In the regression analysis in which anxiety scores were used as the dependent variable to test the hyothesis that encounter attitudes would be positively associated with feelings of anxiety, the 8% of the variance explained by racial identity attitudes was significant, F(4,161) = 3.53, p < .05. Encounter atti- tudes were significantly related to feelings of anxiety, but in a negative direction. In addition, both preencounter and immersion attitudes were positively related to anxiety, although no specific hypotheses about these attitudes have been proposed.
The test of the hypothesis that feelings of anger would be positively related to immer- sion attitudes revealed no significant effect due to the combination of racial identity attitudes, F(4,161) = 1.32, p > .05. How- ever, the Immersion attitude scale was a significant predictor of anger (Hostility scale). The direction of the beta weight suggested that problack-antiwhite attitudes were likely to be associated with feelings of anger or hostility as predicted.
The overall regression model, testing the hypothesis that feelings of self-acceptance would be predicted by internalization atti- tudes, barely missed significance, F(4,161) = 2.42, p = .06. Examination of its beta weight indicated that internalization atti- tudes were not significantly related to feel- ings of self-acceptance. Because the overall model was nearly significant (an F of 2.425 was necessary for significance at the .05 level), we also examined the beta weights for the other attitudes. Preencounter attitudes were inversely related to self-acceptance, indicating that prowhite-antiblack attitudes were indicative of difficulty in accepting oneself in spite of the absence of identifica- tion with one’s ascribed racial group. En- counter attitudes were positively related to feelings of self-acceptance, indicating that making a decision to question previously held negative assumptions about blackness may be indicative of emerging feelings of self-acceptance. The final hypothesis that obsessiveness would be related to encounter attitudes was not supported by the obtained results, F(l, 161) = 0.43, ns; the overall model also lacked significance, F(4, 161) = 1.14, ns.
In summary, it seems that with the ex- ceptions of anger, obsessiveness, and possi- bly self-acceptance, affective states were
BLACK STUDENTS’ RACIAL IDENTITY 437
predicted by linear combinations of the ra- cial identity attitudes, with particular atti- tudes being differentially related to specific affects, though not always as expected.
Secondary Analyses
Secondary analyses were conducted to explore the extent to which racial identity attitudes could be predicted from demo- graphic characteristics. Several additional regression analyses were conducted. In these analyses, social class indicators, racial self-designation, class level, age, and sex served as successive predictor variables, and mean scores on the four racial identity atti- tude scales served as the dependent vari- ables.
Results of the regression analyses indi- cated that racial identity attitudes were not significantly predicted by social class indi- cators, academic class, racial self-designa- tion, or age (all Fs < 1). However, sex sig- nificantly predicted preencounter, /3 = -0.26, F(l, 64) = 12.53,p < .005, and inter- nalization attitudes, /3 = 0.16, F(l, 164) = 4.5, p < .05, but not encounter or immer- sion-emersion attitudes. The direction of the beta weights suggests that black men were more likely to endorse preencounter attitudes and less likely to endorse inter- nalization attitudes than black women.
To explore the nature of these two ob- served sex differences further, one-way analyses of variance comparing men and women on each of the dependent and inde- pendent variables were performed. Means, standard deviations, and F ratios for these analyses are shown in Table 2. The analyses of variance revealed that in addition to lower preencounter and higher internalization attitudes, black women also exhibited sig- nificantly higher levels of inner directedness than did black men.
Discussion
Since the early 1970s, black scholars have speculated about the relation between racial identity attitudes and self-actualization, or the nigrescence process, and affective states that are presumably related to each stage of racial identity (e.g., Butler, 1975; Cross, 1971; Thomas, 1971). The bulk of existing theo-
retical literature seems to suggest that an individual’s progression from Stage 1 (preencounter) to Stage 4 (internalization) is marked by transitions from feelings of inferiority to self-acceptance and from non-self-actualizing to self-actualizing atti- tudes and behaviors. The results of the present study, which is the first to examine empirically the relation between racial identity attitudes, self-actualization tendencies, and affective states, suggest that the racial identity process either may be more complex then previous authors had speculated or it may be a more difficult process to operationalize for diagnostic purposes than one might anticipate.
