Therapist Ethnicity and Treatment Orientation Differences in Multicultural Counseling Competencies

Abstract

This study examined the relationship between therapist characteristics, therapeutic orientations, person-level and agency-level practices with cultural competency among 221 Los Angeles County community mental health clinicians. Results from an online survey indicated that compared to White therapists, ethnic minority therapists were more personally involved in communities of color, more likely to use a cultural framework in clinical practice, and perceived their agencies to be more culturally sensitive. Ethnic minority therapists also reported greater multicultural (MC) awareness and better MC counseling relationships with their clients than White therapists. Personal involvement in communities of color accounted for ethnic differences in MC awareness and MC counseling relationships. Compared to therapists with a strictly non-behavioral (psychodynamic or humanistic) orientation, therapists with an eclectic (or integrative) therapy orientation reported having a higher level of community knowledge. Therapists with an eclectic orientation reported greater MC awareness than therapists with a non-behavioral orientation, while both eclectic and behavioral (cognitive behavioral or behavior modification) therapists recounted better MC counseling relationships with their clients than therapists with a non-behavioral orientation. Community knowledge mediated eclectic vs. non-behavioral therapeutic orientation differences in MC awareness. Agency resources/linkages and outreach both moderated the relationship between therapeutic orientation and MC skills. Results suggest that if therapists become more personally involved with diverse populations, they will feel more culturally aware and feel like they have a better relationship with ethnic minority clients.

In 2001, the U.S. Surgeon General reported that 17% of Asian Americans with a psychiatric illness sought professional services, but less than 6% sought help from a mental health provider (U.S. Department of Health and Human Services, 2001a). The report highlighted similar trends in underutilization among African Americans, American Indians and Alaska Natives, and Hispanic Americans. Asian Americans and Pacific Islanders comprise approximately 4.9 percent of the U.S. population and were the fastest growing racial group in the U.S. between 2000 and 2010 (Humes, Jones, & Ramirez, 2010). Among ethnic populations however, Asian Americans and Pacific Islanders are the least likely to utilize mental health services (National Alliance on Mental Illness, 2003), even after age, gender, and geographical location are taken into account (Lu, 2002). The shortage of culturally competent personnel is one of many possible explanations for low utilization and high drop-out rates among Asian Americans (U.S. Department of Health and Human Services, 1998).

One strategy to address the underutilization of mental health services by Asian Americans is to consider different therapist factors that may contribute to more culturally competent services for Asian clientele. The vast majority of mental health service providers are White (87.5%), whereas only 3.6% are Hispanic, 2.7% are Black, 1.7% are Asian/Pacific Islander, and fewer than 1% are Native American (American Psychological Association Center for Workforce Studies, 2009). A mere 2% of social workers, 1.5% of psychologists, .01% of marriage and family therapists and 0% of psychiatric nurses are Asian American, Native Hawaiian or Pacific Islander (National Asian American Pacific Islander Mental Health Association, 2009). Moreover, few mental health service providers are fluent in Asian languages (National Alliance on Mental Illness, 2004). Due to relatively low levels of Asian American mental health service providers, many Asian American clients may be served by ethnically and linguistically mismatched mental health professionals. Therefore, it is imperative that we train and develop culturally competent practitioners to serve the mental health needs of the growing Asian American population.

Over the last decade, researchers and practitioners have urged the adoption of culturally competent, culturally responsive/sensitive, or multicultural skills when working with Asian Americans and other ethnic minority clients as a means to reduce mental health disparities (Brach & Fraser, 2000; S. Sue, 2006). Local, state, and federal agencies and organizations have adopted cultural competency guidelines specifying the necessary skills and practices therapists should develop in order to address the needs of ethnic minorities (American Psychological Association, 2003, 2006). As a result, public mental health agencies have long been encouraged to develop and implement services that are responsive to the needs of culturally diverse clientele (Segal, Bola, & Watson, 1996; Singleton-Bowie, 1995). For many individuals with severe mental illness, community mental health services are more accessible, tailored to specific needs, and effective than mental hospitals (World Health Organization, 2007). More than 18 percent of Asian Americans and Pacific Islanders are uninsured (DeNavas-Walt, Proctor, & Smith, 2010), therefore understanding the cultural competency attitudes of “front line” community clinicians who deliver public mental health care to underserved populations, regardless of their ability to pay, is essential to reduce mental health disparities. Despite increased efforts to prioritize and promote cultural competency, there continues to be a dearth of research on how therapist characteristics and practices (at both the individual and agency level) are related to multicultural counseling competencies, particularly among community mental health care providers.

Efforts to promote and implement culturally competent mental health care must progress from broad conceptual definitions to the articulation of specific strategies involving therapist attitudes, behaviors, and skills (S. Sue, 2006). D. W. Sue et al. (1982) proposed the most commonly used conceptualization of cultural competency, which consists of three parts: (1) awareness of one’s own biases, assumptions, and values, (2) knowledge about the cultural values and historical background of diverse populations, and (3) specific skills and counseling techniques that can be used to increase effectiveness when working with diverse clientele. A fourth domain, the therapeutic relationship between therapists and clients, was later identified by Sodowsky, Taffe, Gutkin, and Wise (1994). A significant amount of research has been conducted on this tripartite model, which remains the foundation of the most commonly-used cultural competency assessment scales [i.e., Multicultural Counseling Inventory (MCI; Sodowsky et al., 1994), Multicultural Awareness-Knowledge-Skills Survey (MAKSS: D’Andrea, Daniels, & Heck, 1991), Multicultural Counseling Awareness Scale – Form B (Ponterotto et al., 1996; Ponterotto, Sanchez, & Magids, 1990), Cross-Cultural Counseling Inventory-Revised (LaFromboise, Coleman, & Hernandez, 1991)] (Abreu, Chung & Atkinson, 2000). Despite these advances, little research examining multicultural counseling competency has been conducted on community mental health professionals, with most of the extant research conducted on graduate students, trainees, and university counseling center staff.

Factors Associated with Cultural Competency

A number of factors have been found to be associated with multicultural counseling competencies. Among university counseling center staff, there is some research to suggest that ethnic minority counselors evidence higher multicultural counseling competencies than White counselors (Pope-Davis & Ottavi, 1994; Sodowsky, Kuo-Jackson, Richardson, & Corey, 1998). Pope-Davis and Ottavi (1994) theorized that ethnic minority counselors may be more culturally competent due to their personal experiences as minorities. Moreover, counselors serving more ethnic minority clientele in their caseloads reported higher multicultural counseling competencies (Sodowsky, 1996; Sodowsky et al., 1998), providing additional support that experience makes a difference. Pope-Davis, Reynolds, Dings, and Ottavi (1993) found gender and age of university counseling staff and trainees correlated significantly with multicultural counseling competencies, while other studies indicate therapist age, gender, educational level, and experience are not strongly related to multicultural counseling competencies (Pope-Davis & Ottavi, 1994; Sodowsky, 1996).

