Mental health counselor qualities for a diverse clientele: linking empathy, universal-diverse orientation, and emotional intelligence.

We examined how culturally relevant variables, including universal-diverse orientation (UDO), an attitude characterized by awareness and acceptance of similarities and differences between self and others, and emotional intelligence, the ability to accurately monitor emotions, might be related to empathy. Two-hundred-and-eleven counseling graduate students completed measures of these variables, as well as a demographic sheet. Hierarchical regression analyses indicated that UDO and emotional intelligence, along with gender, significantly explained variance in empathy. Implications for effective mental health counseling across diverse settings are discussed.

**********

Definitions and measures of empathy have existed since the late 19th century (Duan & Hill, 1996). Empathy is generally referred to as the ability or process of placing self in others' shoes, "as if one was the other person" (Rogers, 1959, p. 210). In a review of research on empathy, Duan and Hill (1996) reported that considerable debate has occurred regarding the definition and nature of empathy. For example, early theorists proposed that empathy was primarily an emotional response involving care and concern for others (Allport, 1961), whereas others emphasized cognitive components of empathy, involving the intellectual understanding of others (Barrett-Lennard, 1962).

More recent theorists have attempted to delineate discrete components and processes of empathy. For example, Davis (1983a; 1996) proposed a multidimensional model of empathy, involving a combination of both cognitive and affective components. Regarding cognitive components, Davis described perspective-taking, the spontaneous ability to adopt the viewpoint of others, and fantasy, a tendency to imagine the feelings and actions of fictional figures, such as those found in movies or plays. Davis also proposed two distinct emotional components of empathy, empathic concern, involving feelings about others, such as sympathy and concern, and personal distress, feelings of anxiety or tension arising from interpersonal situations. Davis (1983a) demonstrated the usefulness of specifying discrete components of empathy, finding differential relations of these components with interpersonal functioning, self-esteem, and social competence. Davis found, for example, that perspective taking was positively linked with better functioning and higher self-esteem, whereas personal distress was linked with lower functioning and lower self-esteem.

Despite the lack of a common definition or theoretical approach, many professionals continue to regard empathy as a core component of effective mental health counseling (McLeod, 1999). Indeed, Pope and Kline (1999) found that expert counselors listed empathy among the top five personal characteristics of effective mental health counselors. Not surprisingly, developing empathy continues to be a primary goal of training in most graduate programs in mental health counseling (Ivey & Ivey, 2003; Okun, Shepard, & Eisenberg, 2000).

The importance or relevance of empathy in light of the increasingly diverse social world of today has received little theoretical or empirical attention. That is, while some scholars have called for a consideration of the cultural contexts of empathy (Duan & Hill, 1996; McLeod, 1999), little theory and research on empathy exists that shows the relevance or effectiveness of this construct across cultures. Patterson (1996) asserted that empathy is one of five universal counselor qualities essential to all counseling relationships. Sue and Sue (1999) concurred and acknowledged that understanding the world of the client is a key component of counseling that may transcend culture. However, other scholars have questioned whether empathy is truly possible toward people from different cultures (Patterson, 1996). Thus, further research and theory on the cultural implications and applications of empathy seems warranted.

Recently, the construct of universal-diverse orientation (UDO) was defined as a social attitude of awareness and acceptance of the similarities and differences that exist among people, as based, for example, on race, gender, and abilities (Miville et al., 1999). Miville et al. proposed that UDO measures an essential component of effective multicultural counseling, an overarching attitude of openness and acceptance of the various cultures to which clients belong (Vontress, 1996). Such an attitude presumably is based on the philosophical framework that human beings share commonalities with each other (i.e., universal), while at the same time having important differences (Miville et al., 1999). UDO has been correlated with positive racial identity, empathy, healthy narcissism, feminism, androgyny, homophobia and dogmatism (these latter two correlations being negative; Miville et al., 1999) as well as wellness variables, such as self-efficacy and coping skills (Miville, Romans, Johnson, & Lone, 2004). Previous researchers have found UDO is a critical aspect of multicultural competence, being a significant variable explaining variance in both multicultural knowledge and awareness among counselors (Constantine et al., 2001; Fuertes & Probst, 2002; Munley, Lidderdale, Thiagarajan, & Null, 2004; Yeh & Arora, 2003) as well as clients' expectations regarding effective multicultural counseling (Constantine & Arorash, 2001).

