Continued demographic shifts in the U.S. population whereby the proportion of racial/ethnic minorities is increasing (U.S. Census Bureau, 2000), coupled with more attention to social injustices for typically oppressed groups (e.g., racial/ethnic minorities, women, sexual minorities), illuminate that the counseling profession is becoming more complex. To this end, professional counselors must adapt their clinical decisionmaking processes to accommodate a more diverse population. Clinical decision making refers to the intricate decisions professional counselors make when they assess the degree of severity of a client's symptoms, identify a client's level of functioning, and make decisions about a client's prognosis.

Prevalence data for racial/ethnic minorities, women, and sexual minorities show disproportionate rates of assessed mental illness. For example, Hispanics have been diagnosed with schizophrenia 1.5 times the rate of Whites, and African Americans are more likely than Whites to be diagnosed with schizophrenia, substance abuse, and/or dementia (see Block, 1984; DelBello, Lopez-Larson, Soutullo, & Strakowski, 2001; Jones & Gray, 1986; Mukherjee, Shukla, Woodie, Rosen, & Olarte, 1983; Strakowski, Shelton, & Kolbrener, 1993). Women are disproportionately diagnosed with personality disorders (Eriksen & Kress, 2005; Ford & Widiger, 1989); lesbian, gay, and bisexual clients are diagnosed 3 times and 4.7 times the rate of heterosexual clients for major depression and panic disorder, respectively (Cochran, Sullivan, & Mays, 2003); and clients of lower educational levels are diagnosed more often with schizophrenia than with mood disorders (Paniagua, 2005).

The principal question, then, is whether these differences in diagnostic rates are actual differences among cultural groups or whether they are more reflective of bias in decision making among practitioners. That is, are counselors diagnosing accurately, or is misdiagnosis based on cultural differences the cause of the discrepancy in these rates? Researchers highlighting the disproportionate prevalence of particular diagnoses among individuals for diverse identities argue that these rates result from individuals belonging to typically oppressed groups (e.g., racial/ethnic minorities, women, sexual minorities) being diagnosed with more severe diagnoses. Alternatively, available empirical data suggest that some clients may receive less severe diagnoses if symptomatology does not fit an existing diagnostic criteria set (Eriksen & Kress, 2005; Lin, 1996) or is congruent with cultural roles or stereotypes associated with a particular gender or race/ethnicity (Block, 1984; Lopez, 1989). Furthermore, similar symptoms may be judged less severe depending on either the amount of resources practitioners perceive certain clients to have (Lopez, 1989) or the awareness of the role environmental factors has on the perpetuation of mental illness, such as having a strong situational locus of attribution (Gushue, 2004). An increased understanding of whether diagnostic variance (and thus misdiagnosis) exists disproportionately among culturally diverse groups could address the question of whether disproportionate rates are real or not.

Unfortunately, there is minimal literature available on diagnostic variance in counseling in general and, more specifically, on the role culture plays in variance. Diagnostic variance refers to the notion that professional counselors display variability in how they arrive at clinical decisions (Hays, McLeod & Prosek, 2009). Some forms of diagnostic variance include divergent expressions of symptoms across individuals for mental disorders, differences in the amount and type of data that are collected from clients, varying interpretations of the same data among professional counselors, and the differential application of criteria in the diagnostic process (Gigerenzer, 2002). Even with the predominant use in the counseling profession of the Diagnostic and Statistical Manual of Mental Disorders (4th ed., text rev.; American Psychiatric Association, 2000) outlining behavioral criteria to assist in clinical decision making, counselors may be making inaccurate client assessments, thus potentially misdiagnosing client symptomatology and adaptive functioning.

Client cultural characteristics and cultural bias among helping professionals have been investigated as a factor of diagnostic variance. Empirical data show that mental health professionals often rely on racial/ethnic (Gushue, 2004; Rosenthal, 2004) and gender (Ford & Widiger, 1989) stereotypes in clinical decision making, leading to more severe diagnoses for members of oppressed statuses and underdiagnosis of those who are not oppressed. Moreover, symptoms seem to be viewed more negatively for racial/ethnic minority clients, even when they receive similar diagnoses as Whites (Jenkins-Hall & Sacco, 1991).

