Abstract
This study aims to shed light on how civically oriented community communication and active engagement in local migrant communities—defined in this study as geographical communities built on online and offline social networks and dynamic, multimodal interactions among migrants—affect effective adaptation of migrants. This inquiry is theoretically guided by communication infrastructure theory (CIT). This study is based on face-to-face interviews with 506 Korean Chinese migrants who live in Seoul. We conducted structural equation model to test a hypothetical model. The model testing result imply that active engagement in a local migrant community for collective problem-solving based on civically oriented community storytelling would facilitate, not mitigate, integrative adaptation among the Korean Chinese in Seoul.
Previous intercultural communication studies on migrant communities, acculturation, or intercultural adaptation have considered an individual migrant’s engagement in their migrant communities as a factor that slows down the process of effective adaptation in the host society. Although local migrant communities—communities that are defined by both shared geography and shared migrant identity—can play certain positive roles for adaptation at an early stage of migration by, for example, providing venues to escape adaptation stress at least temporarily, previous studies have suggested that individual migrants ultimately need to rearrange (mostly weaken) their connections to local migrant communities to develop significant relationships with the host society. In general, previous studies have demonstrated that there is a negative relationship between engagement in a migrant community and effective adaptation in the host society. This study reconsiders this rather well-established proposition.
This study suggests differentiating the varying types of migrant community engagement. As an example, passive and active engagement can be compared; the former refers to engaging in a local migrant community to withdraw from the host society, whereas the latter involves engaging in a local migrant community to actively change the conditions and solve the problems faced in the host society. Most of the previous intercultural communication studies on adaptation or acculturation seemed to have focused more on the passive types of engagement than the active ones. These studies explain that individual migrants, who are engaged passively in their migrant communities, use migrant media to obtain news, information, and entertainment concerning their home countries and not the places in which they currently live; most of the topics they talk with other fellow migrants are related to their countries of origin and not their current local migrant communities; even when they talk about the problems they face in their local migrant communities, they do it as victims of the problems rather than as active participants to solve them. Accordingly, past research seemingly considers these passive types of migrant community engagement as factors that weaken effective adaptation in the host society.
Passive engagement in the migrant community is not the only option for migrants. They can also participate actively in their local migrant communities while practicing rights and responsibilities as citizens in the host society. However, previous studies on intercultural adaptation and acculturation have paid limited attention to these types of engagement, causing a lack of knowledge on the relationship between active civic engagement in local migrant communities and effective adaptation in the host society. Hence, one of the core questions of this study is whether active engagement, unlike passive engagement, contributes to effective adaptation.
This study aims to shed light on how civically oriented community communication and active engagement in local migrant communities—defined in this study as geographical communities built on online and offline migrant social networks and dynamic, multimodal interactions among individual migrants and migrant organizations—affect effective adaptation of migrants. This inquiry is theoretically guided by communication infrastructure theory (CIT) (Ball-Rokeach, Kim, & Matei, 2001; Y. C. Kim & Ball-Rokeach, 2006a, 2006b). CIT claims that access to communication resources for producing and sharing community stories is a critical factor in active community engagement. More generally, CIT explains that having communication resources to talk about one’s own local issues and problems—including negative ones—with other community members is among the most important resources to mobilize collective actions for active problem-solving (therefore, active community engagement) in a local community. Previous studies that used CIT have validated such claims in many migrant communities. However, few CIT-based studies have addressed the issue of intercultural adaptation in a direct way, except for Katz’s (2014) study on the kids of Latino immigrants as negotiators between the host society and their families. The current study is among the first attempts to connect intercultural communication studies on adaptation to CIT studies on communication and urban local community. It tests whether the CIT-defined concepts of civically oriented place-based communication (i.e., community storytelling) and active community engagement (i.e., collective efficacy and civic participation) can be facilitating factors in effective adaptation.
The sample of this study comprises Korean Chinese migrants in Korea, whose population has significantly increased since the 1990s. They now account for about 1% of the population of the country (Korea Immigration Service, 2016). Most of them are third- or fourth-generation descendants of Korean ancestors who had to move to the northeastern regions of China, which are adjacent to the northern border of the Korean Peninsula, mainly because of political or economic reasons during the late 19th and early 20th centuries. These Korean Chinese migrants have moved back to South Korea, the home country of their grandparents or great-grandparents, to find jobs, earn money, and pursue their “Korean dreams” (Hong, Song, & Park, 2013; Seol & Skrentny, 2009). However, having grown up in China with different political, economic, and educational systems, they have had to suffer from cultural bias, marginalization, otherization, suspicion, and hostility from the host society (Seol & Skrentny, 2009), as well as feeling of frustration, exclusion, isolation, and resistance within their own migrant communities (Hong et al., 2013). The Korean Chinese migrants have built their own migrant community in the Southwestern districts in Seoul. Effective adaptation to the Korean society is a critical and challenging task for the Korean Chinese migrants who are connected to this local migrant community for various reasons such as residence, business, jobs, social relationships, legal or medical services, religious activities, organizational memberships, migrant media, and cultural events.
