counseling and psychotherapy has gained an impetus only ... with the relationship between cultural and counseling vari- ...... Barriers to effective cross-cultural.
Journal of Counseling Psychology 1994, Vol. 41. No. 3, 280-287
Copyright 1994 by the American Psychological Association Inc 0022-0167/94/$3.00
Individualism-Collectivism, Social-Network Orientation, and Acculturation as Predictors of Attitudes Toward Seeking Professional Psychological Help Among Chinese Americans Shiraz Piroshaw Tata and Frederick T. L. Leong Several culturally based variables were used to predict the patterns of help-seeking attitudes among a sample of Chinese-American students (TV = 219) in a large midwestern university. Cultural values operationalized by H. C. Triandis, R. Bontempo, M. J. Villareal, M. Asai, and N. Lucca's (1988) Individualism-Collectivism Scale, social support attitudes operationalized by A. Vaux's (1985) Network Orientation Scale, and the continuous variable of acculturation operationalized by the Suinn-Lew Asian Self-Identity Acculturation Scale (R. M. Suinn, K. Rickard-Figueroa, S. Lew, & P. Vigil, 1987) were selected as predictors of attitudes as measured by E. H. Fischer and J. L. Turner's (1970) Attitudes Toward Seeking Professional Psychological Help Scale. Each of the 4 independent variables were found to be significant predictors of attitudes towards seeking professional psychological help. The counseling and research implications of the results are discussed.
Research on cross-cultural issues that have an impact on counseling and psychotherapy has gained an impetus only since the 1970s (Atkinson, 1985). Atkinson (1985) examined reviews that had critically summarized studies dealing with the relationship between cultural and counseling variables (Atkinson, 1983; Harrison, 1975; Sattler, 1977). Three important characteristics of the studies conducted before 1985 stood out. First, most of the research had focused on relations between African Americans and Caucasians and had not examined other racial and ethnic groups. Second, the settings in which a large proportion of the studies had been conducted were mental health outpatient clinics or psychiatric hospital inpatient units, not college counseling centers. Third, either a survey or an archival research design had been used in a majority of the studies. These observations can also be made about a large number of studies that have examined the role of ethnic and racial variables in areas of psychology other than those of counseling and psychotherapy. Atkinson (1985) recommended that researchers study a number of different ethnic and racial groups, investigate within-group variables, such as self-designated identity and cultural commitment, and do more research in college settings. These recommendations are valuable for studies on a number of psychological issues and processes. Some of these issues include attitudes, values, experiences, expectations, beliefs, and social-support systems of members of different ethnic groups (Root, 1985). These factors affect a number of behaviors, one of which is the act of seeking professional psychological help. Research has shown that members of non-Caucasian ethnic groups
underuse professional psychological help services (Leong, 1986; Sattler, 1977; S. Sue, 1977; S. Sue & Sue, 1974; Yamamoto, 1978). The act of seeking professional psychological services may, in turn, be influenced by attitudes toward seeking such help (Halgin, Weaver, Edell, & Spencer, 1987). Recent studies (Atkinson, Maruyama, & Matsui, 1978; Dadfar & Friedlander, 1982; Haviland, Horswill, O'Connell, & Dynneson, 1983; Sanchez & Atkinson, 1983) have shown that within-group differences on such factors as racial similarity between counselor and counselee and cultural commitment influence students' willingness to seek help. A better understanding of these factors would probably enable mental health service providers to understand how to make counseling more accessible and effective for members of non-Caucasian ethnic and racial groups. An ethnic minority group in the United States that does not seem to have received as much social and political attention as other such groups is that of Asian Americans. As Leong (1985) pointed out, there is also a dearth of research in the social sciences on Asian Americans. Contrary to the myth that Asian Americans are the model minority (Crystal, 1989; Ho, 1976) with few problems, Asian Americans do experience psychological and social problems (Leong, 1986; S. Sue & McKinney, 1975; Tracey, Leong, & Glidden, 1986). The factors that influence the attitudes of Asian-American students toward seeking professional psychological help constitute an important area for future research (Leong, 1986). Atkinson and Gim (1989) found that acculturation or cultural identity had a direct relationship with attitudes toward help seeking among Asian-American students at a university on the West Coast. In this study, we also hypothesized that Chinese-American students who were more acculturated would have more positive attitudes toward seeking professional psychological help as compared with Chinese-American students who were not as highly accul-
Shiraz Piroshaw Tata, Counseling and Student Development Center, Northern Illinois University; Frederick T. L. Leong, Department of Psychology, Ohio State University. Correspondence concerning this article should be addressed to Frederick T. L. Leong, Department of Psychology, Ohio State University, 142 Townsend Hall, 1885 Neil Avenue Mall, Columbus, Ohio 43210-1222. 280
SPECIAL SECTION: CHINESE AMERICANS' PREDICTORS OF HELP SEEKING
