The Role of Shame as a Mediator between Anti-black Racial Identity Attitudes and Negative Affect in a Sample of African American College Students

A sample of 168 African American undergraduates was surveyed to clarify past findings demonstrating a consistent relationship between endorsing negative attitudes about being African American and experiencing negative affect. Specifically, shame was tested as a mediator between participants' endorsement of preencounter attitudes (i.e., anti-Black attitudes) and both feelings of depression and global self-esteem, respectively. As predicted, shame acted as a complete mediator in both equations. Limitations and implications are also discussed. INTRODUCTION Personality theorists in recent decades have focused increasing attention on various aspects of African American personality and racial identity development (Baldwin & Bell, 1985; Parham & Helms, 1981; Sellers, Smith, Shelton, Rowley, & Chavous, 1998; Vandiver, Fhagen-Smith, Cokley, Cross, & Worrell., 2001). One of the major contributors to this area of research is William E. Cross, Jr. (1991), who proposed the theory of Nigrescence. Nigrescence theory attempts to explain how African Americans who experience racial oppression come to develop positive attitudes and identification with African American communities. Cross has formulated a series racial identity statuses instantiated by attitudes ranging from Black self-hatred (preencounter status) to Black pride and acceptance (immersion-emersion and internalization statuses; Cross, 1991; Vandiver et al., 2001). Cross (1991) developed this theory as a response to the proliferation of research asserting that because African Americans are evaluated negatively by the dominant culture (Correll, Park, Judd, & Wittenbrink, 2002; Dovidio, 2001; Mendes, Blascovich, Lickel, & Hunter, 2002), they would internalize such attitudes (for reviews, see Baldwin, 1979; Banks, 1976; Gray-Little & Hafdahl, 2000). Cross (1991) argued that the main studies upon which these theories were based were flawed because they failed to properly define self-concept (i.e., self-concept is a multifaceted construct consisting of both personal identity and reference group orientation). Thus, while personal identity is defined as those aspects of personality that are universal among all humans and putatively independent of socialization (e.g., anxiousness, self-esteem, introversion-extroversion, etc.), reference group orientation is that aspect of persona fabricated by cultural scripts (e.g., gender, racial, and class identities). Nigrescence was a response to decades of conceptual synecdoche in which these facets of self-concept were assumed to represent the whole. Consequently, many studies operationalized African American self-concept by using either measures of personal identity or reference group orientation, but not both. This has resulted in confusion concerning the particular mechanism by which racial group orientation might be associated with both personal identity and psychopathology. The present study is an initial attempt to clarify this confusion. Of most relevance to the current study is Cross's preencounter status. Although it has undergone several iterations, the consistent theme throughout these revisions has been that a core attitudinal stance of this status is one of Black self-hatred (Vandiver et al., 2001). Preencounter Attitudes and Negative Affect How are preencounter attitudes associated with negative psychological states? Carter (1991) found that “preencounter attitudes were significantly positively related to self-reported anxiety, memory impairment, paranoia, hallucinations, alcohol concerns, and global psychological distress" (p. 111). Munford (1994) and Pyant and Yanico (1991) found that preencounter attitudes appear to be positively correlated with depressive symptoms (r =.33 and .38, respectively). Why would endorsing negative attitudes about one's reference group be associated with negative affect? Having an immutable, ascribed membership in a group one does not value would likely create a sense of dissonance in individuals. Such status incongruence has been linked empirically (Higgins, Klein, & Strauman, 1985; Lundberg, Kristenson, & Starrin, 2009) and theoretically (Kaufman, 1996) to feelings of shame; however, research is lacking with regard to empirically examining the association of shame with preencounter attitudes Thus, the present study is a first step toward this end. Shame and Negative Affect What is shame? From the pudor (Kaster, 1997) of the Romans to the Chinese concept of face (Yau-fai Ho, 1976), shame has generally been defined as the feeling of wanting to disappear from view or hide from others (Lewis, 1987a; Tomkins, 1963), or as the experience of losing social attractiveness (Van Vliet, 2008). How is shame associated with negative affect? Extensive theoretical (Kaufman, 1996; Lewis, 1987b) and empirical (De Rubeis & Hollenstein, 2009; Gilbert, McEwan, Bellew, Mills, & Gale, 2009) research supports a strong relationship between depressive affect and shame. The correlation between shame and depression has been found to range from .59 to .79 (Cook, 1994). Preencounter Attitudes, Shame, and Depression Shame and the preencounter stance both appear to tap heavily into themes of self-disaprobation; however, while shame is a global indictment of the self, preencounter attitudes represent a more circumscribed facet of shame (i.e., racial identity, rather than the whole self, is reproached). Further, while both shame and preencounter attitudes are associated with depressive symptoms, the correlation between shame and depression is generally much larger than that between preencounter attitudes and depression. Consequently, I hypothesized that shame would act as a mediator between the endorsement of preencounter attitudes and the experience of depressive affect. Preencounter Attitudes, Shame, and Global Self-Esteem The relationship of African American racial identity attitudes with global self-esteem is a complex one (Cross, 1991); however, past research has demonstrated a consistent negative correlation between endorsement of preencounter attitudes and self-esteem (Collins & Lightsey, 2001; Munford, 1994; Pyant & Yanico, 1991; Speight, Vera, & Derrickson, 1996). With regard to the relation of shame with global self-esteem, research has consistently demonstrated a strong negative correlation between these variables (Brown & Marshall, 2001; Thompson, Sharp, & Alexander, 2008; Yelsma, Brown, & Elison, 2002). However, Cook (1994) asserts that while low levels of self-esteem ought to be associated with increasing experience of shame, low levels of shame are not necessarily indicative of higher self-esteem. Working from this theory, shame and global self-esteem would likely be more strongly associated with each other than either would be with preencounter attitudes. Why? If Cross's (1991) definition of selfconcept is correct, then shame and global self-esteem are both facets of the common superordinate construct of personal identity. Conceptually, this suggests a much stronger relation between these variables. Contrastingly, preencounter attitudes are part of the reference group construct, and therefore ought to be related to, but more distal from, personal identity. Finally, why place shame as a mediator between preencounter attitudes and global self-esteem rather than reversing this relationship (i.e., placing self-esteem as a mediator between shame and preencounter attitudes)? Assuming that shame and self-esteem are subsumed under a common construct of personal identity, then this begs the question: How are preencounter attitudes associated with personal identity? Intuitively, the answer appears to lie in the fact that both shame and preencounter attitudes tap into a general negative self-evaluation construct (i.e., there is no affectively "positive" aspect to the endorsement of either stance). However, global selfesteem has both negative and positive qualities (Rosenberg, Schooler, Schoenbach, Rosenberg, 1995). Thus, if we accept Cook's (1994) assertion that shame represents the negative side of global self-esteem, and we remember that in Cross's (1991) theory preencounter attitudes reflect a negative evaluation of being Black, then the logical link become clear: shame bridges the gap between preencounter attitudes and self-esteem through the theme of negative self-evaluation. To review, the specific hypotheses of this study are as follows: Hypothesis 1: Shame was expected to mediate the relationship between preencounter attitudes and depression. Hypothesis 2: Shame was expected to mediate the relationship between preencounter attitudes and global self-esteem.