عزت نفس و زنان معلول

This study examines the sense of self of women with physical disabilities in terms of self-esteem, self-cognition (perceptions of how others see them), and social isolation. It was hypothesized that these variables mediate the relation of precursor variables (age, education, severity of disability, and childhood experiences, including overprotection, familial affection, and school environment) and outcomes (intimacy, employment, and health promoting behaviors). Data were gathered from a sample of 881 community-dwelling women in the USA, 475 with a variety of mild to severe physical disabilities, and 406 without disabilities. Correlation analyses indicated that the women with disabilities had significantly lower self-cognition and self-esteem, and greater social isolation than the women without disabilities, as well as significantly less education, more overprotection during childhood, poorer quality of intimate relationships, and lower rates of salaried employment. Path analysis indicated that each of the sense of self mediators was significantly related to the outcome of intimacy, that both social isolation and self-esteem were significantly related to health promoting behaviors, and that only self-esteem was significantly related to employment. Respondents who were older, less disabled, less educated, less over-protected, and had more affection shown in the home tended to feel that others saw them more positively. Women with positive school environments, less over-protection, and more affection in the home experienced less social isolation; age, education, and disability severity were not significantly related to social isolation. Older respondents with less disability, a more positive school environment, less over-protection, and more affection in the home tended to have greater self-esteem; education was not significantly related to self-esteem. Older respondents tended to report less intimacy. Younger, more educated, and less disabled respondents were significantly more likely to be employed. More highly educated respondents reported engaging in more health promoting behaviors.

Disability is a stigmatizing phenomenon. Its effects can be profound when combined with women's social devaluation. Yet, clinical experience shows that many women who acquire disability at birth or later develop and maintain high self-worth. The literature on the self-esteem of people with disabilities fails to explain these differences and the connections between self-esteem and health-related outcomes, particularly as related to gender. The proportion of women with physical disabilities in the United States population is substantial and growing. The 1992 Census reports that 26 million women have disability-related work limitations, comprising 20% of the total population of women (McNeil, 1993). Recent analysis of 1994–1995 Census data shows that 16% of women have at least one limitation in physical functioning (National Center for Health Statistics, 2002). Our study of women with physical disabilities revealed that this population experiences problems associated with low self-esteem, such as depression, unemployment, social isolation, limited opportunities to establish satisfying relationships, and emotional, physical, and sexual abuse (Nosek, Howland, Rintala, Young, & Chanpong, 2001). Is disability associated with poorer outcomes among women with disabilities? Disability in this study is defined as a physical limitation in activity. Limitations in activity are frequently associated with preventable secondary health-related problems such as depression, fatigue, difficulties with sleeping, pain, and anxiety (Seekins et al., 1999). These secondary conditions have been associated with primary disabling conditions such as multiple sclerosis, neuromuscular disorders, polio, joint and connective tissue diseases, and spinal cord injuries. Depression, which is clinically and theoretically linked with self-esteem (Mruk, 1995), is a prominent secondary condition among women with physical disabilities (Coyle, Santiago, Shank, Ma, & Boyd, 2000; Hughes, Swedlund, Petersen, & Nosek, 2001). Coyle et al. (2000) reported an average of 12 secondary conditions in the previous year among a sample of women with physical disabilities. Moreover, with increased severity of disability and numbers of functional impairments, women with disabilities tend to report lower levels of physical, mental, and social health status (National Center for Health Statistics, 2002). One of the most notable outcomes among women with physical disabilities is that of low economic status which often translates into lack of medical insurance and/or access to medical care and health services. A woman's health and well-being may be unnecessarily compromised by lack of access to services, inaccessible medical equipment, inadequate public transportation, and lack of disability-related training among health care and other service providers (Nosek, 2000). Women with disabilities share the work-related problems of women in general, including low wages and occupational segregation (Schaller & DeLaGarza, 1995). They may, however, also experience restricted career aspirations as a result of the nature of their disabilities, gender plus disability socialization experiences, and a lack of role models or mentors (Patterson, DeLaGarza, & Schaller, 1998). Women with physical disabilities, like women in general, may be more likely than men to experience stress related to social isolation, poverty, violence and other forms of victimization, and chronic health problems (McGrath, Keita, Strickland, & Russo, 