) to measure communication of cultural diversity at home or on campus. Both multicultural communication scales in the family and campus contexts had the same item with different settings. The first adaptation on this scale was presented by Suryani (2016) and consisted of nine items The multicultural attitudes scale used was adapted from the Multicultural Attitude Scale by Breugelmans and Van de Vijver (2004). This scale was designed to measure personal attitudes toward a multicultural society and the social norm. The scale also examines the support for multicultural society. All the adapted items were found reliable.
Data Analysis. Data in this study were analyzed using SPSS 25. This study used linear regression analysis with entry method. We also conducted simple correlation to study correlation between all variables as a prerequisite for regression analysis. This will allow us to explain the relationship between national identity and the rest of all predictors.

Results
Here is descriptive result of all five variables measured in this study.

Discussion
We hypothesized that multicultural attitudes, ethnic identity, multicultural communication on campus and in the family predict national identity. Regression analysis with entry method supported our hypothesis (F(4, 399) = 15.647, p < 0.05) with an R 2 of 0.137). However, multicultural communication in the family and on campus did not significantly predict national identity. Pearson correlation also showed a negative correlation between these variables and national identity. This result demonstrated the complexity of understanding multicultural communication and identity formation as Indonesian.
Multicultural communication in the family and on campus, in this research, did not predict national identity. It means someone's pride as an Indonesian can not be determined by communication or socialization in regard to multiculturalism received within the family or educational setting (see Table 5). The authors' initial assumption was that individuals raised within families who imparted multicultural socialization would support their positive national identity formation. Within the context of participants from Bali, their existence as people from Bali as an ethnic group among others with various ethnicities will be a matter of pride. However, the research result did not support the hypothesis.
The complexity of communication and processing information can possibly explain this finding. Critical race theory says that ideology communication that causes or maintains racism can be done quietly (Kelly, Stair, & Price, 2013). This latent communication or subtle racism can happen anywhere including within the family and at college/on campus (Gniewosz & Noack, 2015;Picca, 2015). It means participants' perceptions regarding multicultural communication are separated from the hidden message. For example, a family can positively reinforce multiculturalism while at the same time support the necessity to study abroad; or universities campaign about diversity but also stress the importance of studying a foreign language. These may not be seen as part of multicultural communication but they subtly shape national pride and, moreover, national identity.
Other past research studied the account of cultural socialization that tied in to the participants' own ethnic identity (Cohen & Yefet, 2019;Hernández, Conger, Robins, Bacher, & Widaman, 2014). These have placed this research in stark contrast to previous ones. Cultural socialization usually centers on educating individuals to their cultural background and information related to their own ethnic origin. In this research, however, multicultural communication (scale) address the occurrence of interaction related to racism or discrimination. This also may explain why the variables could not predict national identity. This phenomenon may also be closely related to the group status of participants. Social dominance orientation is the perceived position of their group within a social context. A different orientation or group status leads to different perception of the world (e.g., perceived threat) (Morrison, Fast, & Ybarra, 2008;Poch & Roberts, 2015). Participants, while completing the questionnaire, must have perceived the words "racism" or "discrimination" differently depending on their own group status. Participants from strong or dominant ethnic groups can perceive "racism" as a grevious event happening outside their ethnic group. Participants from a weaker group can perceive themselves as the victims of racial discrimination practices. Thus, authors had not addressed this factor in matching the participants ethnic group (Bali), and how they perceived the items on multicultural communication.
On the other hand, national identity is correlated with multicultural attitudes (r = 0.139) and ethnic identity (r = 0.305). This finding echoed similar studies as previously narrated in the Introduction. The authors have to point out that the correlation is rather weak (r <0.500). One of the possible reasons for these numbers is the simplification of viewing identity as purely Indonesian.The authors conceptualized national identity as a unidimensional construct. Indonesia itself is a big (and diverse) country. A complete and thorough understanding about its national identity might be needed. For example, research on national identity in Australia divided the construct into four groupings of identity: 1) dogmatic nativists; 2) literal nativists; 3) civic nationalists; and 4) moderate pluralists (Jones, 2000). Thus, in a multicultural society, Indonesia national identity might need to take into account the identities of nativist, nationalist, religious, and purist for example. Brown et al. (2013) studied how variables such as patriotism, nationalism, and religiosity contribute to attitudes toward diversity. Interestingly this study discovered that different countries had different predictors of positive/negative attitudes. For example, in Malaysia, patriotism, nationalism, and verbal aggression were related to a pro-diversity attitude. It means that to finally understand the relationship between national identity and multicultural attitudes, we have to consider other personal factors contributing to the formation of identity and attitudes.
In regard to ethnic identity, prior studies have usually been based on immigrants as participants (Andreouli & Howarth, 2012;Cebotari, 2015). In contrast, our research measured ethnic identity and national identity among Indonesian participants. As they are members of the nation, they are also members of a particular ethnic group. Linking ethnic identity and national identity in this research is a complex task. Beyond defining national identity, there are several approaches to set boundaries for participants to select their ethnicity. July 2019 ½Vol. 23 ½ No. 1 One method to measure ethnic identity is asking respondents about their identity with the majority group and the respondents' ethnic group (Epstein & Heizler (Cohen), 2015). Another approach is ethnosizer (Constant, Gataullina, & Zimmerman, 2009). Rather than merely asking respondents, ethnosizer measures information such as language, culture, interaction, migration, and ethnic identification. Its result will cluster participants into four categories: 1) integration, 2) assimilation, 3) separation, and 4) marginalization.The current study utilizes a scale that measures respondents' participation in cultural activities and their pride in their own ethnicity and also a questionnaire asking respondents about their ethnic group affiliation.
Based on these ideas, the authors feel that ethnic identity and national identity are closely related constructs, especially when the study was conducted with participants whose identity is both as a citizen and a member of an ethnic group, such as in Indonesia.
Together with the ethnosizer perspective and social dominance theory, it is not only important to categorize participants' assimilation or separation to their identity; we also need to calculate how blended the language, cultures, and interaction of one particular ethnic group are within the scope of Indonesia. Through the ethnosizer, we can calculate integration or marginalization of an ethnic group, which will put said ethnic group into a particular position or place (social dominance). These research scales measured participants' perception regarding their ethnic group and national identity. What authors see as lacking is an attempt to measure participants' perception regarding their ethnic group position, compared to other ethnic groups or within the scope of a nation.
Another interesting finding that we observe in Table 6 is that multicultural communication in the family and on campus were consistently and negatively correlated with positive multicultural attitudes. Participants who rated higher on receiving multicultural communication also reported a more negative attitude toward multiculturalism. Derlan, Umaña-Taylor, Toomey, Jahromi, and Updegraff (2016) offered their research, which investigated cultural socialization behavior by parents and their attitudes toward the process. These were measured through the parents' perspective and the context that they were communicating about their own cultural background. They suggested that in addition to measuring behavioral aspects of socialization at home, researchers also need to examine the attitudes of parents regarding the practice. Thus, our research explains how someone who received higher multiculturalism communication may actually hold negative attitudes toward the socialization process itself. In the end, it makes them have a negative attitude toward the multiculturalism ideology.
National identity, as we have introduced earlier in this paper, is a complex concept. It is tied in with other aspects of social identity (e.g., ethnic identity, racial identity). To fully understand national identity, we also need to examine the binding values that unite Indonesia. Iwamoto and Liu (2012) discovered that cultural values (Asian values) can moderate the relationship between social identity and well-being. It means as participants identify themselves as Indonesian, they also have shared values that may alter their perception and reasoning. Kim, Atkinson, and Yang (1999) outlined some of the Asian values, such as as avoidance of shame, filial piety, conformity, hierarchical relationship, and emotional self-control. This concept was built through a classical test on an Asian sample. To put it simply, there might be values shared by Indonesians. These values could moderate the interaction between national identity and other predictors.
Authors found that there was a discrepancy between Western perspectives and the Indonesian context while measuring multiculturalism, diversity, ethnicity, and identity. Current research offers a different outlook while we also need to carefully address the following question: did we measure the variables accurately? All in all, we are happy to be able to discover the uniqueness of this research as it was translated within the scope of Balinese who are also Indonesian.
On a wider spectrum, we wish that the findings can lead to a better understanding of Indonesian identity and also make a cultural impact toward productivity through research-led policy. McShane (2016) investigated how a policy can influence the economic productivity of a nation. The correlation between between ethnic identity and multicultural attitudes toward national identity create an opportunity for government to compose a culture-/ethnic-sensitive policy that is oriented toward celebrating diversity and increasing productivity at the same time.

