Effect of Therapist Color-Blindness on Empathy and Attributions in Cross-Cultural Counseling

Empathy and attributions of client responsibility for the cause of and solution to a problem were examined for 247 psychologists who were identified as having low, moderate, and high color-blind racial attitudes. Participants responded to 1 of 4 vignettes that controlled for client race (i.e., African American, European American) and client attributions regarding the cause (i.e., depression, discrimination) of a problem. Analyses revealed that the therapists’ level of color-blindness was directly related to their capacity for empathy and also to their attributions of responsibility for the solution to the problem with an African American client but not with a European American client. No relationship was found between therapist color-blindness and attributions of responsibility for cause of the problem. Implications of these results for counseling practice, training, and research are discussed. NOT THE PUBLISHED VERSION; this is the author’s final, peer-reviewed manuscript. The published version may be accessed by following the link in the citation at the bottom of the page. Journal of Counseling Psychology, Vol. 51, No. 4 (October 2004): pg. 387-397. DOI. This article is © American Psychological Association and permission has been granted for this version to appear in e-Publications@Marquette. American Psychological Association does not grant permission for this article to be further copied/distributed or hosted elsewhere without the express permission from American Psychological Association. 2 The original conceptualization of cross-cultural counseling competencies identified 11 specific competencies related to counselor awareness of her or his attitudes and beliefs, knowledge, and skills (D. W. Sue et al., 1982). Counselor awareness of racism and oppression in oneself and the effect of racism on others, for example, are considered fundamental to multicultural counseling (e.g., Neville, Worthington, & Spanierman, 2001; D. W. Sue et al., 1982; D. W. Sue & Sue, 2003). Without such awareness, therapists' own biases may perpetuate oppression against racial-ethnic minority clients during the psychotherapy process (Helms, 1984). Despite the theorized importance of therapists' awareness of their own racial attitudes and beliefs, examination of the relationship between therapist racial attitudes and certain kinds of decision making in psychotherapy has received limited empirical attention. In what little research that has been done in this area, researchers have found that therapists who rated themselves as high in negative racial attitudes also rated themselves lower in multicultural counseling competency (Constantine, 2002; Constantine, Juby, & Liang, 2001). Although these studies establish a correlational relationship between therapists' selfreported racial attitudes and cross-cultural counseling competencies, we do not know whether these findings may be extended to therapists' reactions and decisions in counseling. Thus, an examination of the effect of therapists' racial attitudes on their reactions and decisions in counseling may be an important focus for research and could have implications for multicultural counseling competence as well as therapist training and practice. In addition to affecting multicultural counseling competency, therapist racism toward culturally different clients may also explain some of the discrepancies found in research on clinical judgment about, and service delivery to, White clients and ethnic-racial minority clients. For example, research suggests that in comparison to their White counterparts, racial-ethnic minority clients' level of psychopathology is overestimated (Gynther, 1972, 1979), their prognosis is often underestimated (Butcher, Braswell, & Raney, 1983), they are engaged in shorter periods of treatment (Yamamoto, James, Bloombaum, & Hattem, 1967), they are referred to more restricted forms of treatment (Gynther, 1979; Krebs, 1971), and they have higher termination rates from treatment (S. Sue, Fujino, Hu, Takeuchi, NOT THE PUBLISHED VERSION; this is the author’s final, peer-reviewed manuscript. The published version may be accessed by following the link in the citation at the bottom of the page. Journal of Counseling Psychology, Vol. 51, No. 4 (October 2004): pg. 387-397. DOI. This article is © American Psychological Association and permission has been granted for this version to appear in e-Publications@Marquette. American Psychological Association does not grant permission for this article to be further copied/distributed or hosted elsewhere without the express permission from American Psychological Association. 