Because treatment plans for clients are based significantly on the clinical diagnosis they receive, it is imperative that counselors accurately diagnose culturally diverse clients. In essence, counselors must be aware of how cultural factors (e.g., client's cultural identity, client's oppression experiences, cultural match between counselor and client) relate to diagnostic variance. There is limited information available about the role of culture, including counselor attitudes toward a client's cultural identity, in the clinical decision-making process. To date, research on diagnostic variance and more specifically, misdiagnosis (i.e., differential prevalence rates) has involved quantitative studies largely outside the counseling profession. Additionally, these studies investigated what diagnoses White clinicians assigned primarily hypothetical African American clients of lower socioeconomic status (SES; presented in either hypothetical case vignettes or archival case files).

The purpose of this study was to explore the process by which culture affects the clinical decision-making process by examining a more diverse population of both clients and participants using a mixed methodological design. A mixed methodological approach allowed for gathering quantitative data regarding cultural bias and aspects of case conceptualization as well as for obtaining qualitative data on the process of clinical decision making in the context of the cultural identities of both clients and participants. Culture is defined in this study as attitudes, behaviors, and cognitions related to the identities of race/ethnicity, gender, and SES. Case conceptualization is defined as attendance to and interpretation of client presenting symptoms and diagnoses. Cultural bias is defined as the degree of awareness of oppression across cultural identities. Finally, perceived level of functioning refers to assessment of clients' level of adaptive functioning (e.g., social, occupational) and prognosis.

The overarching research question addressed in our study was as follows: How does culture influence the clinical decision-making process among counselors and counselor trainees? The three subquestions were (a) How does a client's cultural identity influence case conceptualization? (b) What impact does the cultural match between the participant and client have on case conceptualization? and (c) What is the relationship between cultural bias, perceived level of functioning, and the clinical decision-making process?

*Method

Participants and Procedure

In keeping with maximum variation sampling methods (Patton, 2002), research team members (see the following section for details) selected participants who were culturally diverse, held various credentials, and had a broad range of years of clinical experience in community mental health settings (e.g., community agencies, hospitals, private practice). Of the 41 participants (33 women, 8 men), 28 were White, 10 were African American, 1 was Hispanic, 1 was biracial/multiracial, and 1 self-identified as "other." The age of participants ranged from 25 to 59 years old with a mean age of 40 years old. Participants reported their sexual orientation as heterosexual (n = 39), lesbian/gay (n = 1), or bisexual (n = 1). The mean length of clinical experience was 67.80 months (SD = 71.84, range = 262). Twenty-nine participants reported that they had attended a multicultural workshop.

Research team members randomly assigned participants one of six packets of client materials that included a narrative summary and an intake report. These six packets contained information on hypothetical clients who varied in cultural identities but had identical presenting symptoms. Participants were asked to read the materials and provide a multiaxial diagnosis of the client. The mean length of time to diagnose was 50.54 minutes (SD = 30.64, range = 108; 4 participants did not report length of time to diagnose). Next, participants were interviewed by a member of the research team. Upon completion of the interview, participants completed a survey packet assessing perceived level of client functioning and cultural attitudes.

Research Team

The research team for this study consisted of 11 individuals from community mental health and school counseling backgrounds with some clinical experience working with adults. The team included one counselor educator (first author), two doctoral students (including the third author), and eight master's-level counseling students (including the second author). The primary researcher (first author) provided a training seminar on qualitative research, assessment, and multicultural issues for the research team members prior to the onset of the study. The team was made up of culturally diverse individuals with a variety of clinical experiences and perspectives on diagnosis.

Through bracketing, the research team discussed common assumptions about the clinical decision-making process and how it may relate to diagnostic variance and cultural bias. These assumptions included the following: (a) Diagnostic variance may be a common occurrence in the counseling community, especially with underdiagnosis from clinicians with negative attitudes toward the diagnostic process and overdiagnosis from clinicians with negative attitudes toward minority and oppressed populations, and (b) participants may neglect to consider culture or SES when considering diagnoses for a client perceived as having a more severe disorder.

Data Sources

Demographic sheet. The survey packet included a one-page demographic sheet assessing for participants' cultural makeup, including race/ethnicity, gender, and SES. The demographic sheet also solicited for counseling experience, such as educational level, credentials, and participation in multicultural counseling workshops.

