These findings may be limited in generalizability because the sample of undergraduate students at a major university were likely more liberal than others from their respective groups in the community. Spence and Helmreich (1978) found that individuals tend to be more liberal in attitudes toward women while in college. However, the findings from this study add to our knowledge of perceptions of intimate violence by expanding the exploration of Asian American within-group differences. Future studies should focus on community samples that might uncover more variation among and within Asian American subgroups.
Survey designs are useful in describing and making assertions about groups. However, survey designs may be limited in capturing the subtleties and nuances of culture. Large-scale survey studies present difficulties in fully understanding the context of social life (Agbayani-Siewert et al., 1999). If the intent is to understand cultural factors, it may be prudent to supplement surveys with more qualitative methods such as focus groups, ethnographic investigations, and in-depth interviews.
The conventional assumption that Filipino students are more similar to Chinese students and less like Hispanic and white students was not supported by this study. Although Filipinos share similarities with Chinese students, the findings indicate that Filipinos have more similarities with white students than with Chinese students. Chinese students had the least similarities with the other groups. Although the Chinese and Filipinos in this study exhibited similarities (for example, both were egalitarian overall), statistical findings indicate that Filipinos were slightly more egalitarian in their attitudes toward women, tended to justify acts of violence less often, and defined physical aggression as violence more often than Chinese students. In their definitions of physical violence, Filipino students were more similar to Hispanic and white students. The Chinese students had a somewhat narrower view of what constitutes physical violence against women as a means to correct perceptions of wrongful behavior (Yick & Agbayani-Siewert, 1997). Justifying violence in dating relationships may also reflect the cultural norm of "saving face" in situations where the man experiences humiliation because of the woman's behavior, such as loosing emotional control, drunken behavior, and sexual infidelity (Foo & Margolin, 1995).
Within-group gender differences in attitudes toward women were found for all four groups in this study, suggesting an etic universalistic component of culture. For example, Chinese and Filipino women had more similarities with one another than with their male counterparts. Research has consistently reported that women have more liberal attitudes toward women, across cultures, than do men (Braun & Chao, 1978; Chia et al., 1985; Chia et al., 1994; Loo & Logan, 1977; Nelson, 1988; Ullman et al., 1977; Williams & Best, 1990). Although men in this study were egalitarian, they were significantly less so than women and may thus be less motivated to achieve gender equality. Social exchange theory postulates that individuals pursue social relations and interactions that they perceive as offering the greatest reward with the least possible costs (Leonard, 1996). Men may not perceive gender equality as instrumentally or functionally beneficial, but rather as a loss of available resources. Economic, educational, and occupational resources are associated with power and status. A complete disruption of the status quo may not be perceived by men as being in their best interest.
Gender similarities may be the result of changes in traditional cultural beliefs, attitudes, and ideologies regarding gender and women's place in society, most notably in Chinese societies. Uba (1994) argued that the idea of traditional Asian values versus contemporary values and beliefs is inaccurate. The notion that Asian cultures are static, while Western cultures are ever-changing, often permeates and distorts our perception of ethnic and racial minority social and cultural realities. Minority cultures evolve just as other value systems and cultures do. For example, findings from this study and others suggest that in addition to a shift from male dominance (Helmreich, Spence, & Gibson, 1982), Chinese perceptions of women have changed to a more liberal perspective (Chia et al., 1994; Ho, 1990). Because no empirical data have been collected on Filipino American gender-role beliefs and attitudes, it is not clear what, if any, changes have occurred.
With regard to definitions and justifications of violence, the findings were mixed, suggesting a more emic explanation of culture. Within the Chinese and Hispanic groups, men were more conservative in their definition of sexual violence and tended to justify violence more than the women, whereas Filipino and white men and women held similar perspectives.
A conventional model that compares only Asian and white racial categories would not have uncovered similarities and differences found in this study between the two Asian groups. An elaborated model incorporates both social factors (for example, gender and immigration) and cultural factors (for example, beliefs and perceptions about gender roles and dating violence).
Implications for the Social Workers
Ethnic differences within the category Asian American have implications for prevention and interventions. Practitioners should focus on how to assess and understand the unique cultural configurations of Asian American subgroups. Practitioners should begin by evaluating their assumptions about Asian American similarity to avoid stereotypes that might hinder understanding and the client-worker relationship. For example, common sense practice that recommends that the eldest male be addressed first when working with Asian families may be interpreted as offensive by Filipino women and their family members (Agbayani-Siewert & Jones, 1997). The worker can also avoid stereotyping by careful client evaluation using assessment tools and interview methods designed to assess pertinent cultural material. For example, Leigh's (1998) ethnographic interviewing is designed to empower the client to assume the role of educator to inform the practitioner of his or her cultural context. Several brief measures of gender role attitudes, such as the Attitudes Toward Women Scale (Spence et al., 1973) and the Sex Role Traditionalism Scale (Peplau, Hill, & Rubin, 1993), are also recommended. These paper and pencil tests are easy to administer and take approximately five minutes to complete.