UDO seems particularly relevant in examining the ability to empathize with culturally different people. Openness to and acceptance of the cultural perspectives of others seems critical to the ability to understand more personal perspectives (Duan & Hill, 1996; McLeod, 1999). Indeed, perceptions of the similarities and differences that exist between self and others was recently described by Auger (2004) as an implicit assumption held by many mental health counselors about their clients that affect the counseling relationship. UDO may be a relevant construct for assessing empathy since it refers to the degree to which individuals believe that people are both similar and different from others, rather than rigidly assuming one or the other. UDO thus represents a variable potentially relevant to the establishment of professional helping relationships, particularly in a diverse social world. Exploring whether UDO significantly explains variance in empathy may help reveal important culturally relevant dimensions of empathy not represented in current theoretical models.

Another variable, emotional intelligence (Mayer & Salovey, 1993), refers to the ability to monitor and discriminate the emotions of self and others. Emotional intelligence has been significantly related to several personality variables, such as attention to feelings, clarity of feelings, and openness to experience; therapists have scored higher in emotional intelligence than therapy clients (Shutte et al., 1998). Moreover, emotional intelligence has been found to significantly explain variance in multicultural knowledge among school counselors (Constantine & Gainor, 2001).

Emotional intelligence is related to the ability to perceive others accurately and to interact with others appropriately, ft is reasonable to suggest that this ability may be helpful to mental health counselors in order to accurately perceive and relate well with others of varying cultural backgrounds. It seems likely that counselors who are better at "reading" the affective experiences and expressions of others might be more empathic, and that exploring whether emotional intelligence explains variance in empathy might point to culturally relevant dimensions of empathy not represented in current theoretical models. That is, emotional intelligence may be a helpful quality or skill for mental health counselors who must learn to empathize with others who are from similar or different cultures.

An additional focus of the study was to explore how training experiences also might explain variance in empathy; that is, does empathy improve during graduate training? Previous research has shown that formal training experience has a significant positive impact on a variety of relevant counselor variables, including multicultural awareness and knowledge (Carlson, Brack, Laygo, Cohen, & Kirkscey, 1998; Coleman, 1998; Kiselica, Maben, & Locke, 1999) and empathy (Ottens, Shank, & Long, 1995; Ridgway & Sharpley, 1990). However, no research has been conducted examining how training experience together with UDO and emotional intelligence might explain variance in empathy. Demographic training variables (e.g., number of practicum courses) thus were examined with respect to their relations to the variables of interest.

In sum, we examined whether UDO, emotional intelligence and training experience variables significantly explained variance in empathy. Research questions included: (a) What is the relationship between empathy and UDO among counseling graduate students? (b) What is the relationship between empathy and emotional intelligence? (c) What is the relationship between empathy and training experience? and (d) How does a linear combination of UDO, emotional intelligence, and training experience explain variance in empathy? The following relationships were posited: higher levels of empathy would be related to higher levels of UDO and emotional intelligence and greater training experience.

METHOD

Participants

Participants included 211 students in graduate level counseling programs (85% master's degree, 15% doctoral degree) from five academic programs throughout the southwestern United States. The participants were primarily female (80%) and ranged in age from 21 to 57, with a mean age of 32 years (SD=9.05). Mean grade point average was 3.78, and ranged from 2.2 to 4.0 on a 4.0 scale (SD = .28). Participants were predominantly European American (79.2%), the remainder of the sample self-identifying as Native American (5.9%), Hispanic (5.3%), African American (2%), Multiracial (1.5%), Asian American (1%), and Other (5.1%).

Participants presented a wide range of practicum experience: 50% not enrolled in a practicum; 39% completed 1 to 3 semesters; and 11% completed between 4 to 12 semesters. The sample varied in the number of graduate credit hours earned, ranging from 0 to 140 credit hours, with a mean of 32 hours.

Measures

Three instruments were used in the study: the Interpersonal Reactivity Index (IRI; Davis, 1980, 1983a); the Miville-Guzman Universality-Diversity Scale, Short Form (MGUDS-S; Miville et al., 1999; Fuertes, Miville, Mohr, Sedlacek, & Gretchen 2000); and the Emotional Intelligence Scale (EIS; Schutte et al., 1998). Participants also completed demographic questions on age, gender, and race/ethnicity. Items pertaining to the participants' graduate training experience also were included (i.e., number of practicum courses, number of graduate hours in counseling completed, program type [doctoral or master's degree], and GPA).