Case. The case template (i.e., narrative summary and intake report) was developed and adapted by the first author from a client within a hospital setting where she had worked. A narrative summary of a 32-year-old client was presented to participants. Each summary provided equal information, but a manipulation of client race (i.e., White, African American), ethnicity (i.e., Hispanic, non-Hispanic), gender, and SES created six hypothetical client conditions. The six clients were a White man of lower SES (Client A), a White woman (Client B), an African American man (Client C), a Hispanic woman (Client D), a Hispanic man of lower SES (Client E), and an African American woman of lower SES (Client F). Participants received detailed information in the summary of symptomatology, including frequency and duration of symptoms. An intake report included demographic information, such as age, race/ethnicity, gender, occupation, relationship status, number of children, and residential status. The intake report also provided participants with details regarding mental status, presenting symptoms, family history, environmental stressors, medical information, and treatment history.

Interview protocol. Research team members were instructed to follow a semistructured interview style (see Appendix) to assess the participant-assigned diagnosis. Participants were asked whether cultural aspects of the case played a role in clinical decision making. Several of these questions were adapted or modeled from Trierweiler, Muroff, Jackson, Neighbors, and Munday's (2005) interview protocol.

Global Assessment Scale (GAS; Endicott, Spitzer, Fleiss, & Cohen, 1976). The GAS is a rating scale in which participants assess a client's lowest level of functioning on a hypothetical continuum of 1 (gross impairment needing constant supervision) to 100 (superior functioning with no mental illness symptoms).

Prognostic Scale (PS; Friedlander & Stockman, 1983). The PS is a rating scale in which participants are instructed to "rate the highest level of adaptive functioning that could be expected for the client described in the case, given sufficient motivation for change, a good therapeutic relationship, and adequate time for whatever form of treatment is adopted" (Friedlander & Stockman, 1983, p. 641). The scale ranges from 1 (superior) to 10 (grossly impaired).

Privilege and Oppression Inventory (POI; Hays, Chang, & Decker, 2007). The POI is a 39-item self-report measure designed to assess individuals' awareness of privilege and oppression associated with race/ethnicity, sexual orientation, religion, and gender. Each item is responded to on a 6-point Likert-type scale ranging from 1 (strongly disagree) to 6 (strongly agree). The POI yields an overall score as well as scores for four areas: White Privilege Awareness (racial/ethnic advantage, power, and access), Heterosexism Awareness (disproportionate advantage, power, and access among individuals on the basis of their sexual orientation), Christian Privilege Awareness (positive portrayal of Christianity in history and contemporary times), and Sexism Awareness (double standards and devaluation of femininity). The POI demonstrates high internal consistency, with a total score alpha of .95 and alphas for the subscales ranging from .79 to .92. In addition, the POI demonstrates high test-retest reliability (r = .91, p < .01) and strong content and construct validity. For this sample, the POI demonstrated high internal consistency (α = .96).

Data Analysis

We used grounded theory procedures (Corbin & Strauss, 2008) to analyze the qualitative data. Ten randomly selected interview transcripts were analyzed independently, and saturation was reached. We then divided the remaining 3 1 interview transcripts and coded them using an initial codebook developed from the first 10 interviews. During the open coding process, we allowed for continued revisions to the codes to develop the most articulate definitions and descriptions to the processes of the participants that resulted in an initial codebook. Each participant's data were managed and conceptualized using within-case displays (Miles & Huberman, 1994); data were used to continually revise the codebook. We met to develop axial codes in which causal conditions and interactions among open codes were outlined. Furthermore, axial coding procedures considered quantitative findings. Quantitative data from the participant demographic sheet and survey data were analyzed using SPSS (Version 14.0).

*Results

Attention to Client's Cultural Identity in Case Conceptualization

Qualitative data revealed that participants identified various cultural identities yet integrated them differentially in their clinical decision making. That is, participants considered the cultural identities of the clients relevant to case conceptualization at varying degrees. In some cases (n = 5), awareness of cultural bias led participants to defer more severe diagnoses.