turated. More recently, Solberg et al. (1994) replicated a study in the Midwest that Gim, Atkinson, and Whiteley (1990) had done at a West Coast university. Both studies explored the effects of acculturation as measured by a five-item cultural identity index with ethnicity and gender on problems experienced during college and subjects' willingness or likelihood to seek help for these problems as variables. There have been conflicting results with regard to whether gender is a significant predictor of attitudes toward seeking professional psychological help among AsianAmerican students (Atkinson & Gim, 1989; Atkinson, Ponterotto, & Sanchez, 1984; Solberg et al., 1994). Hence in this study, a specific hypothesis was not put forth with regard to gender. Gender was included as a predictor variable in this study because preliminary analyses indicated that there is a significant difference between men and women in terms of attitudes toward seeking professional psychological help. In this study, an effort was made to expand the body of knowledge on the relationship between cultural variables and attitudes toward seeking professional psychological help. Gender and acculturation have been found to be important in prior research. The impact of two additional variables on attitudes toward seeking professional psychological help was also examined: the effect of certain values inherent in the constructs of individualism and collectivism (Triandis, 1989) and the impact of social-network orientation (Tolsdorf, 1976). A number of barriers have been hypothesized to hinder counseling with persons who emigrate to the United States from East Asian countries. D. W. Sue and Sue (1977) identified the variables of language, class-bound values, and culture-bound values as barriers. Some of the values that have been traditionally associated with counseling seem to be in opposition to the values associated with persons from East Asian countries, such as China, Korea, and India. Theoretical analyses of values of persons from different cultures have proved to be valuable (Hofstede, 1983; Triandis, 1989). However, no studies published to date have examined the relationship among Asian Americans between the values underlying individualism and collectivism and attitudes toward help seeking. The findings of such studies could be used as a basis to provide some important suggestions to providers of professional psychological help. In summarizing the findings of his studies, Triandis (1989) defined individualists and collectivists. He stated that individualists give priority to personal goals over ingroup goals, whereas collectivists either do not make a distinction between personal and group goals or, if they do, they give high priority to group goals. Collectivists tend to be high in power distance, tend to be concerned about the results of their behavior on members of their in-groups, usually share their resources with in-group members, and feel involved in the lives of in-group members. Individualism has been found to be more prevalent in the United States than in East Asia (Hofstede, 1980; Triandis et al., 1988). Chinese students are less likely to perceive personal problems as being important enough to seek help for com-
281
pared with Caucasian students (Tracey et al., 1986). Hence, we hypothesized that students who scored higher on individualism would have more positive attitudes toward seeking professional psychological help. Cultural norms, such as heavy familial responsibility and concepts of privacy, have been found to have an impact on the help-seeking behavior of Asians in North America (Lin & Lin, 1978). Such behavior seems to be related to the general social-network orientation that Asian Americans have. Social-network orientation was defined by Tolsdorf (1976) as a set of beliefs, attitudes, and expectations concerning the potential usefulness of network members in helping an individual cope with a life problem. We hypothesized that social-network orientation would be a significant predictor of attitudes toward seeking professional psychological help and that there would be an inverse relationship between negative social-network orientation and positive attitudes toward seeking professional psychological help. Therefore, this study examined the effects of acculturation, social-network orientation, and the values underlying the constructs of individualism and collectivism on the attitudes toward seeking professional psychological help among a group of Chinese-American college students in the midwestem United States. More specifically, we hypothesized that acculturation, level of individualism, and negative social-network orientation would have a significant direct relationship with attitudes toward seeking professional psychological help as measured by the total score on the Attitudes Toward Seeking Professional Psychological Help (ATSPPH) Scale. The relationship between gender and the criterion variable was also explored. Method
Subjects The subjects for this study were Chinese-Americans students enrolled at the University of Illinois at Chicago in the fall of 1991. The surveys used in this study were sent to 640 students who were thought to be Chinese American. All of the 219 students (117 women and 102 men) whose data were used for analysis identified themselves as being Chinese American on the survey used in this study. The mean age of the sample was 24.73 years (SD = 7.74, range = 17-60). The modal age was 20 years, and approximately 70% of the sample fell in the age range of 19 to 28 years. The distribution of the sample with regard to year in school was fairly even, with 15% being freshmen, 14% being sophomores, 17.5% being juniors, and 21.5% being seniors. Of the 67.5% of the participants who were undergraduates, the proportion of students in their senior year was greater than was the proportion of students in each of the other undergraduate years. Of the sample, 14% were graduate students, and almost 14% were professional students.