Unfortunately, besides the examination of demographic factors, there continues to be a lack of research on other factors that could potentially be related to cultural competency (i.e., whether therapist’s theoretical orientation makes a difference). Research suggests that many therapists utilize an eclectic or integrative approach that combines a diverse range of key principles, concepts, and techniques from multiple theoretical orientations to fit the unique needs of clients (Norcross, 2005). Norcross and Karpiak’s 2010 survey of American Psychological Association’s Division of Clinical Psychology (Division 12) members found that cognitive and eclectic orientations were the two most popular theoretical orientations, at 31% and 22% respectively (Norcross & Karpiak, 2012). Although there is a widely held belief by experienced clinicians that an integrative or eclectic approach improves the effectiveness of psychotherapy (Wolfe, 2001), this assumption has yet to be supported or fully explored by empirical research (Schottenbauer, Glass, & Arnkoff, 2005). A related question is whether an eclectic or integrative approach is associated with greater multicultural counseling competencies. Asian Americans evidence positive therapeutic outcomes when engaged in culturally responsive and congruent treatment plans (U.S. Department of Health and Human Services, 1998), and an eclectic or integrative oriented therapist may be more likely to consider and discuss multiple concepts and counseling interventions in the development of a client’s treatment plan, thereby culturally-tailoring care to the client’s needs. The potential contribution of therapist theoretical orientation to multicultural counseling competencies is an important area to explore because many community mental health practitioners report utilizing an eclectic or integrative approach with their clients (Von Ranson & Robinson, 2006; Von Ranson, Wallace, & Stevenson, 2013).

Some researchers note that minimal empirical research has tested potential mediators and moderators of D. W. Sue et al.’s (1982) tripartite model (Atkinson & Israel, 2003). Therefore, another important question is whether personal behaviors and agency or institutional practices can explain or facilitate therapist cultural competency. A growing body of work suggests that personal experience and engagement can contribute to the knowledge, attitudes, and skills necessary for cultural competence (Astin, Voglegesang, Ikeda, & Yess, 2001; Ngai, 2006). Individual involvement in outreach programs can result in increased sensitivity to community concerns and strengthen one’s commitment to addressing them (Astin et al., 2001; Goodman et al., 2004; Ngai, 2006; Speight & Vera, 2004). Furthermore, if the mental health service agency promotes the adoption and awareness of culturally competent practices, will that improve the therapists’ ability to work with diverse clientele? There is some research to suggest that ethnic specific services (i.e., community agencies that specialize in treating specific ethnic minority groups) are more successful at treating ethnic minorities than mainstream clinics (Flaskerud & Hu, 1994; Lau & Zane, 2000; Takeuchi, Sue, & Yeh, 1995; Zane, Hatanaka, Park & Akutsu, 1994). An important agenda then is to identify why that is, to determine whether individual behaviors and organizational priorities can explain or enhance multicultural counseling competencies, and to identify which practices are most related to culturally competent care so that the use of these mechanisms can be encouraged at other public mental health clinics. For example, what about these therapists and clinics contributes to better outcomes? Is it the hiring of ethnically matched staff, offering of bilingual services, increased number of cultural competency trainings, community input and feedback, the culturally accepting environment, a self-selection of culturally minded therapists to such organizations, or a combination of factors and accommodations that improve culturally competent care (Siegal et al., 2011)?

Current Study

The goal of this study was to identify specific therapist characteristics, person-level behaviors, and agency-level factors associated with multicultural counseling competencies (skills, awareness, knowledge, and relationship) among community mental health practitioners. Much of the previous work on multicultural counseling competencies is based on university trainees and staff who may not be exposed to high rates of diverse underserved populations and low-income, uninsured clientele. Therefore, the generalizability of findings from previous research to community mental health providers is unclear. The current study builds upon previous literature by testing common demographic factors (e.g., ethnicity, gender, and age) and examining additional factors such as therapist experience, ethnic minority caseload, and theoretical orientation. Several person-level behaviors (community knowledge, personal involvement, clinical practice) and agency-level practices (resources/linkages, agency staffing, agency organizational climate, policy, and outreach) were tested as potential mediators and moderators in exploratory analyses to assess their contribution to cultural competency outcomes.

Method

Sample

Participants in this study included 221 mental health providers (170 women and 51 men) from Los Angeles County, California. The racial composition of therapists consisted of 15 (6.8%) African Americans, 37 (16.7%) Asian Americans/Pacific Islanders, 41 (18.6%) Hispanic/Latinos, 113 (51.1%) White Americans, and 15 (6.8%) therapists who reported being multi-ethnic or some other ethnicity not listed. Therapists ranged in age from 24 to 63 years (M = 35.42, SD = 8.97), with the majority born in the United States (73.8%). The education level included 106 (48%) with a Master’s degree, 108 (48.9%) with a Ph.D./Psy.D. or M.D., and 7 (3.2%) holding non-advanced degrees. The three most widely reported types of training were Psychology (48.0%), Marriage and Family Counseling (24.9%) and Social Work (23.1%). Therapists were classified into one of three therapeutic orientations based on the specific therapy modalities they reported using: behavioral (cognitive behavioral or behavior modification) only (n = 60; 27.15%), non-behavioral (psychodynamic or humanistic) only (n = 38; 17.19%), and eclectic (integrative or more than one therapeutic approach) (n = 123; 55.66%). On average, therapists reported having 4.97 (SD = 5.42) years of post-degree clinical experience and a current ethnic minority caseload of 86.98% (SD = 14.76).

Design and Procedures

Fliers and sign-up sheets for the study were mailed out to 168 mental health agencies providing outpatient care within Los Angeles County. Agencies contacted were either a Los Angeles County agency or an agency contracted by Los Angeles County to provide mental health services. Agencies providing inpatient day or residential treatment were not included in this study. A response was not expected from the majority of these agencies, however the initial attempt was made to be inclusive and thorough. Small agency units, agency units not adequately staffed, and non-functioning agency units did not respond, and the final sample came from 30 mental health agencies. Upon agreement from agency administrators, recruitment fliers were posted for their staff to view. The fliers included a brief description of the study and instructed interested individuals to provide their name, therapist ID code, and email address on the study sign-up sheets. Only therapists providing outpatient psychotherapy services and those with a caseload of at least 20% ethnic minority clientele were eligible to participate. Respondents were emailed the link to the online-survey and were each given a Study ID Code and password in order to log into the study web page. Therapists were informed that the survey would take about 40 minutes to one hour to complete and were encouraged to complete the survey in one sitting outside of work hours. They were paid $100 to participate in the study.

