Intimate partner violence (IPV) has been recognized as a serious problem for decades, threatening women’s mental and physical health. Extensive research has examined the prevalence, nature, and consequences of IPV, which has helped to inform policymakers and practitioners developing and implementing programs and services for victims and their families (Catalano, 2008; Sherman & Berk, 1984; Straus & Gelles, 1990; Tjaden & Thoennes, 2000). However, only a limited number of studies have been conducted on IPV among racial and ethnic minorities (Ingram, 2007; Leung & Cheung, 2008), and in particular, Asian Americans. Although Asian Americans are one of the fastest growing populations in the US (United States Census Bureau, 2010), they have been underrepresented in most national data collections due to their small numbers. Furthermore, Asian Americans are not a homogeneous group, but are comprised of a variety of ethnic subgroups that are diverse in language, culture, and immigration history (Reeves & Bennett, 2004). Thus, treating them as a homogeneous group is not only an oversimplification of their diversity within Asian American communities, but is also misleading in informing policies and practices targeted to specific subpopulations, such as Chinese, Filipinos, and Vietnamese. Limited knowledge of IPV among Asian Americans and their subgroups presents challenges for service providers in the many communities in which there are an increasing number of Asian Americans. This study narrows the gap between the available knowledge and much needed information on IPV among Asian Americans by identifying risk factors for IPV and examining the ethnic differences within Asian Americans subgroups.

IPV among Asian Americans

Major national studies estimate the prevalence of IPV from 0.4% to 12% of the population (Catalano, 2008; Straus & Gelles, 1990; Tjaden & Thoennes, 2000). Literature suggests there are racial differences in the prevalence and contexts of IPV. In their studies in 1975 and 1985, the National Family Violence Survey (NFVS)—which collected the first national data on IPV in the US—showed that African and Latino Americans perpetrated IPV more often than Caucasians. They excluded Asian Americans from the racial comparisons, possibly due to their small numbers (Straus & Gelles, 1990). Similarly, the National Crime Victimization Survey (NCVS), which has collected data annually since 1973, found the highest IPV rates among Native Americans. It is important to note that Asian Americans were also excluded from this analysis because of the small number of cases (Catalano, 2008). The National Violence Against Women Survey (NVAWS), conducted from 1995 to 1996, was the first national data that reported Asian Americans as a separate group. The NVAWS found that Native Americans experienced more IPV, while Asian and Latino Americans experienced less IPV, than other racial groups (Tjaden & Thoennes, 2000). Finally, the Collaborative Psychiatric Epidemiology Studies (CPES) is the most recent national data on IPV with data on mental health and IPV being collected in the early 2000s (Pennell et al., 2004). An analysis of the CPES found that Asian Americans experienced less IPV, whereas African Americans experienced more IPV than other racial groups, with no difference between Caucasians and Latino Americans (citation excluded for anonymity). Apparently, study results are not consistent in their identification of racial differences in IPV. Various methodological differences (e.g., contexts of the survey administration and counting methods) seem to be related to the discrepancies across the studies (Cho, 2009; Tjaden & Thoennes, 2000).

Although the national studies have rarely considered Asian Americans—not to mention their ethnic subgroups—as a separate group, some small community and clinical studies have reported information on the prevalence and characteristics of IPV among ethnic subgroups. Subgroups included in these studies are Chinese (Hicks, 2006; Yick, 2000), Filipinos (Hoagland & Rosen, 1990), Koreans (Kim & Sung, 2000; Lee, 2007), Bangladeshi (Rianon & Shelton, 2003), Vietnamese (Bui, 2003; Morash, et al. 2007), Japanese (Yoshihama, 1999), Nepali (Thapa-Oli, et al. 2009), Cambodians (Yoshioka & Dang, 2000), and South Asians (Hurwitz, et al. 2006; Raj & Silverman, 2002). While the annual prevalence rates reported by these studies range from as low as 3% to as high as 40%, the different measurements used for identifying IPV, as well as a lack of comparable groups, make it difficult to generalize these results to all Asian Americans across the nation. One exception, however, is a study that utilized the recent CPES national data on mental health and IPV in Asian and Latino Americans, which is also used for the current study. It found that the prevalence rates of minor IPV among Asian American women were 10%, and those of severe IPV were 1.5%. Vietnamese reported the lowest prevalence rates, compared to Chinese, Filipinos, and other Asians (Chang, et al. 2009). In sum, these study results indicate that Asian Americans seem to be victimized by IPV less often than other racial groups, and that differences in IPV exist across ethnic subgroups. A few studies on IPV among Asian American subgroups recently began to emerge, but more research is necessary to enhance our understanding.

