1.1 Contextualizing the Study

In various academic disciplines, researchers have studied the transitional experiences of international students and immigrants.Numerous publications have documented approaches, findings, insights, and reflections regarding the nature, means and outcomes of acculturation.For example, theories have been developed on what directions acculturation may take, how many stages of it can be identified, and what impacts it has on human beings in terms of cognitive, behavioral and attitudinal changes.Empirical evidence has been accumulated about what leads to successful and smooth adaptation and what might cause bumpy and painful maladaptation.Nations and organizations are advised on policy making that could help people in cultural transition, foster multicultural integration, and resolve conflicts caused by cultural collisions.Reflections and debates on the academic issues and related social practices in turn led to more research.The current study is set out in this context awkward.

Among all the efforts, a research paradigm started to form about four decades ago centering around the bi-dimensional framework proposed and continuously elaborated by John Berry and his colleagues（Berry, 1999, 2003; Berry&Annis, 1974）.It is acknowledged by many as the most instrumental theoretical guide（Chirkov, 2009a; Nguyen&Benet-Martínez, 2013; Ward, 2008, 2013; Ward&Kus, 2012）and completely changed the way acculturation was perceived and understood.

Previously, acculturation was treated as unidirectional with the end goal of complete assimilation into the host culture in all life domains.But Berry and his colleagues have pointed out that other alternatives are possible as well.In their framework, heritage cultural maintenance and host cultural participation are treated as two orthogonal dimensions.So, a choice made on one dimension does not affect that on the other, which results in a two by two model structure with four options of acculturation strategies or modes.Those who choose cultural contact or participation over maintenance are categorized into the assimilation mode, which used to be the only option allowed in the uni-dimensional thinking about acculturation. The reverse of such choices（i.e., maintaining heritage culture and avoiding other cultural contacts/adaptation）put people into the separation group, evident in the ethnic enclaves, such as Chinatown, Koreatown, little India, and little Saigon, etc., found in many host nations.Those who endorse both are termed as taking an integration strategy.The rest are considered as the marginalized as they do not have a strong affiliation with either culture.

Compared to the uni-dimensional models, the b-i dimensional understanding of the acculturation process has several advantages.It offers an expanded explanation framework that could account for alternative approaches observed in immigrants and sojourners'acculturation experiences.It is theorized in a straightforward and concise manner that allows for easy adoption in empirical studies.It emphasizes the equal involvement of the acculturating group and the host group in the process by putting the interaction（i.e., orientation and actual behavior）between them at the center of the theory.

Demographic research in the United States finds that ethnic minorities（i.e., African Americans, Hispanics, and Asians）in metropolitan areas are more integrated into the residential areas than before（Frey &Farley, 1996; Logan, Stults, &Farley, 2004; Logan&Zhang, 2010）.The used-to-be bilateral relations between the host and one ethnic group in the neighborhood becomes a multilateral one with many groups.This is the socioecological context of acculturation today.In face of this context, the conventional application of the bi-dimensional framework seems insufficient and sometimes even inadequate.The current case study, in particular, is interested in developing a theoretical understanding of acculturation that better suits the culturally mingled societies today.It highlights two issues that are left out in the old paradigm and proposes that network thinking and analysis can help address such issues.

The first issue concerns the assumption that only two cultures are in contact and each is monolithic within the larger social environment.The community of interest in this research is at the East-West Center（hereafter referred to as the EWC or the Center）in Honolulu, Hawaii.It is a non-profit organization established by the U.S.Congress in 1960 to foster mutual understanding and cooperation among the governments and peoples of the Asia Pacific region and the United States through research and education（http://www.eastwestcenter.org/about-ewc/mission-and-organization）.Annually, it hosts about 300 graduate students from over 30 countries in its two residential buildings on the campus of the University of Hawaii at Mānoa.Both international students and American students from the mainland and Hawaii stay together and interact on a daily basis on various social events.Given its unique geographic location and the multicultural policy the Center endorsed, it is hard to define the host culture as the mainstream American culture.This created a problem for operationalizing the host-culture dimension in the framework.It is possible to bifurcate the two issues as one heritage culture versus all the other cultures, but field observation reveals that such an aggregation does not capture the essence of the acculturation experience in this particular community.Participants do not hold the same attitudes towards different cultural groups they interact with.In addition, such an operationalization might only confirm the integration ideology and miss the nuances between different cultural groups.

The second issue poses a more fundamental challenge to the b-i dimensional framework.It is useful and helpful that the understanding of each heritage-host bicultural relation separately and in comparison, but the sum of them together does not constitute a fair representation of the acculturation phenomenon when there are more than two cultures in play.So, only a study that considers all types of dyadic relations between different cultural groups can provide a more accurate picture of the acculturation phenomenon than studies focusing on heritage-host pair relations only.Although the bi-dimensional framework has been instrumental in guiding the latter kind of studies, its typology does not leave room to theorize about the other types of relations, which are the key differentiating acculturation in a multicultural setting from that in a bicultural context.“Interpreting the lived experiences of such people through a bi-dimensional acculturation lens threatens a considerable loss of important information”（Doucerain, Dere, &Ryder, 2013, p.686）.