3 & Zane, 1991; Terrell & Terrell, 1984). Although this research is dated, it is, in fact, the most recent such research available. Nevertheless, these investigations do highlight the evidence that suggests the existence of bias in the treatment of racial-ethnic minority clients. Not all investigations, however, support this implied hypothesis of bias. Garb (1997), for instance, conducted a review of the literature on bias in clinical judgment, concluding that “for many important tasks, bias related to the collection and/or integration of information does not seem to be present or, if present, seems to be in a direction that favors groups that have historically been discriminated against” (p. 113). Some researchers have suggested that methodological confounds (e.g., use of analogue studies, social desirability effects, not controlling for level of psychopathology) in the research on bias in therapy may account for these discrepant findings in the literature (Abramowitz & Murray, 1983; Garb, 1997). Although methodological confounds may well be a concern in research on therapist bias in therapy, it is also important to note that none of the investigations cited in these reviews included a measure of therapist racial bias in the research design. For example, prior investigations on therapist bias would typically assign a study participant to an African American or European American client condition, assess the participant on a typical task in therapy (e.g., diagnosis, goal setting), and then compare the results found in the African American client condition with that of the European American condition (e.g., Bamgbose, Edwards, & Johnson, 1980; Pavkov, Lewis, & Lyons, 1989). If differences were found, bias was often an explanation, and in cases in which differences were not found, researchers would often conclude that bias did not occur. Conceptually, it becomes problematic to draw a conclusion about therapist bias, when in fact therapist bias was not measured and included as a variable in the research design. To date, research has not directly examined the effect of therapists' racial biases on psychotherapy processes (Atkinson & Lowe, 1995; Ponterotto, Fuertes, & Chen, 2000). It is this omission that the present study seeks to address. Before we examine the possible relationship of racism to therapist actions in counseling, we must clarify the definition of racism NOT THE PUBLISHED VERSION; this is the author’s final, peer-reviewed manuscript. The published version may be accessed by following the link in the citation at the bottom of the page. Journal of Counseling Psychology, Vol. 51, No. 4 (October 2004): pg. 387-397. DOI. This article is © American Psychological Association and permission has been granted for this version to appear in e-Publications@Marquette. American Psychological Association does not grant permission for this article to be further copied/distributed or hosted elsewhere without the express permission from American Psychological Association. 4 used in this study. Early theories conceptualized racism as overt attitudes and behaviors that were used to limit the rights and freedoms of African Americans (McConahay, 1986). Kovel (1970) wrote that a racist individual “acts out bigoted beliefs—he represents the open flame of racial hatred” (p. 54). In contrast, contemporary theories of racism reflect the multidimensional nature of racial prejudice and acknowledge not only overt bigotry, as identified above, but also more subtle and less overt forms of prejudice (Burkard, Medler, & Boticki, 2001). These later conceptualizations of racism assert that individuals may simultaneously have contradictory feelings of egalitarianism and negativity toward racially-ethnically diverse groups, a contradiction that may result in expressed attitudes of equality that are contrasted by behaviors that demonstrate racial bias (Gaertner & Dovidio, 1986; McConahay, 1986). One way of reconciling this contradiction (i.e., between feelings of egalitarianism and concurrent negative attitudes toward racially diverse people) is to deny that racial issues matter in people's lives, or to be “color-blind” (American Psychological Association [APA], 1997; Schofield, 1986; Williams, 1997). Thus, although racial prejudice is admittedly a multidimensional construct, in this study we were interested in examining a more subtle form of racism, that of color-blind racial attitudes, for two reasons. First, measures of overt racism have been highly susceptible to social desirability attitudes (e.g., Gaertner & Dovidio, 1986; McConahay, 1986). However, measures of more subtle forms of racism (e.g., colorblind racial attitudes) appear to be less susceptible (Burkard et al., 2001; Ponterotto et al., 1995). Thus, we sought to include a measure that would be less susceptible to the effects of social desirability. Second, measures of subtle racism have focused predominately on the attitudes of European Americans, whereas color-blindness is believed to be applicable to many racially and ethnically diverse groups (Neville, Lilly, Duran, Lee, & Browne, 2000). Thus, we sought to include a measure of subtle racism (i.e., Color-Blind Racial Attitudes Scale; COBRAS; Neville et al., 2000) that would provide for the opportunity to replicate this study with psychologists from diverse racial and ethnic backgrounds. Color-blind racial attitudes are defined as “the belief that race should not and does not matter” (Neville et al., 2000, p. 60). This NOT THE PUBLISHED VERSION; this is the author’s final, peer-reviewed manuscript. The published version may be accessed by following the link in the citation at the bottom of the page. Journal of Counseling Psychology, Vol. 51, No. 4 (October 2004): pg. 387-397. DOI. This article is © American Psychological Association and permission has been granted for this version to appear in e-Publications@Marquette. American Psychological Association does not grant permission for this article to be further copied/distributed or hosted elsewhere without the express permission from American Psychological Association. 