February 1999 Volume 5 Number 4
Straddling Different Worlds: The Acculturation of Vietnamese Refugee Children -- Min Zhou
THE ACCULTURATION OF VIETNAMESE REFUGEE CHILDREN
University of California, Los Angeles
Forthcoming in Rumbaut, Rubén G. and Alejandro Portes. Eds. Ethnicities: Coming of Age in Immigrant America. Berkeley. University of California Press.
The Vietnamese are one of the largest refugee groups to have settled in the United States since the mid-1970s. The emergence of this new ethnic group in the American scene, along with approximately half a million other Southeast Asians, is primarily the result of U.S. military involvement in the region. The story of Vietnamese Americans is one of very rapid growth, from a population of insignificant size in the early 1970s, to one that numbered over 615,000 by 1990, when it made up almost 10% of the nation's Asian American population. And even this figure understates the true size of the Vietnamese-origin population, since it excludes no fewer than 200,000 sino-Vietnamese (ethnic Chinese), who fled Vietnam and arrived in the United States as part of the larger refugee outflow from Southeast Asia. (Rumbaut 1995).
The children of Vietnamese refugees are thus the newest of the new second generation. As of 1990, 52% of all Vietnamese American children under 18 years of age were U.S. born, 27% arrived in the United States prior to the age of five, 17% arrived between the ages of five and 12, and only 4% arrived as adolescents.
For many Vietnamese refugees, the journey to America and their adjustment to the new land has been extremely hard. With the exception of the relatively small elite group evacuated at the fall of Saigon, most of the refugees lacked education, job skills, and measurable economic resources. They also suffered from the trauma of war and flight and from the severe emotional distress that they experienced at refugee camps. Once arrived in the United States, they were powerless to decide where and when they would be resettled, with almost all of the refugees starting their American life on public assistance.
Growing up in America has been difficult for the children of the refugees as well. The parents' low socioeconomic status makes it hard for the children to succeed, even though both parents and children desperately want to get ahead. The environment in which the children find themselves further limits the chances: too many live in neighborhoods that are poor and socially isolated, where local schools do not function well and the streets are beset by violence and drugs.
To all these difficulties are added the generic problems of second generation acculturation, aggravated by the troubles associated with coming of age in an era far more materialistic and individualistic than encountered by immigrant children in years gone by. Today's second generation often finds itself straddling different worlds and receiving conflicting signals. At home, they hear that they must work hard and do well in school in order to move up; on the street they learn a different lesson, that of rebellion against authority and rejection of the goals of achievement. Today's popular culture, brought to the immigrants through the television screen, exposes children to the lifestyles and consumption standards of American society, raising their expectations well beyond those entertained by their parents. As a result, the children are not as "willing" as their parents to work at low-paying, low-status jobs; but at the same, many may not have the education, skills, or opportunities to do better. This mismatch between rising aspirations and shrinking opportunities can either lead to "second-generation decline" (Gans 1992) or provoke "second-generation revolt" (Perlmann and Waldinger 1997).
So do immigrant children manage to traverse the difficult social terrain they encounter? Or do they fall into one of the many traps that afflict young people - especially those of minority background - in contemporary America? This chapter inquires into these questions through the prism of Vietnamese American adolescents in San Diego. I seek to explore the complex, multi-dimensional process of acculturation, as understood and experienced by these children, and as captured by the San Diego portion of the Children of Immigrants Longitudinal Surveys (the CILS). Before delving into the specific case study, I first furnish a historical account of why the Vietnamese fled their country, how they resettled in the United States, and how resettlement affected the adjustment of the second generation.
THE RESETTLEMENT OF VIETNAMESE REFUGEES IN THE UNITED STATES
The Vietnamese arrived under circumstances quite different from those encountered by today's typical legal immigrant. Unlike most other contemporary immigrants, the Vietnamese were pushed out of their homeland, forced to leave without adequate preparation, and with little control over their final destinations. Many of them had minimum formal education, few marketable skills, little English-language proficiency, and scant knowledge of the ways of an advanced society -- assets that would ease the passage into America. They also lacked a preexisting ethnic community that could help out with assistance of varying sorts. Instead, they had to initially depend almost entirely upon the government and individual or institutional sponsors to determine where they would resettle.
The Flight from Vietnam
The initial flight from Vietnam was touched off by the withdrawal of American troops from Vietnam and by rumors and fears in the face of an uncertain future. Given the bitterness of the war in Vietnam, the suddenness of South Vietnam's defeat in the spring of 1975, and rumors about the Hanoi government's intent to execute all former South Vietnamese civil servants, policemen, and other officials, as well as all those who had served the Americans in any capacity, many people left the country, by sea, land, and air.
Before 1977, a total of 130,000 refugees fled Vietnam. This initial wave of refugees were mostly members of the elite and the middle-class who either had access to the evacuation arranged by the American military or could afford their own means of flight. After the airlift at the fall of Saigon, thousands of refugees fled Vietnam by boat from the end of 1975 to 1978. But the phrase "boat people" came into common usage as a result of the flood of refugees casting off from Vietnam in over-crowded, leaky boats at the end of the 1970s and the beginning of the 1980s. By 1979, an estimated number of 400,000 refugees, known as the second wave of flight, escaped Vietnam in boats, thus the name "boat people," to Thailand, Malaysia, Indonesia, and Singapore (Caplan et al. 1989; Tran 1991). This mass exodus was disproportionately made up of ethnic minorities, particularly the sino-Vietnamese.
It seems relatively easy for most Americans to understand why many South Vietnamese fled out of the country in the early days of the fall of Saigon. But it is more difficult to grasp why the refugees kept fleeing for so many years after the Vietnam War ended, especially considering that the Hanoi government did not plunge the South into a bloodbath as so many had once feared. As Carl L. Bankston and I argued in Growing Up American (1998), several reasons account for the lengthy flow of refugees from Vietnam. First, political repression continued to make life difficult for those individuals who were detained at or released from education camps as well as for their family members. Second, economic hardships, exacerbated by natural disasters and poor harvests in the years following the war, created a widespread sense of hopelessness. Third, incessant warfare with neighboring countries further drained Vietnam's resources for capital investment and development. These severe adversarial conditions, triggering the second and third exodus of Vietnamese "boat people" in the late 1970s and early 1980s, continued to send thousands of refugees off the rugged journey to a better life.
