The Transracial Adoption Paradox

Abstract

The number of transracial adoptions in the United States, particularly international adoptions, is increasing annually. Counseling psychology as a profession, however, is a relatively silent voice in the research on and practice of transracial adoption. This article presents an overview of the history and research on transracial adoption to inform counseling psychologists of the set of racial and ethnic challenges and opportunities that transracial adoptive families face in everyday living. Particular attention is given to emergent theory and research on the cultural socialization process within these families.

The older I get, the more I realize I can’t avoid being Korean. Every time I look into the mirror, I am Korean. When I look at family pictures, I feel that I stand out. I guess it shouldn’t bother me, but sometimes it does. Even though I may seem very American ...I want to be distinctly Korean. I know I’m not in terms of having all the Korean traditions, but I don’t want people to see me and say, “Because she grew up in a Caucasian family, and because she is very Americanized, she’s white.” That’s not what I want anymore.

Janine Bishop (1996, p. 309)

The opening passage by Janine Bishop (1996), a 20-year-old Korean adoptee college student, illustrates the transracial adoption paradox that confronts racial/ethnic minority children who are adopted by White parents. Namely, adoptees are racial/ethnic minorities in society, but they are perceived and treated by others, and sometimes themselves, as if they are members of the majority culture (i.e., racially White and ethnically European) due to adoption into a White family. This set of contradictory experiences that are nevertheless true has been of particular interest to adoptive families, adoption professionals, and researchers in the United States and Europe over the past 50 years (Fanshel, 1972; McRoy & Zurcher, 1983; Simon & Altstein, 2000; Tizard, 1991).

The purpose of this article is to address some of the psychological and cultural questions raised by the transracial adoption paradox: What are the psychological consequences of growing up in a transracial adoptive family? How do the unique experiences of transracial adoptees shape racial/ethnic identity development? Do parents’ and children’s efforts to overcome racial and ethnic differences relate to psychological adjustment? A brief review of the history and controversies surrounding transracial adoption in the United States is presented and followed by a selective review of the empirical literature on transracial adoption. Drawing on the reviewed research, a cultural socialization framework is proposed to understand the psychological and cultural dynamics pertinent to transracial adoptive families. The article concludes with ways in which counseling psychology can contribute to the improvement of transracial adoption research and practice.

MODERN HISTORY OF TRANSRACIAL ADOPTION

Transracial adoption is defined as “the joining of racially different parents and children together in adoptive families” (Silverman, 1993, p. 104) and occurs through various forms of domestic adoption (e.g., foster care, private, and stepchildren in interracial marriages) and international adoption (i.e., children adopted from another country). It is considered the most visible of all forms of adoption because the physical differences between adoptive parents and adoptee are more apparent and immutable (Grotevant, Dunbar, Kohler, & Esau, 2000). In the majority of the adoptions, White parents adopt children who are considered racial/ethnic minorities in this country. These racial/ethnic differences between parents and children have led to social and political controversies and to changes in the processes of domestic and international adoption of racial/ethnic minority children (Chimezie, 1975; Simon & Altstein, 2000; Zamostny, O’Brien, Baden, & O’Leary Wiley, 2003 [this issue]).

Domestic transracial adoption

Among the earliest examples of intentional domestic transracial adoption was the Indian Adoption Project, which occurred between 1958 and 1967. The project was a collaboration between the Bureau of Indian Affairs and the Child Welfare League of America (CWLA) and was designed to remove Indian children from their families on reservations in an effort to assimilate them into mainstream society (Fanshel, 1972). By the 1960s, child advocacy groups in the United States and Canada initiated other programs to find adoptive families for orphaned African American children. These types of programs, however, were soon met with resistance from the racial/ethnic minority communities. The National Association of Black Social Workers (NABSW), for example, argued that transracial adoption was, in essence, a form of race and cultural genocide (i.e., children will not develop proper skills to survive in a racist society), and the NABSW passed a resolution in 1972 calling for an end to the transracial adoption of African American children. Native American opposition to the Indian Adoption Project on similar grounds led to its eventual dissolution with the Indian Child Welfare Act (ICWA) of 1978 (Simon & Altstein, 2000). Social service agencies and organizations, including the CWLA, responded quickly by revising their standards for adoption to a preference for same-race families. The policy change led to a sharp decrease in the number of Black-White adoptions from 2,574 in 1971 to an estimated 1,400 in 1987 (Bachrach, Adams, Sambrano, & London, 1990; Simon & Altstein, 2000). There are no reliable past or present estimates for the number of domestic transracial adoptions that are not Black-White.

Today, national surveys suggest that Whites and African Americans have mixed feelings regarding domestic transracial adoption. Using data from a CBS News public opinion poll, for example, Hollingsworth (2000) found that African American women (84%) and Caucasian/White men (72%) were less likely to approve of transracial adoption than African American men who served as the reference group in the logistic regression analyses. The 1995 National Survey of Family Growth found that among ever-married women who were considering or planning to adopt, 51% of White women preferred to adopt a White child, but 73% to 87% were willing to accept adopting a non-White child (i.e., Black or other race). Similarly, 52% of Black women preferred to adopt a Black child, but 86% to 89% were willing to accept adopting a non-Black child (i.e., White or other race). Interestingly, a minority of White (9%) and Black (12%) women preferred to adopt a non-White/non-Black child, presumably either a child of another race or from another country (Chandra, Abma, Maza, & Bachrach, 1999). Much less, if anything, is known about the attitudes and opinions of Asian Americans, Native Americans, and Hispanics regarding domestic transracial adoption.

A current public policy concern is the overrepresentation of racial/ethnic minority children in the foster care system. The U.S. Department of Health and Human Services (2001), for instance, found that African American, Hispanic, Asian American, and Native American children represented 60% (75,722 out of 127,000) of the children in foster care waiting to be adopted in 1999. To facilitate the adoption of these children in need, a series of federal legislative acts were passed in the last decade that reject the use of racial preferences in adoption among adoption agencies that receive federal assistance (viz., Multi-Ethnic Placement Act of 1994 and the Interethnic Adoption Provisions of 1996). It is now estimated that 15% of all foster care adoptions can be considered transracial adoptions or approximately 5,400 out of 36,000 in 1998, according to the National Adoption Information Clearinghouse (2003).

International adoption

International transracial adoption in the United States reflects a convergence of social and political factors at home and abroad. In particular, wars, poverty, lack of social welfare, and social upheaval in other countries have played a large part in the availability of children for overseas adoption. For example, thousands of war-orphaned Korean children and biracial children whose mothers were Korean and fathers were American military personnel were adopted shortly after the Korean War. It is estimated that there were more than 110,000 children adopted from South Korea to the United States between 1955 and 2001 (Evan B. Donaldson Adoption Institute, 2002), which is approximately 10% of the present-day Korean American population. By the 1960s and 1970s in the United States, White couples, who were usually older and infertile, began to consider international adoption as more feasible than domestic same-race adoption and less controversial than domestic transracial adoption. Today, Americans, still predominantly White, are adopting more than ever before infants and young children from more than 40 countries worldwide. Annual adoption rates, for instance, have risen dramatically from 8,102 in 1989 to 19,237 in 2001 with the majority of adoptions from Asian countries (U.S. State Department, 2001). International adoptions also account for approximately 85% of all transracial adoptions based on estimates of past and present adoption figures of nonrelated racial/ethnic minority children (Bachrach et al., 1990; Simon & Altstein, 2000; U.S. State Department, 2001).