Measures

Demographics

The online-survey questionnaire collected demographic information, including the respondent’s self-reported ethnicity, age, gender, type of degree and training, therapy orientation [i.e., non-behavioral (psychodynamic or humanistic), behavioral (cognitive behavioral or behavior modification), or eclectic (integrative or selected more than one therapeutic orientation)], estimated current ethnic minority caseload (%), and amount of post-degree clinical experience.

Person-level behaviors and agency-level practices

Mason’s (1995) Cultural Competence Self-Assessment Questionnaire (CCSAQ) was developed based on various facets of the Child and Adolescent Service System Program Cultural Competence Model (Cross, Bazron, Dennis, & Isaacs, 1989) and assesses agency and provider cultural competence tendencies along four dimensions: attitude, practice, policy, and structure. The CCSAQ has been used in service delivery programs across the country and is consistent with national standards on Culturally and Linguistically Appropriate Services (CLAS; U.S. Department of Health and Human Services, 2001b). Using a 4-point scale, respondents indicate the degree to which items describe their beliefs (1 = not at all, 4 = very well). Higher average scores represent higher levels of perceived cultural competence. The CCSAQ consists of 8 subscales: Community Knowledge (29 items; e.g., How well are you able to describe the communities of color in your service area?), Personal Involvement (9 items; e.g., Do you interact socially with people of color in your service area?), Resources and Linkages (25 items; e.g., Does your agency have linkages with advocates for communities of color who can provide reliable information regarding community opinions about diverse and important issues?), Staffing (12 items; e.g., Are there people of color represented in direct service positions?), Organizational Climate (9 items; e.g., Does your agency provide training that helps staff work with people of color?), Clinical Practice (19 items; e.g., How well do you use cultural strengths and resources when planning services to clients of color?), Policy (15 items; e.g., As a matter of policy does your agency specifically consider culture in service plans?), and Outreach (9 items; e.g., Does your agency participate in cultural, political, religious, or other events or festivals sponsored by communities of color?).

Mason (1995) conducted an extensive review of relevant research- and theory-based literature, and held focus group consultations with experts to define the CCSAQ subscales, item content identification, and item wording refinement, to ensure the questionnaire is a valid measure of cultural competence. Psychometric analysis on the CCSAQ established that all subscales of the CCSAQ had reliability coefficient alphas at or above .80 with the exception of personal involvement, which has an average reported reliability coefficient of .60 (Mason, 1995). Cronbach’s alphas for each subscale ranged from .79 (Personal Involvement) to .93 (Community Knowledge) in a previous study (Chow, 2008), indicating good to excellent internal consistency. Reliability coefficients for each of the subscales in this study were α = .94 (Community Knowledge), α = .79 (Personal Involvement), α = .92 (Resources and Linkages), α = .86 (Staffing), α = .85 (Organizational Climate), α = .77 (Clinical Practice), α = .85 (Policy), and α = .81 (Outreach), respectively.

Therapist multicultural counseling competencies

The Multicultural Counseling Inventory (MCI; Sodowsky et al., 1994) is a 40-item scale that assesses therapists’ multicultural counseling competencies. Using a 4-point Likert scale, therapists report the degree to which the items accurately describe their work as mental health care providers (1 = very accurate, 4 = very inaccurate). The MCI consists of four subscales: Multicultural Awareness (10 items; e.g., When working with minority clients, I have experience at solving problems in unfamiliar settings), Multicultural Counseling Knowledge (11 items; e.g., When working with minority clients, I use innovative concepts and treatment methods), Multicultural Counseling Skills (11 items; e.g., When working with minority clients, I form effective working relationship with the clients), and Multicultural Counseling Relationship (8 items; e.g., When working with minority clients, I perceive that my race causes the clients to mistrust me). Once the appropriate items are recoded and summed, higher scores indicate greater multicultural competence. The MCI total and subscale scores have produced good internal consistency in previous research (from .68 to .87; Constantine & Ladany, 2001). Ponterotto and Alexander (1996) found that the four subscales of the MCI were reliable across two studies: Multicultural Awareness (.83 and .81), Multicultural Counseling Knowledge (.79 and .78), Multicultural Counseling Skills (.83 and .81), and Multicultural Counseling Relationship (.71 and .72) (p. 665). Sodowsky et al. (1994) established the content validity through expert evaluation of item clarity and rater’s classification of items into the correct subscale categories.

Research demonstrates students and counselors with greater multicultural training or more experience score higher on the MCI (Ponterotto & Alexander, 1996; Sodowsky et al., 1994), providing support for the criterion validity of the scale. Both exploratory and confirmatory factor analyses support an oblique four-factor model, providing moderate support for the construct validity of the MCI (Sodowsky et al., 1994; Sodowsky, 1996). The MCI total score correlates substantially (from .61 to .73) with other multicultural competency measures. Although multicultural counseling competency measures are prone to social desirability (Constantine & Ladany, 2000; Pope-Davis & Dings, 1995; Sue, 1996), MCI total and subscale scores have been found to be insignificantly correlated with social desirability (Constantine & Ladany, 2000). While the Relationship subscale has been found to be correlated with social desirability, the scale is newer and less research has been conducted on its reliability and validity. In the current study, the subscales were internally consistent, with an alpha of .63 for Awareness, .76 for Knowledge, .77 for Skills and .59 for Relationship.

Social desirability

Earlier studies by Sodowsky et al. (1994) and Ponterotto et al. (1996) found non-significant relationships between multicultural counseling competence scales and measures of social desirability, yet more recent research recommends that social desirability be controlled for when examining multicultural counseling competencies (Constantine & Ladany, 2000; Sodowsky et al., 1998). The Balanced Inventory of Desirable Responding (BIDR; Paulhus, 1991) evaluates social desirability along two subscales: Self-Deceptive Enhancement (e.g., My first impressions of people usually turn out to be right), and Impression Management (e.g., I always obey laws, even if I’m unlikely to get caught). Self-deception involves denying self-threatening thoughts, whereas impression management involves conscious attempts to deceive others. Individuals respond using 7-point scales (1 = not true, 7 = very true). The Self-Deception and Impression Management scales each have 20 items and yielded alpha coefficients of .77 and .83, respectively, in our sample. The validity and reliability of this instrument have been demonstrated in numerous studies (see Linden, Paulhus, & Dobson, 1986).