Measures Attitudes Toward Seeking Professional Psychological Help Scale. Fischer and Turner (1970) developed and published the ATSPPH Scale. This 29-item Likert scale has four subscales: Recognition, covering the recognition of need for professional
282
SHIRAZ PIROSHAW TATA AND FREDERICK T. L. LEONG
psychological help; Tolerance, covering the tolerance of stigma associated with seeking professional psychological help; Interpersonal, covering interpersonal openness; and Confidence, covering confidence in the mental health profession. Each of the 29 items on the ATSPPH Scale is scored on a 4-point scale ranging from strongly disagree (1) to strongly agree (4). A high total score on the scale indicates a positive attitude toward seeking professional help for psychological problems. A total score was obtained by summing the item scores. Acceptable reliability and validity estimates were obtained from several studies (Fischer & Cohen, 1972; Fischer & Turner, 1970). Adaptations of this scale have been used with Vietnamese-American students (Atkinson, Ponterotto, & Sanchez, 1984) and with ChineseAmerican, Japanese-American, and Korean-American students (Atkinson & Gim, 1989). The modified version of the ATSPPH Scale that was used in this study is similar to the modified versions used by Good, Dell, and Mintz (1989) and Atkinson and Gim (1989) in their studies with college students. The changes from the original version of the ATSPPH Scale involved replacing the words mental problems with emotional/personal problems or difficulties, the term mental health clinic with counseling center, and the term mental hospital with psychiatric inpatient unit. Thus, the modifications were made mainly to make the scale more relevant to college students and to the present time. In this study, a brief description of the titles of counselor and psychologist and the kinds of challenges with which they are professionally trained to help people was developed and inserted into the directions for this scale. Suinn-Lew Asian Self-Identity Acculturation Scale (SL-ASIA). The SL-ASIA (Suinn, Rickard-Figueroa, Lew, & Vigil, 1987) is a 21-item, multiple-choice scale that assesses four content areas. These areas are language familiarity, usage, and preference; ethnic identity; cultural behaviors; and ethnic interactions. Participants are asked to choose the one response that best describes them. When the scale is scored, a total value is obtained by summing the scores across answers for all 21 items. An acculturation score is obtained by dividing the total score by the number of items on the scale. Thus, acculturation scores can range from 1 to 5. Suinn et al. (1987) indicated that persons with scores close to 1 are considered Asian-identified, or low in level of acculturation, persons with scores around 3 are considered bicultural, and persons with scores around 5 are considered western-identified, or high in level of acculturation. Suinn et al. (1987) reported the reliability coefficient of the scale to be .88. They also confirmed the criterion validity of the scale by using three different approaches. They found a direct relationship between acculturation and the variables of length of stay in the United States, generation, and respondents' self-rating of cultural identity on a 5-point scale ranging from very Asian (1) to very Anglicized (5). The modified version of the SL-ASIA that was used in this study is almost identical to the one used by Atkinson and Gim (1989). The reliability coefficient for their modified version of the SLASIA was .89. The only difference was that Item 20 of the original SL-ASIA, which asks respondents to rate themselves as very Asian, mostly Asian, bicultural, mostly Anglicized, or very Anglicized, was included in this study's survey. However, this item's score was not used when the total acculturation score was computed. Rather, the score on this item was used to provide some validity information on the SL-ASIA with a sample of students from the midwestern United States. Network Orientation Scale (NOS). The NOS (Vaux, 1985) is a 20-item scale that was developed to measure the construct of network orientation (Tolsdorf, 1976). The three factors that emerged from the factor analysis of the items on the NOS were
