Research Article: Self-Esteem Among Amateur Bodybuilders

Self-esteem among amateur bodybuilders: A quantitative examination
Assistant Professor Jeanne Mekolichick
Department of Sociology and Anthropology,
Radford University,
Radford, VA 24141,
U.S.A.;
e-mail: jmekolic@radford.edu.

Abstract
This study quantitatively examined the self-esteem of amateur bodybuilders. Using info from the membership of the National Amateur Bodybuilding Association, USA (NABBA, USA) obtained via mail survey, two hypotheses are advanced concerning the self-esteem of amateur bodybuilders: (1) Amateur bodybuilders will possess a mean level of global self-esteem that is statistically significantly below the arithmetic mean for the Rosenberg Self-Esteem (RSE) scale even when differences for race, class, and gender are taken into account; and, (2) Amateur bodybuilders will possess a mean level of global self-esteem that is statistically significantly below the mean of other social groups. Social groups for comparison were chosen via a Boolean search of Sociological Abstracts 1963-present using the search terms 'Rosenberg' and 'self-esteem'; articles chosen measured the RSE on a Likert scale. Employing t-tests and a bivariate correlation, the data do not indicate support for either hypothesis. On the contrary, the self-esteem of the bodybuilders in the study was statistically significantly above the arithmetic mean of almost all comparison groups. Speculations are made regarding these results, suggesting differences between the elite amateur and professional levels and 'lower' amateur levels. Recommendations for future research are proposed.

Introduction
The concept of self-esteem has not been a focus of general concern in the bodybuilding literature, however, statements from in-depth interviews and participant observations conducted in previous sociological and anthropological research suggest that elite amateur and professional bodybuilders have low self-esteem (Fisher, 1997; Fussell, 1991; Heywood, 1998; Klein, 1993; Roussel & Griffet, 2000). This research, however, has been exclusively qualitative in nature and therefore limited in number and sample size, while consisting primarily of professional bodybuilders. Researchers have yet to examine the self-esteem of bodybuilders using well-established, quantitative measures that have been proven reliable and valid, and which can be compared with other social groups. As such, one area of the bodybuilding literature that begs further analysis in this way is the issue of self-esteem.
This paper begins to address the apparent void in the literature by quantitatively examining self-esteem among a group of amateur bodybuilders. The value of this project is therefore primarily two-fold. Firstly, by using a reliable and valid quantitative self-esteem scale, a standard measure of bodybuilders' self-esteem can be obtained, thus adding generally to the body of knowledge in this regard. Secondly, by conducting a study of self-esteem among other social groups, this project also adds to the research base of self-theory.
Given this rationale, the related goals of this research are threefold: (1) to review the literature on bodybuilders' self-esteem, (2) to empirically examine self-esteem among a national sample of amateur bodybuilders using a well-established self-esteem scale, and (3) to compare the self-esteem of amateur bodybuilders in the sample to other social groups where self-esteem was similarly measured. From these goals, two general research questions were advanced, namely, using a reliable and valid quantitative measure of global self-esteem, how do the results compare with other qualitative findings? And, how does the global self-esteem of amateur bodybuilders compare with other social groups who had their global self-esteem similarly measured? The two specific hypotheses that emerged out of these questions were:
1. that the amateur bodybuilders in this sample would possess a mean level of self-esteem that was statistically significantly below the mean of other social groups. The specific social groups for comparison were chosen via a Boolean search of Sociological Abstracts 1963-present and comprised of those defined as smokers, the elderly, military individuals, women formerly on welfare, and college students.
2. that the amateur bodybuilders in this sample would possess a mean level of self-esteem that was statistically significantly below the arithmetic mean for the RSE scale even when differences for race, class, and gender were taken into account. These social structural variables were introduced because there is some research to suggest that they may be influential in the assessment of global self-esteem.
Global self-esteem refers to one's overall sense of value or worth, and is usually referred to simply as 'self-esteem' (Rosenberg, 1979). As such, in this paper, 'global self-esteem' and 'self-esteem' are used interchangeably. While several quantitative instruments have been constructed to measure self-esteem, the Rosenberg Self-Esteem scale (RSE) is considered a valid well-constructed measure of global self-esteem, and has been used to measure the self-esteem of many social groups (see Wylie, 1979; Blascovich & Tomaka, 1991). Given its stable psychometric properties and its frequency of use, the RSE is used here to quantitatively assess bodybuilders' self-esteem.