Risk Factors for IPV among Asian Americans

Previous studies have identified several risk factors associated with IPV in the general population. Low socioeconomic status (SES)—low income, low education, unemployment, etc.—has been suggested as one of the most prominent risk factors (Bassuk, et al. 2006; Straus & Gelles, 1990). This is particularly relevant to some Asian American communities, with a significantly lower reported SES. In general, Asian Americans are known to have a higher SES than any other racial group (US Census Bureau, 2010). This may be true for some groups such as Chinese and Filipinos, who are likely to have lived in the US for multiple generations, and established themselves as successful immigrants; thus they tend toward a higher SES. Other groups, however, such as Vietnamese and Laotians, are more likely to have arrived in the US recently as refugees and are more likely to have a low SES—they are often less educated, under- or unemployed, and have low incomes (Chang et al., 2009). Thus, the latter is more likely to be vulnerable to IPV than the former. Aside from SES, interpersonal factors, such as marital satisfaction, also seem to affect IPV. Previous study results consistently show that low marital satisfaction is strongly associated with an increased risk for IPV (Stith, et al. 2008).

While some factors, including SES and marital satisfaction, seem to be associated with IPV for all populations, other factors may be particularly relevant to immigrants. For instance, acculturation is definitely relevant to immigrants, but hardly meaningful to non-immigrants. Theories of acculturation assume that when individuals with a certain cultural background come into contact with another culture, their cultural identity can change through gradual acceptance of the language, beliefs, values, and behaviors of the dominant society (Berry & Kim, 1988; Ramirez, 2007). Acculturation is often measured using birthplace and language preference. For example, those born in the US and who speak English at home are supposed to be more acculturated than those who were foreign born and prefer non-English languages. Contrasting levels of acculturation between partners are believed to be strongly associated with risk for marital dissatisfaction among immigrants (Hovey, 2000). Thus, it may be misleading to suppose that the relationship between marital satisfaction and IPV found among non-immigrants would be the same as that for Asian Americans, unless acculturation is taken into consideration. However, the relationship between acculturation and IPV is not as clear as that of marital satisfaction and IPV. Some studies found no relationship between acculturation and IPV (Kaufman Kantor, et al. 1994; Perilla, et al. 1994; Ramirez, 2007), while others reported some association (Champion, 1996; Kim & Sung, 2000). Directions of relationship have been also inconsistent. Some research has found that low acculturation increased the risk for IPV (Kim & Sung, 2000), but others reported the opposite (Jasinski, 1998; Sorenson & Telles, 1991). More research is needed, with a bigger sample, to examine whether—and how—acculturation is associated with the risk for IPV.

Perceived discrimination, which is related to acculturation, is another immigration-related factor that may influence IPV in Asian Americans. Since perceived discrimination only recently emerged as an important factor in understanding IPV in racial and ethnic minorities, information on the relationship between discrimination and IPV is scarce. Nevertheless, previous studies with different populations have reported some associations between discrimination and IPV. For instance, African American women who perceived themselves as being discriminated against showed an increased risk for IPV (Waltermaurer, et al. 2006). However, it is not clear why perceived discrimination affects IPV. Discriminated minority women may perceive mainstream society as being hostile toward them, and thus, be reluctant to seek outside help and be more willing to endure IPV within their family and community (Hampton, et al. 2003). By the same token, men who have faced discrimination may cope with stresses related to their experiences by perpetrating IPV (Flores-Ortiz, 2000; Sugihara & Warner, 2002).