5 definition is founded on the notion that all racial-ethnic groups are and should be socially and economically equal. Individuals who hold colorblind racial attitudes tend to deny the individual, institutional, and cultural manifestations of racism and believe that race has little meaning in people's lives. In contrast to these beliefs, however, various population statistics suggest that racially diverse groups are neither socially nor economically equal in the United States. For example, the United States has continued to see a rise in the formation of hate groups (Serrano, 1998), unemployment rates among African Americans consistently remain double those of European Americans (U.S. Department of Commerce, 1998), and annual incomes of people of color remain disproportionately lower than those of European Americans (U.S. Department of Commerce, 1998). Those who hold color-blind racial attitudes, however, would likely continue to believe that race has no part in these disparities and instead may explain these differences as a lack of effort by racial-ethnic minorities. Without a conscious awareness of racism and privilege, counselors who hold such beliefs may unwittingly perpetuate racism in the psychotherapy process. As an illustrative example, counselors may minimize or ignore reports of discrimination and prejudice in school, at work, or in the community by clients of color. Furthermore, if they remain unaware of the effects of racism, they may be unlikely to address issues of racism in psychotherapy and may not intervene on a client's behalf with systems that are racist and oppressive. In essence, counselors who endorse high levels of color-blind racial attitudes may be more inclined to be unresponsive to racial issues and to the effects of racism on clients during counseling sessions. Although the effect of color-blind racial attitudes on the therapeutic process has not been directly studied, research evidence from client reports suggests that counselors who were culturally responsive (perhaps a reflection of the degree to which a therapist is color-blind) were rated as more credible and culturally competent (Atkinson, Casas, & Abreu, 1992; Gim, Atkinson, & Kim, 1991; Pomales, Claiborn, & LaFromboise, 1986), reported higher client satisfaction and lower client attrition rates from counseling (Wade & Bernstein, 1991), and had clients who self-disclosed more intimately (Thompson, Worthington, & Atkinson, 1994) than did counselors who NOT THE PUBLISHED VERSION; this is the author’s final, peer-reviewed manuscript. The published version may be accessed by following the link in the citation at the bottom of the page. Journal of Counseling Psychology, Vol. 51, No. 4 (October 2004): pg. 387-397. DOI. This article is © American Psychological Association and permission has been granted for this version to appear in e-Publications@Marquette. American Psychological Association does not grant permission for this article to be further copied/distributed or hosted elsewhere without the express permission from American Psychological Association. 6 were culturally unresponsive. Furthermore, Thompson and Jenal (1994) found that African American women became more frustrated and exasperated with counselors, regardless of their racial background (i.e., African American, European American), when the counselors actively avoided racial content in counseling. These prior investigations focused on client perceptions of the counselor after cultural sensitivity training as well as clients' perceptions of counselor cultural responsiveness and unresponsiveness in cross-cultural counseling. Although there may be other explanations for counselors' cultural unresponsiveness (e.g., counselors' discomfort discussing racial issues, counselors' level of multicultural training), perhaps unresponsiveness to cultural issues and decreased cultural sensitivity are manifestations of counselors' color-blind racial attitudes. This potential conceptual link between therapist color-blindness and their reactions and actions with culturally diverse clients in therapy is worthy of further examination. For this study, then, we proposed that therapists' color-blind racial attitudes would directly affect their reactions (i.e., empathy, attributions) in therapy. In this investigation, we were interested in two types of therapist reactions in multicultural counseling. First, we examined therapists' feelings of empathy toward racially diverse clients. Empathy is a foundational disposition in counseling (Bohort, Elliot, Greenberg, & Watson, 2002) and has been found to be one of the most important factors in discriminating effectiveness in therapy (Bohort et al., 2002; Lafferty, Beutler, & Crago, 1989). Empathy is also considered essential to cross-cultural counseling (Ho, 1992; Ridley & Lingle, 