Once the early refugee waves established communities in the United States, leakage information from America to Vietnam provided impetus for a continuing outward flow. Upon resettlement in the United States and other Western countries, many Vietnamese refugees rebuilt overseas networks with families and friends. Letters frequently moved between the receiving countries and Vietnam, providing relatives in the homeland with an intricate knowledge of the changing refugee policies and procedures of resettlement countries.
Since the mid-1990s, however, immigration from Vietnam has begun to assume a different shape. Though a substantial proportion continues to be admitted as refugees, an increasing number have been entering the United States as family-sponsored immigrants, a flow that will surely grow in years to come. As the refugee influx ebbs, family reunification can be expected to dominate Vietnamese immigration into the next century.
The Resettlement in the United States
Unlike most regular immigrants who are sponsored either by close families or by U.S. employers and can make decisions as to where to settle in the United States, refugees are often sponsored by the government or voluntary agencies of the receiving country and cannot choose their places of resettlement. In the case of the Vietnamese and other Southeast Asian refugees who did not have established ethnic communities in the United States, the U.S. government of resettlement agencies, known as VOLAGs, almost entirely decided where the refugees would settle. Initially, the United States admitted Vietnamese refugees as a response to a special emergency, rather than as part of an on-going process of resettlement. Thus, the official resettlement policy aimed at dispersing refugees to minimize the impact on local receiving communities and integrating refugees into the American economy and society as quickly as possible.
Minimizing the impact of refugees on the local communities meant an initial policy of scattering Southeast Asians around the country. The early attempts to scatter refugees around the nation led this new ethnic group to establish a presence even in those Midwestern and mountain states least populated by recent immigrants. However, as time went by, distinctive Vietnamese concentrations emerged, through secondary migration, in large metropolitan areas that were the most popular destinations for many recent immigrants of varying nationalities. Geographically, the Vietnamese were highly concentrated in California. As of 1990, almost half of America's Vietnamese population lived in California (up from 27% in 1980). Within California, the Vietnamese also clustered in just a few metropolitan areas with the state, with over 70% of California's Vietnamese population living in just four metropolitan areas -- AnaheimÂSanta Ana, Los AngelesÂLong Beach, San Jose, and San Diego.
Integrating refugees into American society meant the development of a comprehensive program of support and preparation. While a chief goal of resettlement was to help refugees achieve economic independence as quickly as possible, they benefited from generous government aid assuring a basic level of well-being until the refugees became self-sufficient. Many Vietnamese refugees started their American life at the bottom of the socioeconomic ladder, a fact reflected in their high rates of labor force non-participation, unemployment, poverty, and dependency on public assistance. In fact almost all of them began on welfare. The three cash assistance programs most commonly utilized by Vietnamese refugees were Aid to Families with Dependent Children (AFDC), Supplementary Security Income (SSI), and Refugee Cash Assistance (RCA). The first two programs were forms of public assistance, or welfare, available to U.S. citizens. AFDC was for low-income families with children and SSI was for the elderly or disabled poor. Refugees not eligible for either of these forms of assistance generally received RCA during their first six to eighteen months and were also able to apply for food stamps.
While early official policies toward Vietnamese refugees were generally sympathetic, the American public was ambivalent. As early as 1975 at the time of the initial entry of the Vietnamese, a Gallup Poll indicated that a majority of Americans then preferred that the Vietnamese be kept out of the United States (Kelly 1986). Many Americans felt that such culturally different people would create additional difficulties for acculturation, and some saw the Vietnamese as new economic competitors, as well as reminders of an unpleasant war. In the late 1970s, Americans at lower socioeconomic levels, those most likely to have to compete with the new arrivals for jobs, were found to show significantly higher levels of prejudice than individuals of higher socioeconomic status, against the Vietnamese (Cotter and Cotter 1979). The entry of Vietnamese people into the fishing industry of Gulf Coast in the late 1970s and early 1980s provoked resentment from native-born fishermen who felt threatened by the competition (Starr 1985). Prejudice against the Vietnamese resulted in some incidents of violence. In 1983, for example, a Vietnamese high school student in Davis, California was taunted by a group of white high school students and then stabbed to death. In 1989, in Raleigh, North Carolina, a Chinese American who was mistaken for a Vietnamese was beaten to death by men who were angry over the Vietnam War. In 1990, in Houston, a young Vietnamese American was stomped to death by skinheads. These were, of course, exceptional occurrences, but they could represent the most extreme manifestations of an incomplete acceptance of the Vietnamese by American popular opinion.
The Adaptation to the U.S. Society
Despite severe exit conditions -- the traumatic flight combined with poor human capital and economic resources -- and unfavorable contexts of reception -- a lack of pre-existing community ties, high levels of dependency, and an ambivalent ad sometimes hostile public, Vietnamese refugees have made progress into American society after a decade or so of adjustment. Even with a continuously large refugee influx, the 1990 census data showed a number of quite striking improvements over the pattern observed ten years earlier (see Table 1): English proficiency rate was 39%, up from 27% in 1980 (and exceeding the level attained by the U. S. foreign born population overall). The proportion of college graduates among adults aged 25 and over was 17%, up from 13%. Labor force participation rate among males aged 16 or over was 72%, up from 66% 1980. And ethnic entrepreneurship burgeoned at 7% as compared to 3% in 1980. As their human capital and labor force status steadily improved, so too did their economic well-being. By 1990, the median household income of the Vietnamese stood at $33,500, above the average of $30,000 for all American households; home ownership was 49%, up from 37% in 1980; and the poverty rate stood at 25%, down from 28% in 1980 but still substantially higher than the national average. San Diego's Vietnamese population generally revealed a similar pattern, with two notable exceptions: during the 1980s, levels of educational attainment slightly declined and the poverty rate substantially increased, with both changes likely to have been affected by the continuous influx of low-skilled refugees into the metropolitan area as well as by the exodus of more affluent Vietnamese to Santa Anna and San Jose (Rumbaut and Ima 1988). Notwithstanding improvement, the Vietnamese still lagged behind their American counterparts socioeconomically.