Consistent with previous theory were the obtained relations between preencounter attitudes and the other personality variables. That is, the findings that preencounter at- titudes were related to lower levels of time competence and higher levels of other di- rectedness (i.e., self-actualizing tendencies, according to Shostrom, 1963) as well as to feelings of inferiority, inadequacy, hyper- sensitivity, anxiety, and lack of self-accep- tance are consistent with theoretical dis- cussions in which the preencounter stage is described as least mentally healthy (e.g., Butler, 1975; Cross, 1971).
The relation between immersion attitudes and affective states and self-actualizing tendencies was least consistent with con- temporary theory about the developmental process. Instead of indicating that these attitudes reflect a positive sense of self due to the acceptance of one’s blackness, the re- sults of the present study suggest that im- mersion attitudes (and by implication the immersion stage) are affectively similar to preencounter attitudes. The only difference between the two types of attitudes and the only relation that was anticipated by prior theory was that anger was related to im- mersion attitudes. However, use in the present study of the SCL-90 Hostility scale to operationalize anger feelings does not permit one to determine whether the anger was directed inward (i.e., toward oneself in response to one’s previous identity resolu- tions) or outward (i.e., toward society in re- sponse to discrimination). The latter form presumably is more healthy, and to the ex- tent that outward-directed anger typifies
438 THOMAS A. PARHAM AND JANET E. HELMS
immersion attitudes, it is still possible that immersion attitudes represent somewhat healthier adjustment than preencounter attitudes.
Cross (1978) speculated that encounter attitudes are a muted form of immersion attitudes and that both are alike in their positive orientation toward blackness. The results of the present study suggest that the positive perspective may be more charac- teristic of encounter attitudes than of any of the other attitudes, including immersion. In fact, the present findings that encounter attitudes were predictive of feelings of per- sonal adequacy, self-acceptance, and low levels of anxiety indicate that encounter at- titudes, as operationalized in the present study, may capture the euphoric feelings about becoming black that Cross discussed, but not the feelings of guilt and anxiety that have also been considered part of the en- counter stage. It is possible that entry into the encounter stage is a uniformly positive experience rather than a mixture of positive and negative; it is also possible that current measures may not be sensitive enough to capture the subtle nuances of affective states making up this stage.
At first glance, it is surprising that inter- nalization attitudes were not significantly related to any of the measures of affective states or self-actualizing tendencies. How- ever, a closer examination of the regression analyses indicates that these attitudes seemed to be related to the other measures in the same direction as were encounter at- titudes, though not significantly so. One possible explanation for the lack of signifi- cant relations is that internalization atti- tudes may represent a muted form of en- counter attitudes, that is, encounter atti- tudes with the emotion removed. In his early descriptions of internalization, Cross (1971) described it as a stage governed pri- marily by the intellect rather than affect. Because all of the measures used in the present study were measures of affect of some sort, it is possible that they were not suitable for capturing the rational focus of internalization attitudes. Further studies, which should include measures of cognitive style, affective state, and racial identity at- titude, might be useful in providing addi- tional insight about the internalization stage and consequent attitudes.
Although not all of our hypotheses were confirmed, the present findings suggest that emotions not only are present but also may be a vital part of the conversion experience. Also, to the extent that one can infer stages from attitudes, speculation that the domi- nant affect varies at different stages of the process (e.g., Pugh, 1972) appears to have received some support from the obtained results. Nevertheless, the results do raise some interesting theoretical, methodological, and counseling practice issues in addition to those already discussed.
An interesting theoretical possibility is that cognitive aspects of the racial identifi- cation process and affective aspects may not evolve at the same rate or by the same pro- cess. Studies of attitudes in other areas of psychology have often reported lack of con- gruence between cognitive, affective, and behavioral elements (e.g., Kutner, Wilkins, & Yarrow, 1952; La Piere, 1934; Weitz, 1972), though no commonly accepted explanation for discrepancies seems to exist. In the present instance, it is possible that cognitive aspects of the racial identification process such as attitudes and perceptions may evolve by a stagewise linear model, as Hall et al. (1972) found, but that affective states evolve by a different model. If such is the case, then one possibility is that a typology con- sisting of healthy and unhealthy affective states might be most useful in interpreting the relation between racial identity attitudes and affect. Thus, on the basis of the data at hand, a predominance of preencounter and immersion attitudes might predict un- healthy affective adjustment and a pre- dominance of encounter and internalization attitudes might predict healthy affective adjustment. If this typology is accurate at all, then it may provide a diagnostic frame- work by which the counselor can decide whether to intercede in the black client’s self-actualization process to promote better adjustment. That is, clients demonstrating a preponderance of preencounter or im- mersion attitudes might require such inter- cession, whereas clients demonstrating a preponderance of encounter or internaliza- tion attitudes might not.