1996). It is believed that therapists' expression of empathy in a culturally responsive manner may be a determining factor in culturally diverse clients' perception of therapist cultural sensitivity during assessment and treatment (Fischer, Jome, & Atkinson, 1998). Although research is limited, one investigation found that counselors high in affective empathy were more aware of cultural factors in conceptualizing client issues than were counselors low in affective empathy, and counselors high in affective and cognitive empathy demonstrated increased capacity to conceptualize cultural issues in client treatment (Constantine, 2001). These findings suggest that empathy is potentially an important component in the counseling of culturally diverse clients. NOT THE PUBLISHED VERSION; this is the author’s final, peer-reviewed manuscript. The published version may be accessed by following the link in the citation at the bottom of the page. Journal of Counseling Psychology, Vol. 51, No. 4 (October 2004): pg. 387-397. DOI. This article is © American Psychological Association and permission has been granted for this version to appear in e-Publications@Marquette. American Psychological Association does not grant permission for this article to be further copied/distributed or hosted elsewhere without the express permission from American Psychological Association. 7 Although Constantine's (2001) research has indicated that empathy may be an important factor in cross-cultural counseling, we do not know what specific therapist characteristics may influence their capacity to be empathic in multicultural counseling settings. The present study thus sought to examine the effect of one such characteristic (i.e., color-blind racial attitudes) on therapists' capacity for empathy toward an African American or a European American client when the client attributed his or her problem to depression (i.e., internal cause) or racial discrimination (i.e., external cause). A second type of therapist reaction we examined was therapists' attributions of clients' responsibility for causing and solving their problems. Attributions of cause of the problem have been defined as the responsibility and control one has for the origin of a problem and specifically refer to whether the individual or the environment is responsible. Attributions of solution to the problem involve an assessment of whether the individual or the environment is responsible for and in control of any future events related to the resolution of the problem. Therapist attributions regarding cause of and solution to client difficulties in counseling are important because they may affect the process and outcome of therapy (Brickman et al., 1982). For example, clients have been found to terminate counseling prematurely when their explanation of the problem cause did not match that of the therapist (Tracey, 1988). Although research has not examined the reason for the lack of match between client and therapist attributions, theorists in multicultural counseling believe the difference between therapist and client attributions of cause of and solution to client problems may be a result of differences in client and counselor worldviews (D. W. Sue & Sue, 2003). Some clients' cultural value systems (e.g., Asian American, Native American), for instance, emphasize fatalism or external causes for client difficulties (LaFromboise, 1988; Leong, 1985; Root, 1998). In contrast, research indicates that therapists often are more inclined to emphasize an internal locus of control for client problems (Donnan & Pipes, 1985; Tracey, 1988), unless there is a clear external precipitant for the client's concern (Hayes & Wall, 1998; Worthington & Atkinson, 1996). These contrasting perspectives may lead to therapists encouraging clients to accept responsibility for their actions, whereas clients who have a more fatalistic perspective of the world may seek to understand the external causes of their difficulties. Furthermore, therapist NOT THE PUBLISHED VERSION; this is the author’s final, peer-reviewed manuscript. The published version may be accessed by following the link in the citation at the bottom of the page. Journal of Counseling Psychology, Vol. 51, No. 4 (October 2004): pg. 387-397. DOI. This article is © American Psychological Association and permission has been granted for this version to appear in e-Publications@Marquette. American Psychological Association does not grant permission for this article to be further copied/distributed or hosted elsewhere without the express permission from American Psychological Association. 