Table 1 about here
Progress was also noticeable among the 1.5 or second generation growing up in America. As of 1990, Vietnamese adolescents were less likely than their American peers to drop out of high school, and Vietnamese young adults were more likely than their American peers to attend college. For example, in the Los Angeles metropolitan region, the dropout rate among U.S. born Vietnamese aged 16-19 was 5% compared with 8% among whites; the high school dropout rate among U.S. born Vietnamese adults aged 18-24 was 9% compare with 11% among whites; and the college attendance rate among U.S. born Vietnamese was 50% compared with 38% among whites (Cheng and Yang 1996). These trends, should they continue, foreshadow a catch-up and eventual surpassing of the U.S. educational norm. However, the socioeconomic circumstances in which most Vietnamese children find themselves are still greatly disadvantaged, comparable only to those encountered by children of the most underprivileged native minority group. If Vietnamese children are to succeed, first in school and then, in their careers, they cannot depend on the class resources of their families, limited as they are.
Though the record mainly highlights progress, one can also find troubling trends. The 1990 census data revealed that Vietnamese adolescents were disproportionately more likely than their other Asian counterparts to be institutionalized, constituting a quarter of all Asian institutionalized adolescents, though in absolute numbers relatively few of them were confined to correctional institutions of various sorts. In terms of rates of institutionalization, Vietnamese adolescents ranked second among racial/ethnic groups (210 per 100,000), after blacks (695 per 100,000), and higher than all other Asian groups (93 per 100,000). Noticeably, this phenomenon was a problem of youth: while the rate of institutionalization for all Vietnamese was 140 per 100,000, the rate for minors under 18 was 210 per 100,000. This contrast took on an additional meaning in that while institutionalized Vietnamese adults were almost all foreign-born, the delinquent youths were the products either of refugee flight (as many were unaccompanied minors) or of the U.S. experience (Rumbaut and Ima 1988; Zhou and Bankston 1998). The evolving Vietnamese youth delinquency problems are so real and sometimes life-threatening that they have become the number one concern in Vietnamese communities across the nation. A 1994 Los Angeles Times poll showed that the greatest number of Vietnamese in Southern California named crime, street violence, and gangs as their chief community problems (Los Angeles Times 1994). The bifurcated trends suggest that although acculturation pressures put many Vietnamese on the road to success, they lead others to go astray, for reasons to be explored in greater detail shortly.
The Impacts of Refugee Resettlement on Children
Young people old enough to remember flight and resettlement continue to be haunted by the traumas of sudden exile. The younger ones or those born in the United States might be free of horrific memories of life in Vietnam, of the flight from the ancestral land, or of life in refugee camps. But they are still deeply affected by family histories and stereotypes about their ethnicity in the host country. The abrupt nature of the move from Vietnam to America means that life in Vietnam is still an immediate reality, even for young people whose entire lives have spent in the United States.
The unique circumstances of refugee resettlement exact a heavy toll on the children. As mentioned previously, the movement of refugees is involuntary, lacking adequate preparation, control over final destinations, and a long-term orientation toward settlement, making it a process qualitatively different from that experienced by regular immigrants. Dislocation and disorientation of the parent generation combined with disadvantaged socioeconomic status have affected children's adjustment in several key aspects. At home, the children of refugees are often caught in a role reversal due to social and economic isolation of parents. As children increasingly adopt the role of family spokespersons, parental authority declines, which further intensifies generational conflicts. The issue of conformity to or rebellion against parental authority and homeland cultural ways, is probably one that is faced by most young people with immigrant parents. But for Vietnamese children, the fact that their parents are not simply immigrants, but refugees, adds a unique dimension to their outlooks on life. Hardship in Vietnam and the process of exile have become a central family myth, a shared story that shapes understanding, behavior, and identity formation. The difficulties suffered in Vietnam and in the movement from Vietnam to America have given Vietnamese children a strong sense of their own identity. The traumas of repression and the pains of exile are not just individual biographical episodes, but defining experiences for a whole people. Some young people, born or raised in the United States, may reject or react against these experiences, but they still grow up with them as ever-present parts of their own lives. Thus, one of the greatest challenges facing second-generation, or 1.5-generation, Vietnamese Americans is whether they will respect their family histories and conform to parental expectations or reject these family histories and expectations.
At the neighborhood level, the impact of the parents' social and economic marginality on the children is profound. Most adult refugees came here with little preparation for beginning life in the new country. Although their labor force participation increases and economic dependence decreases over time, the Vietnamese are still heavily concentrated in minimum wage occupations, and still disproportionately rely on public assistance to survive. The young people, keenly aware that they belong to a group that is "marginal" to American society, will sometimes admit to feelings of discomfort toward the relatively low-status jobs performed by their parents and dependence on public assistance. These young people are not simply agonized by the embarrassment of welfare dependency and the negative view of their parents; they have also suffered from real hardships: their families provide them with limited resources and they frequently live in low-income communities subject to adversarial influences of social isolation and poverty.
VIETNAMESE ADOLESCENTS IN SAN DIEGO
San Diego has been a major resettlement center for Vietnamese refugees since 1975. Part of the reason is the location of Camp Pendleton, one of the four military camps to which the 1975 refugees were evacuated from Vietnam (Rumbaut 1989). As of 1990, the metropolitan area ranked the 6th largest metropolitan area for Vietnamese settlement in the United States, following Anaheim-Santa Ana, Los Angeles-Long Beach, San Jose, Houston, and Washington D.C. Of the Vietnamese residing in this metropolitan area, 81% are concentrated in the central city, making up the fourth largest racial/ethnic minority group following Mexicans, blacks, and Filipinos. Of the 15 census tracts where Vietnamese were overrepresented (at least 5%), 12 of them had a high concentration of immigrants (ranging from 30% to 44%); 7 of them had an extremely high poverty rate (over 25%) and 4 had a poverty rate above the national average (16% to 24%). The Vietnamese initially clustered in a neighborhood called Linda Vista ("LV" which later referred to as "Little Vietnam"), just north of Highway 8. During the 1980s, a much larger Vietnamese enclave took shape in East San Diego south of Highway 8, but the more resourceful Vietnamese moved north to Mira Mesa, and further north to Rancho Penasquitos (Rumbaut and Ima 1988).