International adoption, however, is not without controversy (Tizard, 1991). A recent public opinion survey of 1,416 people, for example, found that 47% of respondents believed international adoptees have more medical and behavioral problems than domestically adopted children (Dave Thomas Foundation for Adoption, 2002). International concerns about baby selling, kidnapping, and forced labor also have led some countries to discontinue overseas adoptions and, at other times, have led the United States to disallow adoption from specific countries. Third-world advocates similarly have argued that international adoption is a new form of colonialism and cultural imperialism that treats children as economic commodities (see Tessler, Gamache, & Liu, 1999, for a brief review). These public concerns and protests resulted in the establishment of international rules for adoption (e.g., Hague Convention on Protection of Children and Cooperation in Respect of Intercountry Adoption of 1993) and federal legislative policies (i.e., Intercountry Adoption Act and Child Citizenship Act of 2000) that make international adoption more standardized.

TRANSRACIAL ADOPTION RESEARCH

Empirical research on transracial adoption, in large measure, began as a response to the social and political controversies surrounding domestic transracial adoption in the late 1960s and 1970s. Later, the research expanded to include children adopted from other countries, as the rate of domestic adoption declined and the popularity of international adoption increased. The bulk of transracial adoption research, which emerged from these controversies and trends, occurred in the fields of social work and sociology between the 1970s and 1980s (e.g., Fanshel, 1972; Feigelman & Silverman, 1983; Grow & Shaprio, 1974; Kim, 1977; McRoy, Zurcher, Lauderdale, & Anderson, 1982; Simon & Altstein, 1977). Four integrative reviews were published in the 1990s that summarized much of this earlier research on transracial adoption. Alexander and Curtis (1996), for example, exclusively critiqued the research on African American transracial adoptees. Tizard (1991), likewise, exclusively reviewed the intercountry adoption research literature in the United States and Great Britain. Rushton and Minnis (1997) and Friedlander (1999) reviewed both domestic and international transracial adoption research that was conducted in the United States and Great Britain.

The present review of transracial adoption research focuses on empirical studies from 1990 until the present in the fields of psychology, psychiatry, social work, and sociology that address the racial and ethnic issues faced by transracial adoptees and their families. Studies initially were identified from computer searches on PsycINFO, an electronic database of materials in psychology and related fields. The search terms included all variants of transracial adoption (e.g., Black-White), international adoption (e.g., intercountry), and racial and ethnic minority groups (e.g., African American, Korean). Additional studies were located in the reference lists of the articles identified through the computer searches and through online adoption-related websites. I also included pre-1990 research studies that are considered seminal to the field of transracial adoption, as well as recent research from Europe, where there is a high prevalence of international adoption. When possible, I compared and contrasted the racial, ethnic, and psychological experiences of domestic and international transracial adoptees. In some instances, however, domestic and international transracial adoptees were aggregated together in studies and, as such, group comparisons were not possible.

Nearly all of the reviewed research on transracial adoption can be classified as descriptive field studies on either the psychological outcomes or the racial/ethnic identity development of transracial adoptees. Outcome studies focus specifically on the psychological problems and adjustment of transracial adoptees without direct consideration of racial and ethnic experiences. Racial/ethnic identity studies focus on the relationship between the racial and the ethnic experiences of transracial adoptees and identity development. More recently, efforts have been made to bring the two types of research together in the form of empirical studies on cultural socialization. Collectively, these different types of research studies attempt to answer the questions raised earlier: What are the psychological consequences of growing up in a transracial adoptive family (outcome studies)? How do the unique experiences of transracial adoptees shape racial/ethnic identity development (racial/ethnic identity studies)? Do parents’ and children’s efforts to overcome racial and ethnic differences relate to psychological adjustment (cultural socialization studies)? The appendix provides a selective summary of transracial adoption research published between 1990 and 2003.

Outcome studies

Outcome-based studies typically compare transracial adoptees with either same-race adoptees or nonadoptees on measures of psychological adjustment. An underlying assumption of the research is that the transracial adoption paradox is not a problem for transracial adoptees if there are no significant group differences on psychological adjustment (e.g., Verhulst & Versluis-den Bieman, 1995; Versluis-den Bieman, & Verhulst, 1995). Across a wide range of studies on domestic and international adoption, the research demonstrates that transracial adoption itself does not necessarily place a child at higher risk for emotional and behavioral problems. Specifically, approximately 70% to 80% of transracial adoptees had few serious behavioral and emotional problems, a rate that was comparable to same-race adopted and nonadopted children (Benson, Sharma, & Roehlkepartain, 1994; Bimmel, Juffer, van Ijzendoorn, & Bakermans-Kranenburg, 2003; Lindblad, Hjern, & Vinnerljung, 2003; Versluis-den Bieman, & Verhulst, 1995). Transracial adoptees also did not differ dramatically from same-race adoptees and nonadoptees in levels of self-esteem and social adjustment (Bagley, 1993a; Benson et al., 1994). In studies where transracial adoptees had more serious and long-term behavioral and emotional problems, researchers found the effect sizes to be small and also identified mitigating factors, such as birth country of origin, age at adoption, gender (with boys at greater risk), adverse preadoption experiences, and adoptive family functioning (Benson et al., 1994; Bimmel et al., 2003; Cederblad, Hook, Irhammar, & Mercke, 1999; Fanshel, 1972; Lindblad et al., 2003; Sharma, McGue, & Benson, 1996; Verhulst, Althaus, & Verluis-den Bieman, 1990a, 1990b).

A persistent problem in this type of outcome research, however, is the failure to directly measure the racial and ethnic experiences of adoptees and its potential contribution to psychological adjustment. Hjern, Lindblad, and Vinnerljung (2002), for example, conducted one of the most comprehensive epidemiological studies to date on the outcome of transracial international adoptees. Using Swedish national registry data for 11,320 adoptees, 2,343 Swedish-born siblings, 4,006 immigrant children, and a general population of 853,419 Swedish-born residents (all born between 1970 and 1979), they found that the vast majority of adoptees had no serious psychiatric problems (e.g., hospitalizations, suicide attempts) and social maladjustment problems (e.g., substance abuse, criminal offenses). Adoptees, however, were at least 2 to 3 times more likely to have serious psychiatric and social maladjustment problems than their siblings and the general population, but adoptees were not much more likely to have problems than immigrants in Sweden. This latter finding suggests that race and discrimination may have played a role in the overall adjustment of adoptees and immigrants alike. Unfortunately, it was not possible to determine the actual effects of race and discrimination, because these experiences were not measured. In addition, the findings may not necessarily generalize to the United States, which is less racially homogenous than Sweden.

Racial/ethnic identity studies

Racial/ethnic identity studies typically examine the extent to which transracial adoptees use racial/ethnic self-descriptors and are proud or comfortable with their race and ethnicity. An underlying assumption of the research is that the manner in which adoptees negotiate the transracial adoption paradox is best evinced in their racial/ethnic identity development and that transracial adoptees with positive and secure racial/ethnic identities will be psychologically well adjusted. Unfortunately, as the following review reveals, the actual relationship between the racial and the ethnic experiences of transracial adoptees and their psychological adjustments is not directly addressed in the studies. In addition, none of the studies used reliable and valid measures of racial/ethnic identity. Instead, researchers relied on projective measures of racial preference, ad hoc selfreport items, open-ended questions, and on occasion, parent reports of the children’s interest and involvement with the birth culture. In the context of these limitations, the research suggests that transracial adoptees exhibit a great deal of variability in their racial/ethnic identities.

Hollingsworth (1997), for example, conducted a meta-analysis on six cross-sectional and longitudinal studies that compared the racial/ethnic identity of domestic transracial adoptees with non-White, same-race adoptees. Racial/ethnic identity was measured in a variety of ways in these studies (e.g., Twenty-Statements Test, Clark Doll Study), but they all assessed the extent to which children used racial/ethnic self-descriptors. She found that transracial adoptees had significantly lower racial/ethnic identities than same-race adoptees (d = -.52). This lower racial/ethnic identity finding corresponds with other research showing that domestic and international transracial adoptees were highly acculturated to the majority culture (Andujo, 1988; Bagley, 1993a, 1993b; Kim, 1977; Vroegh, 1997; Wickes & Slate, 1996).