Independence-Advisability, History, and Mistrust (Vaux, 1985). Thus the NOS measures people's orientation to having a social network by assessing their feelings about the advisability or usefulness of seeking help, their past history of actually having sought help, and the extent to which they feel that others cannot be trusted. In addition to designing the NOS and putting it through factor analysis, Vaux, Burda, and Stewart (1986) assessed its reliability and validity by conducting studies with five samples from different ethnic and cultural groups. The NOS was found to have good criterion validity with regard to personality variables and social support, but this level of validity was not as evident with respect to self-disclosure and interpersonal modes of coping. Reliability of the NOS was assessed in terms of the homogeneity of the scale and in terms of its stability over time. The good internal consistency of the scale is evident by the fact that the coefficient alphas in Vaux et al.'s (1986) study ranged from .88 for Sample 2 to .60 for Sample 5, which later showed a higher alpha of .74. In this study, subjects responded to each item on a scale that ranged from strongly agree (1) to strongly disagree (4). As recommended by Vaux (1988), the scores were reversed for the positive items before the item scores were summed to obtain a negative network-orientation score. Therefore, a higher score was indicative of a more negative social-network orientation, and a lower score was indicative of a more positive social-network orientation. Measure of individualism-collectivism. Triandis et al. (1988) recommended the use of the combination of two of their measures of individualism to measure this construct in the United States. These measures include a 29-item scale and a measure with three scenarios (Triandis, Leung, Villareal, & Clack, 1985) to which subjects are asked to respond. Triandis et al. (1988) developed a measure, the IndividualismCollectivism Scale, with 29 items for individualism in their attempt to operationalize the constructs of individualism and collectivism and to find a way to best measure them. In the process of developing this measure, they found that U.S. individualism is reflected in three main factors. The names given to the subscales of their measure of individualism or idiocentrism reflected these factors: (a) Self-Reliance With Competition, (b) Low Concern for In-Groups, and (c) Distance From In-Groups. We decided to use this measure in this study because Triandis and his associates have done the most extensive work on the constructs of individualismcollectivism at a cultural level and of idiocentrism-allocentrism at an individual level. Reliability and validity studies done on the scale have also indicated that the scale has good construct and criterion validity (Triandis et al., 1985). For example, Triandis et al. (1985) found that their measure of individualism-collectivism had an acceptable level of reliability (Cronbach's alpha = .74) and was significantly correlated to certain values in the theoretically expected direction. High levels of collectivism were correlated with values such as cooperation, equality, and honesty, whereas high levels of individualism were correlated with hedonistic values such as a comfortable life, pleasure, and social recognition. In terms of scoring the scale, before the addition of all the item scores for a total individualism score, all items were scored so that a higher score would be indicative of a higher level of individualism. Given the lack of information about the discriminant validity of the subscales, only the total score on individualism was used as a predictor variable in this study.
Procedure Subjects were asked to respond to four questionnaires, which were integrated into one mail survey instrument. The question-
SPECIAL SECTION: CHINESE AMERICANS' PREDICTORS OF HELP SEEKING naires included adaptations of the ATSPPH Scale and the SLASIA. The scales that were used in their original form were the NOS and the 29-item scale of individualism-collectivism recommended by Triandis et al. (1988). Three mailings, the second being a mailing of reminder postcards, were conducted to try ensure that there would be an adequate response rate (Dillman, 1978). A cover letter was included with the mail surveys. Of the 640 students to whom the surveys were sent, 274 responded by returning completed surveys (response rate = 42.81%). The 55 surveys not used as part of the data analysis were completed by persons who, on the basis of the demographic information they provided, were determined to not be Chinese-American students.
