Mapping the Cultural Identities of Youths in Hong Kong from a Social Capital Perspective

With its unique geopolitical status and multicultural setting, Hong Kong has harbored different youth groups generated from cross-border migration with mainland China who are tied to different cultural values and identifications. This study aims to investigate how social capital embedded in the family, school, and community influences the cultural identities across three groups of Chinese youths in the educational system: local students; cross-border students (born in Hong Kong, living in the neighbor city of mainland China but attending schools in Hong Kong on daily commute); and new immigrant students (born in mainland China but living in Hong Kong for less than seven years). Using data from a cross-sectional survey with 2180 fourth- to ninth-grade students in Hong Kong, the logistic regression results suggest that family and community social capital play significant roles in shaping the cultural identity of youths. Implications of the research findings are discussed.


Introduction
The anti-extradition bill protest in 2019 has brought the young generation of Hong Kong to the eyes of the world through daily headlines in the media. Given the considerably large group of participants in this social movement, the underlying reasons for these young people to take up such an active role have raised great concern and increasing attention from researchers, government, and society at large. Research has started to probe into the deep social causes that lead to such an unprecedented social crisis as well as underlying factors that motivate the youngsters to become key actors in the event. Cultural identity is among the many factors that are worth exploring. Indeed, as seen from their slogan and expressed demands, cultural identification of these youths serves as a critical factor that determines their stand, attitude, and action. From one-to-one communication to collective meaning negotiation, previous research suggests that individual differences in cultural identity affect interpersonal interactions and interpretation of social events (Benet-Martínez et al. 2002;Wan et al. 2007). Therefore, it is timely and meaningful to investigate the cultural identities of youths in Hong Kong and what factors contribute to their possession of different cultural identities.
Identity is crucial to positive development during adolescence, and there is no exception for collective identities such as ethnic and cultural identity (Erikson 1968). The majority of existing studies focus on ethnic identity, and the findings are believed to be applicable to cultural identity. In some studies, the two concepts are used interchangeably, since ethnic identity is viewed as social construction and social mobilization of blood relationship and culture. Like Bill Cross argues, at the individual level, racial, ethnic, and cultural identity overlap in one's life experience (Cross and Cross 2008). On the one hand, studies of ethnic identity in research of acculturation and adaptation have revealed the significant role of cultural identification for immigrant adolescents. In addition to the common issues during adolescence concerning biological, cognitive, psychological, and social development, young immigrants often suffer from stressors brought by their minority status in the host society (Berry et al. 1987;Phinney 1990). Fortunately, growing empirical evidence demonstrates that, for ethnic minority adolescents, the strength and orientation of one's ethnic identification with the mainstream society may eliminate or adjust the negative influence of stressors on one's psychosocial adjustment (Berry et al. 2006;Sam and Berry 1995). As for Chinese immigrant adolescents, previous research reports that strong identification with both Chinese and host society predicts better psychological adjustment (Eyou et al. 2000). On the other hand, for local people living in a multicultural society, multicultural attitude has been suggested to be positively associated with adjustment among individuals of the majority culture (Chen et al. 2008;Van Der Zee and Van Oudenhoven 2000). Results from a local study on cultural identification and psychological adjustment among adults (e.g., mainland Chinese immigrants, Hong Kong local people, Filipin sojourners) suggest that bicultural identification matters for local Hong Kong residents as well (Chen et al. 2008). Regardless of minority status, there is much reason to believe that cultural identity is crucial for the adaptation and development of youths. Despite the numerous studies focusing on cultural identities of immigrants and ethnic minorities in Western cultural contexts, a limited amount of research has explored the identity issues of children and youths growing up in a multicultural and multilingual society in the Eastern context like in Hong Kong.
According to Bronfenbrenner's (1994) ecological theory, the developmental processes and outcomes of individuals are largely influenced by the immediate and larger social contexts in the ecological system. Previous research has tried to examine the effects of social contexts on youth development using the umbrella concept of social capital, which captures the operation of a range of contextual factors on various developmental outcomes (Coleman 1990;Dorsey and Forehand 2003;Drukker et al. 2003;Zhou and Bankston 1994). Recent studies have also constructed an integrative framework of social capital to investigate simultaneously how various dimensions of social capital, namely family, school, and community social capital, function on the education, mental health, and psychosocial adjustment of children and youths (Wu 2017;Wu et al. 2010Wu et al. , 2011Wu et al. , 2015. We argue in the present study that, similar to its effect on other developmental outcomes, various dimensions of social capital embedded in a range of social contexts may be influential for the formation of cultural identities among youths. Individuals' cultural knowledge evolves through social interactions (Mead 1934), which are embedded at different levels of social networks. Resources inherent in those social relationships and social networks have been demonstrated to shape the cultural identification of youth towards a certain cultural system or society. However, to date, few studies, if any, have attempted to examine the impact of social capital on the cultural identities of youths in Hong Kong, a multicultural society that facilitates the formation of cultural identity in multiple directions.