Interpersonal Reactivity Index. Empathy was measured using the IRI (Davis, 1983a), a 28-item scale measuring four aspects of empathy reflected in the following subscales: perspective taking (PT), empathic concern (EC), fantasy (FS), and personal distress (PD). Perspective taking measures the extent to which individuals appreciate the perspectives of other people's points of view. Empathic concern is an affective measure of the ability to feel compassion and concern for others who have negative experiences. Fantasy is a measure of the ability to identify with fictitious characters in movies and books. Personal distress is a measure of the extent to which individuals share the negative emotions of others. Each scale is composed of seven items on a five-point Likert-type scale that are summed for a total subscale score. Alpha coefficients for the respective subscales obtained here were somewhat low to adequate: .65 (PT), .66 (EC), .80 (FS), and .78 (PD), and the alpha coefficient for the total IRI was .73.

Davis (1983a) found support for the construct validity of the IRI through predicted significant relationships of the scale with interpersonal functioning, social competence and self-esteem. Davis (1980) also conducted an exploratory factor analyses on the IRI, using an oblique rotation, yielding a four-factor solution that matched the four subscales. Subsequent researchers also have conducted exploratory factor analyses yielding factor structures supporting the multidimensional nature of the IRl among samples from different racial/cultural groups (Escriva, Navarro, & Garcia, 2004; Siu & Shek, 2005). Other authors, however, have found evidence for a unidimensional structure of the IRI via confirmatory factor analyses (Alterman, McDermott, Cacciola, & Rutherford, 2003). Intercorrelations of the IRI subscales for the current sample are listed in Table 1, correlations were moderate and in generally predicted directions.

Although evidence for the reliability and validity of the IRI subscales seems mixed, we chose to utilize the IRI subscales for our analyses, rather than the total score, for several reasons: (a) the IRI subscales were theoretically derived; (b) the total IRI alpha coefficient was only adequately acceptable, and lower than the alpha coefficients of two subscales, FS and PD, and (c) existing evidence of the construct validity of the IRI seems to point more to the multidimensional, rather than unidimensional, structure of the IRI.

Miville-Guzman Universality-Diversity Scale, Short Form. The MGUDS-S was used to measure UDO, a social attitude of appreciation and acceptance for the similarities and differences that exist among people (Miville et al., 1999). Items of the MGUDS-S measure attitudes expressing relativistic appreciation of self and others, comfort with differences, and diversity of contact with others. Miville et al. developed the original 45-item M-GUDS (Long Form) over a series of studies, and provided evidence demonstrating its reliability and validity. Alpha coefficients for the M-GUDS ranged from .89 to .95. With respect to the construct validity of the M-GUDS, Miville et al. found that the scale significantly correlated in theoretically predicted ways with measures of racial identity, healthy narcissism, feminism, androgyny, homophobia, and dogmatism (the last two correlations were negative). The M-GUDS also displayed discriminant validity by failing to correlate with SAT (formerly known as the Scholastic Achievement Test) verbal scores, although mixed results were obtained with social desirability.

The MGUDS-S (i.e., Short Form) (Fuertes et al., 2000) consists of 15 items on a six-point Likert-type scale. A total score is computed by summing responses to all items. The alpha coefficient for the MGUDS-S in the current study was .74. Fuertes et al. (2000) developed the MGUDS-S from an exploratory factor analysis of the Long Form of M-GUDS (Miville et al., 1999) conducted on a racially diverse sample of college students. Highest loading items on the M-GUDS were used as the basis of the short form. As evidence of the construct validity of the MGUDS-S, Fuertes et al. found that the short form highly and positively correlated with the long form (r = .77, p < .001). As well, the MGUDS-S correlated in theoretically predicted ways with demographic items dealing with religious tolerance and choice of friends based on race or sexual orientation.

Fuertes et al. conducted two sets of confirmatory factor analyses on the MGUDS-S. These analyses supported the factor structure of the MGUDS-S [[chi square] (85, N = 206) = 143.84, p < .001; NNFI = .94; GFI = .92; CFI = .95 and [chi square] (87, N = 184) = 123.43, p < .01; CFI = .95]. The CFA results were each based on testing several models of the MGUDS-S in which post hoc modifications allowed errors to correlate. However, the authors noted that the respecification processes did not seem to affect the assumptions underlying the original model (e.g., in the first CFA, errors associated with two items were allowed to correlate given their similarity in content and phrasing, whereas in the second CFA, the final model was based on the factor structure and loading patterns obtained from the original exploratory factor analysis and tested in the first CFA). Unfortunately, Fuertes et al. only reported one goodness of fit index for the second CFA. While the CFA's supported a three-factor model (Relativistic Appreciation, Diversity of Contact, and Comfort with Differences) which we had hoped to incorporate here, we instead used the total score because of the very low reliability coefficients we found for two of the three subscales in the current study (below .60).