Integrative Adaptation and Migrant Community
As people moving across national boundaries, with different motives, have become more prevalent than before, the patterns and processes of migrants’ engagement in the host societies have captured the attention of scholars from various disciplines (Y. Y. Kim, 2008). With the introduction of new communication and transportation technologies, globalization, and changes in immigration policies of different countries, the traditional and unidimensional disconnection–connection model, which suggests that one must disconnect from the country of origin to be connected effectively to the host society, has been criticized owing to its inability to describe everyday dynamic experiences of migrants today. As alternatives to the unidimensional model, multidimensional ones have been proposed, with aims to better describe dynamic two-way flows of people, information, and things between the host societies and the home countries of migrants (Abe-Kim, Okazaki, & Goto, 2001; Ryder, Alden, & Paulhus, 2000).
Berry’s acculturation framework, which is among the most widely used multidimensional approaches, proposes four adaptation strategies based on the assumption that the connection to the host society and that to home place must be assessed separately (Ben-Shalom & Horenczyk, 2003; Ward & Kennedy, 1993). Migrants are faced with two fundamental questions in their experience of migration (Berry, 1974, 1984, 1992): whether they can and want to maintain their existing connections to original cultural heritage and social relationships and whether they want to engage in intercultural contacts with people in the host society. Many empirical studies have confirmed that these two questions are answered by migrants as independent questions rather than a zero-sum issue (Berry, 1997), and individual migrants would be able to build and maintain connections to two or more places simultaneously if they have both enough motivation and capacity (or resources) to do it. According to Berry (1997), there are four different strategies of cultural adaptation depending on how individual migrants answer the two questions: integration (high connections to both the host society and the home country), assimilation (high connection to the host society and low connection to the home country), separation (high connection to the home country and low connection to the host society), and marginalization (low connections to both the host society and the home country).
Among the four types of strategies, integration (or integrative adaptation 1 ) has been considered as the strategy that provides the most desirable coping results for migrants (Berry, Kim, Minde, & Mok, 1987; Farver & Lee-Shin, 2000; Y. Y. Kim, 2008; Y. C. Kim & Ball-Rokeach, 2009; S. Lee, Sobal, & Frongillo, 2003; Marin & Gamba, 1996; Ying, 1995), especially in a rapidly changing communication environment where individuals have more capacity to manage their social connectedness (Benkler, 2006; Castells, 2000; Wellman, 1990). For example, immigrants who choose integrative adaptation strategies are less likely to suffer from acculturation stress than those who choose other adaptation strategies, especially the marginalization ones (Berry et al., 1987). Y. Y. Kim (2008) explained that migrants who choose integrative adaptation strategies can have “intercultural personhood” developed through a cumulative stress–adaptation–growth dynamic process. Those with intercultural personhood can conduct an open, a self-reflective, and a dynamic congruence of their connections to multiple cultures, customs, and place attachments; they are also more likely to “establish and maintain a relatively stable, reciprocal, and functional relationship with the environment” (Y. Y. Kim, 2008, p. 363).
Previous studies have found that integrative adaptation among migrants is influenced by political and sociocultural factors, such as immigration or related policies (Bourhis et al., 2010), political atmospheres, and labor market conditions, and individual-level factors, such as individual migrants’ education, income, language fluency, length of stay, self-esteem, and perceived level of social support (Berry et al., 1987; Stone Feinstein & Ward, 1990; Ward & Kennedy, 1993; Ward & Searle, 1991). Few previous studies have systematically examined individual migrants’ civic engagement experience in their local migrant communities as a factor in integrative adaptation.
Since the works of the Chicago School scholars in the early 20th century, migrants’ connections to their own migrant communities have been considered as barriers to integrative adaptation that intensify the possibility of separation-type adaptation. However, the recent changes in social and communication environment, with globalization, increasing population diversity, and new information and communication technologies (ICTs), have demanded new perspectives toward the roles of local migrant communities. Many migrants are now able to leverage their simultaneous connections to multiple places, including the local migrant community, the host society, and the country of origin, to maximize their access to social, financial, political, or cultural resources (Y. C. Kim & Ball-Rokeach, 2009). In such environments, we need to have a more dynamic view on migrant communities. Many intercultural communication scholars have already pointed out the need to consider the multiplicity of the roles of a migrant community (Y. Y. Kim, 2001), that is, as a bridge to the host society, a buffer against acculturation stress, or a classroom where migrants can acquire basic knowledge and skills needed to survive and grow in the new social environment.
This study is particularly concerned with the role of individuals’ connections to communication and social resources available in their local migrant communities, either as a facilitator or an inhibitor of integrative adaptation to the host society. Certain communication and social resources may discourage integrative adaptation, whereas other types of these resources may encourage it. For example, if one lives in a migrant community where the migrant media talks actively of various problems and issues concerning the local migrant community (as opposed to only delivering news regarding the country of origin), one’s connection to the migrant media may increase the likelihood of active community engagement that would facilitate integrative adaptation.