Literature Review
Self-esteem
Self-esteem refers to one's sense of worth or importance; it is an attitude-positive or negative-held about the self, and includes both cognitive and affective elements (Rosenberg, 1979; Gecas, 1982). Self-esteem continues to be the most studied aspect of the self (Wylie, 1979; Gecas & Burke, 1995). Research on self-esteem demonstrates, among other things, that it is associated with occupational success and the domain of physical appearance (Baumeister, 1993; Luck & Heiss, 1972). More specifically, high self-esteem has been associated with body shape satisfaction, body image satisfaction, and perceived appearance (Abell & Richards, 1996; Baumeister, 1993; Gray, 1993; Monteath & McCabe, 1997; Thomas, 1989; Usmiani & Daniluk, 1997). Low self-esteem, on the other hand, has been associated with higher levels of eating disorders, including anorexia and bulimia, a distorted body image, and a sense of inadequacy in domains where success is valued, such as one's occupation (Baumeister 1993; Boskind-Loodahl 1976; Joiner & Kashubeck 1996; Luck & Heiss, 1972).
Social variables such as class, race, and gender have been implicated in global self-esteem variations among individuals (Bush, Simmons, Hutchinson, & Blyth, 1978; Demo & Savin-Williams, 1983; Offer & Howard, 1972; Rosenberg, 1979; Rosenberg & Pearlin, 1978; Rosenberg & Simmons, 1972; Simmons & Rosenberg, 1975; Simmons, Blyth, Van Cleave, & Bush, 1979; Wiltfang & Scarbez, 1990), although not consistently (Gecas, 1982; Maccoby & Jacklin, 1974; Rosenberg & Pearlin, 1978; Wylie, 1979). With regard to social class, higher social class is reported to positively influence self-esteem, although these findings are inconsistent and the relationships are weak (Demo & Savin-Williams, 1983; Rosenberg, 1979; Rosenberg & Pearlin, 1978; Wiltfang & Scarbez, 1990). Similarly, with regard to race, either no differences have been reported, or the self-esteem of African-Americans has been shown to be slightly higher than that of Whites (Hoelter, 1983; Rosenberg & Simmons, 1972; Rosenberg & Rosenberg, 1989; Taylor &Walsh, 1979; Yancey, Rigsby, & McCarthy, 1972). The findings regarding gender differences have also been inconsistent. Some studies have reported males having higher self-esteem than females, while other studies have found no differences (Hoelter, 1983; Maccoby & Jacklin, 1974; Wylie, 1979).
Bodybuilders' self-esteem
Research conducted on bodybuilders to date has suggested that their self-esteem is comparatively low (Fisher, 1997; Fussell, 1991; Heywood, 1998; Klein, 1986; 1993; Mekolichick, 1999; Roussel & Griffet, 2000). Klein (1993), for example, adopts a psychoanalytic interpretation of the professional bodybuilding subculture. He focuses on why individuals became involved in bodybuilding and then deduces from these explanations their sense of global self-esteem. He argues that "recruitment to the subculture...stems from real, felt unmet needs and personality deficiencies. In short, poor self-image lurks in the background of most bodybuilders" (Klein, 1986, p.129). Fisher (1997) similarly argues that the self-esteem of professional bodybuilders, particularly female bodybuilders, is low. She writes:
Although the female bodybuilders interviewed found motivation and achievement through manipulation of their bodies, that achievement was sought as a result of an impoverished sense of self, a feeling of personal inadequacy...Co-participants described how their sense of self-esteem was not very good at all, either currently or prior to beginning competitive bodybuilding (p.151).
Paralleling the findings of Klein and Fisher, Fussell (1991), in his autobiographical account of his life as a professional bodybuilder, speaks of his own feelings of inadequacy and recounts similar stories of his lifting partners. Further, Heywood (1998) suggests that "competitive bodybuilding...with its emphasis on isolated individual achievement and winning at all costs may...contribute to an individual's sense of valuelessness" (p. 205). Qualitative data gathered by Mekolichick (1999) using amateur bodybuilders is also supportive of this position. In this respect, one of her bodybuilding subjects recounted: "I began to body build for one reason alone, insecurity. I continue to body build for the same reason" (Personal correspondence, September 11, 1998). Finally, Roussel and Griffet (2000), in a study of experienced female bodybuilders in France claim that "the analysis of our interviews showed that before becoming bodybuilders, the majority of the women we talked to had low self-esteem" (p.146). In short, therefore, from the available qualitative data assessing the self-esteem of largely professional and elite amateur bodybuilders, the findings suggest low levels of global self-esteem.