8 responsibility attributions have been found to be relatively unaffected by client responsibility attributions (Hayes & Wall, 1998). The combination of these research findings seems to suggest that therapists typically attribute internal causes for client difficulties and may be unwilling to alter their perspectives even when presented with an alternative explanation by their clients. Such conflicts in client and therapist worldview and attribution styles may result in poor therapy outcomes for the client (e.g., premature termination, lack of compliance with therapy goals) and may provide culturally diverse clients with reasons to distrust counseling services (Ridley, 1995; D. W. Sue & Sue, 2003). Consequently, identifying specific therapist characteristics (i.e., color-blindness) that moderate their attributions in multicultural counseling appears to be important to study. Of concern to the multicultural counseling process is the influence that color-blind racial attitudes may have on therapists' attributions, because these attitudes may interfere with therapists' ability to accurately discern the difference between internal (i.e., depression) and external (i.e., racism) causes for a client's problem. Or perhaps, color-blindness may interfere with therapists' acknowledgment of the validity of the client's attributions. As an illustration, traditional theories of career development typically emphasize individualistic and self-actualizing perspectives regarding a client's career choice and behaviors rather than considering the effect of racial discrimination (Leung, 1995; Smith, 1983). Therapists who fail to recognize the effect of racial discrimination on career behavior may be especially vulnerable to therapeutic errors, particularly when working with clients of color. For instance, a therapist may attribute an African American's unsuccessful attempts to obtain employment to a lack of effort or lack of skills rather than to biased hiring practices. Consequently, if therapists do not make an accurate assessment of the problem cause, particularly when the client's presenting concern involves racism, the therapist's misattribution may result in a poor therapeutic outcome for the client (e.g., increased distress, premature termination from counseling). In the present investigation, we examined the effect of therapist color-blind racial attitudes on their attributions of client responsibility for the cause of and solution to a problem when the client attributed the problem to depression (i.e., internal) or racial discrimination (i.e., external). NOT THE PUBLISHED VERSION; this is the author’s final, peer-reviewed manuscript. The published version may be accessed by following the link in the citation at the bottom of the page. Journal of Counseling Psychology, Vol. 51, No. 4 (October 2004): pg. 387-397. DOI. This article is © American Psychological Association and permission has been granted for this version to appear in e-Publications@Marquette. American Psychological Association does not grant permission for this article to be further copied/distributed or hosted elsewhere without the express permission from American Psychological Association. 9 Research Questions Therapists' color-blindness, feelings of empathy, and attributions of client responsibility for the cause of and solution to a problem may each have important implications for cross-cultural counseling, and the relationship among these factors is worthy of investigation. Theory on color-blind racial attitudes presumes that individuals who are high in these beliefs will deny the role and influence of race in people's lives and that these beliefs will result in actions that may be more consistent with racial prejudice rather than with social justice (Neville et al., 2001). Consistent with this perspective, we hypothesized that those therapists who were higher in color-blind racial attitudes would show less empathy for an African American client than for a European American client. Furthermore, we expected that the client's attributions of the source (i.e., depression, discrimination) of the problem would qualify the ratings of empathy with the African American client but not with the European American client. Second, we predicted that therapists' color-blindness would be related to their attributes of responsibility for cause of the problem for the African American client and would not effect their attributions for the European American client. Again, we expected that these effects would be moderated by the client's attributions of the source (i.e., depression, discrimination) of the problem. Finally, we predicted therapists' color-blind racial attitudes would also be related to their attributions of responsibility for the solution of the problem for the African American client and not the European American client. We expected that these effects would also be moderated by clients' attributions of the source (i.e., depression, discrimination) of the problem. Because research has indicated that social desirability is a potentially confounding variable in multicultural studies (Constantine & Ladany, 2000), we also examined social desirability as a covariate for this study. NOT THE PUBLISHED VERSION; this is the author’s final, peer-reviewed manuscript. The published version may be accessed by following the link in the citation at the bottom of the page. Journal of Counseling Psychology, Vol. 51, No. 4 (October 2004): pg. 387-397. DOI. This article is © American Psychological Association and permission has been granted for this version to appear in e-Publications@Marquette. American Psychological Association does not grant permission for this article to be further copied/distributed or hosted elsewhere without the express permission from American Psychological Association. 10 Method