1990). According to a recent analysis of the National Health Interview Study (National Center for Health Statistics, 2002), younger women with three or more functional limitations may be substantially less likely to be employed than women in general (14% versus 63%). Participation in the labor market is 33% for women with disabilities compared with 69% for men with disabilities (Danek, 1992; US Bureau of the Census, 1989). This gender disparity is further compounded by women's lower disability benefits from public programs, a factor related to women becoming disabled at a younger age, having fewer years in the workforce, and being compensated at lower levels (Kutza, 1985). The present study attempts to document how self-esteem and other aspects of the self affect these outcomes in women with physical disabilities. Is disability associated with lower self-esteem among women? In the current study, self-esteem is equated with an individual's sense of worthiness, adequacy, and self-respect (Rosenberg, 1979). While there have been a few studies addressing self-esteem among women with specific disabilities, a review of the literature failed to identify an investigation of self-esteem among a sample of women with various physical disabilities. However, the literature addressing self-esteem and women and men with disabilities strongly suggests that it is not disability per se, but rather the contextual, social, physical, and emotional dimensions of the impact of disability that may influence self-esteem and other aspects of the self (Barnwell & Kavanagh, 1997; Brooks & Matson, 1982; Craig, Hancock, & Chang, 1994; Walsh & Walsh, 1989). People develop their identities, in part, based on their interpretations of how others evaluate them, similar to the phenomenon that Cooley (1902) called the “looking glass self”. In other words, we look into the eyes of the other to get to know ourselves and evaluate our self-worth. This aspect of the self that is based on external feedback and approval or affection from significant others has been linked to self-esteem (Adler, 1979; Bednar & Peterson, 1995; Mead, 1934). Our earlier qualitative study of women with physical disabilities suggested that negative messages such as being a burden to the family or positive expectations regarding a woman's potential profoundly influenced the women's self-esteem. Women with disabilities must continually cope with assaults on their self-esteem generated by negative societal attitudes that they are “ill, ignorant, without emotion, asexual, pitiful, and incapable of employment” (Perduta-Fulginiti, 1996, p. 298). As a person with a disability, a woman's self-worth may be compromised by internalizing the negative personal and social devaluation that society tends to equate with physical impairment, a devaluing phenomenon that Goffman (1963) termed “stigma”. In the context of disability and chronic illness, diminished self-esteem has been associated with increased pain and fatigue (Cornwell & Schmitt, 1990; Krol et al., 1994) and greater functional limitation (Burckhardt, 1985; Taal, Rasker, & Wiegman, 1997). Losing the ability to perform activities of daily living can threaten one's sense of self. Some studies, however, fail to show consistent associations of self-esteem with either severity (Brooks & Matson, 1982) or duration of disability (Barnwell & Kavanagh, 1997; Fuhrer, Rintala, Hart, Clearman, & Young, 1992; Magill-Evans & Restall, 1991). Intimate relationships and other sources of social connection and support can offer an important validation of one's worth among persons with disabilities (Crisp, 1996). Conversely, social isolation is widely associated with health problems and mortality (Berkman & Syme, 1979). Relationships often furnish positive social support which itself serves an important function in the lives of persons with physical disabilities (Patrick, Morgan, & Charlton, 1986), yet social isolation is one of the most common secondary conditions associated with any primary disability (Coyle et al., 2000; Ravesloot, Seekins, & Walsh, 1997). Physical restrictions, such as chronic pain and fatigue, may discourage people from being socially integrated. With environmental barriers and a lack of positive messages and opportunities, a woman may become disconnected and isolated from sources of support systems and intimacy, employment opportunities, and health promotion. Finally, self-esteem and health behaviors have been positively linked in the general population (Hurst, Boswell, Boogaard, & Watson, 1997). Self-esteem and perceived health status have been associated among women with disabilities (Cornwell & Schmitt, 1990). Nosek (1998) underscores that, in the context of disability, health status and health promoting practices are critical factors in the person's ability to live independently. In this study, our focus of investigation was the mediating role of self-esteem in the relation of disability and contextual factors on pro-self-esteem behaviors, that is, specifically, intimate relationships, employment, and health-promoting behaviors among a sample of women with physical disabilities. The purpose of this study was to examine the mechanisms of self-esteem in women with physical disabilities. First, it examined the research question: What are the differences in self-esteem for women with physical disabilities compared to women without disabilities? Secondly, it tested the following hypothesis: Self-esteem mediates the relation of disability and developmental experiences with outcomes in the areas of employment, intimate relationships, and health-promoting behaviors.