Communication Infrastructure for Local Migrant Community
CIT was designed to explain the importance of community-based communication resources as factors affecting residents’ participation in problem-solving behaviors in their local communities. Originally, this theory was developed in large-scale studies on new and old immigrant communities in Los Angeles (Ball-Rokeach et al., 2001). These studies covered the “old immigrants” in the United States, such as Whites or African Americans, and “new immigrants,” such as Mexican Americans, Chinese Americans, or Korean Americans; they found that the connectedness to various communication resources for community storytelling (especially talking with neighbors, using the local media, and participating in local community organizations to produce, share, and consume stories of the local community) would help individual residents have a sense of belonging in their neighborhood (Ball-Rokeach et al., 2001), and a high-level collective efficacy and community participation (Y. C. Kim & Ball-Rokeach, 2006a).
CIT provides unique conceptual frameworks that are employed to define community-based communication resources. In the CIT framework, communication infrastructure has been defined as a community storytelling network that is set in a communication action context (Ball-Rokeach et al., 2001; Y. C. Kim & Ball-Rokeach, 2006a, 2006b). A community storytelling network is made up of community-based agents who can produce, share, and use community stories. CIT has identified three key storytelling agents, namely, the local media, community organizations, and residents. The three storytelling agents form a community storytelling network by not only telling stories of their community but also facilitating other agents to do the same. Meanwhile, communication action context comprises ecological components that either facilitate or impede the forming and functioning of a community storytelling network.
According to CIT, the variance among individual residents, even in the same local community, in terms of their community engagement is explained by the degree to which individuals are connected to the local community storytelling network. CIT has conceptualized this as integrated connectedness to community storytelling network (ICSN). Those with high ICSN are more likely to be informed of what is happening in their local community, pay attention to common issues in their neighborhood, have access to necessary information and resources to decide on what they could and must do to solve community problems. Previous studies have found that ICSN is a consistently strong positive factor in determining sense of belonging (Y. C. Kim & Ball-Rokeach, 2006a, 2006b), collective efficacy (Y. C. Kim & Ball-Rokeach, 2006a, 2006b), participation in community activities (Y. C. Kim & Ball-Rokeach, 2006a, 2006b; Y. C. Kim, Jung, & Ball-Rokeach, 2006), community-level disaster prevention actions, (Y. C. Kim & Kang, 2010), and access to health information and knowledge (Y. C. Kim, Moran, Wilkin, & Ball-Rokeach, 2011; Literat & Chen, 2014; Matsaganis, 2008; Wilkin & Ball-Rokeach, 2006).
In the context of local migrant communities, the kinds of effects ICSN have on the migrants’ adaptation process and outcomes are topics that are important to be examined. Few of the previous studies that used CIT have addressed this issue directly. Most of the studies that tested CIT have only focused on the effects of ICSN (as a communication factor) on individuals’ engagement within their local communities, including migrant communities. To fill the gap in research, this study examines whether ICSN would bind individuals to their migrant networks in the local communities in the host society, thereby facilitating separation as an adaptation strategy; or, whether it would help migrants connect to not only their migrant networks but also other people in the host society, thereby facilitating integrative adaptation.
This study proposes that ICSN would facilitate integrative adaptation more than the other types of adaptation, including separation, because of the following reasons. First, the most critical information a migrant would receive from the local storytelling network seems to be about not only how they take advantage of resources in their local migrant communities but also how to acquire a range of resources, such as jobs, welfare, education, and legal services, from the host society (E. Kim, 2013). Second, certain key community storytellers in the local migrant communities often function as bridge organizations, working between the migrant communities and the host society. They include the local government, various government-funded local agencies, and local religious organizations. In addition, the local migrant media that is owned and managed by migrants themselves often have direct or indirect relationships with various institutions, organizations, or the mainstream media of the host society. Therefore, the connection to a community-based storytelling network would provide individual migrants with stories not only about their migrant communities but also relevant to the lives of the migrants from outside the migrant communities. Third, migrant community storytellers (i.e., individuals, organizations, or the local migrant media) talk about not only what is happening in their local community but also how their community is received and perceived by the host society. ICSN would not only strengthen individual migrants’ connections to their local migrant community but also help migrants build and rebuild their relationships with those outside their migrant community.
“Internal” and “External” Collective Efficacy as Mediators Between ICSN and Integrative Adaptation
In addition to testing the relationship between ICSN and integrative adaptation, this study tests the effect of one cognitive construct, that is, collective efficacy, which may work as a mediator between communication factors and social outcomes or, more specifically for this study between ICSN and integrative adaptation. Collective efficacy is defined in this study as individuals’ perception of the community-level capacity to solve collectively their community issues and problems (Hackler, Ho, & Urquhart-Ross, 1974; Sampson, Raudenbush, & Earls, 1997). Although an individual’s self-efficacy (Bandura, 1997) measures one’s belief of having the capability to do what he or she wants to do, collective efficacy is a community-level belief that “we” can do what “we” want or must do. Sampson and his colleagues (Sampson et al., 1997; see also Sampson, Morenoff, & Earls, 1999) highlighted two dimensions in their collective efficacy measure, namely, informal social control and social cohesion. In their study, the items in the informal social control dimension asked residents of the likelihood that their neighbors could be counted on to act if, for example, children were skipping school and hanging out on a street corner or the fire station closest to home was threatened with budget cuts. The items in the social cohesion dimension asked the respondents concerning local neighbors’ helpfulness and perceived trust, local harmony, and homogeneity. Sampson et al. (1997) combined these two dimensions into a summary measure of collective efficacy.