Conclusion
Multicultural attitudes and ethnic identity can predict national identity. People who hold a positive atittude toward multiculturalism tend to be prouder of his or her identity as Indonesian, part of a multicultural nation. As part of an ethnic group in diverse Indonesia, on the one hand, the higher sense of belonging and attachment toward their ethnic group may relate to higher national pride and identity. On the other hand, multicultural communication within the family and on campus did not predict national identity.
We strongly suggest that future research should study the response process of participants filling in the scales. The response process is currently under the spotlight by psychologists, as it deals with how people interpret the scale (Mihura, Dumitrascu, Roy, & Meyer, 2018). We were relying on the content and theoretical foundation of identity, attitude, and communication scales. Investigation on how participants interpret and give meaning to the items/scales should be factored in to truly answer the research question using an appropriate construct and scale.
In the Discussion section the authors exposed all the differences of the nature of the scales. Future researchers may want to consider grounding the perspective to be more localized according to the participants' background. It could also be considered to add different layers or perspectives such as cultural values, cultural socialization behavior and attitude, and ethnic group position. An integration of culture specific to particular ethnic groups needs to be calculated within the equation.
In the end, we hope this research can lead other transdisciplinary researchers to study national identity and its unique perspective in Indonesia. We also hope the findings can stimulate a national policy, at the same time respecting ethnic differentiation.