Hong Kong is an ideal place for exploration and examination of the link between social capital and cultural identities. The unique geographic and political location has fostered a multilingual and multicultural society, which creates the base for shaping various cultural identities. As a so called "sibling" culture, it shares with the mainland Chinese culture some similarities in language, customs, and Confucian values, yet exhibits differences in living habits, social norms, lifestyles, and the like. These differences have become even more salient with the intensifying sense of Hong Kong identity among local people (Ma 2009). The unique geopolitical context also leads to diversities of residential status among the younger Chinese generation. In addition to the local young people who are viewed as within the cultural majority, there are other subgroups of youths who carry various residential and cultural backgrounds, such as new immigrant youths and cross-border youths. Since the 1997 handover and with increasing cross-border marriages, there has been a soaring number of immigrants and their offspring entering Hong Kong. According to the official definition, immigrants born in the mainland and have stayed in Hong Kong for less than seven years are called new immigrants, of which people who are under 15 outnumber 25,000 as reported by official statistics (Hong Kong Census and Statistics Department 2018). Recognized as cross-border students, the number of school-age children residing in nearby cities of mainland China (i.e., Shenzhen) yet attending school daily in Hong Kong has increased rapidly (Yuen 2011). These three groups of youths, namely local, new immigrant, and cross-border, are growing up in various social systems (families, schools, communities) and are thus likely to formulate different cultural identities, but little is known about the impact of resources inherent in social relationships embedded in a range of social contexts on their cultural identification.
Taken together, the relationship between social capital and cultural identity is far less than clear and requires empirical examination. Moreover, given the unique geopolitical position of Hong Kong and the roles and actions of youths in recent social events, there is an urgent need to understand the association between social capital and cultural identity in an effort to better understand the underlying motives that have shaped the youth behaviors. Drawing upon data from a recent survey covering students from various residential backgrounds, the present study aimed to investigate how various dimensions of social capital associate with the cultural identity of cross-border, new immigrant, and local youths in Hong Kong. To our knowledge, this is the first study of its kind in this multicultural context.

Cultural Identities of Youths in Hong Kong
The unique geopolitical position of Hong Kong has led to complexity in its population composition, which also manifests in the student population. In addition to local children and youths born and growing up in Hong Kong, there are at least two other groups in the student body: cross-border students and new immigrant/newly arrived students. Starting in the mid-1980s as a result of the "Open Door" policy of China, many male Hong Kong citizens started traveling to mainland China for business opportunities, which naturally resulted in higher frequency of romantic liaisons (Lang and Smart 2002), cross-border marriages (Lin and Ma 2008), and children born into these unions (Leung 2012). The "One-Way Permit" (OWP) scheme under Article 22 of the Basic Law of HKSAR allows mainland Chinese residents to come to Hong Kong for resettlement with a daily quota of 150, mainly for the purpose of family reunion. New immigrants coming from mainland China through this OWP scheme have largely contributed to the population growth of Hong Kong and constitute a considerable proportion of Hong Kong's entire population. Children coming from these families who have been living in Hong Kong for less than seven years are thus considered new immigrant students or newly arrived students at school. According to official statistics, there were a total of 26,368 new immigrant students in Hong Kong by the end of 2016 (Hong Kong Census and Statistics Department 2018). On the other hand, the more recent "Individual Visit Scheme" (IVS) implemented since 2003 also created another unique child population, children born to non-resident parents in Hong Kong who possess legal residency, thus receiving education in Hong Kong. They mainly live in Shenzhen, the border city between Hong Kong and the mainland, and cross the border through daily commute to attend school in Hong Kong. Although the policy that women through the IVS scheme can give birth in Hong Kong and children born through this channel can automatically obtain permanent residency was called to an end in 2013, the number of cross-border children born in earlier years has grown to a large extent and continues to constitute a considerable proportion of the student population. By the end of 2017, the total number of cross-border students in Hong Kong reached 28,280 (Hong Kong Planning Department 2019).
