![]()
Assessment of Self-Esteem Among Individuals with Severe Mental Illness: Testing Two Dimensions of Self-Esteem Theory and Implications for Social Work Practice
Sang Kyoung Kahng
Sang Kyoung Kahng, PhD, is Assistant Professor, Department of Social Welfare, College of Social Science, Seoul National University.
Carol Mowbray
Carol Mowbray, PhD, is affiliated with School of Social Work, University of Michigan, Ann Arbor, MI.
Address correspondence to: Sang Kyoung Kahng, PhD, Department of Social Welfare, College of Social Science, Seoul National University, San 56-1, Sillim-Dong, Gwanak-Gu, Seoul, 151-742, Korea.
SUMMARY. This study examines whether the factor structure of the Rosenberg Self-Esteem Scale consists of two dimensions (Owens, 1993), and whether the two dimensions are associated with different predictive factors and behavioral outcomes among individuals with severe mental illness who are served by community-based, psychosocial rehabilitation agencies. Confirmatory factor analyses of data from these individuals
indicated that the traditional global self-esteem scale does reflect two dimensionsāi.e., self-enhancement and self-deprecation. Bivariate and multivariate analyses revealed that factors associated with self-enhancement differ from factors associated with self-deprecation. These findings support the validity of two dimensions of self-esteem. Implications for social work practice and research are presented.
[Article copies available for a fee from The Haworth Document Delivery Service: 1-800-HAWORTH. E-mail address: <[email protected]> Website: <http://www.HaworthPress.com> Ā© 2005 by The Haworth Press, Inc. All rights reserved.] KEYWORDS. Self-esteem, measurement, mental illness, psychosocial rehabilitation, social work practice
Rosenberg (1965) defined self-esteem as a positive or negative attitude toward the self. As certain attitudes toward an object frequently entail corresponding psychological, social, or behavioral responses to the object (Petty et al., 1997), a personās self-esteem is closely associated with his/her psychological, social, and behavioral outcomes. Individuals with high self-esteem tend to adapt better when they experience major stressors because they have more efficient coping strategies (Carver, Scheier, & Weinstraub, 1989). Self-esteem plays a role as an effective defense against stressful consequences and can prevent negative health and mental health outcomes (Pearlin, 1987; Lazarus & Folkman, 1984). It is, therefore, important for individuals to maintain certain levels of self-esteem to successfully cope with stressors and to achieve optimum outcomes. Thus, the measurement of self-esteem is very relevant to social work research and practice.
In the social work literature, there is substantial evidence that high self-esteem is associated with positive outcomes. For homeless people, lower self-esteem relates to high depressive symptoms and poor health status (Diblasio & Belcher, 1993). Robinson (2000) found that among ethnic minority (i.e., African Caribbean) adolescents, higher self-esteem was related to positive racial identity attitudes. In a study examining the relationships between self-esteem instability and sexual behavior among gay and bisexual men, Martin and Knox (1997) found that those who presented unstable self-esteem were more likely to report avoidance coping behaviors, loneliness, and lower social support compared to those whose self-esteem was relatively stable. In addition, individuals with unstable and lower self-esteem reportedly engaged more in risky sexual behaviors (e.g., unprotected sex with strangers).
Self-esteem has been found to be associated with welfare use. Kunz and Kalil (1999) found that individuals with lower self-esteem early in life were more likely to receive welfare (i.e., AFDC or TANF) by the age of 28, based on data from the National Longitudinal Survey of Youth. Savaya (1998) examined the associations between self-esteem and use of professional services among Arab women living in Israel and found that, on average, those who used professional services (e.g., doctor and hospital, prenatal care, mental health, family and marital counseling, employment, or social welfare) presented lower self-esteem than those who did not use these services. The self-esteem of those who received services from employment or social welfare agencies was especially low.
These findings from the social work literature suggest that self-esteem may be of critical interest for social workers because it is associated with individualsā psychological outcomes (e.g., depression, loneliness), social outcomes (e.g., social support), and behavior (e.g., sexual behavior or human services use). Social work research has focused on self-esteem as a key variable in evaluating the effectiveness of interventions. For example, in an evaluation of the long-term effectiveness of group counseling for women who were survivors of child sexual abuse, self-esteem was a major outcome variable (e.g., Bagley & Young, 1998).
The previously described social work literature on self-esteem utilized the Rosenberg Self-Esteem Scale (Rosenberg, 1965). Reflecting its position as a major measurement tool, the scale has been translated into American sign-language and its internal consistency has been assessed for individuals with hearing disabilities (Crowe, 2002). Consistent with its original development, the scale has been used by social work researchers as a unidimensional measure. However, recent research indicates that measures of global self-esteem combine two different dimensionsāpositive (i.e., self-enhancement) and negative (i.e., self-deprecation) (Owens, 1993, 1994). Owens (1993) argued that a more precise understanding of negative self-esteem in terms of its development and maintenance has been hindered because researchers tend to overemphasize global self-esteem. He offered empirical evidence of the two dimensions, showing the differential impact of negative and positive self-evaluations on emotional and social well-being among adolescents. For example, self-deprecation was strongly related to depression, whereas self-enhancement was not (Owens, 1994). By contrast, both self-deprecation and self-enhancement were related to adolescentsā grades in school. Thus, correlates of the two dimensions of self-evaluation are not always the same. Despite its validated differential impact, sub-dimensions of global self-esteem have not been examined for social work clients and the validity of the two dimension self-esteem measure has not been tested among individuals with severe mental illness.
Current Study
This study aims to address the gaps in previous social work literatureāthat is, the lack of studies testing the validity of the two-dimensional theory of self-esteem among social work clients. Participants in this study are individuals with serious mental illness, being served in psychosocial rehabilitation agencies. Due to deinstitutionalization, increasing numbers of individuals with mental illness seek services from community agencies, where social workers serve as the principal professional group providing mental health services (Stromwall, 2002). Individuals with mental illness experience various stresses related to their mental illnessānot only symptoms but also the stigma of mental illness (Link et al., 1989, 1997) which often negatively influences their self-esteem (Link et al., 2001). Thus, in this study, we also consider the relationship of self-esteem to psychological, social, and behavioral variables for these individuals.
As a means of testing the validity of the two-dimensional self-esteem measure, this study first examines the factor structure of the 10-item Rosenberg Self-Esteem Scale (Rosenberg, 1965), through confirmatory factor analysis, to test the theory of bi-dimensional self-esteem (Owens, 1993, 1994). After analyzing the factor structure, we present bivariate associations of both the global self-esteem measure and the two dimensions of self-esteem with measures of socio-demographics, mental health,...