Methods
Subject selection
Five national amateur bodybuilding organizations in the United States were contacted regarding their participation in the study. Only one organization, the National Amateur Bodybuilding Association, USA (NABBA, USA) agreed to participate. Thus, the sample used in this project is a non-probability sample. The use of a purposive sample is appropriate for these purposes because the investigation was focused on a specialized population, i.e. amateur bodybuilders (Kalton, 1983; Kish, 1965).
Measures
Self-Esteem. The Rosenberg Self-Esteem scale (1979) was used to measure global self-esteem. The scale consists of 10 items answered on a five-point Likert scale from 'strongly agree' to 'strongly disagree'. Negative items were reverse coded so that a high score continued to indicate high self-esteem. The possible range for this scale is 10 to 50. Consistent with previous research, the internal reliability for this scale within the current study was high-alpha = .82 (see Wylie, 1979 and Blascovich & Tomaka, 1991). This measure was chosen because it has been used-and continues to be used-among various populations and is widely accepted as having good psychometric properties (Baker & Gallant, 1984; Blascovich & Tomaka, 1991).
Social Structural Variables. Gender was measured by having respondents select 'male' or 'female'. Race was measured by having respondents select from the following categories: African-American, Asian, Caucasian, Hispanic, Native American, and Other. Social class was measured by combining respondents' level of education and total household income. Respondents indicated their highest level of education by responding to the following categories: 'less than high school', 'high school', 'some college', 'technical school', 'B.A./B.S.', 'some graduate work', 'M.A./M.S.', 'Ph.D.' or 'professional degree'. Total household income was measured in $10,000 increments. The participants responded to the categories: under $10,000, $10,001-$20,000, $20,001-$30,000 and so on. The scores for both of these items were then combined to create a single score where a higher score indicates a higher social class. The possible range for this scale is 2 to 18. This variable is used as a continuous variable, not differentiating specifically between social classes.
Procedures
A complete and current list of the entire national membership of NABBA, USA was obtained from the president of the organization in late June of 1998; this list contained 615 members. A mail survey was conducted using Dillman (1978) as a guide. Following Dillman, a pre-mailing postcard was initially sent, a cover letter from the president of the organization was included with the mailed instrument, with a follow-up mailing being implemented in order to obtain a better response rate.
Specifically, in early September of 1998, 615 postcards were sent to the members of NABBA, USA informing them of the project and that they would be receiving a survey within a week or so about their participation in bodybuilding. A week after sending out the postcards, the instrument was sent to all 615 members of NABBA, USA including a cover letter from the president of NABBA, USA as well as a cover letter from the researcher introducing the project again, informing them of their rights as possible respondents, and requesting their participation in the project. Four weeks after sending out the first wave of the instrument, 120 responses were received-about 20% of the sample. As such, a second wave of the instrument was sent to the NABBA, USA membership roughly five weeks after the first wave. Instruments were received from respondents for about four weeks after the second mailing, then the responses dropped off. Given financial constraints, a third wave of instruments was not distributed. Due to a number of 'dead' addresses, the practical sampling universe was lowered from 615 to 553. A total of 193 instruments were returned, 190 of which were usable, yielding a 35% response rate. Thus, 190 subjects took part in the study.
To obtain the comparison groups, a Boolean search using the terms 'Rosenberg' and 'self-esteem' was conducted of the database Sociological Abstracts 1963-present. This search produced 67 entries. The search was further limited by excluding association papers and dissertation abstracts containing insufficient data for the purposes here. These limitations yielded a total of 56 articles. The abstracts were then screened, with entries using subjects under the age of 18 being excluded. Following these stipulations, a total of 22 entries remained. The articles were then further screened for the application of the Rosenberg Self-Esteem scale (RSE). Hence, if a Guttman scale was used, or if no group means were reported, the article was deleted. This final screening process yielded a total of 7 remaining articles. The resulting groups from these articles were listed as 'smokers', 'the elderly', 'military individuals', 'women formerly on welfare', and 'college students'.