Of course, it is also possible that Cross’s (1971) model is no longer an accurate de- scription of black people’s reactions to the social conditions that they face. It is
BLACK STUDENTS’ RACIAL IDENTITY 439
tempting, for instance, to hypothesize that the model accurately characterized the ni- grescence process of black people in the late 1960s and early 1970s hut that present day hlacks, struggling to find their identity, are influenced hy a different set of personal, social, and environmental factors; as a con- sequence, they may have learned to adapt differently than did their predecessors. If such is the case, then perhaps a stagewise progression of the cognitive aspects of identity development is no longer accurate either. However, the questions of whether the Cross model should be modified in the manner discussed and whether the model continues to be useful and accurate can only be answered through additional empirical investigations involving measurement of racial identity attitudes in conjunction with various operationalizations of adjustment. In addition, longitudinal studies of identity development are greatly needed.
In interpreting the results of the present study, it is important to take several meth- odological issues into consideration. First of all, only one aspect of a person’s iden- tity—his or her adaptations to race and ra- cism—was investigated. Thus, even when the regression analyses were significant, only 8% (interpersonal sensitivity and anxiety) to 18% (time competence) of the variance was explained by linear combinations of racial identity attitudes. This range of effect sizes compares favorably with the median effect size of 8% of explained variance reported by Haase, Waechter, and Solomon (1982) in their review of univariate analyses reported in counseling research, but one still wonders what other factors might contribute to a person’s identity. Perhaps other demo- graphic characteristics (e.g., social class) may influence the person’s global identity to some extent. Nevertheless, the results of the present study suggest that where racial as- pects of that identity were concerned, only sex seemed to be a significant demographic predictor. Black women exhibited lower levels of preencounter attitudes and higher levels of internalization attitudes and inner-directed self-actualizing tendencies than black men. It is possible that because black women experience less diversity of experiences in white culture than black men do, they are more likely to rely on themselves for self-definition and are less likely to be-
lieve that their life situation can be improved by identifying with white attitudes and val- ues (Hooks, 1981; Jackson, 1973). In any case, it is difficult to form concrete conclu- sions about the relation between racial identity attitudes and demographic factors because although the sample size used in the present study was adequate for the analyses used, it was not of sufficient size or diversity to permit separate within-group analyses on the basis of various demographic charac- teristics (e.g., sex, age). Further studies involving samples of different ages, educa- tional levels, and socioeconomic statuses are needed.
The manner in which the different vari- ables were operationalized in the present study is also open to debate. For instance, the Racial Identity Attitude Scale (Parham & Helms, 1981), used to assess subjects’ racial identity attitudes, may require some modi- fications. Although the reliabilities of the subscales are comparable with those of other personality instruments, the measure’s usefulness could possibly be enhanced by improving the reliabilities (Anastasi, 1982). Such scale refinement seems particularly important because the Racial Identity At- titude Scale is one of only a few instruments designed to measure black personality characteristics (cf. Milliones, 1980; Snowden & Todman, 1982), and for it to become a commonly accepted tool in counseling as- sessment, it probably must be shown to be a marked improvement over the more general measures that already exist.
In addition, one might argue about the use of nonblack measures to operationalize af- fective states and self-actualizing tendencies. For example, for a person to obtain a high score on the POI Self-Acceptance Scale, he or she must endorse individualistically ori- ented items. However, theorists such as Akbar (Luther X, 1974) suggest that such an orientation is antithetical to healthy black development. As a result, one may not find the expected relations between variables that evolve from black personality theory because the available personality measures are not consistent with such theory.
Regardless of the theoretical and meth- odological issues raised, the results of the present study do offer some possibilities for understanding the dyamics of black people and for counseling those who are displaying
440 THOMAS A. PARHAM AND JANET E. HELMS
different levels of the various racial identity attitudes. In general, counselors who work with black clients who are struggling with racial identity issues should explore both the cognitive and affective aspects of their con- cerns because the two may not be related in an obvious manner. The goal of the coun- selor should be to help black clients integrate the various aspects of their identity by be- coming aware of what they think about their racial identity as well as how they feel about it. In other words, counselors should avoid merely inferring adjustment from stated attitudes and should actively explore the client’s emotional adaptations. Both counselors and clients may also need to be aware that although some of the feelings associated with particular racial identity attitudes are unpleasant and may require resolution, such feelings may be a natural part of the nigrescence process. Therefore, mental health workers (and researchers) should be cautioned against inferring serious pathology where none exists.