Data Analysis

Exploratory factor analysis

Close examination of the CCSAQ measure revealed that some subscales asked respondents to evaluate their individual behaviors, while other subscales inquired more specifically about agency practices. To test for the presence of underlying higher-order factors, an exploratory factor analysis (EFA) was performed on the 8 CCSAQ subscales, using principal axis analysis as the method of extraction with Promax rotation. The Scree Test and the corresponding eigenvalues were used to verify the number of retained factors (Costello & Osborne, 2005). Results from the EFA are presented in Table 1. The eight subscales of the CCSAQ yielded two factors, explaining 67.08% of the variance. The CCSAQ factor structure is composed of: (1) Agency-level practices factor (α = .85), composed of five subscales representing institutional resources/linkages, staffing, organizational climate, policy, and outreach, and (2) Person-level behaviors factor (α = .78), composed of three subscales representing individual community knowledge, personal involvement, and clinical practice. For theoretical purposes, person-level behaviors were examined as potential mediators while agency-level practices were considered in moderator analyses.

Table 1.

Summary of exploratory factor analysis results for CCSAQ subscales using Promax rotation (n = 221)

CCSAQ Subscale	Factor loadings
	1	2
Community Knowledge	−.12	.85
Personal Involvement	−.04	.79
Clinical Practice	.19	.60
Organizational Climate	.84	−.09
Resources and Linkages	.78	.11
Outreach	.71	.14
Policy	.71	−.06
Staffing	.63	−.06
Eigenvalues	3.98	1.38
% of variance	49.80	17.28

KMO = .829, χ² = 775.510, df = 28, p < .001; 1—agency-level practices; 2—person-level behaviors. Bold values indicate high loading to the factor.

Descriptive analysis

A series of one-way ANOVAs were conducted to assess the relationship between therapist demographic variables, treatment orientation (behavioral, non-behavioral, or eclectic), person-level behaviors (community knowledge, personal involvement, and clinical practice), agency-level practices (resources/linkages, agency staffing, agency organizational climate, policy, and outreach), and four types of multicultural counseling competencies (awareness, knowledge, skills, and relationship). If the multivariate analyses revealed significant effects, Bonferroni corrections were utilized in post-hoc tests to guard against inflated Type I error rates.

Multiple mediation analysis

Multiple mediation analyses were then conducted to identify person-level behaviors that may contribute to culturally competent care. We estimated direct and indirect effects of the multiple mediators (community knowledge, personal involvement, and clinical practice) using the non-parametric bootstrapping procedure recommended by Preacher and Hayes (2008). Multiple mediation not only determines if an indirect effect exists, but it also offers information on how to separate out the distinct mediating effects of several potential mediators with overlapping content (West & Aiken, 1997). This statistical procedure has several advantages: (1) it allows multiple mediators to be tested simultaneously; (2) it allows for non-normality of the sampling distribution; (3) it has higher power than the Sobel test and other traditional tests of mediation; and (4) it reduces Type 1 error because fewer inferential tests are required. The inclusion of multiple mediators also tests the relative magnitudes of the specific indirect effects of all mediators within a single model. Analyses were conducted using the SPSS macros provided by Preacher and Hayes (2008).

To thoroughly evaluate the three treatment orientations in multiple mediation analyses, we created two dichotomous dummy-coded variables to compare eclectic vs. non-behavioral (0 = Non-behavioral; 1 = Eclectic) and behavioral vs. non-behavioral (0 = Non-behavioral; 1 = Behavioral) orientations. The creation of the dichotomous variables permitted us to conduct separate analyses comparing various groups of interest (eclectic vs. non-behavioral, behavioral vs. non-behavioral orientations) while discarding or dropping the remaining data. Although eclectic and behavioral therapists did not significantly differ from each other on study variables, combining the two orientations obscured the study’s findings therefore we did not merge the eclectic and behavioral treatment orientations in our analyses.

Moderator analysis

Agency-level practices were considered in exploratory moderator analyses. A moderated regression framework (Aiken & West, 1991; Baron & Kenny, 1986) was used to examine whether the effect of therapist factors on multicultural counseling competencies varied as a function of agency-level practices.

Results

Correlational and Descriptive Analyses

Correlations between therapist demographics, person-level behaviors, agency-level practices, and multicultural counseling competencies are displayed in Table 2. Being an ethnic minority (non-White) therapist was positively correlated with personal involvement in communities of color and the use of a cultural framework in clinical practice. Ethnic minority status was also positively correlated with employment at an agency with culturally sensitive policies, multicultural awareness, and multicultural counseling relationships. An eclectic treatment orientation (compared to a non-behavioral orientation) was positively correlated with knowledge about communities of color, multicultural awareness, and multicultural counseling relationships. A behavioral treatment orientation (compared to a non-behavioral orientation) was positively correlated with multicultural counseling relationships. Ethnicity and therapeutic orientation were significantly correlated with multicultural counseling competencies, therefore they were considered for further analyses.

Table 2.

Bivariate Correlations among Study Variables

Variable	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16	17	18	19
1. Ethnic Min. (vs. White)	-	.03	.21**	.22*	.12	.18**	.11	.11	.25***	.14*	.08	.03	−.04	.15*	.12	.30***	.02	−.10	.20**
2. Female (vs. Male)		-	.14	.28**	.23**	.01	.06	.08	.06	−.01	−.07	−.00	−.09	−.07	−.11	.14*	.10	.06	.02
3. Eclectic (vs. Non-Beh)			-	n/a	.18*	.10	−.02	.21**	.10	.14	.00	−.08	.02	.03	−.03	.19*	.07	.12	.25**
4. Behavioral (vs. Non-Beh)				-	.26*	.20	.05	.08	.04	−.01	−.12	−.19	−.06	−.07	−.17	.13	−.01	.09	.25*
5. % Minority Cases					-	.00	.13	.14*	−.01	.03	−.02	.00	−.10	−.06	−.04	.13	.01	−.02	.06
6. Self-Deception						-	.40***	.22**	.25***	.13*	.09	.03	.09	.16*	.04	.21**	.10	.23**	.27***
7. Impression Management							-	.12	.01	.02	.03	−.06	.03	−.02	.01	.03	.01	.09	.15*
8. Community Knowledge								-	.60***	.54***	.38***	.10	.25***	.21**	.32***	.51***	.33***	.36***	.33***
9. Personal Involvement									-	.54***	.41***	.25***	.23***	.20**	.37***	.54***	.22**	.31***	.34***
10. Clinical Practice										-	.42***	.31***	.39***	.31***	.50***	.52***	.46***	.38***	.20**
11. Resources & Linkage											-	.50***	.65***	.59***	.68***	.29***	.17*	.13	.09
12. Staffing												-	.55***	.30***	.47***	.17*	.11	.00	.03
13. Organizational Climate													-	.56***	.54***	.23***	.12	.08	−.01
14. Policy														-	.56***	.27***	.12	.01	−.06
15. Outreach															-	.25***	.23***	.10	.04
16. MC Awareness																-	.48***	.48***	.37***
17. MC Counseling Knowledge																	-	.54***	.23**
18. MC Counseling Skills																		-	.30***
19. MC Counseling Relationship																			-

Note. MC = Multicultural.