Limitations
In considering the conclusions drawn from the study's findings as discussed in the upcoming section, one must reflect on the limitations of this study. There are at least two areas of concern: firstly, the use of non-random sampling and the comparison groups used in the analyses. The most serious limitation of non-probability sampling is the fact that its use limits the application of statistical theory to examining the properties of the sample estimators (Kalton, 1983; Kish, 1965). While this is a limitation, the goal of this preliminary study was to provide more insight into the under-researched field of the global self-esteem of amateur bodybuilders. Considering that self-esteem has never been examined using quantitative measures for this population, the principal goal was exploration not prediction or generalisation consequently, this limitation was not considered destructive of the results obtained.
A second limitation of the study relates to the comparison groups used in the analysis. As previously noted, only those studies where the Rosenberg self-esteem scale was employed and measured with a Likert scale among adult populations, published in English, and registered in Sociological Abstracts 1963-present were used in determining the comparison groups. Hence, one should be aware that other valid and reliable measures of global self-esteem are available as are findings pertaining to other social groups for comparison.

Results
Sample Characteristics
The average age of the respondents was 33.9 years, ranging from 17 to 76 years old, with the standard deviation of the sample being 9.55 years. 72% of the sample was male, and 28% female. The racial make-up of the sample was largely Caucasian (82%), with 11% being African-American, 3% being Native American, 1% being Hispanic, 1% being Asian while 2% declared themselves as 'Other'.
In terms of their education, 16% of the sample had at least a high school education, 29% had some college, 13% went/had been to technical school, 19% had earned a bachelor's degree, 8% had done some graduate work, 12% had earned a master's degree, and 4% had earned a doctorate or its equivalent. The average yearly total household income of the sample was between $40,000 and $60,000, with a mode of between $40,000 and $50,000, and a median between $40,000 and $50,000.
With reference to their bodybuilding activities, the amount of times the respondents had participated in a competition ranged from zero (5%) to sixteen or more times (18%). The modal category of the number of times competed was one to three (33%). Respondents competed at various levels, ranging from local (13%), to state (35%), to national (26%), to international (11%).
Table 1
Two-tailed t-test for difference of bodybuilder group mean and RSE mean
RSE Bodybuilder’s M Arithmetic M
43.48 29.5

t value 31.96
df 186
t significance level .00
n=187
Table 1 reports the self-esteem scores for the bodybuilders as a group. The mean level of self-esteem for this group was 43.48, with a median of 45 and a mode of 50. The standard deviation for this sample was 5.98. The arithmetic mean for the Rosenberg self-esteem scale measured on a five point Likert scale was 29.5. A one-sample two-tailed t-test indicated that the mean level of self-esteem among NABBA, USA members was statistically significantly above the arithmetic mean (p < .00). When taking into account possible differences based on the social structural variable of gender, a two-tailed t-test indicated that there were no statistically significant differences in the self-esteem of males and females in this study, as reported in Table 2.
Table 2
Two-tailed t-test for difference of RSE means by gender
RSE Women Men
M 42.96 43.70

t value .75
df 183
t significance level .46

Women n = 49
Men n =136
Examining the effect of race on self-esteem within the study's sample, a two-tailed t-test indicated that the self-esteem of Non-Whites was not statistically significantly higher than Caucasians' self-esteem (p>.05) as can be seen in Table 3. Table 4 examines the effects of social class. In examining the effect of social class on the level of self-esteem within the study's sample of respondents, a bivariate correlation indicated a non-significant correlation between social class and self-esteem (p>.05). In short, the data indicated that the mean level of self-esteem among this group of bodybuilders was statistically significantly above the arithmetic mean. When gender, race, and social class were examined as mitigating variables, the mean of the bodybuilders' self-esteem remained above the arithmetic mean. As such, hypothesis one was not supported by the data.