Other studies on domestic and international transracial adoptees present more mixed results regarding racial/ethnic identity development (Huh & Reid, 2000; Simon & Altstein, 2000; Westhues & Cohen, 1998). To illustrate this point, Brooks and Barth (1999) found that all 244 Asian and African American transracial adult adoptees in their study could be classified as having secure or strong racial/ethnic identities, but roughly half of them still reported discomfort over their racial appearances. Similarly, in a national survey of 289 domestic and international transracial adolescent adoptees, Benson et al. (1994) found that almost half (41%) reported frequent feelings of racial pride, some (20%) wished they were a different race, and very few (3%) felt frequently ashamed or embarrassed of their racial backgrounds. Unfortunately, it is not known whether the findings are similar or different from racial/ethnic minorities who are not adopted.

Only two studies were identified that separately measured racial identity and ethnic identity. In a Canadian study of 155 international, transracial, adolescent, and young adult adoptees, Westues and Cohen (1998) found that men were more likely to identify ethnically as Canadian/Quebecois than members of their birth cultures (51% vs. 35%), whereas women were equally likely to identify as Canadian/Quebecois and as members of their birth cultures (40% vs. 43%). By contrast, a majority of the sample (79% males, 73% females) racially self-identified as members of their races (e.g., “Oriental [sic],” “brown [sic],” “black [sic]”) and about 1 in 10 (11% males, 17% females) racially self-identified as White. In a small, convenience sample of domestic and international transracial adult adoptees in the United States, Baden (2002) found that ethnic identity and racial identity, using modified scales derived from Phinney’s (1992) Multigroup Ethnic Identity Measure, were highly correlated (r = .80), which suggests that some transracial adoptees may construe the two concepts of identity as synonymous.

Some researchers found that the variability in racial/ethnic identity development for transracial adoptees may be attributable to different extrinsic factors, such as age at adoption, race, and geography. Transracial adoptees placed at a later age, for example, identified more strongly with their ethnicities and races than did adoptees placed at a younger age (Wickes & Slate, 1996). African American and Hispanic transracial adoptees appeared to have greater pride and comfort in their ethnicities and races than did Asian transracial adoptees (Benson et al., 1994; Brooks & Barth, 1999; Feigelman & Silverman, 1983). Racial/ethnic identity also appeared to be weaker among transracial adoptees living in more racially homogenous (i.e., White) communities (Cederblad et al., 1999; DeBerry, Scarr, & Weinberg, 1996).

Racial/ethnic identity development for transracial adoptees also may vary according to the social and emotional development of the children. Transracially adopted younger children, for example, tend to identify racially or ethnically with their birth cultures, as measured by racial/ethnic preference (DeBerry et al., 1996; Huh & Reid, 2000; Simon & Altstein, 2000). By adolescence and adulthood, however, some transracial adoptees’ sense of race and ethnicity diminishes (DeBerry et al., 1996) or becomes more ambivalent (Brooks & Barth, 1999; Benson et al., 1994). For other transracial adoptees, race and ethnicity become more salient as the adoptees enter adulthood. In a survey of 163 Korean adult adoptees, Freundlich and Lieberthal (2000) found that, while growing up, 36% described themselves as Caucasian and 42% described themselves as ethnically Korean/Asian. As adults, only 11% of the Korean adoptees described themselves as Caucasian and 78% described themselves as Korean/Asian. Research also suggests that greater awareness of prejudice and discrimination during adolescence and adulthood may lead transracial adoptees to experience a range of conflicting emotions regarding notions of race and ethnicity (e.g., denial, shame, pride, discomfort) (Freundlich & Lieberthal, 2000).

Cultural socialization outcome studies

Cultural socialization outcome research is an emerging area of study that attempts to understand the racial and ethnic experiences of transracial adoptees and their families that promote or hinder racial/ethnic identity development and to examine directly the relationship between these experiences and psychological adjustment. An underlying assumption of the research is that healthy psychological development is contingent on positive racial and ethnic experiences. Cultural socialization outcome research represents a bridge between outcome studies and racial/ethnic identity studies and is a more appropriate methodology to examine how adoptees and families approach and overcome the psychological and cultural challenges related to transracial adoption.

Using an epidemiological sample of 211 internationally adopted adolescents and young adults in Sweden, Cederblad et al. (1999) found that negative racial and ethnic experiences, as measured by perceived discrimination and ambivalence in one’s ethnic identity, were related positively to behavioral problems, emotional distress, and lower self-esteem above and beyond the effects of family functioning, family structure, and friendship support. This subsample was drawn from the same age cohort group that served as the basis for the larger epidemiological study conducted by Hjern et al. (2002), which suggested that race and discrimination may explain the higher rates of psychiatric and social maladjustment among transracial adoptees and immigrants. The two studies together provide preliminary evidence that negative racial and ethnic experiences can have serious psychological consequences for transracial adoptees and their families.

There also is emerging evidence that positive racial and ethnic experiences contribute to the psychological adjustment of transracial adoptees. DeBerry et al. (1996) and Yoon (2001), for example, found that racial/ethnic identity, as measured principally by ethnic pride, was related positively to psychological adjustment in studies of African American and Korean transracial adolescent adoptees in the United States. Yoon also found that Korean adolescent adoptees, whose adoptive parents actively promoted their children’s ethnic cultures, had more positive racial/ethnic identity development and, in turn, better psychological adjustment. Moreover, the direct and indirect effects of these racial/ethnic experiences were above and beyond the effects of general family functioning as measured by parental warmth and positive parent-child communication.

In a longitudinal study of 88 African American transracial adoptees, DeBerry et al. (1996) found that nearly half of all adoptive parents were likely to encourage biculturalism in the upbringing of their children during childhood, but adoptive parents were more likely to deny and deemphasize race and have ambivalent feelings about cultural socialization when their children reached adolescence. These changes in culture-specific parenting corresponded with a decrease in the adoptees’ identification with African American culture between childhood and adolescence, but the direction of influence between changes in parenting and identity development is not known. DeBerry et al., however, did find that African American adoptees, whose parents actively promoted their children’s races (e.g., learn about African American heritage), had more positive racial/ethnic identity development and, in turn, more positive adjustment.

Aside from these few published studies, there is little empirical evidence that directly examines the link between specific aspects of cultural socialization and psychological adjustment in transracial adoptees. There is, however, a limited but growing body of theory and research that focuses exclusively on the cultural socialization process itself in transracial adoptive families. This latter literature sheds critical insight into how families currently attempt to address various aspects of the transracial adoption paradox.

CULTURAL SOCIALIZATION PROCESS RESEARCH

Cultural socialization is a lifelong developmental process that enables individuals and families to have greater adaptability and competence in a given cultural milieu (Harrison, Wilson, Pine, Chan, & Buriel, 1990). For racial/ethnic minorities, it specifically entails the transmission of cultural values, beliefs, customs, and behaviors from parents, family, friends, and community to children that foster racial/ethnic identity development, equip children with coping strategies to deal with racism and discrimination, and encourage prosocial behavior and appropriate participation in society. An underlying assumption of the research is that parents and others play an active and integral role in the psychological development and cultural competence of children.

For transracial adoptive families, the process of cultural socialization is complicated by the apparent and immutable racial and ethnic differences between parents and children that form the basis of the transracial adoption paradox. In addition, White adoptive parents are less likely to have the first-hand knowledge and experience to teach their children about life as racial/ethnic minorities in society. Given these differences, traditional views of cultural socialization have been modified to take into account the unique racial and ethnic dynamics within transracial adoptive families. Specifically, four cultural socialization strategies of relevance to transracial adoptive families are summarized, along with the limited amount of available empirical research on each of them. The strategies characterize typical ways in which transracial adoptive families (specifically, parents and children) might approach the transracial adoption paradox in the lives of adoptees and in the family.