In San Diego, the Children of Immigrants Longitudinal Study interviewed 363 Vietnamese 8th and 9th graders whose median age was 14 in 1992 (Time 1) and 304 of the same group in 1995 (Time 2), who had then become high school juniors or seniors. About 48% of these respondents were female; 15% were U.S. born, 46% were 1.75ers (the foreign born arriving before age 6), and 39% 1.5ers (the foreign born arriving before after age 6). Table 2 shows the contextual factors influencing Vietnamese adolescents. At the neighborhood level, most the Vietnamese families lived in mixed neighborhoods with mostly other immigrants (41%) and with non-whites (56%). Over 60% of the Vietnamese adolescents attended schools where white students were the numerical minority, 49% were enrolled in inner city schools, and 48% were in schools where over half of the students were qualified for receiving subsidized lunch, an indicator of poverty. The average size of the schools was 1,447, though the size of many inner-city schools in San Diego varied from year to year. One of the inner-city schools included in the survey provides an illustrative example. In 1992, Hoover high school had an enrollment of a little over 2,000 students, of whom 16% were white, 17% black, 33% Latino, and 34% Asian. About 66% of the students were so poor that they qualified for free school lunches. In 1998, a Los Angeles Times article showed that, of Hoover High School's 1,900 students, only about 5% were white, 20% black, roughly 50% Latino, and 20% Asian; and that the majority of students qualified for free school lunches (Woo 1998). Moreover, the Vietnamese adolescents, like other immigrant children in school, were relatively isolated within their own coethnic circle. The majority of them reported having friends from Vietnam, and the most frequently used language with friends was Vietnamese.
Table 2 about here
These patterns bear significant sociological implications on acculturation. Living in underprivileged inner city neighborhoods, in particular, immigrant children attend schools where they meet native-born peers, many of whom have little hope for the future and have developed a oppositional outlook of the mainstream society. They also meet foreign-born peers of other national origins, many of whom came from recent immigrant families with few socioeconomic resources. Such contact is likely to alter the pattern of acculturation, leading immigrant children adopt goals and aspirations different from those of the mainstream middle class and of their own parents. How these children negotiate and navigate around their immediate social environment between neighborhood and home shapes their identity and acculturation experience. Next, I take a close look at the Vietnamese adolescents in San Diego from the CILS data to examine the complex process of acculturation and the particular ways by which acculturation affects children's adjustment.
ACCULTURATION AND ITS VARIANTS
The classical assimilationist perspective conceptualizes acculturation as a linear and progressively irreversible process that leads to the reduction of cultural heterogeneity and convergence in patterns of language, thinking, feeling and behavior. Albeit Gordon's multidimensional typology of assimilation (Gordon 1964), the ideology behind this conceptualization is the normative expectation that diverse immigrant groups from underprivileged backgrounds should eventually abandon their old ways of life and become completely "melted" into the mainstream culture as indistinguishable non-ethnics, or "Americans," through residential integration and occupational achievement in a sequence of succeeding generations. Thus, models measuring acculturation often operate under the assumptions that there is a natural process by which diverse immigrant-ethnic groups come to share a common culture and gain equal access to the opportunity structure of society; that this process consists of gradually deserting old cultural and behavioral patterns in favor of new ones; and that, once set in motion, this process moves inevitably and irreversibly toward cultural homogeneity (Park 1928; Warner and Srole 1945).
From the assimilationist standpoint, distinctive ethnic traits such as native languages, ethnic identity, ethnic institutions, and ethnic social relations are assumed to be sources of disadvantages. These disadvantages negatively affect acculturation, but the effects are greatly reduced in each of the successive generations, since native-born generations adopt English as the primary means of communication and become more and more similar to the earlier American population in language use, outlook, and behavior. Although complete acculturation to the dominant American culture may not ensure all immigrants full social participation in the host society, immigrants must free themselves from their old cultures in order to begin rising up from marginal positions (Gordon 1964). How acculturation is conceptualized has been a central theme of inquiry among students of immigration. As acculturation intrinsically interacts with dynamic social environments in which it takes place, the process often entails multiple dimensions with segmented outcomes (Portes and Zhou 1993). In the following pages, I focus on examining several key measures of acculturation and delineating patterns of acculturation among the children of Vietnamese refugees using the CILS data.
Language Adaptation and De-Ethnicization
Two changes, in particular, denote acculturation: the first occurring when immigrants or their descendents abandon the native tongue in favor of English; the second transpiring when they identify themselves in new ways, either linking home and host countries or dropping the home country link altogether. Assimilation models predict that language shifting and identificational changes will both be positively affected by time. Put somewhat more formally, the longer the U.S. residency, the more complete the switching from foreign, to English-language monolingualism, and the more likely that self-categorization will take the form of a non-hyphenated American identity.
To what extent does the experience of Vietnamese adolescents approximate the assimilationist prediction? The CILS data allow for a closer examination. On language proficiency, the CILS data include a set of survey items asking a respondent's ability to speak, understand, read, and write English as well as his or her mother tongue. The response scales range from "very well," "well," "not well," to "very little." From these language proficiency items, I have constructed three new variables: English monolingualism (coded 1 for those who reported speaking English only), fluent bilingualism (coded 1 for those reporting speaking, understanding, reading, and writing English very well while at the same time speaking and understanding Vietnamese well), and preference to speak English (code 1 for those selecting "English" as the language they prefer to speak most of the time). In terms of identity, the CILS data include a question asking respondents to identify which ethnic/national origin category (out of 76 given categories) they will use to identify themselves. Using this information, I have constructed an ethnic identity variable by collapsing all possible answers into five categories: "Vietnamese," "Vietnamese-American," "Asian-American," "Other," or "American."
Table 3 shows the patterns of language adaptation among Vietnamese adolescents in San Diego at Time 1 and Time 2 of the survey. The data indicate a general tendency of convergence toward English corresponding with a divergence from Vietnamese. The change in language proficiency conforms to what is predicted by assimilationist models -- English language proficiency increases over time with decreasing proficiency in one's mother tongue. The trend in language use also reveals a similar pattern, thought not as clear-cut. English monolingualism, representing only a small fraction of the group, increases only slightly, but fluent bilingualism decreases quite significantly, with more adolescents preferring to speak English as of Time 2.