Previous studies have demonstrated that individuals with high ICSN were more likely to have a stronger belief in collective efficacy, and this belief leads to a more active participation in community problem-solving activities (Y. C. Kim & Ball-Rokeach, 2006b). Similar to self-efficacy, collective efficacy is what one has to learn from his or her symbolic environment through communication (Bandura, 1997). In this sense, participation in storytelling is a key to building collective efficacy among migrants. The relationship between collective efficacy for community problem-solving and actual participation in community activities is also well documented by previous studies (e.g., Y. C. Kim & Ball-Rokeach, 2006b). Individual’ belief in local community’s collective problem-solving capacities (i.e., collective efficacy) would facilitate his or her own participation in problem-solving processes (i.e., community participation). Previous studies based on CIT as well as other theoretical models using efficacy (e.g., models based on social cognitive theory or theory of planned behavior) suggest an approach modeling a process of communication → perceived efficacy → social outcomes. The current study generally follows this approach.
This study asks whether the mediating role of collective efficacy can be extended to the case of integrative adaptation and whether collective efficacy can be proposed to work as a bridge between ICSN and integrative adaptation. This study assumes that collective efficacy would be related to integrative adaptation, as the former would give migrants the opportunities to not only develop relationships with other migrants but also increase the chances to meet and negotiate with people outside their migrant networks in their efforts to solve community issues and problems.
The collective efficacy concept mentioned above is basically individuals’ beliefs regarding their migrant community’s capacity to solve “internal” problems within their local community context. Therefore, we may call this “internal collective efficacy” in this study. Then, we conceptualize “external collective efficacy” as a parallel concept of internal efficacy. External collective efficacy refers to individuals’ beliefs in the migrant community’s capacity to address how the host society treats them. In other words, external collective efficacy concerns the migrant community’s capacities to influence the links to the outside world. This study proposes that internal collective efficacy, which is strengthened by ICSN, may facilitate external collective efficacy, and then external collective efficacy will affect integrative adaptation as an independent factor. With a high-level external collective efficacy, individuals are more likely to believe that they can positively change how they are related to the host society and how the host society perceives them. This belief will help individuals build stronger relationships with both the host society and their own migrant community. The role of external collective efficacy in the context of intercultural adaptation can be discussed using certain existing theories, such as anxiety/uncertainty management theory (AUM; Gudykunst, 2005) and integrated threat theory (ITT; Stephan & Stephan, 2000), both of which address the issues of intercultural conflicts. Both theories do not mention directly a concept related to external collective efficacy. However, within the framework of AUM, external collective efficacy can be considered a critical factor that influences the likelihood of anxiety or uncertainty management for effective communication of migrants with the host society. ITT focuses on threats (both realistic and symbolic ones), anxiety, and stereotypes, which are all negative perceptions, as factors in how individuals perceive and behave in the context of intercultural encounters. In other words, ITT focuses on various structural constraints on intercultural perceptions and behaviors. The concept of external collective efficacy provides ITT-based studies with a novel approach to migrant-host society interactions by focusing on the capacity of individuals to overcome such structural constraints.
Based on the discussion on ICSN and the two collective efficacy concepts, this study hypothesizes that ICSN, internal collective efficacy, and external collective efficacy will have direct effects on integrative adaptation; internal collective efficacy will be a mediator between ICSN and integrative adaptation; and external collective efficacy will be a mediator between internal collective efficacy and integrative adaptation.
Community Participation and Integrative Adaptation
Among the core issues that need to be resolved in this study is whether community participation will help integrative adaptation. As indicated earlier, many of the previous studies have suggested that participation in ethnic or migrant communities would conflict with integrative adaptation and, at the same time, intensify the likelihood of separation adaptation (Espiritu, 1992; Glazer & Moynihan, 1975). However, a few other studies have shown different results, indicating positive relationships between community participation in migrant communities and integrative adaptation to the host society (e.g., London, 1975; London & Hearn, 1977). These studies have suggested a view that migrant communities can work as places where migrants learn about how they live in their new homes (E. Kim, 2013). Focusing on Korean Chinese in Seoul, this study hypothesizes that the latter research results make more sense than the former, especially when an active type of community engagement is considered. Active participation in migrant community as responsible citizens would not only strengthen migrant networks but also build a range of connections to the host society.
Based on the discussions above, this study proposes the following hypothetical model (Figure 1):

Hypothetical model.
Model Comparison
This study focuses on integrative adaptation as the dependent variable and hypothesizes that the hypothetical model will work best for integrative adaptation compared with other types of adaptation strategies, that is, separation, assimilation, and marginalization. The model assumes that civically oriented communication (i.e., ICSN), active types of community engagement (internal collective efficacy and community participation), and active engagement that influences the host society (external collective efficacy) will strengthen not only the community networks inside the migrant community but also their connections to the host society. This study hypothesizes that the independent variables included in this study will be associated more strongly with integrative adaptation than other types of adaptation, especially separation.