Emotional Intelligence Scale. The EIS measures emotional intelligence, which refers to the abilities to appraise and express emotions, to regulate emotions, and to utilize emotions in problem solving (Schutte et al., 1998). A total score is derived from summing 33 Likert-type items. Previous internal consistency analyses yielded a Cronbach's alpha of .90 (Schutte et al., 1998), as well as a test-retest reliability coefficient over a two-week interval of .78. In the current study, coefficient alpha of the EIS was .90. Evidence for the construct validity of the scale has been demonstrated via significantly different scores on the EIS for groups in predicted directions (i.e., as might be expected, psychotherapists scored higher than prisoners and substance abusers; Schutte et al., 1998). Evidence of construct validity also was demonstrated by correlating EIS scores with variables expected to be related to emotional intelligence. EIS scores correlated significantly and positively with clarity of feelings, attention to feelings, optimism, openness to experience, and nonverbal expressiveness of emotion, and correlated significantly and negatively with depression, pessimism, and impulsiveness. Schutte et al. (1998) also provided support for the discriminant validity of the EIS (i.e., as predicted, EIS scores were not related to SAT scores, suggesting a difference between emotional intelligence and cognitive ability). More recently, Schutte et al. (2001) demonstrated the construct validity of the EIS by reporting significant relations in predicted directions with a number of interpersonal variables, such as self-monitoring, social skills, cooperativeness and marital satisfaction. To date, no factor analysis of the EIS has been published, and no evidence regarding the underlying structure of the EIS is available.

Procedure Upon approval of our research protocol by the university institutional review board, we contacted faculty colleagues at three universities in the Southwest to distribute and collect the research packets. All participants were administered research packets consisting of the IRI, MGUDS-S, the EIS, and a demographic sheet in a counterbalanced format to control for order effects. Participants completed the packets on a voluntary basis either during class time or on an individual basis, with the large majority of packets (approximately 80%) being completed during class time. Each participant signed a consent form and then completed the research packet. Participants were given a debriefing sheet after completing the packet. Participants typically completed the research packet in 20 to 30 minutes.

RESULTS

Preliminary Analysis

A multivariate analysis of variance (MANOVA) was conducted to ascertain whether significant differences on the dependent variables, the IRI subscales (perspective taking, empathic concern, fantasy, and personal distress) existed among the different gender, ethnic and language groups represented in the sample. Also, because there were a large number of novice trainees in the sample who had not yet completed a practicum, we explored whether there were significant differences on the dependent variables among participants based on training experience (i.e., completion of at least one practicum and number of credit hours completed). The results of the MANOVA revealed gender differences among the dependent variables (Wilks's [??] = .84, F(4, 165) = 8.31, p < .01), specifically, on emotional concern (F(1,173) = 17.85, p < .01) and personal distress (F(1,173) = 12.51, p < .01), with women scoring higher on both scales. No other differences were found. Based on the results of the preliminary analysis, gender was included in the principal analyses. Table 1 summarizes the means, standard deviations, and correlation coefficients for the IRI, EIS, and the MGUDS-S; a Bonferroni adjusted alpha level of .05/15 or .003 was used to test the significance of each correlation.

Principal Analyses

To investigate our research questions, the data were analyzed using a series of hierarchical regression analyses. The independent variables included gender in the first step of each equation and universal-diverse orientation (MGUDS-S) and emotional intelligence (EIS) in the second step. The dependent variables were the four empathy subscales of the IRl: perspective taking, empathic concern, fantasy, and personal distress. A Bonferroni adjusted alpha level of .05/4 or .01 was used to test the significance of each regression analysis.