Results Description of sample The most common primary disability type was spinal cord injury (26%), followed by polio (18%), neuromuscular disorders (12%), cerebral palsy (10%), multiple sclerosis (10%), joint and connective tissue disorders (8%), and other (16%). Nearly half of the sample (49%) had disabilities since childhood (0–11 years old), 10% since adolescence (12–17 years old), and 41% since adulthood (18 years and over). Twenty-two percent had severe functional limitations, 52% had moderate disabilities, and 26% had mild disabilities. Eighty-two percent were white, 9% African–American, 4% Hispanic, 2% Native American, 2% Asian and 1% other. Those living in urban or suburban areas comprised 89% of the sample, with 11% living in small towns or rural areas. The sample was well educated, with 53% of the women with disabilities and 42% of the women without disabilities having college degrees. Fifty-nine percent of the women with disabilities were working for a salary part time or full time, compared to 86% of women without disabilities. The median annual personal income of the sample of women with disabilities was $15,000, with a median household income of $25,000, compared to personal income for women without disabilities of $18,500 and household income of $32,000. The sample was more representative of persons who receive services from independent living centers than the general population of women with disabilities. In comparison to population-based statistics (Jans & Stoddard, 1999), women with joint and connective tissue disorders, women with mild impairments, minorities, and women who were living in poverty were under-represented in this sample. Chi-square and t-test analyses were conducted to determine significant differences between women with disabilities and women without disabilities on all variables in the predictive model. In some cases there were significant differences in variances between groups. Unequal variance t-tests were used for these cases. Women with disabilities compared to those without were found to have significantly more education, t(865)=2.31; p=0.02, more overprotection during childhood (reverse scored), Ms=4.36 vs. 4.79, t(848)=4.75; p<0.001, poorer quality of intimate relationships, Ms=3.72 vs. 4.37, t(853)=−5.24; p<0.001, and lower rates of salaried employment, χ2(1,n=881)=77.08; p<0.001. No significant differences were found on childhood school experiences, Ms=4.57 vs. 4.53, affection in the childhood home, Ms=3.70 vs. 3.78, or health promoting practices. Ms=6.14 vs. 6.29. On the measures of self-esteem, women with disabilities had significantly more detrimental self-cognition related to how others see them, Ms=3.81 vs. 2.34, t(828)=14.87; p<0.001, greater social isolation, Ms=2.89 vs. 2.69, t(857)=1.96, p=0.05, and lower scores on the Rosenberg self-esteem scale, Ms=5.36 vs. 5.53, t(855)=−2.06; p=0.04. Mediational model Path analysis was used to investigate the relations among the precursors, hypothesized mediators, and the outcome variables. Path analysis can be formulated as a special case of structural equation modeling where the indicators are used in lieu of latent variables (Bollen, 1989). In this case, the advantage of this approach is that instead of four different analyses to examine mediator effects as suggested by Baron and Kenny (1986), only one analysis is needed. The model used the set of precursors as exogenous variables and the mediators and outcomes as endogenous variables. We allowed the unaccounted for variance for the mediators and for the outcomes to be correlated rather than specifying causal relations among them as this was not necessary for our purpose. The precursors were also allowed to be correlated. The model included the hypothesized paths from the precursors to the mediators and from the mediators to the outcomes. In addition, paths from the precursors to the outcomes were included to allow for direct effects and a test of the mediation hypothesis. The model included age, education, degree of disability, school environment, overprotectiveness of parents (reverse scored), and display of affection in the home as precursors; self-cognition (how others see you), social isolation, and the Rosenberg as mediators; and employment, intimacy, and health promoting behaviors as the outcomes (see Table 1). Because all paths from the precursors to the mediators and outcomes were estimated as well as the correlation among the unaccounted for variance for the mediators, the model was just identified. Thus, no test of goodness of fit for the full model was possible. However, because we were not as interested in model fit as we were in testing our particular hypotheses about mediation, this was not considered a problem. Table 1. Variables in the predictive model Precursors Mediators Outcomes Demographic characteristics Age Ethnicity Education Disability severity Developmental experiences School environmenta Affectionb Self-esteem Self-cognition Social isolation Intimacy Employment Health-promoting behaviors a Socialization experiences growing up in school. b Quality of relationships in family of origin (affection in the home, overprotectiveness). Table options Results for the model indicated