In addition to immigration-related variables, sociocultural factors—gender roles, and family and community values—have been suggested as influencing IPV in Asian Americans. In traditional Asian cultures, men are raised being taught to dominate and control the family with all kinds of power, whereas women are expected to respect and obey them (Huisman, 1996; Kim, et al. 2007). The family and community are emphasized over individual in Asian cultures, which may make women endure IPV, without seeking outside help (Bauer, et al. 2000). Study results are consistent in showing associations between the traditional gender and family values, and IPV, across Asian ethnic subgroups (Morash et al., 2007; Song, 1996; Yick & Agbayani-Siewart, 1997). However, previous studies with different populations have found inconsistent patterns in the relationship between those values and IPV (Mirande, 1997; Sugihara & Warner, 2002). Thus, further research is needed to examine whether, and why, traditional gender and family values may work differently between Asian Americans and other racial and ethnic groups.

In sum, literature on IPV in Asian Americans shows that the prevalence rates for IPV differ across ethnic subgroups, and that various risk factors are associated with IPV. Those factors may account for some differences in prevalence rates across subgroups. For instance, low SES, such as low income and unemployment, may explain the different prevalence rates of IPV across Asian subgroups, given that low SES increases the risk for IPV in the general population (Kessler, et al. 2001; Sorenson et al. 1996). Indeed, a few studies that used national data, such as NFVS and the NCVS, reported that the initial differences in IPV prevalence rates across racial groups disappeared, when controlled for SES and other individual factors (Lambert & Firestone, 2000; Rennison & Planty, 2003; Straus & Gelles, 1990). Studies with Latino Americans also found similar results (Bassuk et al., 2006). To date, there has not been a study that examined these relationships within the Asian American community.

Study Hypotheses

This study used a nationally representative sample to examine the risk factors for IPV among Asian Americans, as well as whether there are ethnic differences in those factors. Specifically, this study tested two hypotheses. Previous studies have suggested that socioeconomic and cultural factors might account for different IPV prevalence rates across racial and ethnic groups. Those analyses do not include Asian Americans. This study hypothesized that (1) IPV prevalence rates would not differ across subgroups within Asian American women, controlling for demographic, interpersonal, and cultural factors. Of those factors, some may be associated with IPV for certain subgroups, but not for others. For instance, traditional gender roles may be associated with IPV for Chinese Americans, but not for Vietnamese Americans. Thus, it was also hypothesized that (2) the association between IPV and risk factors would differ across Asian ethnic subgroups.

Method

Study Sample

This study used data from the National Latino and Asian American Study (NLAAS). The NLAAS collected nationally representative data on mental health and IPV from Latino and Asian Americans in the US, aged 18 or older (Alegria, Takeuchi, et al., 2004). Bilingual interviewers conducted either face-to-face or telephone interviews. After deleting cases with missing values for the study variables, 548 Asian American women were included: 193 Chinese, 167 Filipino, and 188 Vietnamese Americans. Given that Chinese, Filipinos, Vietnamese, Asian Indians, and Koreans constitute 80% of the Asian American population (Reeves & Bennett, 2004), the three distinctive ethnic subgroups included in this study are expected to provide detailed, and somewhat representative, information on IPV and differences in risk factors for IPV among Asian Americans.

Variables

Demographic variables included age, ethnicity, education, financial security, and employment. Ethnicity consisted of three categories: Vietnamese, Filipino, and Chinese. Education referred to the respondents’ years of education and was dichotomously recoded for this study: “Under 12 years” and “12 years or over.” To assess financial security, respondents were asked, “In general, would you say you have more money than you need, just enough for your needs, or not enough to meet your needs?” Those who answered as having either “more than you need” or “just enough for your needs” were coded as being “financially secure,” with others coded “financially not secure.” Employment was measured by three categories in the original survey, and was recoded dichotomously for this study: “employed” was considered as “employed,” while “unemployed” was comprised of those either “not in the labor market” or “unemployed.”