Relevance of cultural identity in case conceptualization. Some participants identified cultural identities as part of the presenting symptoms or diagnostic decision, but seldom both. Gender was the most often mentioned cultural category for influencing presenting symptoms (n = 18) and diagnoses (n = 7). Other identified client cultural identities affecting presenting symptoms included race/ethnicity (n = 12), age (« = 4), and sexual orientation (n = 4). Some participants identified the following cultural identities influencing their diagnostic decision: race/ethnicity (n = 6), age (n = 2), sexual orientation (n = 2), and spirituality (n = T). Furthermore, race/ethnicity was not considered in the diagnostic decision for any of the 1 5 White clients. Several participants stated that culture did not influence a client's presenting symptoms (n = 1 7) or diagnosis (n = 27). This finding was equally evident across participants and clients, regardless of their cultural makeup.

There were some disproportionate diagnoses by cultural makeup that warrant attention. First, although there was a 0.58 ratio of Whites to racial/ethnic minorities among the 41 clients, there was a disproportionate number of racial/ethnic minorities diagnosed as bipolar compared with Whites (0.38). Regarding gender, a disproportionate number of men were diagnosed as having a bipolar disorder (0.54), whereas women were disproportionately diagnosed with major depressive disorder (1.60), with alcohol abuse/dependence (1.67), and as having a personality disorder (2.00). These ratios can be compared with the 0.86 female-to-male ratio for the participant sample.

Relevance of cultural identity independent of case conceptualization. Several participants referenced client cultural group memberships independent of presenting symptoms or diagnoses. The predominant cultural group mentioned was race/ethnicity (n = 15), particularly for racial/ethnic minority clients (i.e., 1 1 of 13). Other cultural groups mentioned were gender (n = 12), family (n = 3), spirituality (n = 2), age (n = 2), and SES (n = 1). It was difficult for some participants (n = 1 1) to understand the term partner as being interchangeable with spouse, particularly for White clients (i.e., 8 of 11).

Degree of Cultural Match and Case Conceptualization

The degree of cultural match between each participant and client assessed by quantitative and qualitative analyses played a role in the extent that culture was considered in case conceptualization. Cultural match between the participant and client was examined for race/ethnicity, gender, and SES. Racial/ethnic match was conceptualized as pairs constituting either both majority (i.e., White- White) or minority statuses (e.g., African American-Hispanic). The racial/ethnic match (for minority pairs) between the participant and client was significantly related to whether race/ethnicity was considered a factor in a client's presenting problem (χ^sup 2^ = 10.76, p = .001) and diagnosis (χ^sup 2^ = 20.5 1, p < .001). Additionally, the gender match between the participant and client was significantly related to whether gender was considered a factor in a client's presenting problem (χ^sup 2^ = 8.81,/) = .003) and diagnosis (χ^sup 2^ = 23.44, p < .001). Moreover, an SES difference between the participant and client was significantly associated with whether the participant identified SES as a factor affecting a client's presenting problem (χ^sup 2^ = 33.39, p < .001).

Cultural match and case conceptualization. Of the 41 possible participant-client pairs for this study, 17 represented racial/ethnic matches, with 10 White-White pairs and 7 minority-minority pairs. Seventeen female participants were assigned female cases, and 6 male participants were assigned male cases.

For the 10 White- White pairs, only 1 pair identified race/ ethnicity as a factor in case conceptualization. For this case, the participant stated, "My mind just automatically thought he was White ... I guess I superimposed him being a White person." The participant described how the symptoms were in keeping with someone of a "White middle-class background" (i.e., Client A). For the 7 pairs matched by racial/ethnic minority status, 3 participants (43%) identified race/ethnicity as a consideration in their assigned client's presenting problem or diagnosis. Additionally, nearly 52% (n = 12) of the 23 pairs matched on gender identified gender as a component of a client's presenting problem or diagnosis.

Cultural differences and case conceptualization. For this sample, 23 participant-client pairs were racially/ ethnically different (i.e., cultural mismatch), with 18 White participant-minority client and 5 minority participantWhite client pairs. Regarding gender, 1 8 participant-client pairs had different genders, with 15 female participantmale client pairs and 3 male participant-female client pairs. There were 3 participant-client pairs in which there was a significant SES difference (i.e., participant with an income of $100,000 or greater and client with an income of less than $20,000).