^*p < .05.

^**p < .01.

^***p < .001. Therapist age and degree (PhD vs. Masters) were not significantly correlated with any of the study variables and have been excluded from the above table.

Results from the one-way ANOVAs comparing ethnic and treatment orientation groups on person-level behaviors, agency-level practices, and multicultural counseling competencies are found in Table 3. Ethnic minority therapists were more personally involved in communities of color in their service area (p < .001, d = .50), more likely to use a cultural framework in their clinical practice (p < .05, d = .29), and perceived their agency to be adopting more culturally sensitive policies (p < .05, d = .30) than White therapists. Ethnic therapists also had greater multicultural awareness (p < .001, d = .62) and better multicultural counseling relationships with their clients (p < .01, d = .42) than White therapists. No ethnic differences were found with respect to multicultural counseling knowledge and skills, community knowledge, resources and linkages, staffing, organizational climate, or agency outreach.

Table 3.

Ethnic and Therapy Orientation Comparisons on Person/Agency-Level Factors and Cultural Competency Outcome Measures

	Ethnic Group			Therapy Orientation
Variable	White (n=113)	Ethnic Minority (n=108)	F	Behavioral (n=60)	Non-Behavioral (n=38)	Eclectic (n=123)	F
	Mean (SD)	Mean (SD)		Mean (SD)	Mean (SD)	Mean (SD)
Control Variables
Female (vs. Male)	76.10%	78.70%	.21	86.70%a	63.20%b	77.20%ab	3.75*
Age	36.42 (9.64)	34.38 (8.15)	2.82	34.52 (9.14)	35.35 (8.11)	35.87 (9.18)	.44
PhD (vs. Masters)	54.46%	46.08%	1.50	49.12%	47.37%	52.10%	.16
% Ethnic Minority Caseload	85.31 (15.37)	88.73 (13.95)	3.00	88.98a (13.07)	81.34b (16.00)	87.75ab (14.83)	3.58*
Self-Deception	88.84 (15.30)	94.44 (16.40)	6.88**	94.15 (14.69)	87.76 (17.20)	91.50 (16.22)	1.86
Impression Management	85.50 (21.30)	90.06 (21.85)	2.47	89.83 (23.70)	87.63 (23.49)	86.74 (20.06)	0.41
Person-Level Behaviors
Community Knowledge	2.57 (0.40)	2.66 (0.44)	2.64	2.54ab (0.41)	2.48b (0.42)	2.69a (0.41)	5.00**
Personal Involvement	2.36 (0.54)	2.63 (0.54)	14.05***	2.46 (0.57)	2.41 (0.52)	2.54 (0.56)	0.95
Clinical Practice	2.71 (0.34)	2.81 (0.36)	4.27*	2.69 (0.36)ns	2.70 (0.39)ns	2.81 (0.33)ns	3.07*
Agency-Level Practices
Resource & Linkages	2.66 (0.53)	2.74 (0.56)	1.34	2.59 (0.56)	2.73 (0.59)	2.74 (0.53)	1.52
Staffing	3.20 (0.47)	3.23 (0.54)	.22	3.13 (0.52)	3.32 (0.50)	3.23 (0.49)	1.87
Organizational Climate	2.84 (0.59)	2.79 (0.65)	.31	2.74 (0.58)	2.82 (0.58)	2.85 (0.65)	0.58
Policy	2.37 (0.64)	2.58 (0.75)	4.93*	2.37 (0.68)	2.47 (0.79)	2.53 (0.68)	1.07
Outreach	2.41 (0.54)	2.55 (0.60)	3.42	2.36 (0.56)	2.56 (0.59)	2.51 (0.57)	1.81
Cultural Competency Outcome Variables
MC Awareness	30.55 (3.91)	32.89 (3.62)	21.28***	31.45ab (3.51)	30.42b (4.43)	32.20a (3.91)	3.19*
MC Counseling Knowledge	36.13 (4.27)	36.30 (4.29)	.08	35.77 (4.27)	35.84 (4.97)	36.54 (4.04)	0.84
MC Counseling Skills	39.04 (3.40)	38.34 (3.64)	2.14	38.63 (3.58)	37.97 (4.02)	38.95 (3.34)	1.13
MC Counseling Relationship	24.99 (3.08)	26.31 (3.25)	9.52**	25.85a (3.20)	24.11b (3.62)	26.00a (2.99)	5.40**

Note. MC = Multicultural. Means in the same row having the same superscript are not significantly different at p < .05 Bonferroni post-hoc tests.

^*p < .05.

^**p < .01.

^***p < .001.

Superscript ns = ended up not significantly different in Bonferroni post-hoc tests.

Therapists with an eclectic orientation had significantly higher community knowledge (p < .01, d = .51) than therapists with a strictly non-behavioral (psychodynamic or humanistic) orientation. Therapists with a behavioral orientation had a significantly higher percentage of minority clients in their caseload (p < .05, d = .52) than therapists with a non-behavioral orientation. Both behaviorally-oriented and eclectic therapists reported better multicultural counseling relationships with their clients (p < .01, d = .51 (beh)/.57 (ecl)) than therapists with a non-behavioral orientation. Therapists with an eclectic orientation also had greater multicultural awareness (p < .05, d = .43) than therapists with a non-behavioral orientation. No treatment orientation differences were found on measures of multicultural knowledge and skills, personal involvement in communities of color, clinical practice, or on any of the agency-level practices.

Multiple Mediational Analyses

We conducted multiple meditation analyses to assess person-level behaviors that may explain ethnic and treatment orientation differences in multicultural awareness and multicultural counseling relationships, the findings of which are presented below. Both facets of social desirability (self-deception and impression management) were positively correlated with multicultural counseling competency outcomes in our study (see bivariate correlations in Table 2). Consistent with recommendations by Constantine and Ladany (2000) as well as Sodowsky et al. (1998), we controlled for both facets of social desirability in all analyses.

Ethnic differences in multicultural awareness

In the multiple mediational analyses examining ethnic differences in multicultural awareness, multicultural awareness was entered as the dependent variable (DV), ethnicity (0 = White; 1 = Ethnic Minority) was entered as the independent variable (IV), and the three person-level behaviors (community knowledge, personal involvement, and clinical practice) were entered as potential mediators. To isolate the effect of ethnicity above and beyond other therapist variables previously considered in the literature on multicultural counseling competencies, gender, treatment orientation, and socially desirable responding were entered as control variables.