Interpersonal factors included two variables—marital satisfaction and marital partnership. Marital satisfaction was assessed with self-ratings of marital satisfaction, using an 11-point scale from 0 (the worst) to 10 (the best). Marital partnership was defined as perceptions regarding how well partners trusted and understood each other. It was measured by asking respondents’ opinions on four statements, such as “How much does your partner/spouse really care about you?” and “How much does your partner/spouse understand the way you feel about things?” Responses were rated on a Likert scale, ranging from 1 (not at all) to 4 (a lot; Alegria, Takeuchi, et al., 2004; Pennell et al., 2004). Answers to the questions were averaged: 1 was considered the lowest partnership and 4 the highest.

Two variables were used to identify immigration-related factors —birthplace and perceived discrimination. Birthplace was comprised of two categories: US-born and foreign-born. Perceived discrimination was measured by asking respondents’ opinions on nine statements regarding their experiences with discrimination. Representative items included “You are treated with less courtesy than other people,” and “People act as if they think you are not smart.” A six-point Likert scale was used, ranging from 1 (never) to 6 (almost every day; Alegria, Vila, et al., 2004). The average score of the nine questions was used for analyses with a higher score indicating higher perceived discrimination.

Family values assessed the sociocultural characteristics. Family values were measured through the average of ten questions that asked for respondents’ opinions on statements regarding family pride and cohesion. Representative items included “We really do trust and confide in each other,” and “Family members feel loyal to the family.” Responses were rated on a Likert scale, ranging from 1 (strongly agree) to 4 (strongly disagree; Alegria, Vila, et al., 2004). The responses were reverse coded so that higher scores represented higher family values.

IPV was measured by the adapted subscales of the Conflict Tactics Scale (Straus, 1979). Respondents were asked how often their partner/spouse had perpetrated IPV on them over the course of their relationship. The severity of IPV was assessed with two categories: less severe (e.g., pushing and spanking) and severe (e.g., kicking and threatening with a gun; Straus, 1979). Respondents who experienced less severe IPV were coded as “less severe IPV,” while those who experienced severe IPV—with or without less severe IPV—were coded as “severe IPV.”

Analysis

The NLAAS data collection used a multistage area-probability sample design. To maximize the utility of this design, unbiased estimates of population statistics and relationships should be computed, using weights and complex survey sample design measures (Heeringa et al., 2004). Statistical Package for the Social Sciences for Windows (SPSS) version 13.0 was used to conduct all analyses. The Taylor series linearization method was applied, to estimate variances from complex sample datasets (Rust, 1985). Descriptive statistics were obtained first, identifying sample characteristics and differences among ethnic subgroups in all the variables. Weighted estimates, unweighted sample sizes, and associated p-values are reported in the results.

Logistic regression analyses were conducted, to identify risk factors for IPV in Asian Americans. IPV victimization, which included both less severe and severe IPV, was the dependent variable, with all others being independent variables. Interaction terms of ethnicity and each of the independent variables were included in the analysis to examine differences in risk factors for IPV across ethnic subgroups. Only statistically significant interaction terms were retained, together with all of the independent variables, in the final analysis. Complex-design-adjusted 95% confidence intervals were reported, along with population-weighted estimates.

Results

Table 1 presents descriptive statistics and differences across ethnic subgroups. The average age of the sample was 44 years of age. Filipino Americans showed the highest educational attainment, followed by Chinese and Vietnamese Americans. On average, 77.6% of Asian Americans completed 12 years of education or more. There was no difference in financial security and employment across subgroups. The vast majority of the sample (82.9%) considered themselves financially secure; the majority (61.3%) were employed. Most Vietnamese Americans (97.4%) were foreign-born, as were 89.1% of Chinese Americans and 79.4% of Filipino Americans. On average, 87.6% of the sample was foreign-born.