In examining the association between racial/ethnic mismatch, approximately 44% (n = 8) of the White participants (i.e., the 18 White participant-minority client pairs) did not identify race/ethnicity as a factor in either the presenting problem or the diagnosis. Nearly 80% (n = 4) of the racial/ethnic minority participants (i.e., the 5 minority participant-White client pairs) did not identify race/ethnicity as a factor in either the presenting problem or the diagnosis. For gender, approximately 56% of the participants (9 of the 15 female participant-male client pairs and 1 of the 3 male participant-female client pairs) did not identify gender as a contributing factor in either the presenting problem or the diagnosis. Finally, approximately 67% (n = 2) of the participants who differed significantly for SES match identified SES as a factor in the presenting problem.

Relationship Between Cultural Bias, Perceived Level of Functioning, and the Clinical Decision-Making Process

Cultural bias and perceived level of functioning. Cultural bias was measured by POI scores, and perceived level of functioning was assessed by GAS and PS scores. The means for the POI, the GAS, and the PS were 4.67 (SD = 0.72), 40.12(SD= 11.81), and3.45 (SD= 1.16), respectively. These scores indicate that participants had a moderate awareness of oppression issues (i.e., lower levels of negative cultural bias), perceived clients to have major impairment or serious symptomatology affecting daily functioning, and projected that clients would be expected to display minimal symptoms as the highest level of adaptive functioning, respectively. To examine the relationship between cultural bias and perceived level of functioning, we conducted correlational analyses. Results indicated that there was a significant relationship (r = -.35, p = .03 1) between increased awareness of cultural bias (i.e., higher POI scores) and higher levels of adaptive functioning (i.e., lower prognostic ratings). Awareness of cultural bias was not significantly associated with ratings for current functioning for the GAS (r = -.02, p = .911), and GAS and PS scores were not significantly associated (r = -.27,p = .091). Additionally, a series of analyses of variance assessing the relationship between participants' demographic variables (i.e., race/ethnicity, gender) and cultural bias and perceived level of functioning in clinical decision making were performed (see Table 1). Data indicated a main effect between race/ethnicity and GAS scores, F(A, 40) = 4.14, p = .008, partial η^sup 2^ = .33.

Relationship between prognosis and locus of attribution. Locus of attribution refers to the degree to which participants view causes or manifestations of symptoms as a result of external (i.e., situational locus) or internal (i.e., dispositional locus) factors. Furthermore, those who endorse a situational locus may view presenting problems and symptoms as a response to stressors (e.g., financial barriers, oppression experiences), whereas those who have a dispositional locus may consider mental health symptoms as biologically based or within their individual control. Although it is likely that participants could have a combined locus of attribution (i.e., viewing client problems as both within and outside the client's control), qualitative data revealed that only 6 of the 33 participants who discussed attribution of symptomatology viewed it this way. Of the remaining participants who discussed attribution of symptomatology, 22 participants endorsed a situational locus and 5 endorsed a dispositional locus of attribution.

[TABLE 1 OMITTED]

Although the majority of participants focused on either a combined or situational locus of attribution, there did not seem to be a clear association for this sample between locus of attribution and assigned prognosis. Prognosis was coded for each case as "good," "fair," "poor," or "uncertain." Most of the participants (n = 30) viewed a client's prognosis as good, with 1 1 of these participants citing positive progress when a client's problems are situationally based and there are adequate personal and professional support networks. Furthermore, clients whose prognoses were rated as fair (n = 1), poor (n = 3), or uncertain (n = 4) represented only minority statuses.

*Discussion

This article highlights ways in which culture - particularly a client's cultural identity, cultural match between a client and counselor, and cultural bias - affects die clinical decisionmaking process. This process relates to how counselors and counselor trainees assess presenting symptoms, develop diagnostic impressions, and define current and potential levels of adaptive functioning. To this end, this study expands on the limited research available that addresses the interface of culture and diagnosis within counseling.

Regarding the relationship between case conceptualization and attention to cultural variables, there were disproportionate diagnoses for particular cultural groups, which is consistent with the literature (i.e., Block, 1984; DelBello et al., 2001; Eriksen & Kress, 2005; Ford & Widiger, 1989; Jones & Gray, 1986; Mukherjee et al., 1983; Strakowski et al., 1993). Also, participants primarily reported that culture did not influence presenting symptoms or diagnoses - even in cases in which participants identified client cultural variables. Often, participants did not mention cultural factors unless asked specifically about them.