Table 3
Two-tailed t-test for difference of RSE means by race
RSE Caucasians Non-Whites

M 43.15 45.30

t value 1.90
df 182
t significance level .06

Caucasians n =151
Non-Whites n =33
Table 4
Zero order correlation between social class and self-esteem
Social Class x Self-Esteem .13 p =.08
n = 172
The data in Table 5 tests the second hypothesis. As can be seen in Table 5, for all other comparison groups where self-esteem was measured using the RSE with a Likert-type scale, the bodybuilders' self-esteem was found to be statistically significantly above the mean of all the comparison groups (p< .05) with the exception of military women. Consequently, hypothesis two was not supported by the data.
Table 5
Two-tailed t-tests for difference in comparison group means and bodybuilder group mean on self-esteem
Author Comparison Group M RSE t-value t-sig
Defleur & Warner 1985a Military men (650) 42.3 2.70 .01
Military women (330) 44.41 -2.12 .04
Civilian men (1310) 41.19 5.24 .00
Civilian women (930) 39.81 8.39 .00
DeMan and Beniot 1982a Women Feminists (31) 42.38 2.53 .01
Men (31) 42.5 2.24 .03
Women Traditionalists (31) 38.7 10.93 .00
(French-Canadians)
Elliott 1996a Women 1980 (3,076) 40.51 6.79 .00
Women 1987 (3,076) 42.08 3.20 .00
(Economically disadvantaged white women)
Kowalski 1997a Adult smokers (50) 37.75 13.10 .00
Adult non-smokers (25) 42.38 2.53 .01
Lee and Shehan 1989a Men 55+ yrs (1,395) 38.46 11.48 .00
Women 55+ yrs (1,609) 38.40 11.62 .00
(Residents of Washington State)
Rowley, et. al. 1998b African-American (176) 40.9 5.90 .00
(College Students)
Singelis, et. al. 1999b U.S. Euro-Americans (232) 40.8 6.13 .00
Hawaii Asian Am. (146) 39.4 9.33 .00
Hong Kong Chinese (271) 36.9 15.05 .00
(College Students)
Note: Numbers in parentheses indicate number of cases. a In these articles RSE was measured using a 4 point Likert scale. The numbers reported here were converted to a 5 point Likert scale to make a direct comparison. bThe authors reported the RSE group means divided by the number of items in the scale (10) for ease of interpretation in their studies. For ease of interpretation here, and to make direct comparisons with other groups, the RSE group means for these studies were multiplied by the number of items in the scale (10).
Both the hypotheses received no support from the data. Indeed, the findings were positively contrary to the predictions made in the hypotheses. Thus, the bodybuilders' global self-esteem was statistically significantly above the mean for the scale used, even when considering race, class, and gender. Furthermore, the bodybuilders' self-esteem was statistically significantly above the mean of all but one of the comparison groups.

Discussion
This project was designed to examine some questions and test related hypotheses regarding the self-esteem of amateur bodybuilders. Given the scant research conducted on bodybuilders concerning self-esteem, this largely explorative project examined this construct using a well-known quantitative measure of global self-esteem (RSE) using a larger population than had been previously undertaken. Two main avenues of investigation were pursued, firstly, to add to the understanding of the self-esteem of amateur bodybuilders, and secondarily, to compare the self-esteem of this group of bodybuilders with other social groups. Below, the findings are summarized, speculations regarding the findings are discussed, and some directions for future research proposed.
The data that resulted from testing the two hypotheses among 190 NABBA, USA members provided no support for the hypotheses. The findings of the study thus concluded that in comparison to the arithmetic mean of the RSE and in comparison with other social groups similarly measured, the bodybuilders in this sample did not exhibit low levels of self-esteem. On the contrary, the amateur bodybuilders in this study exhibited a moderate to high level of self-esteem. This finding contradicts the existing literature on this topic among bodybuilders.
At least two avenues of explanation can be offered regarding these findings. One explanation as to why the findings of this study differed so drastically from previous work is that the respondents in this study were amateur bodybuilders, whereas the bodybuilders studied in previous work were almost exclusively professionals or elite amateur competitors. A second possible explanation is selection bias. These lines of discussion are addressed below.