Method
Data
This study is based on face-to-face interviews conducted between November 2, 2012 and January 30, 2013, in Seoul, Korea. As of 2012, there were about 447,000 Korean Chinese migrants in Korea; the number has been increased to reach more than 650,000 in 2016 (Korea Immigration Service, 2016). More than 80% of Korean Chinese migrants have lived in the Greater Seoul Area. And more than half the Korean Chinese people in the Greater Seoul Area found their new homes in South West Seoul. Our sample was selected from an unofficial list of Korean Chinese people, collected by a nongovernmental organization serving the Korean Chinese community, from many different sources including government sources, mailing lists of Korean Chinese organizations, subscriber lists of Korean Chinese media, and client mailing lists of migrant service organizations. Among 570 interviewed who lived in all of the 25 districts in Seoul, 506 responses were analyzed after excluding incomplete and unreliable ones. The respondents were selected using a quota sampling method considering the percentage of Korean Chinese in each of the 25 districts in Seoul. Only those who had stayed at least 1 year in Korea were selected. We collected our sample not only from the Korean Chinese concentrated area in South East Seoul, but from all 25 districts of Seoul since most Korean Chinese migrants in Seoul considered the concentrated area as their local migrant community no matter where they lived.
Interviews were administered by 31 Korean Chinese bilingual (Korean and Chinese) interviewers, who were recruited and trained for this study to avoid any possible bias that the respondents may have if Korean interviewers were used. Each interview took about 55 minutes to complete The questionnaire was developed in both Korean and Chinese. The interviews were conducted in the language (either Korean or Chinese) that the respondents preferred. The interviews were conducted in various places such as respondents’ homes, workplaces, coffee shops, and local parks, depending on respondents’ preferences. Of the 506 cases, 292 were females (57.7%) and 214 were males (42.3%). The respondents’ ages ranged between 19 and 79, with the average age at 43.1 (SD = 13.0). Education was measured as the highest year of school completed and then recoded into four categories: “less than high school (18.2%),” “high school graduate (50.9%),” “college student or college graduate (24.8%),” and “graduate degree (6.1%).” The length of stay in Korea was recorded into four categories: 1 = less than 5 years (34.8%), 2 = 5-10 years (35.4%), 3 = 10-15 years (17.1%), and 4 = over 15 years (12.8%). There were no official census data of Korean Chinese migrants, to which we can refer to see how representative our sample was. When compared with several government-sponsored surveys, our sample showed similar results (e.g., C. W. Lee & Choi, 2014) especially regarding most demographic characteristics such as gender, age, and length of residence; however, the average education level of our sample was a little higher than those of the other recent surveys.
Measures
Integrative adaptation
The items from the East Asian Acculturation Measure (EAAM; Barry, 2002) were used to measure integrative adaptation. The 29 items in the EAAM covered the following four dimensions of acculturation outlined by Berry (1984): assimilation (eight items), separation (seven items), marginalization (nine items), and integration (five items). The 5-point Likert-type scale was used to measure all items, ranging from “1” (strongly disagree) to “5” (strongly agree). The items to measure integrative adaptation included “I tell jokes in both Korean and Chinese” (M = 3.47, SD = 1.01), “I think in Korean as much as I do in Chinese” (M = 3.44, SD = 1.02), “I feel that both Koreans and Korean Chinese people value me” (M = 3.28, SD = 0.882), “I feel very comfortable around both Koreans and Korean Chinese people” (M = 3.44, SD = 0.968), and “I have both Korean and Korean Chinese friends” (M = 3.38, SD = 0.988). The Cronbach’s alpha test for index scalability was 0.85. The items to measure separation include “most of the music I listen to is Chinese” (M = 2.55, SD = 1.10), “My closest friends are Korean Chinese” (M = 3.93, SD = 1.056), “I prefer going to social gatherings where most of the people are Korean Chinese” (M = 3.21, SD = 1.192), “I feel that Korean Chinese treat me as an equal more so than Koreans do” (M = 3.54, SD = 1.096), “I feel more relaxed when I am with Korean Chinese people than when I am with Korean ones” (M = 3.65, SD = 1.021), “I would prefer going out on a date with a Korean Chinese person to a Korean one” (M = 3.50, SD = 1.256). The Cronbach’s alpha test for index scalability was 0.75.
Integrated connectedness to a storytelling network (ICSN)
Following the procedure explained in the previous studies (Y. C. Kim & Ball-Rokeach, 2006b), ICSN was calculated based on three neighborhood storytelling measures: intensity of interpersonal neighborhood storytelling, scope of connection to community organizations, and local media connection. The intensity of interpersonal neighborhood storytelling was assessed as responses to the question “how often have you had discussions with other people about things happening in your neighborhood” (range = 1-5, M = 2.63, SD = 1.01). For the measure of connections to community organizations, respondents were asked if they took part in any of the eight different types of migrant community organizations during the last 1 year: (a) recreational (43.5%), (b) neighborhood (55.4%), (c) religious (15.2%), (d) sports and cultural (20.9%), (e) political (3.6%), (f) educational (20.9%), (g) volunteerism (19.8%), and (h) others (5.1%). Having a membership or participating in each type of migrant community organizations was coded as “1.” By obtaining the sum of these eight scores, a variable was formed to represent the scope of individuals’ connections to community organizations (range = 0-8, M = 1.84, SD = 1.45). To measure the respondents’ connectedness to local media, they were asked whether they spent any time using a range of different types of local media in the prior week. Then, the sum of the number of connections was obtained to create a new variable reflecting the breadth of connectedness to different types of local media. Local media was defined as either the community media targeted to a particular group, Korean Chinese, or the public media that is oriented toward the area (i.e., Korean Chinese migrant community area). This study also employed what previous studies have developed as types and measures of local media exposure and attention (Ball-Rokeach et al., 2001; McLeod, Scheufele, & Moy, 1999) (range = 0-9, M = 4.39, SD = 2.97).