Table 2 provides a summary of our findings from the four hierarchical regression analyses. With respect to perspective-taking, the first step involving gender was not significant, F (1,208) = .69, p = .41, [R.sup.2] = .003. Once we accounted for the variance explained by gender, we found that that greater emotional intelligence and higher universal-diverse orientation were related to greater perspective-taking ([DELTA] [R.sup.2] = .184), F (1,206) = 23.26, p < .01, [R.sup.2] : .187. With respect to empathic concern, gender accounted for a significant amount of variance in the first step, F (1,208) = 19.25, p < .01, [R.sup.2] = .085. As noted previously, women expressed greater empathic concern than men. In the second step, greater emotional intelligence and higher universal-diverse orientation were related to greater empathic concern, ([DELTA][R.sup.2] = .086), [DELTA]F (1,206) = 10.71, p < .01, [R.sup.2] : .171. With respect to fantasy, the first step involving gender was not significant, F (1, 208) = 2.29, p = .13, [R.sup.2] = .011. The second step involving the addition of emotional intelligence and universal-diverse orientation also was not significant, ([DELTA][R.sup.2] = .018), [DELTA]F (1,206) = 1.90, p = .15, [R.sup.2] = .029. Finally, with respect to personal distress, the first step involving gender was significant, F (1,208) = 12.99, p < .01, [R.sup.2] = .059. Women expressed higher personal distress than men. The second step also was significant, ([DELTA][R.sup.2] = .07), AF (1, 206) = 8.28, p < .01, [R.sup.2] = .129, though only emotional intelligence meaningfully contributed to the additional variance explained in personal distress (see Table 2).

DISCUSSION

We examined whether emotional intelligence and universal-diverse orientation explained a significant proportion of the variance in several dimensions of empathy, a quality/skill long identified as a principal ingredient of effective mental health counseling, over and above variance accounted for by gender. We found that emotional intelligence was moderately related to three dimensions of empathy: perspective taking, emotional concern, and personal distress. Thus, the ability to monitor and discriminate the emotions of self and others was associated with the ability to empathize, particularly taking the perspective of other persons as well as feeling emotional concern for others. Previous researchers (Constantine & Gainor, 2001) found similar results with school counselors, indicating that placing self in others' shoes involves being able to accurately sense and respond to emotions as well as thoughts. Thus, the notion of "deep listening" is relevant; in order for mental health counselors to appreciate the depths of their clients' stories, they must be able to perceive or imagine the emotional reactions, perhaps unspoken by their clients, to the narrative content that is being presented.

Regarding the inverse relationship between emotional intelligence and personal distress, we suggest that mental health counselors who are more emotionally intelligent may be less likely to become distressed by the emotions of others. Given their greater ability to differentiate their emotions from those of others, it makes sense that these counselors might be less likely to over-identify with others in the process of empathizing with them, and would be less likely to confuse others' emotions with their own. Mental health counselors who are more emotionally intelligent may be less likely to become emotionally aroused and reactive as they attempt to understand the emotional experiences of others, and they may be able to stay more focused without getting caught up in others' distress. Since beginning counselors are often concerned with their ability to manage their own emotional reactions and personal distress in the counseling process, these results suggest that helping counseling students learn how to differentiate their own emotions from those of others (i.e., become more emotionally intelligent) may be an important aspect of graduate training, particularly regarding empathy.

Universal-diverse orientation also was significantly, though modestly, related to empathy. The MGUDS-S was positively related to both cognitive (perspective taking) as well as affective (empathic concern) components of empathy (Davis, 1983b). Together, these findings indicate that being aware and accepting of similarities and differences among people may be a somewhat important dimension of the ability to genuinely take others' perspectives as well as to experience warmth and compassion toward others. As well, the more interactions individuals have with diverse individuals, as reflected in UDO, the more likely and able they may be to appreciate truly different perspectives. Finally, dealing with anxiety and discomfort about differences among people, also reflected in UDO, may aid in increasing counselors' abilities to empathize with diverse others by providing a more emotionally genuine context for doing so (Munley et al., 2004). Thus UDO may play some role in a global ability to empathize with individuals from a variety of life and collective experiences. These findings are in line with previous research (Constantine et al., 2001; Fuertes & Probst, 2002; Yeh & Arora, 2003) that seemed to indicate the potential value of UDO as a broad measure of multicultural competence for mental health counselors across a variety of settings and clients. Future researchers might investigate UDO as a measure of implicit assumptions focusing on similarities and differences held by mental health counselors that purportedly affect counseling relationships, particularly with diverse clients (Auger, 2004).