Cultural assimilation

Early studies on domestic transracial adoption found that most parents were likely to engage in parenting behaviors that rejected differences or downplayed the unique racial and ethnic experiences of children (Andujo, 1988; DeBerry et al., 1996; McRoy & Zurcher, 1983). These types of families engaged in the practice of cultural assimilation or the acculturation of the child into the majority culture (Gordon, 1964). Cultural assimilation typically occurs with minimal parental effort because children are immediately and constantly exposed to the majority culture. A variant of cultural assimilation is a humanistic strategy that emphasizes a “colorblind” orientation or a view of humanity without reference to ethnicity and race. In some cases, the child’s ethnicity and race are intentionally denied or deemphasized. It is argued that transracial adoptees exposed to these forms of cultural assimilation are more likely to internalize their adoptive parents’ cultural worldview and identify more strongly with the majority culture than with their ethnic cultures (McRoy & Zurcher, 1983).

Enculturation

Current research suggests that a growing number of White adoptive parents acknowledge differences within the family and specifically promote the enculturation of their children. That is, they make a concerted effort to teach their children about their birth cultures and heritages (Carstens & Julia, 2000; Friedlander, Larney, Skau, Hotaling, Cutting, & Schwam, 2000; Johnson, Shireman, & Watson, 1987; Rojewski & Rojewski, 2001; Tessler et al., 1999; Vonk & Angaran, 2001). Adoptive parents with a belief in enculturation typically provide their children with educational, social, and cultural opportunities to instill ethnic awareness, knowledge, pride, values, and behaviors, as well as to promote a positive ethnic identity. Feigelman and Silverman (1983) and Huh and Reid (2000), for example, reported that international transracial adoptees were more likely to show racial pride when adoptive parents emphasized the children’s racial backgrounds, encouraged ethnic participation, lived in racially integrated communities, and were highly involved in these activities along with their children. Sometimes, however, adolescent transracial adoptees may resist parental efforts at enculturation because they are seeking to belong with their predominantly White peers (Freundlich & Lieberthal, 2000). At other times, adoptees may choose to develop a bicultural or more fluid identity that allows them greater role flexibility across cultural milieus.

Racial inculcation

There is limited empirical research on the extent to which transracial adoptive parents engage in racial inculcation or the teaching of coping skills to help children deal effectively with racism and discrimination. Past descriptive studies have found that transracial adoptive parents varied a great deal in the ways in which they handled racial issues. Parents may downplay racist comments, make derogatory comments about racists, and in fewer cases, take a more active role in the community to promote social justice (Andujo, 1988; Friedlander et al., 2000; Johnson et al., 1987; Scroggs & Heitfeld, 2000). Other parents believe that enculturation is an appropriate, albeit indirect, approach to prepare their children for racism and discrimination (Steinberg & Hall, 2000). Westues and Cohen (1998), for example, found that parents were slightly more likely to read books and attend cultural events that focused on the promotion of a positive ethnic identity than a positive racial identity. In a recent epidemiological study on the adoptive parents of 590 Korean-born children, Lee, Yoo, Weintraub, and Su (2002) found that the majority of parents were racially aware, believed in racial inculcation and enculturation, and spoke with their children about racism and discrimination in school. In addition, they found that parental belief in racial inculcation best predicted whether parents actually talked with their children about racism and discrimination in school. Parental belief in enculturation, however, did not predict whether parents talked with their children about racism and discrimination in school. Surprisingly, there are no known published, empirical studies on how transracial adoptees personally cope with racism and discrimination.

Child choice

Tessler et al. (1999) commented in his study on adoptive parents of 391 children from China that a fourth socialization strategy—child choice—has emerged in recent years. Parents with a belief in child choice initially provide their children with cultural opportunities, but they become more ambivalent about such efforts and adjust their socialization efforts according to the children’s interests and wishes. Unfortunately, Tessler et al. provided only a few anecdotal examples of child choice parenting and did not present additional empirical support for this strategy. DeBerry et al.’s (1996) longitudinal study of African American transracial adoptees provides some evidence for child choice parenting. In the study, many adoptive parents became more ambivalent about engaging in cultural socialization when their children entered adolescence, possibly because the children became less interested or the parents became more uncomfortable. It is not known if the ambivalent parents would respond appropriately to any increased cultural interests from their children. It also is interesting to note that the child choice strategy shifts parenting responsibilities away from the parents and places the burden on the child to determine how to best be raised. In such instances, children may be sensitive to their parents’ ambivalent feelings toward cultural socialization and may suppress interest in their racial and ethnic cultures to maintain family harmony.

Summary

These cultural socialization strategies are not exhaustive, and it is plausible that transracial adoptive families engage in a variety of other strategies to resolve the transracial adoption paradox. Moreover, these strategies are not necessarily mutually exclusive. In some instances, parents may downplay the salience of a child’s race and ethnicity (i.e., cultural assimilation), but in other instances, parents may attempt to teach a child how to protect himself or herself (i.e., racial inoculation). Likewise, a transracial adoptee child may express strong interest in the birth culture one minute (i.e., enculturation) but deny racial and ethnic differences the next minute. Clearly, counseling psychologists must consider the various factors that affect the extent to which transracial adoptive families engage in any of the cultural socialization strategies.

RESEARCH LIMITATIONS AND CONSIDERATIONS

Transracial adoption research attempts to understand the psychological and the cultural issues faced by adoptees who are racial/ethnic minorities in society but are perceived and treated by others as if they are members of the majority culture (i.e., racially White and ethnically European) due to adoption into a White family. As reviewed previously, a number of theoretical and methodological gains have been made in the past decade to help answer the various questions that arise from the transracial adoption paradox. In this section, I highlight some current research limitations and propose a few theoretical and methodological considerations for future research.

Transracial adoption theory

In the majority of transracial adoption research, including current research on the cultural socialization process, there is a lack of a formal theory specific to transracial adoptive families. Benson et al. (1994) attempted to address this issue in their study of same-race and transracial adolescent adoptees and their adoptive families by adapting the shared fate theory of adoption, developed by Kirk (1964) and elaborated on by Brodzinsky (1990), to explain the research findings. According to shared fate theory, appropriate acceptance of differences inherent in adoptive families creates a mutual bond (or “shared fate”) between parent and child that contributes to healthy and adaptive family functioning and psychological adjustment. In their study, Benson et al. found that the majority (64%-69%) of transracial adoptees agreed or strongly agreed that their parents accepted the differences between adoptive parent and child. Far fewer transracial adoptees (6%-24%) reported that their parents either denied or overinsisted that their child was different. Interestingly, transracial adoptees also reported greater acceptance of differences by their parents than did same-race adoptees. Furthermore, Benson et al. found that parents’ acceptance of differences, as perceived by same-race and transracial adoptees, correlated positively with a variety of adoptee mental health indicators (rs = .30 to .35). By contrast, perceptions of parents’ denial of or overinsistence on differences correlated negatively with mental health (rs = -.20 to -.35). Unfortunately, Benson et al. did not define the source of these differences between parents and children (e.g., race, adoption), so it is uncertain if children felt that their parents specifically accepted the racial and ethnic differences.

The Benson et al. (1994) study demonstrates how a traditional theory of adoption can serve as a meta-theory to guide transracial adoption research and, specifically, to understand the role of cultural socialization in addressing and resolving the transracial adoption paradox. Although they did not study cultural socialization directly, it is possible to generate research hypotheses from this adaptation of shared fate theory. Adoptive parents who deny or overlook racial and ethnic differences between parents and child, for example, may be more likely to engage in cultural assimilation parenting strategies, which in turn, may contribute to poorer mental health. By contrast, adoptive parents who acknowledge and accept racial and ethnic differences may be more likely to engage in enculturation and racial inoculation parenting strategies, which in turn, may contribute to more positive racial/ethnic identity development and mental health.