In relation to the former tentative explanation offered, recall that most of the respondents in this sample (33%) had competed only between one and three times. Further, only 26% of the sample had competed at the national level. In addition to recognizing their scant participation in national competitions, it is useful to understand the structure of professional and amateur bodybuilding in the United States and worldwide in this respect. Professional bodybuilders who want to make money and obtain profitable endorsements on the world market join the International Federation of Bodybuilding (IFBB) (Lowe, 1998). The IFBB is a closed organization, which designates one amateur bodybuilding organization per country as its amateur affiliate (Lowe, 1998). In the United States, that designate is the National Physique Committee (NPC) (Lowe, 1998). To be eligible to become a professional, one must either be "the overall winner of the NPC USA championships…[or one of]…the three weight class winners of the national championship" (Lowe, 1998, p.62). Consequently, the participants in this study-who belonged to the NABBA, USA, and not the NPC-did not appear to be moving in the direction of professional status. Given that the majority of respondents in this study had competed so few times, at relatively low levels of competition, and did not belong to the amateur bodybuilding organization that is the most direct route to professional status, the argument can be made that the respondents constituted, in large part, amateurs who were not at the elite level of their sport and/or did not have an interest in participating as professionals. As such, it may be that the level of involvement in the sport influences self-esteem. There are two dimensions to this argument: body image and financial dependence.
The activity of bodybuilding itself invites participants to be highly critical of their physical appearance. The level of physical excellence required for competition is extremely difficult to achieve, and impossible to maintain over time (Bolin, 1992). As such, much of the time the bodybuilder does not meet his or her physical ideal-if it is ever achieved. Thus, it may be that professional and elite amateur bodybuilders are generally not satisfied with their body image. Prior research has focused on the discrepancy between the 'real body image' and the 'ideal body image' as a source of lowered self-esteem; that is, the greater the discrepancy between real and ideal body image the lower the self-esteem (Johnson & Petrie, 1995). Given that the 'ideal' is only temporarily met for bodybuilders, this process of comparison may lead to lowered levels of self-esteem among professional and elite amateur bodybuilders whose sense of self worth is more likely to depend upon their physical appearance, but may not have a similar effect on amateur bodybuilders whose sense of self worth may not be similarly dependent.
In a related area of research, dissatisfaction with body image is documented as being positively correlated with low self-esteem (Abell & Richards, 1996; Gray, 1993; Monteach & McCabe, 1997; Thomas, 1989; Usmiani & Daniluk, 1997). These findings may also inform the results presented here. It may be that the difference in participation level (from elite amateur or professional to low-level amateur) may influence expectations of one's body and therefore impact global self-esteem. In a piece of related research, Guthrie, Ferguson, and Grimmett (1994) studied 13 female bodybuilders at various levels, assessing their eating habits and body image. They found "very positive body images among the sample of bodybuilders" (Guthrie, et. al., 1994, p.275). The authors do not report what percentage of the sample were amateurs, yet knowing that some amateurs were included in the study lends support to the argument that satisfaction with body image among bodybuilders is somewhat dependent upon level of participation in the sport. Additionally, Klein (1986; 1993) reports that the professional bodybuilders in his sample suffered from a poor body image, which further adds credence to the current argument.
Furthermore, the mind-set imposed with regard to physical appearance by elite amateur and professional bodybuilding has been compared with anorexia (Bordo, 1993; Klein, 1993; Pope, Katz, & Hudson, 1993). For both the anorexic and the elite amateur or professional bodybuilder, the individual visualizes what he or she wants his or her body to look like, then works to create that form (Bordo, 1993). Anorexics, like elite amateur and professional bodybuilders, have been shown to exhibit low levels of self-esteem (Boskind-Loodahl, 1976; Button, Sonuga-Baqrke, & Thompson, 1996; Joiner & Kashubeck, 1996). Thus, once again, the disjuncture between the real body image and the ideal body image may be a source of lowered self-esteem. This suggests that the mechanisms involved in sculpting one's body to meet an ideal are similar and that both may create low levels of self-esteem among elite amateur and professional bodybuilders aiming to achieve this ideal.
In an extension of this line of research, Pope, Katz, and colleagues have focused on a disorder they term "reverse anorexia," where "individuals...believed that they looked small and weak, even though they were large and muscular" (Pope & Katz, 1994, p.380). This disorder is built on the discrepancy between ideal and real body image (Pope, Katz & Hudson, 1993). Although their work does not examine the effects of this disorder on self-esteem, it is reasonable to assume, based on the related research addressed above, that low self-esteem could be associated with reverse anorexia.