The complex composition of the youth population thus creates diversity and challenges in the formation of cultural identity. Based on Erikson's (1968) theory of ego identity development, obtaining a secure identity is the central developmental task during adolescence. Previous studies have suggested that failure of completing such a task could have negative psychological consequences for adolescents, including negative self-concept as well as poor physical health and psychological well-being. Particularly for the immigrant population, it is commonly believed that the formation of ethnic identity of immigrants may involve two parallel dimensions-ethnic identity (i.e., maintaining values and practices of heritage culture and retaining a sense of belonging) and national identity (i.e., accepting the values and practices and forming a sense of belonging in the culture of the host society) (Phinney et al. 2001). This process is further complicated in Hong Kong by the fact that it is politically and administratively part of China while culturally maintaining a relatively independent system. Ethnically being Chinese, youths with different residential backgrounds are exposed to different or dual cultural systems in their daily lives, thus facilitating the formation of their cultural identities in different directions. The aforementioned parallel dimensions of ethnic and national identity thus refer to identification with the mainland Chinese culture and the local Hong Kong culture, respectively, in this particular context. New immigrant adolescents, originating from mainland China and thus holding mainland Chinese culture as the heritage culture, have to experience the adaptation to Hong Kong society and its core values as well as the formation of a new local cultural identity. Local adolescents, despite being rooted in the local Hong Kong environment, grow up in a multicultural society and are exposed to numerous inter-cultural contacts, which largely influence their identity development. Cross-border students live within and move between two cultural systems on a daily basis and are thus faced with unique conditions and challenges for identity formation. Waldinger (2008) contends that immigration is a transitional process where many immigrants move back and forth between "home" and "host" society. Such a process is clearly manifested in the life of cross-border youths-a frequently recurring phenomenon of travelling between "home" and "host" societies. As a consequence, they may experience more ongoing intersections and confrontations between two cultural systems than the other two groups of youths.
The long and complicated history of civic education in Hong Kong also plays a role in the formation of cultural identity among youths. In the colonial period, the citizenship education curriculum was designed for specific political goals, to denationalize local people and strengthen their local identity, which also set the tune for the plausible moralization and depoliticization in civic education . After the 1997 handover, the Hong Kong SAR government put effort into promoting national education with the introduction of new citizenship education curriculums while addressing global education at the same time Tsui and Tollefson 2007). Nevertheless, the nuanced value conflicts between national education and global education were never clearly addressed, which resulted in confusing and unbalanced citizenship education in current practice (Hung 2013). A local qualitative study reveals that most frontline Hong Kong civic education teachers in secondary schools generally perceive civic education as moral values education in the private sphere and should not be taught as a formal independent subject (Leung and Ng 2014). The ongoing debates on citizenship education cloud the process of forming cultural identity among school-age Chinese youths, which renders the issue even more complicated in this multicultural context with complex historical backgrounds.
Despite all the unique characteristics of the local setting, previous studies suggest that being exposed to two or more sets of cultural meaning systems, individuals may vary greatly in experiencing and arranging their cultural orientations and loyalties (Hong et al. 2000;LaFromboise et al. 1993;Nguyen and Benet-Martínez 2007). Research on the association between identity formation and exposure to bicultural or multicultural environments has yielded inconsistent findings. In general, acculturation scholars tend to believe that integration into the receiving society is often associated with bicultural identity (Benet-Martínez et al. 2002;Berry 2005;Berry et al. 2006). For instance, LaFromboise et al. (1993) contend that individuals can adapt to two cultures without losing their cultural identity or "having to choose one culture over the other". Kim (2008) suggests that an individual who frequently experiences acculturation may achieve an intercultural identity, described as "an open-ended, adaptive, and transformative self-other orientation" (p. 364). In contrast, Vivero and Jenkins (1999) argue that some multicultural individuals may feel culturally disoriented and experience cultural homelessness because of cross-cultural tensions within ethnically mixed families and culturally different ecological contexts (e.g., cultural differences between family and living environment). Although numerous studies have addressed the link between cultural identity and multicultural contexts, their interplay has yet to be systematically studied and documented, regardless of cultural majority or minority.
The three groups of Hong Kong youths are exposed to two cultural systems differently because of the variations in their multiple social contexts. It is reasonable to hypothesize that their development of cultural identity will vary and may be affected by resources inherent in the social relationships within these contexts. However, virtually no previous studies have been conducted to examine the potential association between social capital embedded in different social contexts and the cultural identities of youths in Hong Kong.