ICSN is computed using the following equation (Y. C. Kim & Ball-Rokeach, 2006a). LC refers to local media connectedness, INS is intensity of interpersonal neighborhood storytelling, and OC is scope of connection to community organizations (range = 3-15, M = 6.61, SD = 2.62).
Internal collective efficacy
Following the research of Sampson et al. (1997), this study measured two variables of “internal” collective efficacy: informal social control and social cohesion. Informal social control was measured with five items, such as “If there is a safety issue that makes people worry about walking at night in your neighborhood, how many of your neighbors participate in activities to solve this problem?” A 5-point scale was used to measure these items, ranging from “1” (no one will participate) to “5” (everyone will participate). Social cohesion was measured by asking the respondents how strongly they would agree with five statements, including “People in my neighborhood are willing to help one another,” and “People in my neighborhood share the same value.” The 5-point Likert-type scale was used to measure these items, ranging from “1” (strongly disagree) to “5” (strongly agree). Then, the sum of the values of these two variables was obtained to construct a value of internal collective efficacy (range = 1-5, M = 3.29, SD = 0.73). The Cronbach’s alpha was .86.
External collective efficacy
Five items of external collective efficacy were constructed based on the results of the five focus group interviews that had been conducted before the main study. External collective efficacy was assessed by asking the respondents to indicate, on a scale of 1 to 5 where 1 represents “strongly disagree” and 5, “strongly agree,” if they agree with each of the following five statements: “The Korean Chinese people have become important members of the Korean society” (M = 3.90, SD = 1.03); “The Korean Chinese community can have a significant impact on Korea” (M = 3.67, SD = 1.14); “The perception of the Korean government and people of Korean Chinese will improve” (M = 4.07, SD = 1.0); “People in Korea are interested in Korean Chinese migrants in Korea” (M = 2.85, SD = 1.09); and “The Korean society care about what the Korean Chinese people want” (M = 2.67, SD = 1.167). Regarding the reliability score, the Cronbach’s alpha of the five items was .66. We deleted the last two items to have a three-item measure of external collective efficacy. The Cronbach’s alpha of the three items was .78. As an alternative measure to Cronbach’s alpha, the composite reliability (CR) was .78. In addition, a confirmatory factor analysis (CFA), with the three items of external collective efficacy measure and 10 items of internal collective efficacy, was conducted to test its factor structure (Byrne, 1998). The results of the analysis showed that all three items converge into one factor that is distinct from the internal collective efficacy items (χ2 = 237.04, p < .001; nonnormed fit index [NNFI] = .92; comparative fit index [CFI], = .91; root mean square error of approximation [RMSEA] = .042). The average variance extracted score, as the measure of discriminant validity, was .53, which is higher than the conventional threshold at .50 (Fornell & Larcker, 1981). Overall, the three-item measure of external collective efficacy developed in this study is reliable and valid enough to be included in the analyses.
Community participation
Community participation was measured by counting the number of activities in which the respondents had participated among nine different community activities adapted from the previous studies (Y. C. Kim & Ball-Rokeach, 2006b): “attending Korean Chinese community events,” “attending public hearings or legislative meetings concerning issues relevant to migrants,” “participating in volunteer works,” “taking part in any political demonstration or protest,” “circulating a petition,” “writing a letter or an email to the editor of the press,” “voting,” “donating,” and “discussing community issues with others.” All of these items were binary measures where respondents were supposed to answer either “yes” (value = 1) or “no” (value = 0). By obtaining the sum of the nine dummy scores, a scope variable of community participation was created and then transformed into a log value.
Control variables
Gender, age, education, income, and years in Korea were used as covariates for the analyses in this study.
Results
Prior to the main analyses, a zero-order correlation analysis, with all the variables, was conducted (see Table 1). All pairs were positively correlated with one another, although they were below the recommended threshold of .70 (Tabachnik & Fidell, 2001). The results of the analysis indicate that the variables in this study do not exhibit significant multicollinearity problems.
Zero-Order Correlations, Means, and Standard Deviations (N = 506).
Note. ICSN = integrated connectedness to community storytelling network; ICE = internal collective efficacy; ECE = external collective efficacy.
Scope of connection to community organizations.
p < .05. **p < .01.
Based on the discussions presented in the literature review section, a hypothetical model was developed (see Figure 1) and tested using LISREL 8.8. Employing the maximum likelihood estimation, the following criteria were used to evaluate the overall fit of the model (Bentler, 1988): (a) the chi-square statistic (= nonsignificant); (b) the NNFI (close to 1.00), (c) the CFI (greater than .90); and (d) the RMSEA (less than .05).