Surprisingly, training experience was not significantly associated with empathy. It is possible that training experience may not have been sufficiently measured in the study. Rather than gross measures such as numbers of semesters and credit hours, future researchers might assess training variables more directly for their relations with empathy, as well as UDO and emotional intelligence. In particular, it would be important to investigate the differential impact that various types of training experiences (e.g., courses, supervision in this area) may have on these relations. Gender was significantly related to several dimensions of empathy, and findings were consistent with traditional sex roles and stereotypes. Men reported less personal distress in their interactions with others than did women, whereas women reported greater empathic concern. This finding indicates that there might be some value in attending to strengths and limitations associated with sex role socialization and stereotypes among counselors. That is, being male or female may be associated with specific strengths and weaknesses that need to be recognized and integrated into training and practice in order for mental health counselors to be more effective in empathizing with others.

Limitation and Directions for Future Research

Limitations of the study included the lack of participants with a wider range of counseling experiences. Our sample included many students who had not yet worked directly with clients. Thus, our findings may have been affected by a restriction of range based on participants' lack of clinical experience. As well, the small number of people of color in the sample likely impacted the statistical power of analyses related to race/ethnicity. To be sure, the participant pool was a fairly representative sample of the racial/ethnic and gender makeup of the profession. However, in the future, researchers might examine more directly the relations of the race/ethnicity of mental health counselors with empathy, UDO, and emotional intelligence. In this study, information on types of clients, particularly clients' racial/ethnic backgrounds, was not collected; future researchers might include this information to help explore another possible link between counselor attitudes and behaviors. Finally, method variance and social desirability associated with the use of paper-and-pencil measures may have impacted our results (Heppner, Kivilghan, & Wampold, 1999). Future researchers might incorporate ratings from instructors, or even examine counselor-client dyads to assess the relations of empathy with UDO and emotional intelligence.

Implications for Practice

Mental health counselors today work with clients from very different backgrounds in an increasingly pluralistic social context (Coleman, 2004; Liem, 2001). Findings here indicate that it may not be sufficient to have specific skills to help deal with these differences. Instead, the personal qualities and values of mental health counselors may be essential for effectiveness with diverse clients (Carlozzi, Bull, Eells, & Hurlburt, 1995). Genuine acceptance of others and the working through of anxiety and discomfort about people who are significantly different seems potentially important to developing genuine empathy for clients from diverse backgrounds. Being able to understand others' frameworks further involves being able to adequately monitor the emotions of self and others. Our findings demonstrate the need for mental health counselors to expand their traditional understanding of empathy to include more global perspectives for working with others, emphasizing awareness and acceptance of similarities and differences among people, as well as honest and intelligent awareness of emotions.

REFERENCES

Allport, G. (1961). Pattern and growth in personality. New York: Holt, Rinehart, & Winston.

Alterman, A. I., McDermott, P. A., Cacciola, J. S., & Rutherford, M. J. (2003). Latent structure of the Davis Interpersonal Reactivity Index in methadone maintenance patients. Journal of Psychopathology & Behavioral Assessment. 25(4), 257-265.

Auger, R. W. (2004). What we don't know CAN hurt us: Mental health counselors' implicit assumptions about human nature. Journal of Mental Health Counseling, 26(1), 13-25.

Barrett-Lennard, G. T. (1962). Dimensions of therapy response as causal factors in therapeutic change. Psychological Monographs, 76, 1-33.

Carlozzi, A.F., Bull, K.S., Eells, G.T., & Hurlburt, J.D. (1995). Empathy as related to creativity, dogmatism, and expressiveness. The Journal of Psychology, 129, 365-373.

Carlson, M. H., Brack, C. J., Laygo, R., Cohen, R., & Kirkscey, M. (1998). An exploratory study of multicultural competence of counseling graduate students: Support for experiential skills building. Clinical Supervisor, 17, 75-87.

Coleman, H. L. K. (2004). Multicultural counseling competencies in a pluralistic society. Journal of Mental Health Counseling, 26(1), 56-67.

Coleman, H. L. K. (1998). General and multicultural counseling competency: Apples and oranges? Journal of Multicultural Counseling and Development, 26, 147-156.

Constantine, M. G., & Arorash, T. J. (2001). Universal-diverse orientation and general expectations about counseling: Their relation to college students' multicultural counseling expectations. Journal of College Student Development, 42(6), 535-544.

Constantine, M. G., Arorash, T. J., Barakett, M. D., Blackmon, S. M., Donnelly, P. C., & Edles, P. A. (2001). School counselors' universal-diverse orientation and aspects of their multicultural counseling competence. Professional School Counseling, 5(1), 13-18.

Constantine, M. G., & Gainor, K. A. (2001). Emotional intelligence and empathy: Their relation to multicultural counseling knowledge and awareness. Professional School Counseling, 5(2), 131-137.