The second theme indicates that cultural match between participants and clients did play a role in the extent that culture affected case conceptualization, but more so for minority pairs. Specifically, race/ethnicity and gender match among pairs was significantly associated with whether each of these variables was discussed in case conceptualization. An interesting finding among the White-White pairs was that only 1 of 10 of the pairs discussed race/ethnicity as important in symptom presentation and diagnostic decisions. Thus, the majority of instances in which race/ethnicity was discussed involved racial/ethnic minority pairs. The majority of participants - regardless of their racial/ethnic identity - did not attend to race/ethnicity when conceptualizing White clients. When there was a mismatch for the pair, cultural factors were less focused on. The exception was that for pairs in which there were SES differences, SES was noted as significantly affecting the client's presenting problem. These findings in conjunction with those of the first theme highlight not only that culture is mostly ignored but also that participants are unaware of how cultural variables - including cultural match - affect their clinical decision-making process.

The final theme highlights that an increased level of awareness of cultural bias (i.e., awareness of privilege and oppression issues) is related to participants reporting more positive prognoses for their clients. This result relates to previous literature indicating that having a situational locus of attribution increases the likelihood of seeing symptoms as less severe and, thus, clients as more adaptive (Gushue, 2004). Another important finding is that participants in this study primarily attributed client problems as either dispositional or situational without consideration of both. Furthermore, participants reported symptoms as attributed more frequently to situational factors. This last finding seems inconsistent because cultural factors were minimized or ignored by most participants.

*Implications for Counselors

This study indicates that there are disproportionate prevalence rates of more severe diagnoses for more oppressed cultural groups, cultural (i.e., racial/ethnic minority and gender) match is related to whether cultural variables are considered in clinical decision making, and participants' awareness of social injustice is related to more positive prognoses of clients. Data provide several implications for counselor educators and supervisors. First, counselors are encouraged to become aware of their own cultural identities as well as cross-cultural dynamics. Culture blindness among counselors and counselor trainees seems to exist: A majority of participants stated cultural factors were not significant to the clinical decisionmaking process, yet findings indicated that culture does matter. In examining a client case, counselors should reflect on the following questions: (a) What cultural characteristics define the client? (b) What symptoms does the client present with? (c) How could the symptoms be caused or perpetuated by cultural characteristics? (d) How might symptom expression be related to environmental factors and situational factors? (e) What diagnosis or diagnoses fit this client? (f) To what degree does the client's cultural characteristics - or the interface between these characteristics and environmental factors - influence my diagnosis or diagnoses? and (g) Would I give the same diagnosis or diagnoses to a client with different cultural characteristics?

Second, regarding cultural match, counselors are encouraged to reflect on how their cultural identities match those of their clients. Because counselors may be more likely to attend to cultural characteristics of which they match with clients, they should pay particular attention to cultural characteristics of which they do not match their clients. For example, if a male counselor is working with a female client, he should especially attend to how the gender mismatch influences his case conceptualization. Unfortunately, however, counselors (especially White counselors) who work with White clients do not typically address race/ethnicity in case conceptualization and clinical decision making. Counselors then are charged with paying closer attention to how being White matters in case conceptualization. According to Lopez (1989), counselors may be unconsciously assigning less severe diagnoses or perceiving White clients to have more resources because of their race/ethnicity. This tendency may lead to the issue of underdiagnosis. Because race/ethnicity also significantly relate to current ratings of functioning, it is especially important for White counselors to consider how race/ethnicity plays a role in symptom expression and judgments of clients' current functioning.

There is a disconnect between reported higher levels of awareness of privilege and oppression issues (decreased cultural bias) and attributing client problems to situational factors and the minimal attention to culture in clinical decision making. Counselors may be stating that environmental stressors are important to consider, yet they are not considering these stressors. Because there seems to be a significant relationship between lower levels of cultural bias and more positive prognostic ratings, attending more to specific environmental stressors and how they may directly apply to client symptomatology could assist counselors to attend to cultural differences and oppression experiences with the client.