Table 3 about here
While language adaptation seems to be linear and straightforward, de-ethnicization appears to be a more complex process. Patterns of self-reported ethnic identity among Vietnamese adolescents shift over time, but that shift is by no means linear. As shown in the upper panel of table 4, in 1992, 44% of Vietnamese identified themselves ubiquitously as "Vietnamese," another 46% as "Vietnamese Americans," almost none as "Asian Americans," and only a few (less than 3%) as "Americans." Three years later, the modal category became the national origin category with over half of the respondents (53%) using "Vietnamese" to identify themselves. The proportion identified as "Vietnamese American" dropped to 32%, the proportion identified as "Asian American" increased substantially to 14%, and the proportion identified as "American" dropped to almost zero (only 1 case). The varied patterns of ethnic identification do not appear to be associated with gender and generational status. With gender and generational status controlled, the self-reported identity associated with national origin or one with a hyphen is still the dominant category. These findings complicate the prospect of acculturation.
Table 4 about here
What is more interesting about this identify shift is that, among those who identified themselves as "Vietnamese" at Time 1 (127 out of 304), 63% (83) remained "Vietnamese" while 21% (27) switched to "Vietnamese-American" by Time 2. In contrast, among those who identified themselves as "American" (10 out of 304) at Time 1, all had switched to either "Vietnamese," or "Vietnamese-American," or "Asian-American" by Time 2. Comparatively, among those who identified themselves as "Vietnamese-American" (146 out of 304) at Time 1, 39% (57) remained so identified while 46% (67) dropped the hyphen by Time 2. Figure 1 visualizes these multidirectional shifts.
Figure 1 about here
It may be argued that self-reported identity among adolescents should not be treated as a reliable measure of ethnic resilience, because teenagers' identity formation is influenced not simply by acculturation but also by the process of growing up as adolescents. The lower panel of table 4 explores to what extent these adolescents take their chosen ethnicity seriously. In CILS, respondents were asked another question immediately after they had chosen their identities -- "how important is this identity to you?" Possible responses were "not important," "somewhat important," and "very important." As shown, 57% of all respondents viewed their chosen identities as very important. Among those who identified themselves as "Vietnamese," almost 70% regarded this ethnic choice as very important, compared with 50% or less among those choosing other identities. This finding may not exclude the possibility of adolescent identity swing. However, given that fact that more adolescents shifted their identities back to the unhyphenated ethnic category (their parents' ethnicity) than the other way around, the chosen national-origin identity is both real and is seriously taken, indicating a strong ethnic resilience among Vietnamese adolescents. Here, the linear prediction of an ideal "ethnic-turned-American" scenario does not seem to fit well with the Vietnamese. The Vietnamese adolescents' experience is definitely not de-ethnicization, but rather ethnic resilience. This "reverse" pattern challenges the traditional notion of acculturation.
Perception, Feeling, and Expectation of Discrimination
At another dimension, acculturation may be measured by an individual's perception, feeling, and expectation of social acceptance or treatment by the host society. The assimilationist models assume a unified host society to which immigrants are expected to assimilate and attribute new immigrants' uneasiness, anxiety, and discomfort with the host society and their perception and feeling of discrimination to the lack of exposure and initial disadvantages. Over time, immigrants, especially their children, are expected to experience greater ease and comfort with American life as they become an indistinguishable part of the host society. For example, upon arrival, immigrants tend to cluster in ethnic enclaves for emotional and instrumental supports provided by coethnics or ethnic institutions because they are structurally and culturally barred from participation in the larger society. As they overcome disadvantages associated with the cultural lag or the lack of human capital, they will face fewer entry barriers. Likewise, their perception of the host society will be increasingly favorable as they become an integral part of that society.
Stratification models provide contrary hypotheses. Because American society is stratified by class and race, perceptions, feelings, and expectations of social acceptance or treatment are inevitably shaped by the system of stratification. Often times, members of the first generation are eager to embrace the American ideals of individual freedom and equal opportunity. Upon arrival, they are busy trying to rebuild their lives in the new land or are simply striving for survival and tend to view their own initial disadvantages or downward mobility as temporary and surmountable. Thus, the less acculturated are less likely to be affected, at least psychologically, by the negative effects of racial/ethnic stratification. As they become more acculturated, their perception of the host society will be inevitably molded by the social or racial strata to which they have become a part. While the less acculturated first generation members may be slow in adopting the same sense of entitlement and deprivation as the native-born, their children, with a birth right and a high degree of acculturation, tend to be more likely than their parent generation to see things through a racial/ethnic lens and thus more likely to attribute certain life and work experiences to factors associated with racial or ethnic group memberships.
Table 5 explores this possibility with regard to perceived, felt, and expected discrimination and makes comparisons between children and their parents. Perception of discrimination was measured by two survey items in the CILS. Children and parents were asked similar questions but in a somewhat different way. Children were asked whether they agree with the following two statements: "There is racial discrimination in economic opportunities in the U.S.," and "Americans generally feel superior to foreigners." Possible answers ranged from "agree a lot," "agree a little," "disagree a little," to "disagree a lot." Parents were asked two comparable questions: "Compared to people of other races or nationalities, you view opportunities for your own job advancement as: a lot less, a little less, same as others, a little more, or a lot more," and "In general, you think that white Americans consider themselves: superior, equal, or inferior."
As shown in the upper part of table 5, the overall perceptions about racial discrimination and native or white superiority were substantial. Almost a third of Vietnamese adolescents held pessimistic views on the overall racial discrimination in economic opportunities, and the decreases in reporting from Time 1 to Time 2 were insignificant. Comparatively, parents held a more optimistic view. Though measures in the parental survey were not exactly the same as those in the children's survey, less than 2% of the parents viewed their own economic opportunities as "a lot less" than people of other race or national origins. In terms of the perceived superiority of Americans, however, children and parents varied widely but the reverse: about a third of the children at Time 2 thought that Americans felt superior to foreigners, up significantly from Time 1 (28%), while over half of the parents thought that white Americans considered themselves superior. One possible explanation for the parent-child disparity in perception is that parents may have formed stereotypical preconceptions about white superiority before migration whereas children's perception may be based more on their actual exposure than preconceptions.