A second dimension of this explanation, based upon level of participation, is the financial dependence predicated on the professional bodybuilder's physique that is not similarly dependent for amateurs. Klein (1993), in his study of the bodybuilding subculture, discussed the struggles of the professional bodybuilder. Included in his discussion is the battle of the bodybuilder to maintain his or her status as a professional. He described a very cut-throat market where "only a select few can manage access to whatever fame and fortune the sports moguls offer" (Klein 1993, p.51). Added to this is the financial aspect of making a living through bodybuilding, as the monetary prizes for winning bodybuilding competitions are quite low when compared with other professional sports. Thus, many professional bodybuilders rely on other jobs to pay their bills because of the unstable income of professional bodybuilding (Klein, 1993). This extremely competitive environment coupled with the financial uncertainty of professional bodybuilding may also add to the lower levels of self-esteem found in other studies.
Future research should examine the self-esteem of professional, elite amateur, and lower-level amateur bodybuilders to compare the construct at different levels of this sport. In addition, the issues of body image and financial insecurity linked to one's physique should also be taken into account as a possible mitigating factor influencing the global self-esteem of bodybuilders.
A second explanation for the contradictory findings is the issue of selection bias. Given that only 35% of the total NABBA, USA membership responded to the study, it is unknown what the global self-esteem is of the rest of the population. It is possible that only those with a positive sense of self responded to the survey; those that did not answer the survey may have opted out of participating in the study because of their low self-esteem. As such, future research should attempt to obtain a larger, possibly random sample of the bodybuilding population at multiple levels to either confirm or refute the current findings.
It is also recommended that future work should go beyond the self-esteem of bodybuilders (both professional and amateur) to examine other aspects of the bodybuilder's self-concept. One other aspect of the self, related to self-esteem, which may also be implicated in the bodybuilder's view of self, is the concept of self-efficacy. Self-efficacy refers to one's sense of self-mastery, competence, or power (Bandura, 1977; Gecas, 1982; 1989). For the bodybuilder, control over the body itself may bring a sense of power. For example, Klein (1993) reports that the women in his project felt "a strong sense of security and control achieved by adhering to the monastic, somewhat obsessive regimen of contest preparation" (p.89). He also reports that these women place importance on "power and self-mastery" (Klein, 1993, p.88). Similarly, Lowe (1998), in her study of professional female bodybuilders, concluded that bodybuilders have extensive control over their bodies. Finally, Heywood (1998) similarly found that "bodybuilding can function...as a declaration of independence" (p.156). These statements suggest that bodybuilders may experience moderate or high levels of self-efficacy as a result of their participation in the activity of bodybuilding. This aspect of the self should be an avenue for future exploration.
A second aspect of the self that should be explored among bodybuilders is role authenticity. Role authenticity has been defined as feeling real in a role (Benson & Trew, 1995; Reid, Epstein, & Benson, 1993; 1994). Research on the concept of role authenticity has not been explicitly conducted on bodybuilders; however, prior research indicates that bodybuilding is, for some, an authentic role (Guthrie, et. al., 1994). For example, one professional female bodybuilder said of bodybuilding: "it's [bodybuilding] made me freer to be who I really am..." (Guthrie, et. al., 1994, p.277). Given the importance of role authenticity to the self, it is another meaningful aspect of the self that should be explored. These and other issues should be the focus of future research on the self-concept of amateur and professional bodybuilders.

References
Abell, S.C. & Richards, M. (1996). The relationship between body shape satisfaction and self-esteem: An investigation of gender and class differences. Journal of Youth and Adolescence, 25, 691-703.
Baker, P.M., & M.J. Gallant. (1984). Self-esteem: Measurement strategies and problems. Humbolt Journal of Social Relations, 12, 36-48.
Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change. Psychological Review, 84, 191-215.
Baumeister, R.F. (1993). Self-esteem: The puzzle of low self-regard. New York: Plenum Press.
Benson, D. E. & Trew, K.J. (1995). Facets of self in Northern Ireland: Explorations and further questions. In A. Oosterwegel, & R.A. Wicklund (Eds.), The self in European and North American culture: Development and processes (pp. 291-307) Amsterdam: Kluwer Academic Publishers.
Blascovich, J., & Tomaka, J. (1991). Measures of self-esteem. In J. Robinson, P. Shaver, & L. Wrightsman (Eds.), Measures of personality and social psychological attitudes (pp. 161-194). New York: Academic Press.