Social Capital and Cultural Identities
Social capital refers to resources that reside in one's interpersonal relationships, which are embedded in different social systems and could be used for achieving specific goals or actions (Kawachi and Berkman 2000). In the literature, with different research focuses, scholars have defined social capital as social resources or a part of social structure that promotes individual or collective benefits (e.g., social trust, social norms, social connections, social networks) (Bourdieu 1973;Coleman 1990;Putnam 1993). In the present study, we follow Coleman's tradition and define social capital as resources inherent in social relationships and social networks that facilitate certain social outcomes (Coleman 1990;Midgley and Livermore 1998). For children and youths, these social relationships and networks are mostly established in their immediate social contexts such as families, schools, and communities. In the family sphere, social capital refers to the bonds between parents and children, mainly reflected by the time spent and efforts made by parents to interact with children and monitor their activities (Coleman 1990). Numerous studies have demonstrated that family social capital is associated with fewer depressive symptoms, higher life satisfaction, and better social adjustment of youths (Dorsey and Forehand 2003;Dufur et al. 2008;Morgan et al. 2012;Wu et al. 2010). In the school context, social capital ensembles resources inherent in relationships among all stakeholders at school, such as interactions between students and teachers, between peer groups, and communications between school and family (Roffey 2010). School social capital is also suggested to be beneficial for youths' academic performance and psychological well-being (Croninger and Lee 2001;Crosnoe 2004;Wu 2017). In the neighborhood environment, community social capital refers to social connectedness, social cohesion, and trust among neighborhood adults and children; informal social control; and civic engagement, all of which provide a base of potential resources that people could draw upon for the benefit of both themselves and the neighborhood as a whole (Coleman 1988;Putnam 1993). There is much evidence in the literature that community social capital promotes better mental health and psychosocial adjustment of youths living in a neighborhood (Drukker et al. 2003;Meltzer et al. 2007;Wu et al. 2011Wu et al. , 2014. Despite the bulk of literature regarding the significant effects of social capital on various developmental outcomes of youths, few studies have included cultural identifications in the discourse (Reynolds 2007). Although both concepts of social capital and cultural identity are related to the collective dimension and have been extensively researched, few studies have paid attention to the potential linkage between social capital and cultural identities. However, it is suggested by some scholars that social connections within family, school, and community are pathways to transmitting values and expectations about perceptions and behaviors, which potentially influence identity formation during the period of adolescence (Wright and Fitzpatrick 2006). Bronfenbrenner's ecological theory (1994) helps understand such linkage. The ecological theory articulates that interactions between individuals and their environments generate constancy and change in individuals' characteristics through the life course. Previous studies have proposed and examined an ecological model of ethnic identity formation in the context of Mexican-origin adolescents and Asian adolescents (Umaña-Taylor et al. 2006). Consistent with this idea, the current study seeks to examine the interface of individual, familial, school, and neighborhood characteristics to better understand youths' cultural identification. Empirical evidence also suggests that social connections within different social contexts are more or less associated with one's cultural identity. In the family context, it has been suggested that parental cultural socialization affects adolescents' development of ethnic and cultural identification (Rivas-Drake et al. 2009;Umaña-Taylor et al. 2006;Umana-Taylor and Guimond 2010). The closeness between parents and children has an influence on children's values and beliefs in their later life. In the school context, close relationships between students and teachers may enhance students' belief in school culture (e.g., normative order, values, and ideology) (Crosnoe 2002;Crosnoe et al. 2004;Hamre and Pianta 2001). In the community context, various community characteristics are suggested to affect the acculturation attitude and process of immigrants. Neighborhood social cohesion also affects immigrants' perceptions towards the host society. To sum up, even though promising in some studies, the association between various dimensions of social capital and cultural identity is equivocal. Therefore, the present study aimed to empirically examine whether and how social capital embedded in the family, school, and community might have an impact on the cultural identity of youths living in a multicultural society with different residential and cultural backgrounds.