The bootstrap analysis indicated that personal involvement in communities of color partially mediated the relationship between ethnic minority status and multicultural awareness, even after controlling for all other potential mediators. Ethnic minority therapists were more personally involved in the communities of color they serve (p < .01), and individuals with more community involvement tended to have more multicultural awareness (p < .01). Community knowledge and clinical practice did not add to the overall model once all the potential mediators were taken into account. Ethnic minority status continued to have a statistically significant influence on multicultural awareness (p < .01) even with the mediators in the model. The results of the bootstrap tests showed that the total effect of ethnic minority status on multicultural awareness was attenuated but remained significant when the mediators were included in the model. Point estimates and bias-corrected 95% confidence intervals are provided in Table 4.

Table 4.

Multiple Mediation of Indirect Effects of Ethnic Minority Status on Multicultural Awareness and Multicultural Counseling Relationship through Person-Level Community Knowledge, Personal Involvement, and Clinical Practice (5,000 bootstrap samples)

Multiple Indirect Effect	MC Awareness			MC Counseling Relationship
	Point Estimate	95% CI		Point Estimate	95% CI
		Lower	Upper		Lower	Upper
Community Knowledge	.092	−.094	.370	.072	−.060	.349
Personal Involvement	.349*	.091	.772	.236*	.035	.560
Clinical Practice	.225	−.013	.607	−.029	−.220	.056
Total	.666	.045	1.277	.279	−.008	.626

Note. MC = Multicultural. Confidence Intervals (CIs) that do not include zero indicate a significant indirect effect (and are marked with an asterisk).

Ethnic differences in multicultural counseling relationship

In the multiple mediational analyses examining ethnic differences in multicultural counseling relationship, multicultural counseling relationship was entered as the DV, ethnicity (0 = White; 1 = Ethnic Minority) was entered as the IV, and the three person-level behaviors (community knowledge, personal involvement, and clinical practice) were entered as potential mediators. To isolate the effect of ethnicity above and beyond other therapist variables previously considered in the literature on multicultural counseling competencies, gender, treatment orientation, and socially desirable responding were entered as control variables.

After controlling for all potential mediators, bootstrap analyses indicated there was an indirect effect of personal involvement in communities of color on the relationship between ethnic minority status and multicultural counseling relationship. Ethnic minority therapists were more personally involved in the communities of color they serve (p < .01), and individuals with more community involvement tended to have better multicultural counseling relationships (p < .05). In contrast, community knowledge and clinical practice did not add to the overall model once all other potential mediators were taken into account. With the potential mediators in the model, the direct effect of ethnic minority status on multicultural counseling relationship was non-significant (p = .08). Point estimates and bias-corrected 95% confidence intervals are provided in Table 4.

Treatment orientation differences in multicultural awareness

In the multiple mediational analyses examining treatment orientation differences between eclectic and non-behavioral therapists in multicultural awareness, multicultural awareness was entered as the DV, eclectic orientation (0 = Non-behavioral; 1 = Eclectic) was entered as the IV, and the three person-level behaviors (community knowledge, personal involvement, and clinical practice) were entered as potential mediators. To isolate the effect of an eclectic (vs. non-behavioral) orientation above and beyond other therapist variables previously considered in the literature on multicultural counseling competencies, ethnicity, gender, and socially desirable responding were entered as control variables.

Bootstrap analyses indicated there was an indirect effect of knowledge about local communities of color on the relationship between having an eclectic treatment orientation and multicultural awareness, even after controlling for all other mediators. Therapists with an eclectic treatment orientation were more knowledgeable about the communities of color they served (p < .05), and individuals who were more knowledgeable tended to have higher multicultural awareness (p < .01). Personal involvement and clinical practice did not add to the overall model once all other potential mediators were taken into account. Point estimates and bias-corrected 95% confidence intervals are provided in Table 5.

Table 5.

Multiple Mediation of Indirect Effects of Specific Treatment Orientations on Multicultural Awareness and Multicultural Counseling Relationship through Person-Level Community Knowledge, Personal Involvement, and Clinical Practice (5,000 bootstrap samples)

Multiple Indirect Effect	MC Awareness			MC Counseling Relationship
	Point Estimate	95% CI		Point Estimate	95% CI
		Lower	Upper		Lower	Upper
Eclectic vs. Non-Behavioral
Community Knowledge	.403*	.042	1.012	.162	−.040	.621
Personal Involvement	−.006	−.433	.404	−.005	−.367	.381
Clinical Practice	.212	−.085	.751	−.057	−.415	.053
Total	.609	−.307	1.55	.100	−.400	.622
Behavioral vs. Non-Behavioral
Community Knowledge	-	-	-	.101	−.285	.723
Personal Involvement	-	-	-	.001	−.255	.271
Clinical Practice	-	-	-	.014	−.089	.312
Total	-	-	-	.115	−.464	.708

Note. MC = Multicultural. Confidence Intervals (CIs) that do not include zero indicate a significant indirect effect (and are marked with an asterisk).

Treatment orientation differences in multicultural counseling relationship

Two separate multiple mediational analyses examined treatment orientation differences (eclectic vs. non-behavioral and behavioral vs. non-behavioral) in multicultural counseling relationship. Multicultural counseling relationship was entered as the DV, treatment orientation was entered as the IV, and the three person-level behaviors (community knowledge, personal involvement, and clinical practice) were entered as potential mediators. To isolate the effect of treatment orientation above and beyond other therapist variables previously considered in the literature on multicultural counseling competencies, ethnicity, gender, and socially desirable responding were entered as control variables. After controlling for all potential mediators, bootstrap analyses indicated no significant mediators of the effect between treatment orientation and multicultural counseling relationship. Community knowledge, personal involvement, and clinical practice did not add to the overall model in either of the two theoretical orientation comparisons. Point estimates and bias-corrected 95% confidence intervals are provided in Table 5.

Exploratory Moderator Analyses

A moderated regression framework (Aiken & West, 1991) was used to examine whether the effect of therapist ethnicity or treatment orientation on multicultural counseling competencies varied as a function of agency practices. All main effects were included in the moderated multiple regression analyses, and separate hierarchical analyses were used to contrast eclectic vs. non-behavioral and behavioral vs. non-behavioral treatment orientations. In our statistical model, we mean-centered and created interaction terms to assess the moderating effect of five agency practices (resources/linkages, staff, organizational climate, policy, and outreach) on the association between therapist ethnicity or specific treatment orientations and the four multicultural counseling competencies (awareness, knowledge, skills, relationship). In hierarchical multiple regression analyses, predictor variables were entered first (Step 1) and followed by product terms in the model (Step 2).

Agency-level practices did not moderate the effect of therapist ethnicity on any of the four multicultural counseling competences. Moreover, the effect of treatment orientation on multicultural awareness, multicultural counseling knowledge, and multicultural counseling relationship did not vary as a function of agency-level practices. However, the effect of treatment orientation (eclectic vs. non-behavioral and behavioral vs. non-behavioral) on multicultural counseling skills was moderated by agency resources/linkages and outreach.