Results Means and standard deviations were computed for each of the primary variables (Table 1) to compare the data of this study with that of previous studies. To determine whether the small changes that were made in the wording of items on the scales affected the reliability of the scales, we calculated internal consistency reliabilities for each of the scales used in this study. We used a t test to compare the differences between our mean score (74.94) on the ATSPPH Scale and that of Atkinson and Gim's (1989), and found that their mean score (77.87) was significantly higher than was ours, f(765) = 3.19, p < .01. However, Atkinson and Gim did not report a total mean score for the ATSPPH Scale but instead reported mean scores for the low-, medium-, and highacculturation groups, and we used an average of those three means (M = 77.87) for comparison. A mean score on the NOS score was not available from Vaux's (1985) study for comparison. In terms of reliability, we tested for the differences between our internal consistency score (r = .80) and that of Atkinson and Gim's (r = .86) for the ATSPPH Scale and found that the z score was not significantly different (z observed = 0.06, p > .05). In addition, the mean internal consistency score (r = .74, range = .60-.88) in Vaux's study was identical to our internal consistency score (r = .74). The
Table 1 Means, Standard Deviations, and Internal Consistency Reliabilities of Variables Variable
M
SD
Cronbach's a
Level of acculturation Total score on the SL-ASIA (16) 47.18 9.56 NA Acculturation score 2.95 0.60 .87 Negative social-network orientation (20) 44.80 5.39 .74 Total individualism score (29) 66.98 7.75 .78 Attitude toward seeking professional psychological help (29) 74.94 7.93 .80 Note. Numbers in parentheses are the number of items on the scale used to measure the preceding variable. On all of the variables except for the Acculturation Scale (the Suinn-Lew Asian Self-Identity Acculturation Scale; SL-ASIA), items were rated and scored with the following scale: 1 = strongly disagree, 2 = disagree, 3 = agree, 4 = strongly agree. NA = not applicable.
283
internal consistency reliability of the acculturation scale used in the present study was .87, which is almost identical to .88, the alpha reported in Suinn et al.'s (1987) initial report on the SL-ASIA. Although there are some sample differences, the scales used in the present study seem to have been as reliable for our sample as when they were used in earlier studies. On the Acculturation Scale (i.e., the SL-ASIA), the distribution of scores obtained by the sample of the present study differed from that of Atkinson and Gim's (1989) study. For the purpose of comparing our sample with that of Atkinson and Gim, the percentage of the two samples that fell into the low-, medium-, and high-acculturation categories was computed. A larger percentage of the sample in the present study fell into the low- and medium-acculturation categories as compared with Atkinson and Gim's sample. We conducted a chi-square analysis to compare the number of individuals who were classified as low (1.00-2.33 mean score on the Acculturation Scale), medium (2.33-3.66) or high (3.66-5.00) on acculturation in both samples. We found that the distribution of the two groups was significantly different, ^ ( 2 , N = 767) = 55.70, p < .001, with the Atkinson and Gim sample having more high-acculturation individuals (40% vs. 12%) and fewer medium- (50% vs. 71%) and low- (9% vs. 16%) acculturation individuals than did our sample. This difference does not seem to be a function of minor changes made in the scale for this study. Asians with higher levels of acculturation are defined as those who are more western-identified in their customs, values, and behaviors; whereas those with lower levels of acculturation are more Asian-identified. Following Suinn et al.'s (1987) approach, the validity of the SL-ASIA was confirmed in the present study by a significant positive correlation (r = .74) between acculturation and the score obtained by subjects on Item 20. This item, which asks subjects to rate themselves on a 5-point scale ranging from very Asian (1) to very Anglicized (5), was used as a validity item by Suinn et al. as well. Thus, the SL-ASIA seems to be a valid measure of acculturation for Chinese-American college students in the midwestern United States, even though it was developed and validated on the West Coast. Because the individualism scales have not been used before, no means and standard deviations were available for comparison. The internal consistency reliabilities for the individualism scales as a whole were good, indicating an acceptable level of stability for these scales to be used as research instruments. The subscales varied in their level of internal consistency. For the Individualism-Collectivism Scale, alpha internal consistency reliability coefficients for the Self-Reliance With Competition and the Low Concern for In-Groups subscales were within acceptable experimental limits, but the reliability estimate of the Distance From In-Groups subscale was minimally acceptable (r = .55). A correlation matrix was also a product of the preliminary analyses (see Table 2). All of the correlations between the independent variables of acculturation, negative network orientation, and values inherent in the construct of individualism-collectivism were low to moderate. This re-
284
SHIRAZ PIROSHAW TATA AND FREDERICK T. L. LEONG
Table 2
Intercorrelations of Variables Variable 1. Acculturation scale (20) -.34 -.20 .16 2. Negative network orientation (20) — .40 -.33 3. Total individualism score (29) — -.33 4. Attitude toward seeking professional psychological help (29) Note. Numbers in parentheses are the number of items on the scale used to measure the preceding variable.