that each of the hypothesized mediators was significantly related to the outcome of intimacy, that both social isolation and self-esteem were significantly related to health-promoting behaviors, and that only self-esteem was significantly related to employment. Fig. 1 shows the significant pathways (z>1.96). Each path is labeled with the standardized coefficient and raw score in parentheses. Since self-cognition was scored so that a higher score meant that the respondents felt people saw them as less capable or desirable, this variable and social isolation were inversely related to intimacy while the Rosenberg was positively related. The Rosenberg was also positively related to employment, indicating that those with higher self-esteem tended to be employed more than those with lower esteem. Both social isolation and self-esteem were positively related to health promoting behaviors, indicating that while those with higher Rosenberg scores reported better health promoting behaviors, so did those who were more isolated socially. Full-size image (11 K) Fig. 1. Final Path-analytic model: Influence of precursor and mediator variables on the outcome variables. Raw scores are provided on paths with standardized coefficients in parentheses; non-significant paths are omitted. Figure options The coefficients relating the precursors to the mediators indicated that all but school environment was related to self-cognition. Respondents who were older, less disabled, less overprotected, and had more affection shown in the home tended to feel that others saw them more positively. Those who had more education, however, tended to feel that others saw them less positively. Positive school environment, lack of overprotectiveness, and more affection in the home were inversely related to social isolation, but age, education, and disability severity were not significantly related. All were related to self-esteem except education. Examination of Fig. 1 indicates that older respondents with less disability, a more positive school environment, less overprotection, and more affection in the home tended to have better self-esteem as measured by the Rosenberg. With respect to the outcomes, of all the precursors, only age was related to intimacy when the mediators were in the model. Older respondents tended to report less intimacy in their relationships. Age, education, and degree of disability were significantly related to being employed, indicating that younger, more educated, and less disabled respondents were more likely to be employed. Of all the precursors, only education was related to health-promoting behaviors when the mediators were in the model. More educated respondents reported engaging in more of these behaviors. The model predicted 32% of the variance in self-cognition, 19% of social isolation, and 16% of Rosenberg self-esteem. With respect to the outcomes, the model predicted 27% of the variance in intimacy, 16% of employment and 9% of health-promoting behaviors. As shown in Fig. 1, several of the precursors were related to each other. Age was positively related to more severe degree of disability, and higher education was associated with parents not being overprotective and with showing more affection in the home. Degree of disability was related to overprotection in that more severely disabled respondents reported being more over-protected. School environment was related to overprotection and affection in the home in that respondents who reported having a more positive school environment also reported parents being less overprotective and more affectionate. Finally, showing affection was associated with less overprotection. All of the hypothesized mediators were related to each other in the predicted direction. None of the outcomes, however, seemed to be significantly related with each other. A second model was estimated, zeroing these three relations. The model showed no significant lack of fit: χ2(3)=5.79; p>0.05; Root mean square of approximation=0.036; goodness of fit=1.00. Thus, the variance in the outcomes not accounted for in the model does not appear to be concomitant. Calculation of indirect effects of precursors on outcomes through the mediating variables was performed by multiplying the two path coefficients (from a precursor to the mediator and from the mediator to the outcome). All precursors showed significant indirect effects, above direct effects, on intimacy largely through self-cognition and/or self-esteem (see Table 2). For example, overprotection led to more detrimental self-cognitions which in turn led to lower intimacy. All precursors, except disability, had indirect effects on employment through the self-esteem mediator, while only age and school environment influenced health behaviors via self-esteem. Table 2. Indirect effects of the precursors on intimacy expressed as standardized coefficients Precursors Mediating variables Total Self-cognition Self-esteem Social isolation Age 0.028 0.16 0.002 0.047** Education −0.031 −0.003 −0.000 −0.035* Disability −0.154 −0.014 0.001 −0.167*** School Environment 0.003 0.034 0.017 0.055** Overprotection 0.093 0.028 0.027 0.149*** Affection 0.023 0.028 0.018 0.070*** *p<0.05. **p<0.01. ***p<0.001.