Table 5 about here
The middle part of table 5 reports the feeling of discrimination. Children were asked a straightforward question: "Have you ever felt discriminated against?" Parents were asked a similar question: "Do you feel that you have been discriminated against?" Answers to both questions were a simple "yes" or "no." As shown, the feeling of discrimination varied widely between children and parents. Children were over twice as likely as their parents to report that they had felt discrimination, and such reporting increased over time. Among children, there were specific measures on how they felt about discrimination by various groups. Noticeably, there was a significant increase over time in their reporting on all groups, except on discrimination by African American students. It is noteworthy that the proportions reporting discrimination by counselors and by Latinos increased over 100%, which should not be taken lightly despite the overall low proportions. School counselors usually work closely with students at a more intimate and less threatening basis than teachers. The significant increase in feelings about counselor discrimination are suggestive of serious problems in the way in which schools treat Vietnamese students. Likewise, the significant increase in Latino discrimination may suggest that intergroup tensions are being built up and accumulated among students in school, since most of the Vietnamese students attend schools where Latino students are overconcentrated.
The lower part of table 5 examines expected discrimination. Respondents were asked whether they thought the statement -- "No matter how much education I get, people will still discriminate against me" -- "very true," "partly true," "not very true," or "not true at all." Over a third of the respondents considered the statement very true or partly true and there was a modest increase over time. There were no directly comparable questions in the parental survey, but parents were asked three yes-no questions about expected discrimination against their children by white Americans: 1) "My child will experience opposition if he/she joins a club of white Americans;" 2) "My child will experience opposition if he/she moves into a white American neighborhood;" and 3) "My child will experience opposition if he/she marries a white American." About 8% responded affirmative to the first and third statement, and only 2% to the second statement. Here, the differences in parent-child views were significantly large. Finally, children and parents differed in attribution; children were much more likely than their parents to single out racial, ethnic, or nationality as the primary reason for discrimination.
Of course, these findings are by no means conclusive. It is not uncommon for adolescents, native-born and immigrants alike, to constantly change or adjust their perception about the world as they experience the pain and tensions of growing up. However, the consistency in reporting patterns over time among adolescents and the persistent gaps between parents and children in their perception, feeling, and expectation of discrimination are beyond the mere agonizing experience of adolescence.
GENERATIONAL DISSONANCE AND CONSONANCE
Conflict between immigrant parents and their U.S.-born or -raised children is a perennial theme in the study of immigrant adaptation. Members of the second generation want to fit in. Because their frame of reference is shaped by the lessons learned from their American peers, from the television screen, and from other forms of mass media, this process of fitting in also means that the children want more than their parents even do. Parents, in contrast, tend to focus on survival as well as economic mobility. More importantly, they retain a dual frame of reference, in which the home society, and not the norm of the host society, provides the benchmark against which to assess their accomplishments. These differences provide the flash points for the conflict between immigrant children and their parents. In the children's eyes, the parents are traditional-minded and old-fashioned, holding tightly to "old world" values, norms and behavioral patterns, which inhibit assimilation. In the parents' eyes, the children are too attracted by American culture and its negative influences.
Differing life experiences between children and parents inevitably widen the generational gap, leading to intense bicultural conflicts that push children and parents into separate social worlds. However, it may be simplistic to assume that the parent generation is reluctant to let go of the old-country tradition and ties and that the younger generation is eager to sever these tradition and ties. For many of today's new immigrants, acculturation began before arrival in the United States, thanks to the influence exercised by American media influence in their homelands, and in the Vietnamese case, a direct American presence (Rumbaut 1997 and 1999). Hence, immigrant parents and children will inevitably share certain experiences in acculturation while differing in many others.
The generational gap between immigrant adolescents' world and their parental world is best captured by what Portes and Rumbaut conceptualize as "generational dissonance and generational consonance" (Portes and Rumbaut 1996: Chapter 7). Generational consonance occurs when parents and children both remain unacculturated, or both acculturate at the same rate, or both agree on selective acculturation. Generational dissonance occurs when children neither correspond to levels of parental acculturation nor conform to parental guidance, leading to role reversal and parent-child conflicts. Past studies of earlier European immigrant family life have shown intense generational dissonance in the family over host-society and homeland interests (Brown 1994; Child 1943; Covello 1972). However, other studies have also found that generational conflicts within immigrant families do not necessarily frustrate successful adaptation to a host society. In fact, many immigrant families today consciously modify and adapt their own values, such as the value on education, to make them more congruent with the host society than with the homeland, laying the foundation for generational consonance (Schulz 1983; Sung 1987; Zhou forthcoming; Zhou and Bankston 1998).
Whether or not generational conflicts lead to negative adaptational outcomes depends largely on the class status of the family as well as that of the community. Today's new second generation comes from diverse socioeconomic backgrounds. Children from middle class families may not like their parents' ways of pushing them to do as the parents expect, but these children are shielded with a comfortable safety net to counter negative cultural influences and are unlikely to reject the expectations of the community in which they live, which are coincided with their familial expectations. Children from poor families, in contrast, lack such a safety net and often encounter in their neighborhoods a gap between familial goals and those prevailing in their local social environment, which are often marginal to, some even at odds with, the larger society. Under this situation, rejecting familial goals on the part of the children can cause a lack of mobility or downward mobility.
Worlds Apart: Generational Dissonance
At what point do children and parents diverge? Language is undoubtedly the foremost point of divergence. Parents' lack of English proficiency not only jeopardizes their own social mobility but also creates intense generational tensions. For example, parents who are unable to speak English well will in many respects become dependent on their children for daily contact with the outside world, hence putting the children who act as interpreters and translators on behalf of the parents in an authoritative position. Such role reversal usually leads to weakening parental authority. Meanwhile, the children are anxious that they may never become "American" because of these duties and intrinsic family and ethnic ties (Zhou 1997a and 1997b; Zhou and Bankston 1998).
Here I examine two aspects of language dissonance: the language proficiency gap and fluent bilingualism. As discussed previously, fluent bilingualism is operationalized in terms of the fluency in speaking, understanding, reading, and writing English and fluency in speaking and understanding one's mother tongue. This measure is designed to capture the extent to which children and parents maintain fluid and regular communication, assuming that continued mother tongue retention would allow parents to maintain effective control over children.