To address this problem, this study turns to social network analysis（SNA）, because it is not only able to handle the two issues summarized above, but also has had an impressive record of applications across disciplines and levels of analysis. This combination of SNA with other methods in social sciences has been called on earlier and expected to lead to a paradigm shift in many fields of study（Borgatti, Mehra, Brass, &Labianca, 2009; Robins &Kashima, 2008）.Particularly for research about acculturation, the promises of SNA have also been hailed for decades（Kim, 1978; Rogers&Kincaid, 1981; Smith, 1997, 1999; Yum, 1988）.The size and ethnic composition of individual social networks have gradually generating increasing academic attention（Bhattacharya, 2011; Domínguez&Maya-Jariego, 2008; Froese, 2010; B.J.Kim, Sangalang, &Kihl, 2012; Smith, 2013）, but these were mainly implemented as ego-network studies.This study takes the step further by implementing a full-network analysis to investigate acculturation, which by itself is worth pursuing.

In the theoretical framework of SNA, a phenomenon of interest is constructed using（human or nonhuman）agents as nodes or actors and relations between them as ties or links.Its focus on relations and positions instead of the agents（Marin&Wellman, 2010）and ability to test multilevel hypotheses in one model（Monge&Contractor, 2003）make it a suitable candidate for conducting this research. Specifically, network study does not assume groups based on membership labels but define them based on the actual interactional patterns suggested by the relational data.It also allows for the visualization of individual behavior rather than the average.These advantages can help reveal the complexity in acculturation.Also, network analysis can take into consideration other intergroup relations present in the acculturation context, so those who are involved in the process（other cultural groups beyond the heritage-host pair）would no longer be absent from the research due to their incompatibility with the theoretical framework.

On the other hand, the application of the network perspective does not require a complete break up with the bi-dimensional framework.All four acculturation modes theorized before can be translated into the SNA paradigm, and the inflexibility relaxed to include more intercultural relations that could matter to those involved.Very unique questions about acculturation in a multicultural setting could be explored in this way.For example, what is the difference between people who are bicultural-oriented（culture-specific）and those who are multicultura-l oriented（culture-general）? Should interaction with host-nationals only and interaction with people from many other cultures be considered as the same or different acculturation strategies? Does the latter（in a multicultural context）still qualify as an integration strategy? Does it make a difference in terms of adaptation whether a person is bicultural or multicultural? These questions can be investigated more thoroughly using the SNA method.

The critiques here are not made to discredit the b-i dimensional framework and models as“no text（no matter how generous the word allocation）nor figure（no matter how complicated）can represent every aspect of the realities of the acculturation process”（Berry, 1997a, p.62）.These critiques are presented with an aim to reflect on the challenges that a unique acculturation case has posed to the conventional analytical framework and to show how an interdisciplinary approach could continue the dialogue in new directions and with new questions.

The following sections move from general discussion of the problems in acculturation research to the description the specific context of the current study. They then lay out the three research questions investigated.Finally, the contributions of this study to theory building and methodology innovation as well as implications for practices are summarized.

First of all, at the East-West Center, people from many different cultures live together and are inevitably involved in each other's acculturation experiences.Such a situation is exactly where the world is heading, which makes it possible to deepen our understanding through studies like this one in a similar context.

Secondly, as Brislin（1981）noted, success stories are worth studying for appreciation of efforts facilitating intercultural relations and motivating continuous attempts.Simply given the Center's impressive world-wide alumni networks as a result of the many intercultural friendships formed there, it is worth documenting and investigating the students'acculturation experience in the community.Admittedly, the acculturation experience here is special given the geographical location and the institution's reputation and tradition.Yet, the goal of promoting intercultural relations is shared by many organizations and nations in the world as cultural diversity becomes the norm.

Lastly, the particular socio-cultural context at the Eas-t West Centerprovides a unique opportunity to tackle the issues facing the dominant theoretical framework for acculturation research.Fong and Shibuya's（2005）review of literature on residential patterns in multiethnic cities noted that there is a stream of studies focusing on the effects of the presence of racial and ethnic groups in the city or neighborhoods on the residential segregation of these minority groups, but“the pairwise measure has circumscribed the theoretical development of residential segregation in a multiethnic context”（Grannis 2002, Reardon & Firebaugh, 2002）.Thus, they argued that“given that residential patterns reflect group relations, the failure to explore residential segregation among minority groups can limit the full account of intergroup dynamics, especially among minority groups, in multiethnic contexts”（Fong&Shibuya,2005, p.288）.A similar issue motivated the current study and its choice of full-network analysis.The same argument is made here:acculturation in a multicultural setting is different by kind, not by degree, from a bicultural context, and a full-network analysis is able to capture the differences.