Table 1 Sample characteristics
Full size table

There were differences in interpersonal and socio-cultural characteristics, and in the prevalence rates of IPV among Asian Americans. Vietnamese Americans showed the highest scores in both marital satisfaction and family values, followed by Filipino and Chinese Americans. Filipino Americans reported the highest level of marital partnership, followed by Vietnamese and Chinese Americans. Filipino Americans reported more instances of perceived discrimination in their daily lives than any other Asian Americans. Average rates of victimization reported by Asian Americans were 9.2% for less severe IPV and 1.6% for severe IPV. Less severe IPV was the least prevalent among Vietnamese Americans, while Chinese Americans showed the least prevalence rates for severe IPV.

Table 2 shows the results of the logistic regression analysis of risk factors associated with IPV victimization. Initial differences in the prevalence rates across ethnic subgroups, as found from descriptive statistics, disappeared when controlling for demographic, interpersonal, and sociocultural characteristics. Thus, the current study supports the first study hypothesis, which did not expect an ethnic difference in the prevalence rates of IPV when controlling for those variables. Of all the variables, birthplace and perceived discrimination were significantly associated with IPV. Foreign-born Asian Americans were 51.8% less likely to be victimized by IPV than those born in the US (OR = .482). Asian Americans who felt discriminated against had a 4.766 times higher likelihood that they were to be victimized by IPV than those who do not have the discrimination feeling (OR = 4.766).

Table 2 Risk factors associated with IPV victimization
Full size table

Specific risk factors definitely accounted for differential IPV victimization in the distinctive Asian American subgroups. While perceived discrimination was a significant risk factor for IPV, the strength of that association differed across the subgroups. Specifically, compared to Chinese Americans, Vietnamese Americans were 79.1% less likely and Filipino Americans were 76.8% less likely to be victimized by IPV (OR = .209 and .232, respectively). Thus, the current results support the second hypothesis, which expected different associations between IPV and its risk factors across Asian ethnic subgroups.

Discussion

The current study found significant differences in the prevalence of IPV and the associated risk factors, across three distinctive Asian American ethnic groups. Descriptive statistics revealed interesting ethnic differences in their sociocultural characteristics. Overall, the vast majority of the sample (87.6%) was foreign-born, which is much higher than the US Census results; in 2008, 64% of Asian Americans in the US were foreign-born (United States Census Bureau, 2010). This seems to be related to the current study sample, which only included married respondents. Unmarried respondents did not answer the questions related to IPV in the NLAAS data collection, and thus were excluded from the sample. Married respondents are likely to be older and foreign born, as compared with the young and unmarried ones. This aspect of the study sample also seems to explain its relatively high average age of 44 years. Thus, the current study results may not be representative of all Asian Americans; it excludes the experiences of the relatively young. Most of the Vietnamese Americans (97.4%) were foreign-born, which was higher than Chinese and Filipino Americans. Interestingly, Vietnamese Americans also showed the highest family values and the lowest perceived discrimination.

Asians and Latinos have been reported as placing a higher value on family over individuals than other racial groups do (Bauer et al., 2000; Huisman, 1996). As foreign-born immigrants are more likely to preserve their cultural values than their US-born counterparts are, it is not surprising that Vietnamese Americans showed higher family values than Chinese and Filipino Americans. Although study results on the relationship between birthplace and perceived discrimination are not conclusive (Yip, et al. 2008), individuals born in the US were often reported to experience discrimination more than their foreign-born counterparts (Kuo, 1995). Thus, Vietnamese Americans having the lowest perceived discrimination might be related to their birthplace.

Descriptive statistics also revealed differing prevalence rates of IPV across ethnic groups, with Vietnamese Americans showing the lowest. Given that they were mostly foreign-born, showed the highest family values, and had the lowest perceived discrimination, compared to other ethnic groups, their low frequency of IPV victimization might be related to sociocultural characteristics. Indeed, logistic regression analysis results have confirmed that ethnicity had no effect on IPV victimization when controlling for demographic, interpersonal, and sociocultural variables. This is consistent with previous studies that reported similar results, either among several racial groups or within Latino Americans (Bassuk et al., 2006; Lambert & Firestone, 2000; Rennison & Planty, 2003; Straus & Gelles, 1990). Thus, this study clearly shows that IPV is an artifact of a complex set of intertwined factors, rather than a result from membership in any certain group.