As presented in Table 2, the chi-square test of hypothesized integrative adaptation model was significant (χ2 = 8.03, p < .05). However, the other indices indicate that the models hold an acceptable fit. NNFI was .93; CFI, .99; and RMSEA, .049. These summary statistics indicate that the proposed model represents a good fit with this study’s data. The result of the hypothetical model testing was compared with the other competing models, including the saturated model. This study found that all of these models were inferior to the hypothetical model, indicating that the hypothetical model is both substantially meaningful and statistically well-fitting (Byrne, 1998).
Fit Indices for Hypothesized Models.
Note. NNFI = nonnormed fit index; CFI = comparative fit index; RMSEA = the root mean square error of approximation.
More specifically, the first set of hypotheses (H1, H2, H3) was regarding the effects of ICSN on internal collective efficacy, community participation, and integrative adaptation. As illustrated in Figure 2, ICSN had significant associations with internal collective efficacy (β = .40, p < .001), community participation (β = .56, p < .001), and integrative adaptation (β = .16, p < .01). Thus, H1, H2, and H3 were supported as expected.

Structural equation modeling results for integrative adaptation.
The next set of hypotheses (H4, H5, H6) proposed that internal collective efficacy would be positively associated with external collective efficacy (β = .30, p < .001), community participation (β = .20, p < .01), and integrative adaptation (β = .32, p < .001). H4, H5, H6 were also supported.
Hypothesis 7 discussed the association between community participation and integrative adaptation. Korean Chinese community participation positively associated with integrative adaptation (β = .14, p < .01). Thus, H7 was supported.
Hypothesis 8 predicted that external collective efficacy would have a direct positive association with integrative adaptation. As presented in Figure 2, external collective efficacy had significant associations with integrative adaptation (β = .23, p < .001), confirming H8.
Hypotheses 9-1 and 9-2 were about indirect effects of ICSN on integrative adaptation through internal collective efficacy (H9-1) and community participation (H 9-2). Hypotheses 10-1 and 10-2 predicted indirect effects of internal collective efficacy on integrative adaptation through external collective efficacy (H10-1) and community participation (H10-2). This study’s model testing results show that all of the mediating effects were significant. To confirm this, Sobel tests were performed for these interaction effects. The results of the Sobel tests are reported in Table 3.
Sobel Test Results for Mediating Effects.
Note. ICSN = integrated connectedness to community storytelling network; ICE = internal collective efficacy; ECE = external collective efficacy.
Last, it was tested whether the hypothetical model with integrative adaptation would work better than the alternative models with other types of adaptation strategy, especially the separation strategy (H11). Structural equation model tests were conducted after replacing integrative adaptation in the hypothetical model (Figure 1) with separation (χ2 = 9.48, p < .001, NNFI = .97, CFI=.95, RMSEA = .078). In this alternative model with separation as the dependent variable, the relationships among the variables were not consistent with the hypotheses of this study (see Figure 3). For example, ICSN was negatively related to separation (β = −.13, p < .01) while other engagement variables (internal collective efficacy, external collective efficacy, and community participation) were not significantly related to separation. This result shows that the model with integrative adaptation, but not the model with separation, supports the hypothetical relationships among ICSN, two collective efficacy variables, community engagement, and adaptation. Therefore, H11 is confirmed.

Structural equation modeling results for separation.
Discussion
The purpose of this study was to test the hypothetical model that contains links among ICSN, internal collective efficacy, external collective efficacy, community participation, and integrative adaptation in the Korean Chinese community in Seoul. This study found that the hypothetical model fits well to the data. ICSN is significantly and positively related to internal collective efficacy, community participation, and integrative adaptation. Internal collective efficacy is positively associated with community participation, external collective efficacy, and integrative adaptation. Community participation and external collective efficacy have positive association with integrative adaptation.
Several important points demand more discussions. First, the most important message from this study is that migrants’ engagement in their local migrant community (measured by ICSN, internal collective efficacy, and community participation) would facilitate, and not mitigate, migrants’ level of integrative adaptation in the host society. As a result consistent to CIT that explains the impact of community-based communication resources (i.e., ICSN) on community participation, the current study first found a significant link between ICSN (and internal collective efficacy) and community participation among Korean Chinese migrants in Seoul. However, this study went beyond existing CIT-based studies to investigate further how migrant community factors affect migrants’ ways of adapting to the host society. This study found a significant positive link between active engagement in migrant communities and integrative adaptation.
Second, the results generally confirmed what the previous CIT studies have found, especially in terms of how ICSN is related to (internal) collective efficacy and community participation. This study found that ICSN is positively associated with the level of internal collective efficacy and community participation. These results are consistent with the previous research, demonstrating positive effects of ICSN on community engagement (Y. C. Kim & Kang, 2010; Y. C. Kim et al., 2011). As in Y. C. Kim and Ball-Rokeach’s (2006b) study, it is found that the concept of internal collective efficacy used in this study worked as a mediator between ICSN and community participation. Therefore, this study confirms the general model of community engagement based on CIT. Even among the Korean Chinese migrants, their connection to community-based resources (i.e., ICSN) to storytell community issues and problems increases their perceived (internal) collective efficacy to solve community issues with others and the likelihood of participating in a range of different community activities in their migrant communities.