Comparison studies

An assumption made throughout the reviewed studies is that the transracial adoption paradox uniquely affects the psychological adjustment of adoptees. Yet only a few researchers compared transracial adoptees with same-race adoptees of the same racial/ethnic background, and the studies were conducted exclusively with Black transracial and same-race adoptees (e.g., McRoy et al., 1982). In the case of Asian transracial adoptees, who constitute the largest population of transracial adoptees, it would be extremely difficult to identify many Asian American same-race adoptive families because of the persistent stigma of adoption in Asian immigrant communities. An alternative research design might be to compare the racial and ethnic experiences of transracial adoptees with nonadopted individuals of the same racial/ethnic background. This latter type of group comparison would help determine the extent to which racial and ethnic challenges faced by transracial adoptees are unique to them or are shared by nonadopted racial/ethnic minorities.

Researchers also typically failed to distinguish between the unique racial and ethnic experiences of domestic and international transracial adoptees. In some studies (e.g., Baden, 2002; Brooks & Barth, 1999), domestic and international transracial adoptees were inappropriately aggregated together, thereby obfuscating potential between-group and within-group variations in the experiences of different racial and ethnic groups. In other studies (e.g., Benson et al., 1994), comparisons were made between domestic and international adoptees without specific theories and hypotheses to explain similarities and differences in experiences. Moreover, race and ethnicity often were treated as synonyms in research studies. However, it is quite possible that some transracial adoptees may identify more strongly with their races and other transracial adoptees may identify more strongly with their ethnicities.

Adoptees as agents of change

It is interesting to note that there remains a tendency to portray racial/ethnic minority adoptees as passive recipients of the transracial adoption paradox. For example, most studies inquired about adoptees’ racial/ethnic self-designations, subjective comfort with their races and ethnicities, or perceptions of discrimination and racism. Far fewer studies examined the ways in which transracial adoptees negotiated and resolved racial and ethnic differences (Meier, 1999). However, the portrayal of racial/ethnic minorities as passive recipients of racial and ethnic experiences is not unique to transracial adoption research. Similar criticisms can be levied against a great deal of racial and ethnic minority research in the field of psychology. Researchers must do a better job at understanding transracial adoptees as active agents of change in their lives. For example, there is a need to study how adoptees personally negotiate their identities and sense of place in society (Grotevant et al., 2000; Meier, 1999). Similarly, it is important tostudy ways in which transracial adoptees and parents manage to cope with discrimination and racism in a healthy and adaptive manner (Freundlich & Lieberthal, 2000; Hjern et al., 2002).

Methodological improvements

Reliable and valid measurement of racial and ethnic experiences remains a persistent problem in all the studies. For example, most studies on racial and ethnic identity in transracial adoption relied on projective measures of racial preference, ad hoc self-report items, open-ended questions, and parent reports. The lack of precise measurement of racial and ethnic identity prevents researchers from answering the basic question on the salience of these forms of identity for transracial adoptees. In some instances, it may be feasible to modify and adapt existing identity measures, such as Phinney’s (1992) Multi-Group Ethnic Identity Measure, which was used in Baden’s (2002) and Yoon’s (2001) studies of transracial adolescent and adult adoptees.

There is a similar need to develop reliable and valid self-report measures and observational methods to assess the different types and qualities of cultural socialization, such as implicit-explicit attitudes and behaviors, proximal-distal parenting practices, and socialization strategies. Traditional self-report measures of acculturation and enculturation are not applicable to transracial adoptive families, because adoptive parents are not members of the racial/ethnic groups of their children. A question that asks if the ethnic specific language is spoken in the home, for example, does not have face validity for transracial adoptive families. Recently developed but still to be validated instruments may prove useful in tapping into certain aspects of cultural socialization, such as the Transracial Adoptive Parenting Scale (Vonk, 2001; Vonk & Angaran, 2001), which was developed to measure cultural socialization parenting beliefs and behaviors.

Finally, it is important to improve on the methodology of cultural socialization research. Previous reviews of the empirical literature on transracial adoption consistently identified the importance of better sampling procedures, more sophisticated research designs, and multivariate statistical analyses (Alexander & Curtis, 1996; Hollingsworth, 1997; Rushton & Minnis, 1997). A major limitation of transracial adoption research is the overreliance on small, convenience samples drawn primarily from adoption agencies and organizations (Waldfogel, 2000). The use of nonrepresentative samples makes it difficult to generalize findings. An epidemiological study across a wide range of ages, using multiple informants (e.g., parents and children), would provide valuable baseline data on the adjustment of transracial adoptees in the United States. The majority of research studies also were cross-sectional and often reported only descriptive statistics. In cases where there were comparison groups of same-race adoptions or nonadopted popu-lations, researchers did not stratify samples on relevant demographic and adoption variables and did not address cohort and contextual effects. A handful of studies employed prospective or longitudinal designs, but these studies used small convenience samples and had high attrition rates that were not always examined to a sufficient degree. Moreover, the research failed to take advantage of multivariate statistical analyses that could control for confounding variables and could account for the complexity of relationships involved in transracial adoption.

PRACTICE CONSIDERATIONS

As with transracial adoption research, counseling psychologists have been a relatively silent voice in transracial adoption practice, advocacy, and outreach—despite the substantial rise in international transracial adoption over the last decade. To assure cultural competency in the delivery of mental health services to transracial adoptees and their families, counseling practitioners must start with an examination of their own attitudes and beliefs about transracial adoption, followed by an understanding of the history and controversy surrounding transracial adoption in this country (Vonk, 2001). In the process, it is critical that practitioners dispel personal myths and stereotypes about transracial adoption that can impair the ability to work effectively with individuals and families. Specifically, practitioners must understand the psychological and cultural issues surrounding the transracial adoption paradox, including the psychological consequences of growing up in a transracial adoptive family, the unique experiences of transracial adoptees that shape racial/ethnic identity development, and the parents’ and children’s efforts to overcome the transracial adoptive paradox.

For example, practitioners must be careful to not essentialize group differences among transracial adoptees, same-race adoptees, and nonadopted racial/ethnic minorities. As evidenced by the reviewed research, there remains a significant amount of variability in the psychological adjustment and racial/ethnic identity development of transracial adoptees. It also is important that racial and ethnic experiences of transracial adoptees and families be differentiated when possible. For example, some transracial adoptees may be more likely to identify as racial minorities than members of a specific ethnic group, whereas other adoptees may perceive race and ethnicity as synonyms. For adoptive parents, some may emphasize enculturation parenting strategies but discount experiences of discrimination and racism. Other adoptive parents may maintain a more colorblind racial attitude and endorse cultural assimilation parenting strategies.

Perhaps most important, practitioners should view transracial adoptees and families as active agents of change in their personal and family lives (Steinberg & Hall, 2000)—that is, transracial adoptees and adoptive parents likely engage in a variety of cultural socialization strategies to manage the complexities of the transracial adoption paradox. At times, however, transracial adoptive family members (i.e., parents and children) may differ in their levels of racial and ethnic awareness, as well as in their expectations regarding cultural socialization, and these differences can lead to intense family conflicts that alienate children from the adoptive family (Freundlich & Lieberthal, 2000). Moreover, as transracial adoptees enter adulthood, they may find themselves facing new racial and ethnic challenges that they are not able to manage effectively on their own (e.g., discrimination and racism, interracial or same-race dating). In these examples, culturally competent practitioners can serve as cultural brokers who help transracial adoptees and adoptive parents identify and resolve these differences. Practitioners also can assist in the cultural socialization process by identifying appropriate resources in the adoptive and racial/ethnic minority communities that will enable adoptees and families to receive appropriate information and support.