Davis, M. (1980). A multidimensional approach to individual differences in empathy. JSAS Catalog of Selected Documents in Psychology, 10, 85.

Davis, M. (1983a). Measuring individual differences in empathy: Evidence for a multidimensional approach. Journal of Personality and Social Psychology, 44, 113-126.

Davis, M. (1983b). The effects of dispositional empathy on emotional reactions and helping: A multidimensional approach. Journal of Personality, 51(2), 167-184).

Davis, M. H. (1996). Empathy:A social psychological approach. Boulder, CO: Westview.

Duan, C., & Hill, C. E. (1996). The current state of empathy research. Journal of Counseling Psychology, 43, 261-274.

Escriva, V. M., Navarro, M. D. F., & Garcia, P. S. (2004). Measuring empathy: The Interpersonal Reactivity Index. Psicothema, 16(2), 255-260.

Fuertes, J. N, Miville, M. L., Mohr, J. J., Sedlacek, W. E., & Gretchen, D. (2000). Factor structure and short form of the Miville-Guzman Universality-Diversity Scale. Measurement and Evaluation in Counseling and Development, 33, 157-169.

Fuertes, J. N., & Probst, K. (2002). Clients' ratings of counselor multicultural competency. Cultural Diversity and Ethnic Minority Psychology, 8(3), 214-223.

Heppner, P. P., Kivlighan, D. M., & Wampold, B. E. (1999). Research design in counseling. Belmont, CA: Wadsworth.

Ivey, A. E., & Ivey, M. B. (2003). Intentional interviewing and counseling: Facilitating client development in a multicultural society (5th ed.). Belmont, CA: Wadsworth.

Kiselica, M. S., Maben, P., & Locke, D. C. (1999). Do multicultural education and diversity appreciation training reduce prejudice among counseling trainees? Journal of Mental Health Counseling, 21, 240-254.

Liem, K. K. (2001). Counseling applications of racial and ethnic identity models: An introduction to the special issue. Journal of Mental Health Counseling, 23(3), 185-192.

McLeod, J. (1999). A narrative social constructionist approach to therapeutic empathy. Counselling Psychology Quarterly, 12(4), 377-394.

Mayer, J. D., & Salovey, P. (1993). The intelligence of emotional intelligence. Intelligence, 17, 433-442.

Miville, M. L., Gelso, C. J., Pannu, R., Liu, W., Touradji, P., Holloway, P., & Fuertes, J. (1999). Appreciating similarities and valuing differences: The Miville-Guzman Universality-Diversity Scale. Journal of Counseling Psychology, 46, 291-307.

Miville, M. L., Romans, J. S. C., Johnson, D., & Lone, R. (2004). Universal-diverse orientation: Linking wellness with social attitudes. Journal of College Student Psychotherapy, 19(2), 61-79.

Munley, P. H., Lidderdale, M. A., Thiagarajan, M., & Null, U. (2004). Identity development and multicultural competency. Journal of Multicultural Counseling, 32, 282-295).

Okun, M. A., Shepard, S. A., & Eisenberg, N. (2000). The relations of emotionality and regulation to dispositional empathy-related responding among volunteers-in-training. Personality and Individual Differences, 28, 367-382.

Ottens, A., Shank, G. D., & Long, R. J. (1995). The role of abductive logic in understanding and using advanced empathy. Counselor Education and Supervision, 34, 199-211.

Patterson, C. H. (1996). Multicultural counseling: From diversity to universality. Journal of Counseling and Development, 74, 227-231.

Pope, V. T., & Kline, W. B. (1999). The personal characteristics of effective counselors: What 10 experts think. Psychological Reports, 84, 1339-1344.

Ridgway, I. R., & Sharpley, C. F. (1990). Multiple measures for the prediction of counsellor trainee effectiveness. Canadian Journal of Counselling, 24, 165-177.

Rogers, C. R. (1959). A theory of therapy, personality and interpersonal relationships as developed in the client-centered framework. In S. Koch (Ed.), Psychology:A study of a science. Student 1. Conceptual and systematic: Vol. 3 Formulations of the person and the social context (pp. 184-256). New York: MrGraw-Hill.

Schutte, N. S., Malouff, J. M., Bobik, C., Coston, T. D., Greeson, C., Jedlicka, C., Rhodes, E., Wendorf, G. (2001). Emotional intelligence and interpersonal relations. Journal of Social Psychology. 141(4), 523-536.