It might be surprising to find that both marital satisfaction and marital partnership did not affect IPV victimization, given that higher marital satisfaction and marital partnerships were consistently reported to be associated with lower IPV prevalence (Stith et al., 2008). Marital satisfaction and marital partnership themselves may not be a strong factor for IPV among immigrants, whose lives are likely to be affected by a variety of events and stresses related to immigration. For instance, marital satisfaction, alone, was reported to be strongly associated with acculturation among immigrants (Hovey, 2000). The current study results seem to support this conjecture. While marital satisfaction was not significantly related to revictimization, there may be moderators between the two variables. It is possible that while marital satisfaction is associated with IPV, birthplace and perceived discrimination, two significant predictors of IPV victimization, account for most of this association. Additional logistic regression analysis, excluding birthplace and perceived discrimination from the independent variables, confirmed this possibility; marital satisfaction and marital partnership were significantly associated with IPV (data not shown). These results suggest that marital satisfaction and marital partnership decrease the risk for IPV in general but, among immigrant populations, birthplace and perceived discrimination confound their effects.

The current study results showed that foreign-born Asian Americans were less likely to be victimized by IPV than their US-born counterparts were. These results are consistent with previous studies that reported similar relationships among Korean and Mexican Americans (Champion, 1996; Kim & Sung, 2000). Since the current study used birthplace as a proxy for acculturation, these results may suggest that the more acculturated immigrants are, the more vulnerable they are to IPV. It is not clear why acculturation influences IPV. One explanation may be that birthplace may influence immigrants’ perceptions of what constitutes IPV. For instance, foreign-born Mexican women rated similar abusive behaviors as less severe than their US-born counterparts (Peek-Asa, et al. 2002). This could be because the people in some countries are likely to accept IPV as more normal than those in the US. However, there are caveats to interpreting the current study results as if they relate to acculturation. First of all, since Asian Americans’ perception of IPV and associated factors may differ from Mexican Americans, the results based on the latter cannot be generalized to the former without further research. In addition, as study results on this relationship are not conclusive among Latino Americans (Kaufman Kantor et al., 1994; Ramirez, 2007), the current results—that acculturation can be a risk factor for IPV among Asian Americans, if not among other racial groups—need to be viewed with caution. Finally, birthplace is just one of many measures of acculturation. These results, therefore, need to be considered in combination with other measures of acculturation, such as language preference, acculturative stress, and assimilation.

The current study results revealed that the more Asian Americans felt discriminated against, the more likely they were to be victimized by IPV. Previous studies have consistently reported similar relationships among African and Latino Americans (Flores-Ortiz, 2000; Sugihara & Warner, 2002; Waltermaurer et al., 2006). The reasons for why and how victims’ perceived discrimination increases their risk for IPV are mostly unknown. One explanation for this relationship is the cultural dilemma racial/ethnic minority women face. They may have been socialized to sacrifice themselves on behalf of the survival and integration of the family unit within a discriminatory society, which makes them endure IPV (Billingsley, 1992).

Minority men face discrimination as well. They may become frustrated by institutionalized discrimination. In response, they may resort to violence as a means of solving disputes, increasing their risk for perpetrating IPV against their partners (Hampton et al., 2003). Since this idea has mostly been used with the African American population, whose experiences are likely to be different from other immigrant populations, generalizing this explanation to other immigrants seems premature. For instance, some immigrant women may have to endure IPV due to their immigration status, which makes them dependent on their partner (Bauer et al., 2000). Thus, future research needs to include immigration-specific factors potentially related to perceived discrimination—such as immigration status, birthplace, and years of living in the US—and explore the multifaceted relationships between those factors, perceived discrimination and IPV.