Third, among the unique contributions of this study is the concept of external collective efficacy. This study conceptualized and proposed this new concept and found that external collective efficacy must be among the critical factors in integrative adaptation. Internal collective efficacy is described as individuals’ perceived beliefs in the willingness of other neighbors to intervene in community problem-solving processes, whereas external collective efficacy refers to the beliefs in the community’s capacity to change positively the ways in which the host society treats them. This study found that external collective efficacy significantly increased the level of integrative adaptation. Korean Chinese migrants in Seoul who had been confident in their community’s capacity to reduce prejudice, stereotype, bias, or hostility from the host society and are able to overcome sense of powerlessness, marginalization, frustration, and feeling of isolation in their lives in the host society are more likely to adopt the integrative adaptation strategy.
In the model tested in this study, internal collective efficacy is related directly and positively to external collective efficacy. Individuals’ belief in their capacity to solve their problems collectively within the community can facilitate their beliefs in their capacity to change how they are treated in the host society. In previous studies (E. Kim & Kim, 2015), Korean Chinese migrants’ connections to the mainstream media in Korea generally increase their perception of the negative stereotypes about their migrant communities in the Korean society, primarily because the Korean mainstream media provides negative images of Korean Chinese to their audience. Conversely, Korean Chinese’ connections to the migrant media reduce the negative perception. This study suggests that individuals’ connections to opportunities to produce, share, or use community stories would first increase their perceived confidence in solving collectively their own community issues, and this perception of collective problem-solving capacity regarding community internal issues would be a critical source of confidence in changing the host society in terms of how they treat them.
A model containing community-based communication and social factors in integrative adaptation was also tested and examined whether it would work for other types of adaptation strategy. This question was verified in the separation type of adaptation. Based on the results of the analyses, we found that the hypothetical model with integrative adaptation worked better than the alternative models with other types of adaptation strategy, especially the separation strategy. Such result indicates that individual migrants’ access to civically oriented community communication resources and active engagement in their migrant communities would facilitate integrative adaptation. Instead, Korean Chinese active engagement in their migrant community, especially their access to community-based communication resources measured by ICSN in this study (e.g., talking with other migrants, using the local migrant media, and participating in community organizations, to share community stories), decreases the likelihood of taking separation as an adaptation strategy. 2
Conclusion
This study shows that active community engagement for problem-solving, based on civically oriented community storytelling of the local migrant community, would facilitate integrative adaptation among the Korean Chinese in Seoul. These results are not totally contradictory to negative relations between migrant community engagement and adaptation reported in the previous intercultural communication studies. By differentiating community engagement between active and passive ones and claiming that the negative relationships are only related to passive engagement, this study contributes to the literature on intercultural adaptation. In addition, it expands the scope of CIT by using the theory to explain the relationship between active engagement in local migrant communities and integrative adaptation in the host society.
There are some practical implications of the findings of this study. Most of all, in order to engage migrants in their new social environment as responsible citizens, there should be efforts to diagnose the conditions of communication infrastructure in migrant communities and invest various resources to strengthen community-based storytelling networks. To achieve these goals, there should be community-level efforts to facilitate storytelling capacities in community organizations, (online and offline) migrant media, and various social media sites. One of the most critical goals of these storytelling networks should be to increase not only migrants’ confidence in their beliefs in collective problem-solving capacities for various community issues but also their confidence in changing how the host society treats them. The local migrant community with a strong communication infrastructure can be a platform for integrative adaptation that will give migrants the leveraging power on using critical resources for survival and growth in their new social environment from multiple places without having to get completely assimilated to the host society or to get stuck in their migrant community.
This study contains several limitations. First, although two types of community engagement were suggested, the different effects of passive and active community engagement on adaptation were not specifically tested. This study only tested the effect of active community engagement on adaptation. Future studies must test the different effects of passive and active community engagement. Second, as this study used cross-sectional data, any causal inference must be made with caution. With longitudinal data, future studies can better make casual inferences. Third, our measure of integrative adaptation was limited to capture only subjective connections to the migrant community and the host society. Future research needs to consider using an additional measure that captures more behavioral dimensions of integrative adaptation. Fourth, we tried our best to obtain a representative sample of Korean Chinese in Seoul, which is a hard-to-reach population, by using bilingual Korean Chinese interviewers, face-to-face interviews, or receiving supports from several institutions and organizations that were trusted by the Korean Chinese. Even with these efforts, the nature of data as a nonprobability sample would limit the authors’ capacity to generalize the findings in regards to the Korean Chinese population. Therefore, extra caution must be taken when one studies other migrant communities in Seoul or those in other places based on the findings of this study. Future research must test the hypothetical model in this study for other migrant communities in contexts other than Seoul, to confirm that the main finding of this study, that is, active civic engagement in migrant community will be a positive factor in integrative adaptation, can be generalized.
Footnotes
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the National Research Foundation of Korea Grant funded by the Korean Government (NRF-2011-327-B00923).