References
Anastasi, A. (1982). Psychological testing (5th ed.). New York: MacMillan.
Atkinson, D. E. (1983). Ethnic similarity in counseling psychology: A review of research. Counseling Psychologist, 11(3), 79-92.
Butler, R. 0. (1975). Psychotherapy: Implications of a black-consciousness process model. Psycho- therapy: Theory, Research and Practice, 12, 407-411.
Cross, W. E. (1971). The Negro-to-black conversion experience: Towards a psychology of black libera- tion. Black World, 20,13-27.
Cross, W. E. (1978). The Cross and Thomas models of psychological nigrescence. Journal of Black Psychology, 5,13-19.
Derogatis, L. R., Rickels, K., & Rock, A. F. (1976). SCL-90 and the MMPI: A step in the validation of a new scale. British Journal of Psychiatry, 128, 280-289.
Gardner, L. H. (1971). The therapeutic relationship under varying conditions of race. Psychotherapy: Theory, Research and Practice, 8. 78-86.
Gynther, M. D. (1972). White norms and black MMPls: A prescription for discrimination? Psy- chological Bulletin, 78, 386-402.
Haase, R. F., Waechter, D. M., & Solomon, G. S. (1982). How significant is a significant difference? Journal of Counseling Psychology, 29, 58-65.
Hall, W.S., Cross, W. E., & Freedle, R. (1972). Stages
in the development of black awareness: An empirical investigation. In R. L. Jones (Ed.), Black psychology (pp. 156-165). New York: Harper & Row.
Hooks, B. (1981). Ain’t I a woman: Black women and feminism. Boston: South End Press.
Jackson, G. G. (1977, summer). The emergence of a hlack perspective in counseling. Journal of Negro Education, 46, 230-253.
Jackson,J.J. (1973). Black women in a racist society. In C. C. Willie, B. Kramer, & B. Brown (Eds.), Racism in mental health (pp. 185-268). Pittsburgh, PA: University of Pittsburgh Press.
Klavetter, R. E., & Mogar, R. E. (1967). Stability and internal consistency of a measure of self-actualiza- tion. Psychological Reports, 21, 422^124.
Knapp, R. R. (1965). Relationship of a measure of self-actualization to neuroticism and extraversion. Journal of Consulting Psychology, 29, 168-172.
Kutner, B., Wilkins, C., & Yarrow, P. (1952). Verbal attitudes and overt behavior involving racial preju- dice. Journal of Abnormal and Social Psychology, 47, 649-652.
LaPiere, R. T. (1934). Attitudes vs. action. Social Forces, 13, 230-237.
Milliones, J. (1980). Construction of a black con- sciousness measure: Psychotherapeutic implica- tions. Psychotherapy: Theory, Research and Practice, 17,175-182.
Parham, T. A., & Helms, J. E. (1981). The influence of black students’ racial identity attitudes on pref- erence for counselor’s race. Journal of Counseling Psychology, 28, 250-256.
Pugh, R. (1972). Psychology of the black experience. Belmont, CA: Brooks/Cole.
Sattler, J. M. (1977). The effects of therapist-client racial similarity. In A. S. Gurman & A. M. Razin (Eds.), Effective psychotherapy (pp. 252-290). New York: Pergamon Press.
Shostrom, E. (1963). Personal Orientation Inventory. San Diego: Educational and Industrial Testing Service.
Smith, E. (1977). Counseling black individuals: Some strategies. Personnel and Guidance Journal, 55,390-396.
Snowden, L., & Todman, P. (1982). The psychological assessment of blacks: New and needed develop- ments. In E. E. Jones & S. J. Korchin (Eds.), Mi- nority mental health (pp. 193-226). New York: Praeger.
Thomas, C. (1971). Boys no more. Beverly Hills, CA: Glencoe Press.
Weitz, S. (1972). Attitude, voice, and behavior: A repressed affect model of interracial interaction. Journal of Personality and Social Psychology, 24, 14-21.
X, Luther (Luther Weems). (1974). Awareness: The key to black mental health. Journal of Black Psy- chology, 1, 30-31.
Received May 4,1984
Revision received November 28,1984 •