Data and Sampling
Data for the study came from a questionnaire survey of 2180 fourth-to ninth-grade students from 12 schools in Hong Kong. The study employed school-based multi-stage cluster sampling to recruit participants. First, three districts from New Territories and Kowloon, which have a great concentration of cross-border and/or new immigrant students, were selected. Second, within each district, a full list of primary and secondary schools was generated according to the records of the Education Bureau. Two primary schools and two secondary schools were then randomly selected from each district. Given the population composition of the selected districts, the schools are expected to include students with various residency statuses, including cross-border, new immigrant, and local students. Third, from each selected school, two average classes (i.e., neither the elite class nor the low-performance class if there was a rank among all classes) from each grade (fourth-to sixth-grade in primary schools and seventh-to ninth-grade in secondary schools) were randomly selected, and all students in the selected classes were invited to participate. This procedure resulted in a total sample of 2180 students from six primary schools and six secondary schools located in three districts of Hong Kong, including 1387 local students, 445 cross-border students, and 348 new immigrant students.
The survey was administered to students in the classroom with research staff being present to answer questions that students might have. Human subject research ethics approval was granted by the first author's institution. Both the students and their parents signed the consent form before the students completed the survey.

Measures
The outcome variable of the study, cultural identity, was measured by a single-item question asking the respondents to select one from five categories that best describes which group they identified themselves with: (1) Hong Kong people; (2) mainland Chinese; (3) both Hong Kong people and mainland Chinese; (4) Unsure; (5) Other identifications. Since participants with other cultural identifications (e.g., ethnic minorities) were not the targeted population of this study, they were excluded from the final study sample. The remaining four responses were recorded as: 1 = Local Hong Kong identity, 2 = Chinese identity, 3 = Dual identity, and 4 = Confused identity.
Family social capital was assessed by three dimensions: (1) child's relationship with mother and father, respectively, measured by a 12-item short version of the Inventory of Parent and Peer Attachment (parent subscale). The 12-item scale was composed of three relational constructs-trust, communication, and alienation. Respondents were asked to evaluate their relationship with their mother or father on each item on a five-point Likert scale ranging from never (1) to always (5); (2) parent-child interaction, measured by an eight-item scale on things done together with parents. Respondents were asked to answer a set of questions about activities they did together with parents in the past month, including shopping, exercising, watching TV, playing, discussing things that happened in school, discussing sad things, discussing people known to both, and visiting friends or relatives.
Each activity scored one point, and added up to a total score to indicate the intensity of parent-child interaction; (3) parental monitoring, measured by a self-designed scale that incorporated eight items from previous research with regard to how often parents were involved in their children's school activities or disciplined children at home (e.g., attending a parent meeting, checking if they have completed homework), as well as parents' knowledge about their children's whereabouts after school (e.g., where they are, who they are with) (Prater et al. 1997;Strayhorn 2010). Respondents were asked to rate each item on a four-point scale (1 = never to 4 = often). The sum score was used in analysis to indicate the quality of parental monitoring.
School social capital was assessed by student-teacher relationships and peer relationships.
(1) relationship with peers was measured by the 12-item peer subscale of the short version of the Inventory of Parent and Peer Attachment Scale (Armsden and Greenberg 1987;Lodder et al. 2016). In the survey, respondents evaluated their attachment to peers with a five-point scale (1 = strongly disagree, 5 = strongly agree) on items like "my friends listen to what I have to say"; (2) relationship with teacher was measured by a seven-item scale used in a previous study (Croninger and Lee 2001) study. Participants rated each item of description about their experience in their relationships with their teacher (e.g., "teachers care about me") on a five-point scale ranging from not true at all (=1) to very true (=5).
Community social capital was measured by an 18-item scale covering three dimensions: (1) social cohesion and trust among adults, measured by a five-item scale assessing the bonds and trust among adult residents in the community (Sampson et al. 1997); (2) social cohesion and trust among children. Since the study focused on adolescents, the bonds and trust among adolescents living in the community were measured by a separate five-item scale targeted at children (Drukker et al. 2003); (3) sense of belonging to the community, measured by an eight-item scale used in previous studies (Buckner 1988). Participants rated items such as "I feel like I belong to this neighborhood" on a five-point Likert scale ranging from strongly disagree to strongly agree. This 18-item scale has been applied and validated in previous studies in the Chinese context (Wu et al. 2015).