In this context, ethnicity may be one of the factors influencing perceived discrimination and IPV. Indeed, the current study showed that while perceived discrimination increased the risk for IPV, the strength of the relationship differed across ethnic groups. Specifically, the effect of perceived discrimination on IPV was weaker among Vietnamese and Filipino Americans than it was for Chinese Americans. Since it is not clear how perceived discrimination affects IPV, as discussed above, it is even harder to explain the reasons for the ethnic differences found in this study. One explanation may be that other factors, such as employment and birthplace, in addition to ethnicity, can moderate or mediate the relationship between perceived discrimination and IPV. It is also possible that other unknown factors confound the relationship. As suggested earlier, the multifaceted relationships between perceived discrimination, IPV, and potential intervening factors need further exploration.

Conclusions

This is one of the few studies to have used the nationally representative sample to examine ethnic differences in risk factors for IPV among Asian Americans. This study found that the prevalence rates of IPV varied across ethnic groups, but those differences disappeared when controlling for demographic, interpersonal, and sociocultural variables. The findings also show that the risk for IPV was higher for US-born Asian Americans, and that perceived discrimination increased the risk.

When viewing the current results, the reader should keep several limitations in mind. First, the study sample consisted of relatively older, foreign-born Asian Americans due to limitations inherent in the study data, effectively excluding younger Asian Americans, and with a smaller percentage of US-born people than is found in the general US population. It is possible that US-born Asian Americans’ experiences might have been underestimated, and coefficient estimates of age might have be biased. Second, the study sample only included three ethnic groups, excluding a variety of other Asian ethnic groups. The NLAAS dataset aggregates those groups into “all other Asians;” these constitute 22% of all Asian Americans. Thus, the current study results only describe the three ethnic groups studied, and cannot be generalized to other ethnic groups not included in this study. Finally, the study findings represent victims’ experiences, not perpetrators’. Since IPV takes place between partners, data collected only from the victim has natural limitations. Furthermore, it is obvious that the risk factors for IPV identified by this study should influence both victims and perpetrators. For instance, perceived discrimination has been also identified as a risk factor for perpetrating IPV (Flores-Ortiz, 2000; Sugihara & Warner, 2002). It is possible that the relationships identified in this study, among birthplace, perceived discrimination, and IPV, may change significantly when perpetrators’ experiences are included.

Notwithstanding the limitations, this study provides implications for future research, policy, and practice. This study revealed that risk factors for IPV varied across ethnic groups, and that interactions between the factors somehow influenced IPV. While this study examined only Asian Americans, similar results have been found among other racial and ethnic groups (Kaufman Kantor et al., 1994; Waltermaurer et al., 2006). Why these interactions take place, and how they affect IPV are mostly unknown. Thus, future research needs to utilize advanced research techniques (e.g., mixed methods and latent variable analysis) to investigate and explain the intertwined relationships between risk factors and IPV, and to develop comprehensive frameworks to better explain the varying relationships across racial and ethnic groups.

Social policies and programs for immigrants have often focused on improving socioeconomic conditions (e.g., employment training and support), and on easing restrictions for immigration (e.g., allowing victims to stay in the US regardless of their immigration status). However, this study found that the risk for IPV increased for US-born Asian Americans and those with higher perceived discrimination. This suggests that some immigrants’ problems may not be “their” problems, but “ours.” Thus, social programs and services for immigrants need to address how to approach assimilation, acculturation, and discrimination in the context of IPV. For instance, employment training may help immigrants gain the economic means that can mitigate the causes of IPV, but culturally insensitive and discriminatory workplaces can increase the risk for IPV, which may cancel the benefits of employment and make immigrants more vulnerable to IPV. While the current study is far from conclusive, it extends the literature on the relationships among ethnicity, IPV, and risk factors, thereby contributing to a better understanding of these issues and leading to the development of appropriate policies and practices for better prevention and intervention of IPV among immigrants.