suit reduced the possibility of having to deal with the problem of multicollinearity in the primary analyses. To determine whether men and women differed significantly in their scores on the primary variables, we performed t tests. A significant gender difference was found on the dependent variable of attitudes toward seeking professional help. Women had a significantly higher total score on the ATSPPH Scale than did men, r(217) = 2.28, p < .05. This result means that women expressed more positive attitudes toward seeking professional psychological help than did men. This significant difference led to the decision to include gender as a predictor variable in the regression analysis. A multiple regression analysis procedure was used to assess the degree of impact that each of the independent variables had on the dependent variable of attitudes toward seeking professional help. A hierarchical regression analysis was conducted with gender, cultural identity, socialnetwork orientation, and individualism as the predictors for the total score on the ATSPPH Scale (Table 3). Gender was entered in the first block because it is a primary demographic variable. Acculturation was entered in the second block before the other predictor variables because previous research had been done on the impact of acculturation on
attitudes toward seeking help. The effect of acculturation was also controlled by entering it on this block. The variables of social-network orientation and individualistic values were entered in the third block because these were the two variables that had not been researched before in the context of acculturation and help-seeking attitudes. Interactions between the above variables were entered in the fourth block because entering interactions in the last block has been suggested by Wampold and Freund (1987). The interaction effects entered in the fourth block included Gender x Individualism, Gender x Social-Network Orientation, Acculturation x Individualism, Acculturation x SocialNetwork Orientation, and Gender x Acculturation. The overall regression model was significant for the criterion variable of attitudes toward seeking professional psychological help, F(9, 209) = 5.05, p < .001, as noted in Table 3. The results of the multiple regression analysis showed that each of the four independent variables of gender, acculturation, social-network orientation, and individualism was a significant predictor of the total score on the ATSPPH Scale. In predicting attitudes toward seeking professional psychological help, none of the interaction effects added unique predictive variance beyond that accounted for by each of the independent variables itself. Gender accounted for 2% of the variance and was a significant predictor. In other words, gender was significantly predictive of attitudes toward seeking professional psychological help, and women were significantly more likely than were men to have positive attitudes toward seeking professional psychological help. Acculturation, which accounted for 3% of the variance, contributed significant incremental variance over gender. The results supported the hypothesis that students who were more acculturated, or Anglicized, would have more positive attitudes toward seeking professional psychological help. Individualism and social-network orientation accounted for 12% of the variance and made a significant contribution to the prediction of the criterion variable. The standardized
Variable
R2
R2 change
F change
"CD
Table 3 Hierarchical Multiple Regression of Effects of Gender, Acculturation, IndividualismCollectivism, and Social-Network Orientation on Total Score of the Attitudes Toward Seeking Professional Psychology Help (ATSPPH) Scale t
.02 5.18* -.15 -2.27* .02 Block 1: Gender .04 .03 5.74* .16 2.39* Block 2: Acculturation Block 3 .17 .12 15.32*** — — Individualism — — -.23 -3.35*** — — — -.21 -2.88** Social-network orientation Block 4: Interactions between variables .18 .01 0.55 — — Note. The overall regression model was significant, F(9, 209) = 5.05, p < .001. For acculturation, a higher score denotes a higher level of acculturation, or more of a Western rather than Asian identification. On individualism, a higher score denotes a higher level of individualism. On social-network orientation, a higher score denotes a negative network orientation, whereas a lower score denotes a positive network orientation. On the ATSPPH Scale and its subscales, higher scores denote higher levels of the variable in question and more positive attitudes toward seeking professional psychological help. Dashes indicate not applicable.
* p < .05. **p