Data Analysis
First, a descriptive analysis of related variables was performed to provide a basic knowledge of sample characteristics. This was followed by a chi-square test to examine the independence of samples (McHugh 2013). Lastly, we applied binary logistic regression models to examine the four categories of cultural identification separately among all the three residential groups. In the analysis, basic individual demographic characteristics were controlled as covariates. Odds ratios (ORs) were used to estimate and interpret regression outcomes of predictors in binary logistic regression. All the above analyses were conducted using Stata version 15.1 (Stata Corp). Table 1 presented the characteristics of the final sample. The percentages of cross-border, new immigrant, and local students in the sample were 19.5%, 15.8% and 64.7%, respectively. The majority of respondents (71.1%) were born in Hong Kong, with 77.1% possessing Hong Kong permanent citizenship. As for cultural identification, 58.5% of respondents only identified themselves as Hongkonger, followed by 30.7% of respondents who identified with bicultural identification of both Chinese and Hongkonger, and 6.4% of respondents who were unsure about their cultural identification. Lastly, 4.3% of respondents only identified themselves as Chinese.

Cultural Identities of Hong Kong Chinese Youths
Further analysis on the intersection between cultural identity and residency status using Person's chi-square test suggested that cultural identity was significantly different across the three groups of students (p < 0.000). The majority of local students (75.9%) identified themselves as Hongkongers only (local Hong Kong identity), while cross-border students (51.7%) and new immigrant students (60.4%) were more likely to hold dual identification; that is, they identified themselves as both Hongkonger and Chinese. Moreover, cross-border students also tended to be more confused about their cultural identity than the other two groups, with 9.1% of them not sure about which cultural group they felt they belonged to (Table 2).  Table 3 displays the odds ratios of all the social capital variables predicting each specific cultural identity across the three groups of youths with different residential backgrounds.

Social Capital and Cultural Identity among Hong Kong Chinese Youths
The results of Models 1 to 3 suggested whether possessing a Hong Kong identification was affected partially by family social capital and community social capital among local and cross-border students. Specifically, higher scores of parent-child interaction predicted a higher probability of identifying as Hongkonger in the cross-border student group (OR = 1.08, p < 0.01). Social cohesion and trust among children in the community predicted a lower tendency of Hongkonger identification (OR = 0.93, p < 0.05), whereas social cohesion and trust among adults in the community predicted a higher likelihood of cultural identity as Hongkonger.
As for the cultural identification as Chinese, the results of Models 4 and 5 indicated that social capital did not exhibit a significant effect in the groups of local and cross-border students. Model 6 suggested that only parental monitoring was positively associated with the Chinese identification among new immigrant students (OR = 1.11, p < 0.05).
Regarding dual identification, results from Models 7 to 9 suggested that only family social capital was partially associated with dual identity among new immigrant students. In particular, parent-child interaction was positively associated with dual identity (OR = 1.07, p < 0.05), whereas parental monitoring was negatively associated with dual identity (OR = 0.93, p < 0.05).
Lastly, as demonstrated by the results of Models 10 to 12, relationship with peers was negatively associated with confused identity among new immigrant students (OR = 0.93, p < 0.05). Table 3. Odds ratios of social capital variables in logistic regression models.

Social Capital Variables
Hongkonger (

Discussion
Applying an ecological framework to investigate how social capital is embedded in a range of social contexts, namely family, school, and community, influencing the cultural identities of Hong Kong youths coming from three different residential backgrounds, the findings of this research suggested that social capital did play a significant role. As reflected by the odds ratios in a series of binary logistic regression models, various dimensions of social capital demonstrated significant effects on youths' single, dual, or confused cultural identification. The social capital effects also varied greatly across different groups of youth with different residency statuses.
For local students, community social capital was of great influence regarding their cultural identity. Community cohesion and trust among children (i.e., participants and their peer counterparts in the community) led to lower likelihood to identify themselves with a single local identity yet led to higher probability of dual identification of both Hongkonger and Chinese. In contrast, a sense of belonging to the community might strengthen their local identification, which echoed findings from previous studies on the formation of ethnic identities (Berry et al. 2006;Eyou et al. 2000). As Amin (2002) argues, for individuals living in a multicultural environment, much of the negotiation of difference occurs at the very local level, where community certainly creates such a local level context for the everyday experience of youths.
For cross-border students, of all the social capital variables, only parent-child interaction was influential for their cultural identity. A higher degree of parent-child interaction predicted higher likelihood of identification as a Hongkonger. This was probably because that one essential motivation driving their parents to arrange their children to receive education in Hong Kong was the hope that they could benefit from the culture of Hong Kong and the advantages brought by the citizenship of Hong Kong (Cheung 2013;Chiu and Choi 2019). The more interactions between cross-border students and their parents, the higher likelihood that these youths internalized such beliefs through intergenerational communication. As a result, though being exposed to different cultures frequently, cross-border students were able to overcome identity ambivalence with their parents' support and reinforcement (Chiu and Choi 2019).