While children are far more likely to acquire English-language proficiency than their parents, and to do so at a more rapid rate, the degree of divergence from parents may vary from one dimension to another. For example, Vietnamese children are more than ten times as likely as their parents to speak, understand, read, or write English very well. This substantial language gap can be viewed as a potential risk for immigrant families, where parents may feel a lose of control over their children and the children may develop resentment their parents' broken or accented English. It can also be a potential risk causing the children to sever ties from their ethnic institutions that give meaning and directions to the lives of the parent generation (Habenstein 1998). Nonetheless, language dissonance is not just an intergenerational problem, it is also a problem among members of the younger generation. Over 70% of Vietnamese adolescents use the Vietnamese language frequently with their friends in school and that, among those whose self-reported identity is "Vietnamese," over 70% are limited bilinguals.
Another point of divergence is the perception, feeling, and experience of the American society. Immigrant children and their parents tend to perceive their host society and their relationships with it from different angles. In terms of perceived discrimination in economic opportunities, felt discrimination, and expected discrimination, there are significant generational gaps (see also table 5). Take felt discrimination for example: children are almost three times as likely as their parents to report feeling of discrimination, not to mention the much wider gaps in perceived and expected discrimination. With respect to American ways, children are three times as likely as their parents to prefer American ways most of the time. Figure 2 visualizes the aspects and degree of divergence.
Figure 2 about here
It should be noted that generational dissonance occurs when children acculturate faster than parents and parents lose the ability to exercise guidance, developments which lead to intensified parent-child conflicts, role reversal, and ultimately the loss of parental authority (Portes and Rumbaut 1996). As we all know, children will almost always acculturate faster, and this form of generational dissonance is assumed to start soon after children enter school and become exacerbated over time (Covello 1972; Habenstein 1998). The question has to do with whether and when the difference in the pace of acculturation leads to intense intergenerational conflicts and the loss of parental control.
Figure 3 addresses to what extent Vietnamese children experience intergenerational conflicts by looking at the longitudinal patterns of children's responses to the following six survey items: in trouble with parents, being disliked by parents, parent's lack of interest in their children, being embarrassed by parents' ways, parent-child conflict, and parent-child split on American ways. As Figure 3 shows, about a quarter of the adolescents report that they get in trouble with parents because of the disagreement in the way of doing things "most of or all the time," but this proportion is not particularly high and it does not change much over time. A small percentage of these adolescents feel that their parents do not like them -- less than 10% respond "partly true" or "very true" to the statement "my parents don't like me" in 1992 -- and that percentage goes up only 2 percentage points in 1995. Relatively more adolescents (about 30%) respond "partly true" or "very true" to the statement "my parents are usually not interested in what I say" in 1992, but that percentage remains the same in 1995. A quarter of the adolescents initially reported that they were embarrassed by their parents' ways, and that proportion fell to 18% in 1995.
Figure 3 about here
Parent-child conflict and parent-child split on American ways are two reconstructed measures for generational dissonance that take into account both the children and their parents views. Parent-child conflict refers specifically to the situation where the children feel embarrassed by parents' ways and at the same time get in trouble with parents because of disagreement in the ways of doing things, or when the children think that their parents either did not like them or were uninterested in what they say. As shown, the proportion of those reporting conflicts is low, and there is not much change from 1992 to 1995. Parent-child split on American ways refers to the situation where the children prefer American ways when their parents dislike American ways. Here, parent-child split on American ways is moderate (less than 30%) and remains similar over time. Overall, these findings imply a relatively low level of generational dissonance among Vietnamese adolescents. Put in somewhat different terms, the generations do not always agree, but the range of disagreement is relatively modest and does not appear to eventuate in sharp conflict. Not surprisingly, then, the survey also shows that parents maintain relatively strong rules. About two thirds of the parents reported that they had specific rules about academic performance (about maintaining certain GPA and homework), watching television or playing video games, and household chores. It would appear that these rules are also successfully enforced, given the relatively low levels of parent-child conflict, as just note above.
Thus, generational dissonance need not always cause a loss of parental control. In this case, parents and children differ in linguistic habits, in perceptions of discrimination, in preference for American ways, and yet parental control appears unchallenged. That disagreement does not eventuate in conflict, I suggest, results from the structure of social relations within the family and the immigrant community, which, as I have shown elsewhere (Zhou 1998; Zhou and Bankston 1998), prevents the displacement of parental authority so often found in immigrant communities.
Worlds Connected: Generational Consonance
At what point do children and parents converge? Children and parents share common educational aspiration and similar familial beliefs. As Figure 4 shows, both generations aim high, with respect to education: over 90% of Vietnamese adolescents report that they want to attain at least a bachelor's degree, an answer echoed by 87% of their parents. Moreover, the children actually maintain higher expectations than their parents: 68% of the children want to earn a graduate degree, compared to only 16% of their parents. Interestingly, the children don't perceive this disparity, as two-thirds think that their parents' expect them to earn a graduate degree. While we have no way of knowing whether this difference is an artifact of the survey or reflects some "real" generational divergence, it may not matter: the crucial point is not so much the correctness of the children's perception as the relationship between those perceptions - whether right or wrong - and their own expectations. In short, the children aspire to those goals which they perceive their parents to hold.
Figure 4 about here
Parents' and children's also intersect on the home front. Family cohesion and familism have been central in the lives of Vietnamese Americans. Many parents fear that, as children become increasingly acculturated, they may be losing their children and that the level of familism may decline. The 1995 CILS data contain three relevant questions on family cohesion and three on familism. Family cohesion measures the attitude and feeling toward family, reflected in the responses to the following statements: "Family members (often or always) like to spend free time with each other;" "Family members (often or always) feel very close to each other;" and "Family togetherness is (often or always) very important." Familism measures the family ideology about duties and obligations, indicated by the agreements to the following statements: "If someone has the chance to help a person get a job, it is always better to choose a relative rather than a friend;" "When someone has a serious problem, only relatives can help;" and "The person who helps you the most with homework is a family member (parents or siblings)."