*Limitations and Future Research Directions

There are several limitations to this study. First, the sample used for this study was relatively small and represented participants from only five states, most of which were in the southeastern United States. The sample size was decided as a balance between available resources and the desire to provide thick description of the phenomenon under investigation. The sample size also resulted in lowered power estimates, creating higher risks for Type II errors (see Table 1). Thus, the study may be conceptualized as an exploratory study that needs to be expanded to further test some of the quantitative subquestions. Future research directions exploring the role of cultural variables in clinical decision making could involve mixed methodological approaches with additional, larger samples of counselors and counselor trainees, as well as participants of other mental-health-related professions.

Second the case used in this shady represented a hypothetical client and was presented in writing. This approach may have limited the diagnostic process for participants. Third there were only six cases; thus, the study did not represent all the possible conditions (i.e., race/ethnicity, gender, SES). Although we were interested in making the client case more varied than those founded in other studies, the six presented conditions contain confounds that prevent the reader from ascertaining the degree to which each cultural variable influenced the clinical decision-making process. It might be assumed that participants will use more visibly oppressed identities to frame case conceptualization and diagnostic decisions, because the more visible identities are likely to outweigh those that are not. Nevertheless, this assumption needs to be tested by creating conditions that allow only one independent variable to vary. Finally, counseling researchers may want to use actual clients or videos of clients so that participants are able to assess more nonverbal behaviors and dynamics between the practitioner and the client. This approach may allow for increased exploration of cultural dynamics in case conceptualization.

Third one of the instruments in this study (i.e., the POI) is a self-report measure susceptible to potentially socially desirable responses. Thus, participants may have overestimated their awareness of cultural bias. Future research might include an additional method for assessing cultural bias, such as counselor educator and supervisor ratings.

An additional area of study examining diagnostic variance and cultural factors may involve examining participants' attitudes toward and experiences with the diagnostic process within the constraints of managed care. This research direction in conjunction with replication studies may allow for optimal training experiences for culturally relevant and accurate diagnoses among helping professionals.

*References

American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders (4th ed., text rev.). Washington, DC: Author.

Block, C. B. (1984). Diagnostic and treatment issues for Black patients. The Clinical Psychologist, 37, 51-54.

Cochran, S. D., Sullivan, J. G., & Mays, V M. (2003). Prevalence of mental disorders, psychological distress, and mental health service use among lesbian, gay, and bisexual adults in the United States. Journal of Consulting and Clinical Psychology, 71, 53-61.

Corbin, J., & Strauss, A. (2008). Basics of qualitative research: Techniques and procedures for developing grounded theory (3rd ed.). Thousand Oaks, CA: Sage.

DelBello, M. P., Lopez-Larson, M. P., Soutullo, C. A., & Strakowski, S. M. (200 1 ). Effects of race on psychiatric diagnosis of hospitalized adolescents: A retrospective chart review. Journal of Child and Adolescent Psychopharmacology, 11, 95-103.

Endicott, J., Spitzer, L., Fleiss, J. L., & Cohen, J. ( 1 976). The Global Assessment Scale: A procedure for measuring overall severity of psychiatric disturbance. Archives of General Psychiatry, 33, 766-771.

Eriksen, K. P., & Kress, V E. (2005). Beyond the DSM story: Ethical quandaries, challenges and best practices. Thousand Oaks, CA: Sage.

Ford, M., & Widiger,T. (1989). Sex bias in the diagnosis of histrionic and antisocial personality disorder. Journal of Consulting and Clinical Psychology, 57, 301-305.

Friedlander, M. L., & Stockman, S. J. ( 1 983). Anchoring and publicity effects in clinical judgment. Journal of Clinical Psychology, 39, 637-643.

Gigerenzer, G (2002). Calculated risks: How to know when numbers deceive you. New York: Simon & Schuster.

Gushue, G. V (2004). Race, color-blind racial attitudes, and judgments about mental health: A shifting standards perspective. Journal of Counseling Psychology, 51, 398-407.

Hays, D. G, Chang, C. Y., & Decker, S. L. (2007). Initial development and validation of the Privilege and Oppression Inventory. Measurement and Evaluation in Counseling and Development, 40, 66-79.