CONCLUSION

The purpose of this selective, integrative review was to address some of the psychological and cultural questions raised by the transracial adoption paradox: What are the psychological consequences of growing up in a transracial adoptive family? How does the unique experiences of transracial adoptees shape racial/ethnic identity development? Do parents’ and children’s efforts to overcome racial and ethnic differences relate to psychological adjustment? The current research literature on psychological outcome, racial/ethnic identity development, and cultural socialization suggests that transracial adoptees—both domestic and international—are psychologically well adjusted, exhibit variability in their racial/ethnic identity development, and along with their parents, engage in a variety of cultural socialization strategies to overcome the transracial adoption paradox. More theory and research, however, are necessary to unravel the specific factors that affect cultural socialization, racial/ethnic identity development, and psychological adjustment, as well as to inform practitioners on how to facilitate the developmental challenges that transracial adoptees and families will encounter in life.

APPENDIX Selective Review of Published Transracial Adoption Research (1990 to 2003)

Author and Date	Sample	Method and Measures	Relevant Findings
Baden (2002)	Included 51 (38 F/13 M) transracial adult adoptees; Mean age of 24 years old; Races include 30 Asian, 6 Black, 1 Latino, 1 White, and 13 Multiracial/Other; Average age at adoption was 22 months.	Cross-sectional survey study; Convenience sample using multiple recruitment methods; Standardized measures included Multigroup Ethnic Identity Measure (modified) and Brief Symptom Inventory.	Adoptees identified more with parents’ culture (M = 3.37) than ethnic culture (M = 2.11). Adoptees’ ethnic identity and racial identity were highly correlated (r = .80). Adoptees’ identification with parents’ culture correlated with less distress (-.24).
Bagley (1993a)	Included 44 Chinese female adoptees in Great Britain; Age between 22 and 28 years old; Age at adoption unknown; Compared with 88 Canadian, female, nonadopted Whites of similar age.	Cross-sectional, comparison survey study; Convenience sample drawn from one orphanage in Hong Kong; Qualitative interviews and standardized measures included Middlesex Hospital Questionnaire, Coopersmith Self-Esteem, Weinreich’ Identity, Ego Identity Scale, and Self-Image Questionnaire; Ad hoc questions on ethnic identity and adoption experience.	Twenty year follow-up study to earlier research on 53 Chinese adoptees (see Bagley & Young, 1981) but not a longitudinal design. Limited statistics were presented in publication. No significant group differences between adoptees and nonadoptees on distress, self-esteem, and self-concept. All self-identified as English and half had strong interest in Chinese culture. Reported positive adoption experiences and no experiences with discrimination and ethnic-identity confusion.
Bagley (1993b)	Included 27 (13 F/14 M) Black or mixed-race transracial adoptees; Mean age of 19 years old; Age at adoption unknown; Compared with 25 (13 W/12 M) White same-race adoptees of similar age and social class.	Longitudinal, comparison survey study; Sample drawn from southern England; Standardized measures included Middlesex Hospital Questionnaire, Coopersmith Self-Esteem, Shostrom’s Personal Orientation, Cattell’s Self-Sentiment Measure; Ad hoc questions on ethnic identity and adoption experience.	Twelve year follow-up study to earlier research on 114 adopted and nonadopted children (including 30 Black or mixed-race transracial adoptees) (see Bagley, Verma, Mallick, & Young, 1979). No longitudinal effects were detected on the outcome measures using variety of predictors from 1979 study. No significant group differences between transracial adoptees and White adoptees on distress, self-esteem, and self-concept. Majority (63%) of transracial adoptees reported best friend is White (vs. 84% for White adoptees). Transracial adoptees reported positive adoption experiences and no experiences with discrimination and ethnic-identity confusion.
Benson, Sharma, & Roehlkepartain (1994)	Included 715 families with 881 (56% F/44% M) adolescent adoptees; Mean age of 14.80 years old (between 12 and 18); Races include 199 Asian, 27 Black, 39 Hispanic, 24 Native American, and 579 White; Children adopted before 15 months old; Compared with 78 (58% F/42% M) nonadopted siblings and large samples from other studies.	Cross-sectional comparison survey sample; Representative sample drawn from 42 Midwest adoption agencies; Standardized measures included Colorado Self-Report Measure of Family Functioning, Bell Global Psychopathology Scale, Youth Self-Report, and Profiles of Student Life, as well as new measures of ethnic identity and adoption experience.	No significant group differences between adoptees, nonadopted siblings and nonadoptees on well-being, high risk behavior, and mental health. Small significant difference on internalizing and externalizing with adoptees slightly more likely to fall in clinical range compared to national data (15% vs. 11%) but likely mitigated by other factors. Of transracial adoptees, 37% reported race made growing up difficult. Transracial adoptees are significantly more likely to report parents accept child differences than same-race adoptees (64%-70% vs. 59%-61%). Of transracial adoptees, 20% wished they were a different race; 41% reported frequent feelings of racial pride.
Bimmel, Juffer, van Ijzendoorn, & Bakermans-Kranenburg (2003)	Included 2,317 internationally adopted adolescents; Age between 12 and 19 years old; Adopted from primarily Korea, South/Southeast Asia, and South America; Age at adoption between 4 and 50 months; Compared with 14,345 nonadopted adolescents.	Meta-analysis of 10 published studies from United States and Europe; Standardized measures included either Child Behavior Checklist, Symptom Checklist, Youth Self-Report, or 90-item health and risk behavior questionnaire.	Internationally adopted adolescents had more externalizing behavior problems than nonadopted adolescents (d = .11). Adopted girls having more total behavior problems than nonadopted girls (d = .10), but differences were small. Effect sizes were larger when looking at differences in clinical cut-off scores than mean score differences. Group differences disappeared when large sample from Versluis-den Bieman & Verhulst (1995) study was excluded. Age at adoption, convenience versus random sampling, and attrition rate significantly predicted effect size differences in internalizing and externalizing behavior problems.
Brooks & Barth (1999)	Included 224 parents of transracial and same race adoptees (128 F/96 M); Mean age of 24.7 years old; Races included 39 Black, 144 Asian, and 41 White; Age at adoption unknown.	Cross-sectional, comparison survey study; Convenience sample drawn from national and local adoption organizations; Standardized measure included Global Assessment Scale; Ad hoc questions on problem behaviors, adjustment, ethnic identity, and discrimination.	Sixteen year follow-up study to earlier research on transracial adoptees (see Feigelman & Silverman, 1984; Silverman, 1980), but not a longitudinal design; Original study had 737 families, but this study recruited only 224 families. Black and White male adoptees had more school and behavioral problems than Asian adoptees, but no group differences on global assessment scale. About 50% of Asian and Black adoptees reported discomfort over racial appearance; exception was Black female adoptees (21%). Black adoptees reported greater pride in race (58%-67% vs. 34%-35%) and more racist comments (53%-63% vs. 24%-31%) than Asian adoptees. All transracial adoptees were classified as having secure ethnic identities.
Cederblad, Hook, Irhammer, & Mercke (1999)	Included 147 parents and 211 internationally adoptees; Age between 13 and 27 years old; Adopted from primarily India, Thailand, and Chile; Age at adoption varied (54% before 7 months old and 30% after 1 year old); Compared with 647 nonadopted Swedish adolescents and young adults.	Cross-sectional, epidemiological survey study; Representative sample drawn from main adoption agency in Sweden; Standardized measure included Child Behavior Checklist, Symptom Checklist, Family Relations Scale, Family Climate Inventory; Ad hoc questions on ethnic identity and discrimination.	No significant group differences between adoptees and nonadoptees on problem behavior and psychological distress, except obsessive-compulsive symptoms (adoptees had higher scores). Adoptees had higher self-esteem than nonadoptees. Pre-adoption experiences (i.e., length of time in orphanage and foster home) had significant effect on behavior problems and psychological distress. Ethnic identity confusion, perceived discrimination, and family dysfunction had significant effect on behavior problems and psychological distress.