Similarities and differences emerged for the groups in this study. Commonalties provide an opportunity for the development of cost-effective programs and services that can be directed to more than one ethnic group. General topics can include information about what dating violence is, the risk factors associated with victimization, and the link between attitudes toward violence and stereotypes. Similarities, however, should not overshadow differences. The practitioner must remain cognizant that Asian Americans speak different languages and have different immigration experiences (for example, refugee versus immigrant), customs and beliefs, and historical pasts that may require different prevention or intervention approaches. For example, Chinese American prevention strategies may need to include a historical examination of the culture's sanction of the subjugation of women, whereas such an account may not be necessary for Filipino Americans.
It is nearly to impossible to obtain substantial knowledge about all Asian American subgroups that reside in the United States. Moreover, inclusion of numerous groups in research is costly. On the other hand, incorrect assumptions lead to costly social welfare policies, programs, and interventions that do little to serve the needs of ethnic minority populations. It is necessary to find a balance between the reality of limited resources and the practical and ethical responsibility of providing effective services to those in need. Developing guidelines to help a social services agency identify which Asian subgroups it should focus its services for would be a first step in this direction. These guidelines should apply to other minority subgroups as well.
One guideline might be to identify a subgroup that has a specific social or health problem. Problem identification should not be based on popular public perceptions of an issue as a problem. Agencies should consider the relative size of the group in its catchment area, knowledge possessed about the group, and information from reliable sources in the affected community such as leaders and community organizations, the justice and judicial systems (for example, rates of domestic violence), health and mental health agencies and hospitals (for example, service use and types of presenting problems), census data (for example, poverty statistics), and other public social services agencies (for example, child abuse agencies). Social services agencies can form a task force to give voice to the community to identify problems.
For example, an agency with a large population of Filipino Americans in its catchment area might uncover the following: There is a paucity of knowledge on Filipino Americans, who have the highest immigration rate into the United States next to Mexicans, are the second largest Asian American group according to the 2000 U.S. Census (Dela Cruz & Agbayani-Siewert, 2003), and appear to be experiencing problems with underemployment and exploitation (Cabeza, Shinagawa, & Kawaguchi, 1986-1987), youth gang involvement (Alsaybar, 1993), mental health, and suicide (Lau, 1995) that would warrant more in-depth knowledge about this group.
Once a group has been identified, practitioner education about that group can take various forms: in-service training on the needs of that group identified by the task force, recruitment of culturally knowledgeable representatives of the identified group, introduction of relevant cross-cultural interventions and skills, access to cultural and demographic literature on ethnic minority groups provided by the worker's agency, or credit for cross-cultural conferences, workshops, continuing education, and materials.
This study provides evidence that assumptions of Asian American similarity based on conventional models of research are unfounded. White American society often serves as the reference point in research with ethnic minority groups that have been clustered into general categories. It is time to move beyond the categorization of disparate groups into four racial categories. Racial categories are not synonymous with cultural values, beliefs, and attitudes and therefore do little to expand our knowledge. Moreover, as this study demonstrates, greater differences may exist among members within a racial group than among seemingly disparate racial groups.
Social work has generally tended to rely on untested assumptions to develop policies and programs for racial minority populations, especially Asian Americans. It is imperative that programs and services be based on knowledge rather than assumptions. In an era of increased outcome accountability, combined with fewer economic resources, it behooves social service agencies to provide relevant and appropriate services to the populations they serve. Social work has a long and rich history of working with vulnerable and op pressed groups and has established some rapport with minority communities. Social work researchers and practitioners have a unique opportunity to take the lead in focusing on an elaborated model that goes beyond conventional racial categories.
Original manuscript received September 1, 1998
Final revision received February 22, 2000
Accepted August 14, 2000
The author thanks the UCLA Asian American Studies Center, Institute for American Cultures, for grant support and Dr. Stuart Kirk, UCLA Department of Social Welfare, and Glenn Omatsu, UCLA Asian American Studies Center, for reading drafts of the article. An earlier version of this article was presented at the Annual Program Meeting, Council on Social Work Education, March 1998, Orlando, FL.
member (yes)(f) 6.1 5.3 5.0 14.7 NOTE: Reference categories in parentheses. (a) All values are percentages unless otherwise indicated. (b) Chinese significantly younger than Hispanics. (c) White Americans significantly older than all other ethnic
groups. (d) Nonwhite Hispanics have significantly more years in the
United States than all other immigrant groups. (e) Chi square = 167.34, alpha = .001, df = 3. (f) Chi square = 14.53, alpha = .01, df = 3.
Table 2 Mean Scores and Standard Errors: Attitudes Toward Women Scale, Contextual Justifications, and Definitions of Dating Violence among Four Ethnic Groups