For new immigrant students, both family and school social capital were associated with their cultural identities. In the family context, new immigrant students with a higher degree of parent-child interaction were more likely to identify themselves with both the local Hong Kong and Chinese identity. This echoed previous research findings that parental cultural socialization in the family sphere did influence adolescents' development of ethnic and cultural identification (Rivas-Drake et al. 2009;Umaña-Taylor et al. 2006;Umana-Taylor and Guimond 2010). However, a higher degree of parental monitoring tended to result in lower probability of Chinese identification in the present study. This was probably because such strict monitoring aroused children's tension and rejection in values and beliefs with their parents. Since their parents were originally from the mainland, these youths might project their objection against parental control and supervision on their attitudes toward Chinese culture, which results in a weaker Chinese identification. In the school context, new immigrant students with better peer relationships were less likely to encounter confusion in cultural identities, which suggested the protective function of peer relationship for new immigrant students.
It has to be noted that interpretation of the study results could be constrained by a few limitations of the study. First, given the cross-sectional nature of the data, it is hard to establish the causal inferences in the association between social capital and cultural identities. Albeit observing significant effects of social capital on students' cultural identification in certain directions, there exists the possibility that youths with particular cultural identities are more capable of developing social capital in particular social contexts. Longitudinal design in future research will help resolve the causality puzzle. Second, the measure of cultural identity in the present study may not reach sufficient accuracy and sophistication. As a broad and abstract concept, cultural identity has been defined and operationalized in various ways. Our measure of cultural identity is a simplified one and cannot capture the full spectrum of all relevant characteristics of this concept. Developing a more sophisticated measure of cultural identity in future research will deepen the understanding of this construct. Third, the current study only examined the effects of social capital without including a broader set of individual and social factors that may also affect youths' cultural identity. Future studies could go further to explore additional predictors so as to advance our understanding of the formation of cultural identities among various youth groups. Last but not least, we are aware of the risk of inflated uniformity among individuals within given group categories. The youth groups are not monolithic, and one's cultural identity is not static; thus, we will further our work with a more detailed design for such nuances in the future.
Despite the above limitations, the findings of this research have a few crucial theoretical and practical implications. From a theoretical perspective, the current study contributes to the linkage between two multifaceted collective concepts with an original exploration of the interplay between social capital and cultural identity in a multicultural society. As Durkheim asserted in The Rules of Sociological Method, "the group thinks, feels and acts entirely different from the way its members would if they were isolated" (Durkheim et al. 1938, p. 129). To fully comprehend the conditions and trends of a society, there is a need to dive into the interplay within and beyond social networks and unveil the dynamic interactive process between individuals and the sociocultural systems. Our investigation of the cultural identities among youths with different residential backgrounds in relation to a range of social capital factors undoubtedly builds on theory development along this line. This study reconciles the earlier mixed conclusions on acculturation and cultural identity with comprehensive accounting for the roles of personal social network in one's embedded contexts. Our findings point to the nuanced roles that interactions between parent and child, peers, and social cohesion perceived by the youth play in the formation of their cultural identities, which manifested differently with variations in exposure to multicultural contexts.
From a policy standpoint, this study would help provide an initial understanding of the current vicious circles of division and distrust in Hong Kong society. To promote mutual support and cooperation among people from all walks of life, it is imperative that policymakers understand the full picture of what and how cultural identity forms differently across culturally and residentially different populations, particularly young people. Hong Kong society has experienced an unprecedented blast since the social movement in 2019. Presumably, the rapidly changing societal atmosphere must have profoundly impinged on the cultural identification of the young generation. Since the data for the current study were collected before this social movement, findings of this research thus provide a baseline understanding of Hong Kong youths and their cultural identities, as well as a foundation for further examination of the formation of their cultural identities in such a shifting and transitioning social environment. Although it is hard to crystalize the key challenges of Hong Kong society in the following years, it is axiomatic that one of the most pressing social needs is to restore social cohesion and to foster a sense of belonging in the younger generation. To begin with, we call for the mobilization of social resources and the collective support of service professionals to develop appropriate programs that help youths struggling with disorientation to cultivate a clear and positive cultural identity for their healthy adaptation and development.

Conflicts of Interest:
The authors declare no conflict of interest.