Since the CILS data do not contain parallel information on the earlier survey, I cannot make a longitudinal comparison. Figure 4 shows a considerably high level of family cohesion: 35% report that family members often or always like to spend free time with each other; 47% report that family members often or always feel close to each other; and 60% report that family togetherness is often or always very important. Familism is even more compelling: 71% agree that they will give preference to a relative over a friend in the help of finding jobs; 70% agree that only relatives can help with serious problems; and 67% agree that the person who helps the most with homework is either a parent or a sibling.
So far, I have shown the variants of acculturation and the dynamics of intergenerational relations. How immigrant children are related to American society is largely contingent upon particular patterns of language use, ethnic identification, perception, feeling, and expectation of daily experiences, and various forms of social relations in and out of the family. At the societal level, it involves cross-cultural exposure and intergroup relations. At the family level, it involves bicultural socialization and intergenerational relations. At either level, immigrant children must constantly straddle between these different worlds -- that of their adolescent peers and that of the adults. In the process of growing up, they are unlike other American adolescents, who simply fight between an adolescent world and an adult world. They also have to struggle to make sense of the inconsistencies between two separate adult worlds: that of the immigrant family or community and that of the larger society. Even within their own adolescent world, they often find inconsistencies centering around hyphenated identities. Next, I focus on examining the effects of some of these acculturation issues on the adjustment outcomes.
THE EFFECTS OF ACCULTURATION
What constitutes successful adjustment? Recent studies of immigrant children have focused on several key measures of adjustment: self-esteem and depression at the psychological level, school suspension at the behavioral level, and educational performance and aspiration at the academic level (Portes and Rumbaut 1996; Rumbaut 1997, 1999, forthcoming; Zhou and Bankston 1998). Accordingly, I examine five reconstructed variables in the CILS indicating different levels of adjustment. Self-Esteem is a composite measure from 10 survey items developed by Rosenberg (1979) with a scale ranging from 1 to 4 and depression a composite measure from four survey items ranging from 1 to 4 (see Appendix 1 for details; also see Rumbaut forthcoming). Average GPA is a continuous variable ranging from 0 to 5. Number of school suspension varies from 0 to 6. Educational Aspiration is categorized into a dummy variable coded 1 for those who expect to earn a graduate degree and 0 otherwise. Table 6 provides descriptive information on the these dependent variables. Over a third of Vietnamese adolescents report low self-esteem and almost a third report high depression. Over time, both self-esteem and depression increased modestly. Over half of the Vietnamese had above-average GPAs. About 80% had never been suspended from school during the study period. Educational aspiration was strong at both Time 1 and Time 2 and the increase was significant.
Table 6 about here
How are the various acculturation measures -- ethnic identity, language use, felt discrimination, generational dissonance, and generational consonance, related to Vietnamese adolescents' adjustment? To examine the effects of acculturation, I use OLS regression models (and a logistic regression model for educational aspiration). I include eight key independent variables in my multivariate analyses. These are: ethnic identity (coded 1 for those who identified themselves as Vietnamese or Vietnamese-American and regarded this identity as very important), fluent bilingualism (coded 1), English monolingualism (coded 1), felt discrimination (a composite measure ranging from 0 to 7), parent-child conflict (coded 1 if adolescents felt embarrassed by parents' ways and at the same time often got in trouble with parents, or thought that their parents either did not like them or were uninterested in what they said), parent-child split on American ways (coded 1 if children preferred American ways when their parents dislike American ways), parent-child agreement in educational achievement (coded 1 for those whose educational aspiration matches that of their parents), and family cohesion (a composite measure ranging from 0 to 15). My multivariate analyses also control for gender (code 1 as female), length of U.S. residence (coded 1 for those who arrived in the United States less than five years), parental education (coded 1 as for having either parent with a high school education or more), and inner-city school (coded 1). Study hours daily is included as an additional control in the model predicting GPA. Appendix 2 provide detailed information on the operationalization of these independent and control variables. Based on the findings in the previous sections, I hypothesize that ethnic identity, fluent bilingualism, and generational consonance to positively influence outcomes, while English monolingualism, felt discrimination, generational dissonance negatively influence outcomes.
Table 7 provides OLS or Logistic regression models predicting five adjustment outcomes. Since space precludes detailed discussion of the individual coefficients, I simply summarize some of the most important findings. First, fluent bilingualism significantly boosts self-esteem, reduces depression, and raises educational aspiration, supporting existing research about the importance of maintaining communication with parents in immigrant families (Bankston and Zhou 1995; Rumbaut 1994). Second, Vietnamese or Vietnamese American identity taken as very important by the adolescents significantly influences self-esteem but does not yield noticeable effects on other outcomes. Third, felt discrimination significantly increases depression, but also raises educational aspiration. Fourth, neither of the two generational dissonance measures have significant effects, though they both point to expected directions. However, parent-child conflict significantly raises rather than reduces educational aspiration. Finally, the effects of both measures of generational consonance are in the expected directions: parent-child agreement in educational achievement significantly affects outcomes at emotional, behavioral, and academic levels, while family cohesion significantly affects the adolescents only at the emotional level.
Table 7 about here
DISCUSSION AND CONCLUSION
For the children of refugees and immigrants, the adaptation to American society involves a constant struggles: learning to speak a new language different from the one spoken at home; developing one's own identity; going to school; meeting societal expectations; and ultimately fitting in. In contrast to their parents, these children are unlikely to use their parents' homeland standards to assess their process in the new land. Instead, they consciously develop and modify their own coping strategies in order to become acculturated and incorporated in American society. Inevitably the children straddle different social worlds. While they often experience conflict with their parents, the children also find ample room for agreement. Thus, this study of acculturation among Vietnamese adolescents in San Diego suggests that acculturation takes a multi-dimensional form.
Though generalization at this point may be premature, here I only summarize some of the most important themes that have emerged from this study and speculate about their implications. First, immigrant adolescents experience much faster linguistic adaptation than their parents, with the language gap between English proficiency and mother tongue proficiency widening with time. While acquiring English proficiency is undoubtedly crucial, maintaining fluent bilingualism is equally important and sometimes even more beneficial. As shown in the analysis, English monolingualism does not have any significant effects on outcomes, but fluent bilingualism significantly raises self-