DeBerry, scarr, & weinberg (1996)	Included 88 (59 F/29 M) matched cases of White parent-African American adoptee; Average age of 17 years old; Average age at adoption was 14 months.	Longitudinal study; Sample drawn from local adoption agency and Minnesota state records on adoption; Data drawn from structured interviews that were coded for content on racial socialization and ecological competence.	Ten year follow-up study to earlier research on transracial adoptees (see Scarr & Weinberg, 1976); Original study had 101 families with 176 transracial adoptees (average age of 7 years old), but this follow-up study focused on subsample of 130 African American adoptees. There were five forms of racial socialization: denial/deemphasis, ambivalent, bicultural, multicultural, and overenthused/overzealous. At Time 1, 42% of families emphasized bicultural; At Time 2, only 20% of parents emphasized bicultural and majority emphasized denial (35%) or ambivalence (40%); At Time 2, adoptees reported that 55% of parents emphasized denial and 30% emphasized ambivalence. At Time 1, according to parents, adoptees’ Africentric orientation (M = 3.42) was nearly equal to Eurocentric orientation (M = 4.26); At Time 2, adoptees’ Africentric orientation decreased (M = 1.75) and Eurocentric orientation increased (M = 5.41); Adoptee self-report data at Time 2 is similar. Path analysis found that racial socialization did not predict Africentric or Eurocentric orientation at Time 1, but it did predict orientation at Time 2 and orientation, in turn, predicted adjustment; Interestingly, adjustment also declined over the 10 years.
Feigelman (2000)	Included 240 parents of transracial and same race adoptees; Mean age of 23 years old; Races included 33 Black, 151 Asian, 19 Latino, and 37 White; Age at adoption unknown.	Drawn from same data set as Brooks & Barth (1999); Standardized measure included Global Assessment Scale; Ad hoc questions on problem behaviors, adjustment, ethnic identity, and discrimination.	See Brooks & Barth (1999), Feigelman & Silverman (1984), and Silverman (1980) for past findings. No significant group differences by race were found on problem behaviors and adjustment problems. For transracial adoptees, discrimination and other negative racial experiences was significantly related to problem behaviors and adjustment problems. Transracial adoptees who lived in White only communities were more likely than adoptees living in racially mixed communities to have discomfort with their racial appearances (51% vs. 25%).
Freundlich & Lieberthal (2000)	Included 167 (82% F/18% M) Korean adult adoptees; Mean age of 31 years old; Median age at adoption was 2 years old.	Cross-sectional, descriptive survey study; Convenience sample drawn from national adoption conference; Open-ended questions on adoption, ethnic identity, and discrimination experiences.	No statistical analyses were conducted on the data. Only descriptive and qualitative descriptors were provided. Thirty six percent identified as Caucasian and only 42% identified as Korean/Asian while growing up; By contrast, 78% identified as Korean/Asian as adults and only 11% identified as Caucasian. Adoptees reported that discrimination was based more on race (70%) than adoption (28%). 54% of adoptees explored ethnic heritage growing up, but only 33% reported parents helped them establish a positive ethnic identity.
Hjern, Lindblad, & Vinnerljung (2002)	Included 11,320 (6,984 F/4,336 M) international adoptees; Average age between 16 and 25 years old (born between 1970 and 1979); Adopted from primarily Korea, India, and Colombia; 74% adopted at age 0 to 1 years, 16% at age 2 to 3 years, and 9% at age 4 to 6 years; Compared with 2,343 nonadopted siblings, 4,006 non-European immigrants, and 853,419 native-born Swedes.	Cross-sectional, epidemiological study; Representative sample drawn from national registry data in Sweden; Outcome variables included suicide death, suicide attempt, psychiatric hospitalization, alcohol and drug abuse, criminal offense, and imprisonment.	Five percent of adoptees had at least one psychiatric and social problems, compared to 2% of native-born Swedes. Using logistic regression analyses, adoptees had 2 to 4 times higher rates of psychiatric and social problems than nonadopted siblings and natives but similar rates as immigrants except suicide (4 times higher among adoptees); These odds-ratios were constant after controlling for sex, age of child and mother, social class, housing, and parents’ psychiatric status. Among adoptees, adoption at 4 to 6 years age and birthplace in Latin America was associated with almost 2 times higher rates of mental health disorders and social maladjustment.
Hollingsworth (1997)	Included 157 transracial adoptees; Age between 3 and 17 years old; Races included Asian, Black, Mexican, and mixed-race; Age at adoption unknown; Comparison group of 121 same-race adoptees and 28 nonadopted siblings.	Meta-analysis of six cross-sectional and longitudinal studies from United States; Standardized measures included racial identity and self-esteem.	Four data sets were drawn from one longitudinal study (see Shireman & Johnson, 1980). Transracial adoptees had lower combined racial identity/self-esteem scores than nonadoptees (d = -.38). Effect size was larger when racial identity was examined separately from self-esteem (d = -.52).
Huh & Reid (2000)	Included 30 transracial adoptive families with 40 Korean children; Median age was 10 years old (9 to 14 years old); Children adopted before 15 months old.	Cross-sectional, descriptive study; Convenience sample drawn from local child welfare agency; Qualitative interviews and ad hoc measure of ethnic identity and ethnic exposure.	Using regression analysis, participation in cultural activities and communication about adoption experience were strongly related to ethnic identity (r2 = .47). Using grounded theory to analyze qualitative data, Korean adoptees recognize differences at ages 4 to 6 years old, ethnic identification at ages 7 to 8 years old, ethnic dissonance or acceptance at ages 9 to 11 years old, and integration of cultures by ages 12 to 14 years old.
Lindblad, Hjern, & Vinnerljung (2003)	Included 5,942 (3,880 F/2,062 M) international adoptees; Average age between 23 and 30 (born between 1968 and 1975); Adopted primarily from Korea, followed by South Asia, Latin America, and Africa; Children adopted before 7 years age; Compared with 1,884 nonadopted siblings, 8,834 European immigrants, 3,544 non-European immigrants, and 723,154 native-born Swedes.	Cross-sectional, epidemiological study; Representative sample drawn from national registry data in Sweden; Outcome variables included family status, employment, health-related benefits, psychiatric illness, and addictions.	One to two percent of adoptees had at least one health-related problem, especially compared to 0.1% to 1.2% of native-born Swedes and 0.1% to 1.5% of nonadopted siblings. Using logistic regression analyses, adoptees were 2 to 3 times more likely to have health-related problems, which is in stark contrast to nonadopted siblings who had similar rates as the general population. Adoptees from outside of Asia (i.e., Latin America, Africa, and South Asia) had 2 times higher risks for psychiatric problems than adoptees from Asia. Adoptees adopted between ages 4 and 6 years old had 2 times higher risk for psychiatric problems.
Sharma, McGue, & Benson (1996)	Included 4,682 (52% F/48% M) adolescent adoptees; Average age of 15 years old; Races included 5% American Indian, 6% Asian, 5.5% Black, 2.4% Hispanic, and 81% White; Age at adoption unknown; Comparison with a matched control group of 4,682 nonadoptees.	Cross-sectional, epidemiological survey study; Representative sample drawn from schools in 35 states; Standardized measure included Profile of Student Life.	Adoptees showed small but consistently lower levels of adjustment than nonadoptees on 8 of the 12 scales (Effect Size (ES) = -.12 to .26); Interestingly, adoptees showed higher prosocial behaviors (ES = .17). Adopted boys had larger effect size differences than adopted girls on illicit drug use and antisocial behavior (ES = .29 to .35). Hispanic adoptees had large effect size on illicit drug use (ES = .70); American Indian adoptees had medium effect size on negative emotionality (ES = .46) and parental nurturance (ES = -.48).