Results show that treatment orientation and agency-level practices did not independently predict multicultural counseling skills (eclectic vs. non-behavioral, ΔR² = .08, ns.; behavioral vs. non-behavioral, ΔR² = .05, ns) (Table 6). Agency resources and linkages moderated the effect of specific treatment orientations (eclectic vs. non-behavioral β = −.60, p < .05; behavioral vs. non-behavioral β = −.69, p < .05) on multicultural counseling skills; therapists with non-behavioral orientations reported greater multicultural counseling skills when their agency had more resources and linkages to the communities they serve (Figure 1). Non-behaviorally oriented therapists also reported better multicultural counseling skills when their agency engaged in fewer outreach activities (eclectic vs. non-behavioral β = .61, p < .05; behavioral vs. non-behavioral β = .71, p < .01) (Figure 2). All other treatment orientation by agency-level moderator effects were non-significant.

Table 6.

Hierarchical Regression Analyses Predicting Multicultural Counseling Skills

	Eclectic (vs. Non-Behavioral)				Behavioral (vs. Non-Behavioral)
Step and Variable	B	SE B	β	ΔR2	B	SE B	β	ΔR2
Step 1				.08				.05
Therapeutic Orientation (TherOr)	.93	.65	.11		.60	.80	.08
Resources and Linkages	1.49	.80	.23		1.53	1.17	.24
Staff	−.52	.69	−.07		−1.08	.99	−.15
Organizational Climate	.82	.64	.15		−.31	1.00	−.05
Policy	−.51	.54	−.10		−.88	.68	−.17
Outreach	−.02	.70	−.00		.27	1.03	.04
Step 2				.05				.17**
TherOr × Resources and Linkages	−4.573	2.15	−.60*		−5.84	2.42	−.69*
TherOr × Staff	2.22	1.81	.27		1.84	2.03	.20
TherOr × Organizational Climate	−1.605	1.77	−.26		−4.05	2.03	−.49
TherOr × Policy	.16	1.16	.03		−.27	1.30	−.04
TherOr × Outreach	4.35	1.81	.61*		6.05	2.07	.71**

Note. This table represents results from two separate hierarchical regression analyses, one for each therapeutic orientation comparison (eclectic vs. non-behavioral and behavioral vs. non-behavioral).

^*p < .05.

^**p < .01.

^***p < .001.

Figure 1.

Moderating effect of agency-level resources/linkages on the relationship between therapeutic orientation and multicultural counseling skills. * p < .05. ** p < .01. *** p < .001.

Figure 2.

Moderating effect of agency-level outreach on the relationship between therapeutic orientation and multicultural counseling skills. * p < .05. ** p < .01. *** p < .001.

In sum, the current analyses yielded six key findings to the study of therapist factors, person-level behaviors, and agency-level practices related to multicultural counseling competencies. (1) Ethnic minority therapists had more multicultural awareness and better multicultural counseling relationships than White therapists and (2) these differences were partially accounted for by more personal involvement in communities of color in their service area, as reported by ethnic minority therapists. (3) Therapists with an eclectic or integrated treatment orientation had more multicultural awareness and better multicultural counseling relationships than therapists from a strictly non-behavioral (psychodynamic or humanistic) treatment approach. (4) Differences in multicultural awareness between the eclectic and non-behavioral therapists were partially explained by more knowledge about local communities of color, as reported by therapists with an eclectic approach. (5) Therapists with a behavioral treatment orientation also reported better multicultural counseling relationships than therapists with a non-behavioral approach, however there were no significant mediators of any treatment orientation differences in multicultural counseling relationships. (6) Agency-level practices moderated the effect of treatment orientation on multicultural counseling skills. Multicultural counseling skills were higher among non-behavioral therapists reporting more agency resources/linkages and less outreach.

Discussion

Amongst the cultural competency training movement, we cannot underestimate the importance of therapist characteristics that may lead to greater cultural competency, particularly among mental health service providers in the public sector who often deliver care to uninsured and underserved individuals. The identification of therapist factors related to culturally competent personnel offers one strategy to address the needs of Asian Americans and Pacific Islanders who greatly underutilize mental health services (National Alliance on Mental Illness, 2003). The study’s findings indicate that therapist ethnicity and therapeutic treatment orientation are associated with multicultural counseling competencies among community mental health professionals, particularly in the areas of multicultural awareness and multicultural counseling relationships.

First, ethnic minority therapists reported increased levels of multicultural awareness and multicultural counseling relationships relative to White therapists. This finding was mediated by more community involvement by ethnic minority therapists, supporting previous research that proposes a developmental process by which providers become more culturally competent through personal experiences and encounters with others at the individual, family, or community level (Kim-Godwin, Clarke, & Barton, 2001). This supports recommendations by Sue, Arredondo, and McDavis (1992) that culturally skilled counselors engage with minority individuals outside the counseling setting, through community or cultural events, social and political functions, and celebrations. These experiences broaden the counselors’ perspective of minority issues beyond a strictly academic or therapeutic perspective. Researchers have also highlighted the importance of community involvement to those receiving services (Siegel, Haugland, Reid-Rose, & Hopper, 2011). Ethnic minority therapists may be more likely to belong to the local community of color than their White therapist colleagues and thus may be more familiar with community factors (e.g., traditions, roles, family structures, hierarchies, values, beliefs) and comfortable interacting with other persons of color. For example, an Asian American therapist may be more likely to reside within the local Asian American community they serve and participate in ethnic festivals or other community events pertaining to their cultural heritage. Pre-existing cultural, social, and personal ties to a particular community of color may contribute to higher levels of involvement within these communities. Thus, it is unclear whether the ethnic minority therapists in our sample were simply more personally involved in and connected to the communities of color they already belong to, or if they spend more time seeking out diverse sociocultural experiences in multiple communities of color than their White colleagues.

In addition to the therapist’s ethnicity, treatment orientation emerged as relevant to multicultural competencies. Therapists using an eclectic treatment orientation reported more multicultural awareness, while both eclectic- and behaviorally-oriented therapists felt they had better multicultural counseling relationships with their clients than those from a strictly non-behavioral treatment approach. Individuals with a more inclusive and diverse treatment orientation may be better prepared to treat individuals from backgrounds with differing attitudes and views about treatment. A pluralistic approach facilitates a collaborative discussion around client preferences, with the end goal of designing a client-specific therapeutic experience (Cooper & McLeod, 2011). Cognitive behavioral therapy (CBT) is a widely used evidence-based approach with substantial evidence supporting its effectiveness (David, Szentagotal, Eva, & Macavei, 2005) that also encourages therapists to adapt interventions to the individual needs of the client. Apart from the therapeutic session itself, clients may also have preferences for the frequency and location of sessions (Mohr et al., 2006). The development of an effective working alliance is particularly important at the start of therapy, when clients actively assess their therapists’ ability to accept, understand and appreciate the client’s life experience (Ward, 2005).