Schutte, N. S., Malouff, J. M., Hall, L. E., Haggerty, D. J., Cooper, J. T., Golden, C. J., & Dornheim, L. (1998). Development and validation of a measure of emotional intelligence. Personality and Individual Differences, 25, 167-177.

Siu, A. M. H., & Shek, D. T. L. (2005). Validation of the Interpersonal Reactivity Index in a Chinese context. Research on Social Work Practice, 15(2), 118-126.

Sue, D. W., & Sue, D. (1999). Counseling the culturally different (3rd ed.). New York: Wiley.

Vontress, C. E. (1996). A personal retrospective on cross-cultural counseling. Journal of Multicultural Counseling and Development, 24, 156-166.

Yeh, C. J., & Arora, A. K. (2003). Multicultural training and interdependent and independent self-construal as predictors of universal-diverse orientation among school counselors. Journal of Counseling and Development, 81(1), 78-83.

Marie L. Millville, Ph.D., is with the Department of Counseling and Clinical Psychology at Teachers College, Columbia University. Alfred F. Carlozzi, Ed.D., is the head of the School of Applied Health & Educational Psychology at Oklahoma State University. George V. Gushue, Ph.D. is with the Department of Counseling and Clinical Psychology at Teachers College, Columbia University. Sara L. Schara and Masafumi Ueda are psychologists. Correspondence concerning this manuscript should be addressed to Marie L. Miville, Department of Counseling and Clinical Psychology, Teachers College, Columbia University, 525 W. 120th St., Box 73, New York, NY 10027. E-mail: [e-mail].

Table 1
Means, Standard Deviations and Correlations of Empathy Subscales,
Universal-Diverse Orientation, and Emotional Intelligence (N=211)

    Variable                    M        SD        1       2

1   Empathic Concern           29.23     3.46     --
2   Fantasy                    24.40     5.46    .39 *     --
3   Personal Distress          15.71     4.60    .05       .15
4   Perspective Taking         27.75     3.47    .33 *     .06
5   Emotional Intelligence    133.55    13.42    .30 *     .15
6   Universal-Diverse
    Orientation                71.84     7.26    .16      -.02

    Variable                    3         4        5

1   Empathic Concern
2   Fantasy
3   Personal Distress           --
4   Perspective Taking        -.29 *     --
5   Emotional Intelligence    -.23 *    .34 *     --
6   Universal-Diverse
    Orientation               -.11      .30 *    .13

* p < .003

Table 2
Summary of Hierarchical Multiple Regression Analyses with
Emotional Intelligence and Universal-Diverse Orientation
as Independent Variables and Empathy Subscales as Dependent
Variables

                               Empathy Subscales

                              Perspective-Taking
Independent
Variable           [R.sup.2]        B       SE B     [beta]

Step 1
                    .003
                   (.001) (1)

  Gender                           -.50     .60      -.06

Step 2              .187 **
                   (.175)

  Gender                           -.25     .55      -.03

  Emotional
  Intelligence                      .07     .02       .31 **

  Universal-
  Diverse
  Orientation                       .13     .03       .26 **

                               Empathy Subscales

                               Empathic Concern
Independent
Variable           [R.sup.2]        B       SE B     [beta]

Step 1
                    .085 **
                   (.08)

  Gender                           -2.5     .58      -.29 **

Step 2              .171 **
                   (.159)

  Gender                           -2.3     .55      -.27 **

  Emotional
  Intelligence                      .06     .02       .25 **

  Universal-
  Diverse
  Orientation                       .06     .03       .13 *

                               Empathy Subscales

                                    Fantasy
Independent
Variable           [R.sup.2]        B       SE B     [beta]

Step 1
                    .011
                   (.006)

  Gender                          -1.43     .95      -.10

Step 2              .029
                   (.015)

  Gender                          -1.22     .95      -.09

  Emotional
  Intelligence                      .05     .03       .13

  Universal-
  Diverse
  Orientation                       .01     .05       .01

                               Empathy Subscales

                               Personal Distress
Independent
Variable           [R.sup.2]        B       SE B     [beta]

Step 1
                    .059 **
                   (.054)

  Gender                          -2.8      .78      -.24 **

Step 2              .129 **
                   (.116)

  Gender                          -3.11     .76      -.27 **

  Emotional
  Intelligence                     -.09     .02      -.25 **

  Universal-
  Diverse
  Orientation                      -.04     .04      -.07

(1) Numbers in parentheses represent adjusted [R.sup.2] values.

* p < .05, ** p < .01