Verhulst, Althaus, & Verluis-den Bieman (1990a)	Parents of 2,148 (1,112 F/1,036 M) international adoptees; Age between 10 and 15 years old; Adopted primarily from Korea, followed by Colombia, India, Indonesia, and others; Age at adoption varied with 24% adopted within 6 months of age; Compared with 933 same-aged sample of nonadopted Dutch.	Cross-sectional, epidemiological survey study; Representative sample drawn from central adoption register of Dutch ministry; Standardized measure included Child Behavior Checklist for parents.	Adoptees had higher total behavior problems than nonadoptees controlling for socioeconomic status (M = 22.21 vs. 19.89). Adopted boys had higher total behavior problems than nonadopted boys (M = 25.51 vs. 21.14), which explains the group difference in the total sample comparisons. Adopted boys (aged 12-15 years old) were twice as likely to have behavior problem scores above clinical cut-off; Adopted girls (aged 12-15 years old) were 1.5 times as likely to have behavior problem scores above clinical cut-off.
Verhulst, Althaus, & Verluis-den Bieman (1990b)	Same as Verhulst, Althaus, & Verluis-den Bieman (1990a).	Same as Verhulst, Althaus, & Verluis-den Bieman (1990a).	For 12- to 15-year-old adoptees, age at placement predicted greater problem behaviors in boys and girls.
Verhulst, & verluis-den Beirman (1995)	Parents of 1,538 (804 F/734 M) international adoptees; Age between 14 and 18 years old; Adopted primarily from Korea, followed by Colombia, India, Indonesia, and others; Compared with Time 1 sample, there was a slight underrepresentation of older and problem behavior children.	Longitudinal, epidemiological survey study; Representative sample drawn from central adoption register of Dutch ministry; Standardized measure included Child Behavior Checklist for parents.	Three-year follow-up study to earlier research on 2,148 international adoptees (see Verhulst, Althaus, & Verluis-den Bieman, 1990a). There was a significant increase in total problem behaviors over time (η2 = 3.8%); Specific increases in withdrawn (η2 = 6.5%), delinquent (η2 = 8.5%), and internalizing behaviors (η2 = 6.3%) had moderate effect sizes; Early pre-adoption experiences (e.g., age at adoption, medical condition, abuse/neglect) were not responsible for increases in problem behavior; No group differences among adoptees born in non-European countries and other European countries. There was a significant decrease in total competence scores over time (η2 = 19.8%), which is considered a large effect size; Specific decreases in activities (η2 = 11.5%), social (η2 = 11.9%), and school (η2 = 8.1%) had moderate effect sizes.
Verslius-den Bieman & Verhulst (1995)	Parents of 1,538 (804 F/734 M) international adoptees and 1,262 international adoptees; Age between 14 and 18 years old; Adopted primarily from Korea, followed by Colombia, India, Indonesia, and others; Compared with parent report of nonadoptees (n = 213) and adolescent nonadoptee self-report (n = 390) who were matched for age and gender.	Drawn from same data set as Verhulst & Versluis-den Bieman (1995); Standardized measure included Child Behavior Checklist for parents and Youth Self-Report for adoptees.	Three-year follow-up study to earlier research on 2,148 international adoptees (see Verhulst, Althaus, & Verluis-den Bieman, 1990a; Verhulst & Versluis-den Bieman, 1995). Small group differences by age and sex were found on specific problem behaviors (η2 < 5.5%); Girls had more self-reported behavior problems (withdrawn, somatic, anxious/depress, internalizing), and parents of boys reported more behavior problems (attention, delinquent, aggressive, externalizing); Early preadoption experiences (e.g., medical condition, abuse/neglect) predicted problem behavior; No group differences among adoptees born in non-European countries and other European countries. Approximately 20% of adoptees had behavior problem scores above clinical cut-off (with a higher percentage of boys), compared to 10% for nonadoptees; Adoptees were 2 to 4 times more likely to have scores in deviant range than nonadoptees.
Vroegh (1997)	34 (16 F/18 M) Black transracial adoptees; Average age of 17 years old; 68% adopted before 4 months; Compared with 18 Black same-race adoptees.	Cross-sectional, comparison survey study; Convenience sample drawn from two private child welfare agencies in Chicago; Data drawn from structured interviews that were coded for content on adjustment, racial family, and peer relationships; Standardized measure included Rosenberg Self-Esteem Scale.	Fourteen year (fifth phase) follow-up of earlier research on 46 Black transracial adoptees (see Shireman & Johnson, 1986), but not longitudinal design. No significant group differences between transracial and same-race adoptees on adjustment. Same-race adoptees were more likely to identify solely as Black compared with transracial adoptees (83% vs. 33%); Racial identification was affected by skin color complexion. Transracial adoptees reported greater frequency of racial incidents; All same-race adoptees had Black friends, but 25% of transracial adoptees did not.
Westhues & Cohen (1998)	Included 155 (112 F/41 M/2 n/a) adolescent and young adult international adoptees; Average age of 17.3 years old (between 9 and 37 years); Adopted primarily from Korea, followed by Bangladesh, Vietnam, Haiti, India, and others; 54% were adopted before 2 years old.	Cross-sectional, comparison survey study; Convenience sample drawn from social service ministry records in Canada; Data drawn from structured interviews with parents, adoptees, and siblings that were coded for content on racial/ethnic identity, discrimination, and cultural exposure.	Men were more likely to identify as Canadian (51%) than as ethnic minority (35%), but women were more equally divided in identification (40% and 43%); 79% men and 73% women racially self-identified as minority, but 11% to 17% racially self-identified as White. Roughly half the men (44%) and women (46%) reported that ethnicity was important to them; 82.5% of men and 70.5% of women were comfortable with their ethnicities. Eighty-five percent of men and 82% of women reported past racial and ethnic discrimination; Most discrimination involved elementary school age peers (69% to 74%), but a small percent involved adults (7% to 9%); Discrimination was not found to be related to ethnic identity or ethnic comfort. Parents exposed children to race and culture primarily through books (62% to 78%) and attending cultural events (36% to 50%).
Wickes & Slate (1996)	Included 175 (138 F/36 M) Korean adoptees; Average age of 24 years old; Average age at adoption was 3 years (2 months to 14 years old).	Cross-sectional, comparison survey study; Convenience sample drawn from adoption agency that specializes in Korean adoption; Standardized instruments included Self-Description Questionnaire, Suinn-Lew Asian Self-Identity Acculturation Scale, and the Asian American Cultural Identity Scale.	Adoptees identified more with an integrated cultural identity (M = 4.26) than ethnic (M = 2.36), marginal (M = 2.81), or sociopolitical (M = 2.44); Adoptees were also highly acculturated (M = 4.00 out of 5-point scale). Age at adoption was significantly correlated with acculturation (r = -.47).
Yoon (2001)	Included 241 Korean adolescent adoptees; Average age of 14 years old; Age at adoption unknown.	Cross-sectional, comparison survey study; Convenience sample drawn from adoption agency that specializes in Korean adoption; Standardized measures included Parental Acceptance-Rejection Questionnaire, Parent-Adolescent Communication Scale, Multi-Group Ethnic Identity Measure, Rosenberg Self-Esteem Scale, State-Trait Anxiety Inventory, Beck Depression Scale, Affect Balance Scale, and Satisfaction with Life Scale, as well as ad hoc measure of parent support of child’s ethnic background.	Parent support of ethnicity correlated with ethnic pride (rs = .20 to .28) and well-being (r = .26); Ethnic pride correlated with well-being (r = .35) and distress (r = .36). Structural equation modeling revealed that the direct effect of parent support of ethnicity on adjustment was mediated partially by ethnic identity.