The relationship between eclectic treatment orientation and multicultural awareness was partially due to an indirect effect of greater knowledge about local communities of color. Therapists with an eclectic approach may use a unique combination of therapeutic techniques reflective of their knowledge about specific factors that affect the communities of color they serve, to best treat their clients. Culturally competent counselors should be aware of relevant discriminatory practices and biases at the social and community level that may affect the mental health of the population they serve (Sue et al., 1992) and the treatment approach they plan to take with a client. This suggests that providing therapists with pertinent information about the minority communities they serve may result in more tailored and appropriate therapeutic techniques and in turn, increased multicultural counseling competencies. For example, therapists working with Asian clients should become aware of and understand the significant heterogeneity among Asian Americans. Certain Asian ethnic groups may have disproportionally low educational attainment (Cambodian, Laotian), limited English proficiency (Cambodian, Chinese, Korean, Taiwanese, Vietnamese), and high rates of poverty (Cambodian, Laotian) (Chang et al., 2010). Knowledge of these variations and an awareness of other ethnic-specific factors can help therapists tailor a treatment plan to the specific needs of their client. Agencies should consider promoting and teaching an in-depth understanding of the communities they serve, for many cultural variables (i.e., gender roles, power distance, face concerns, race relations, politics, religion) may be related to how their clients benefit from the treatment process.

Overall, these results identify potential strategies to improve multicultural counseling competencies that can be applied to therapists from all backgrounds who serve ethnic minorities. For example, the person-level behaviors of increasing personal involvement in Asian communities and cultivating knowledge about Asian communities in one’s service area may contribute to more culturally competent attitudes and practices among therapists working with Asian clients. Mental health service providers and agencies should consider possible avenues of strengthening their knowledge of and engagement in the local communities of color they serve. From the client perspective, Asian Americans may want to actively seek out therapists and agencies that are greatly invested in and integrated into the ethnic communities they serve. Findings from this study suggest that therapists with a greater understanding of an ethnic community, its people, and the cultural factors influencing well-being within that community may be more likely to provide culturally competent care, which can lead to greater client satisfaction and reduced drop-out rates.

It should be noted that the current study did not find ethnic or treatment orientation differences in multicultural counseling skills or knowledge. However, exploratory moderator analyses offer insight into specific agency-level practices that could enhance (or hinder) multicultural counseling skills. Non-behavioral therapists with access to more agency resources and linkages in the community reported greater multicultural counseling skills. Findings suggest that therapists with non-behavioral treatment orientations who take advantage of established partnerships and collaborations with other service programs (e.g., employment training, education, housing, recreation, welfare) in the community feel they are providing more culturally competent care. In addition, non-behavioral therapists employed by agencies with little community outreach reported greater multicultural counseling skills. The directionality of the moderating effect of agency outreach was unexpected. However, a therapist who believes they are already providing culturally competent care may not feel the need to actively engage in outreach to other institutions (e.g., churches, medicine, businesses, media). Therefore, non-behavioral therapists who are satisfied with their clinical skills and practices may not feel the need to outreach to additional programs and providers. Findings from the exploratory moderator analyses suggest that agency-level outreach and collaborations with other programs in a community have the potential to impact therapist multicultural counseling competencies.

It is possible that select domains or areas within the tripartite model (D. W. Sue et al., 1982) are more heavily influenced by therapist background and therapeutic approach than others. The American Psychological Association mandates that all doctoral trainees demonstrate competence in cultural and individual diversity (American Psychological Association, 2009). Higher education and therapist training programs are encouraged to lay a foundation and instill basic multicultural knowledge and skills in their trainees. For all service providers, knowledge about the demographics of a particular ethnic minority group, their cultural value and orientation system, and their attitudes toward mental health can likely be read about in a textbook or taught in a class curriculum, training program, or continuing education workshop. On the other hand, awareness of one’s own prejudices and biases may be more dependent on the life experiences of an individual therapist, suggesting it could be harder to develop through institutional education or alter through basic trainings. Because of these shared lived experiences ethnic minority therapists may be more aware of their own attitudes, biases, and prejudices when working with clients of color than their White counterparts. The lived experiences and awareness of one’s own prejudices and biases among ethnic minority therapists may also contribute to their ability to effectively treat and develop healthy counseling relationships with clients of color.

Findings should be taken into consideration with the limitations of the current study, including the use of self-report and a limited sample size. Socially desirable responding was controlled for in analyses, however additional self-presentation norms and tendencies may persist. The current study identifies therapist factors, person-level behaviors, and agency-level practices related to self-views of multicultural counseling competencies, but did not assess patient outcomes or client perceptions of therapist competence. Thus, future studies should investigate the relationship between cultural competency and treatment outcomes for ethnic minority clients to determine if self-report multicultural counseling competencies are related to actual clinical competencies and enhance mental health care. It also should be noted that the therapist sample was taken from Los Angeles County, a large county in California that serves an extremely diverse caseload of clients, as evidenced by the reported percentages of ethnic minority clientele by participating providers. It is unclear if the same results would emerge among therapists in relatively isolated areas who serve more homogenous groups. Future research would benefit from examining cultural competency outcomes among providers from a diverse set of organizational and demographic environments.

Nonetheless, the results of this study support the major emphasis on developing “front line” therapists as culturally competent service providers to underserved populations, and the findings suggest specific areas to target for professional training and education. Work by Heppner, Multon, Gysbers, Ellis, and Zook (1998) suggests that strong self-efficacy beliefs with respect to counseling diverse clientele should theoretically predict actual behavior. Our study found that ethnic minority therapists and therapists with specific therapeutic orientations feel they have more multicultural awareness and better multicultural counseling relationships with ethnic minority clientele. Ethnic differences in multicultural counseling competencies were partially explained by involvement within the community of color the therapist served, whereas therapeutic orientation differences between the eclectic and non-behavioral therapists were partially accounted for by community knowledge. This suggests multiple opportunities by which therapists from all ethnic backgrounds and agencies can proactively enhance their cultural competency with ethnic minority clients and reduce disparities in utilization. The use of an eclectic or behavioral treatment approach, greater personal involvement in Asian American communities, and the accumulation of knowledge about the Asian American communities in one’s service area may contribute to the development of more culturally competent therapists in the public sector who can effectively treat underserved Asian American clientele. Through these actions, the mental health field can move toward training culturally competent personnel, ultimately increasing utilization and lowering drop-out rates among Asian Americans.