Hays, D. G., McLeod, A. L., & Prosek, E. A. (2009). Diagnostic variance among counselors and counselor trainees. Measurement and Evaluation in Counseling and Development, 42, 3-14.

Jenkins-Hall, K., & Sacco, W P. (1991). Effect of client race and depression on evaluations by White therapists. Journal of Social and Clinical Psychology, 10, 322-333.

Jones, B., & Gray, B. (1986). Problems in diagnosing schizophrenia and affective disorders among Blacks. Hospital and Community Psychiatry, 37, 61-65.

Lin, K. M. (1996). Asian American perspectives. In J. E. Mezzich, A. Kleinman, H. Fabrega, & D. L. Parron (Eds.), Culture and psychiatric diagnosis: A DSM-IV perspective (pp. 35-38). Washington, DC: American Psychological Association.

Lopez, S. R. (1989). Patient variable biases in clinical judgment: Conceptual overview and methodological considerations. Psychological Bulletin, 106, 184-203.

Miles, M. B., & Huberman, A. M. (1994). Qualitative data analysis: An expanded sourcebook (2nd ed.). Thousand Oaks, CA: Sage.

Mukherjee, A., Shukla, S., Woodie, J., Rosen, A. M., & Olarte, S. (1983). Misdiagnosis of schizophrenia in bipolar patients: A multiethnic comparison. American Journal of Psychiatry, 140, 1571-1574.

Paniagua, F. A. (2005). Assessing and treating culturally diverse clients: A practical guide (3rd ed.). Thousand Oaks, CA: Sage.

Patton, M. Q. (2002). Qualitative research & evaluation methods (3rd ed.). Thousand Oaks, CA: Sage.

Rosenthal, D. A. (2004). Effects of client race on clinical judgment of practicing European American vocational rehabilitation counselors. Rehabilitation Counseling Bulletin, 47, 131-141.

Strakowski, S., Shelton, R. K., & Kolbrener, M. (1993). The effects of race and comorbidity on clinical diagnosis in patients with psychosis. Journal of Clinical Psychiatry, 54, 96-102.

Trierweiler, S. J., Muroff, J. R., Jackson, J. S., Neighbors, H. W., & Munday, C. (2005). Cultural Diversity and Ethnic Minority Psychology, 11, 351-364.

U.S. Census Bureau. (2000). The population profile of the United States: 2000 (Internet release). Retrieved November 29, 2007, from http:// www.census.gov/population/www/pop-profile/profile2000.html

Author affiliation:

Danica G. Hays and Elizabeth A. Prosek, both at Department of Counseling and Human Services, Old Dominion University; Amy L. McLeod, Department of Counseling, Argosy University, Atlanta. This study was funded by grants from Old Dominion University and the Association for Counselor Education and Supervision. The authors give special thanks to the following research assistants: Rebekah Byrd, Jessie Guyton, Breyan Haizlip, Donna Henry, Holly Moore, Desaree Murden, Tracy Roberts, and Rebecca Witcher. Correspondence concerning this article should be addressed to Dánica Q. Hays, Department of Counseling and Human Services, Old Dominion University, 110 Education Building, Norfolk, VA 23529 (e-mail: dhays@odu.edu).

Appendix:

Interview Protocol for Assessing Participants' Diagnoses of a Hypothetical Case

1 . What diagnosis or diagnoses would you give this client?

2. How would you summarize the symptoms used to arrive at your diagnosis or diagnoses?

3. Are there other diagnoses that could explain the client's symptoms?

4. What aspects of the case did you use to arrive at your diagnosis or diagnoses?

5. What cultural characteristics, if any, are important to this client's presenting problem? How so?

6. What cultural characteristics, if any, are important in your diagnostic decision? How so?

7. How would you describe this client's level of functioning? How, If at all, do you see the client's level of functioning changing with treatment?

8. What additional information or area of inquiry would have been helpful to you in arriving at a diagnosis? What might make it difficult to get this additional information?

9. As you weigh all the factors influencing your diagnostic decision, what was the single most important factor in your decision about the client?

10. What were your initial impressions of the client? To what degree did they weigh into your diagnostic decision?

11. Were there salient/intense aspects of the client's story? To what degree did they weigh into your diagnostic decision?