Bolin, A. (1992). Flex appeal, food, and fat: Competitive bodybuilding, gender and diet. Play and Culture, 5, 378-400.
Bordo, S. (1993). Unbearable weight: Feminism, western culture, and the body. Berkley, University of California Press.
Boskind-Loodahl, M. (1976). Cinderella’s step-sisters: A feminist perspective on anorexia nervosa and bulimia. Signs, 2, 342-356.
Bush, D. E., Simmons, R.G., Hutchinson, B., & Blyth, D.A. (1978). Adolescent perception of sex-roles in 1968 and 1975. The Public Opinion Quarterly, 41, 459-474.
Button, E.J., Sonuga-Baqrke, E.J.S., Davies, J., & Thompson, M. (1996). A prospective study of self-esteem in the prediction of eating problems in adolescent schoolgirls: Questionnaire findings. The British Journal of Clinical Psychology, 35, 193-203.
De Man, A. F., & Benoit, R. (1982). Self-esteem in feminist and nonfeminist French-Canadian women and French-Canadian men. The Journal of Psychology, 111, 3-7.
Defleur, L. B., & Warner, R. L. (1985). Socioeconomic and social-psychological effects of military service on women. Journal of Political and Military Sociology, 13, 195-208.
Demo, D. H., & Savin-Williams, R.C. (1983). Early adolescent self-esteem as a function of social class: Rosenberg and Pearlin revisited. American Journal of Sociology. 88, 763-774.
Dillman, D.A. (1978). Mail and telephone surveys: The total design method. New York: John Wiley & Sons.
Elliot, M. (1996). Impact of work, family, and welfare receipt on women’s self-esteem in young adulthood. Social Psychology Quarterly, 59, 80-95.
Fisher, L. A. (1997). Building one’s self up: Bodybuilding and the construction of identity among professional female bodybuilders. In P.L. Moore (Ed.), Building bodies. (pp. 135-164). Rutgers University Press: New Brunswick, NJ.
Fussell, S. W. (1991). Confessions of an unlikely bodybuilder. Poseidon Press: New York.
Gecas, V. (1982). The self-concept. Annual Review of Sociology, 8, 1-33.
Gecas, V. (1989). The social psychology of self-efficacy. Annual Review of Sociology, 8, 1-33.
Gecas, V., & Burke, P.J. (1995). Self and identity. In K.S. Cook, G.A. Fine, & J.S. House (Eds.), Sociological perspectives on social psychology (pp.41-67). Allyn and Bacon: Needham Heights, MA.
Gray, E.A. (1993). Women’s body image: A multivariate study. Free Inquiry in Creative Sociology, 21, 103-110.
Guthrie, S. R., Ferguson, C., & Grimmett, D. (1994). Elite women bodybuilders: Ironing out nutritional misconceptions. The Sport Psychologist, 8, 271-286.
Heywood, L. (1998). Masculinity vanishing: Bodybuilding and contemporary culture. In P.L. Moore (Ed.), Building bodies (pp. 165-183). New Brunswick, NJ: Rutgers University Press.
Hoelter, J. W. (1983). Factorial invariance and self-esteem: Reassessing race and sex differences. Social Forces, 61, 834-846.
Johnson, C. E. & Petrie, T.A. (1995). The relationship of gender discrepancy to eating disorder attitudes and behaviors. Sex Roles, 33, 405-416.
Joiner, G. W., & Kashubeck, S. (1996). Acculturation, body image, self-esteem, and eating-disorder symptomalogy in adolescent Mexican American women. Psychology of Women Quarterly, 20, 419-435.
Kalton, G. (1983). Introduction to survey sampling. Beverly Hills, CA: Sage
Kish, L. (1965). Survey sampling. New York: Wiley.
Klein, A. M. (1986). Pumping irony: Crisis and contradiction in bodybuilding subculture. Sport Sociology Journal, 3, 112-133.
Klein, A. M. (1993). Little big men: Bodybuilding subculture and gender construction. Albany, N.Y.: State University of New York Press.
Kowalski, S.D. (1997). Self-esteem and self-efficacy as predictors of success in smoking cessation. Journal of Holistic Nursing, 15, 128-142.
Lee, G. R., & Shehan. C.L. (1989). Social relations and the self-esteem of older persons. Research on Aging, 11, 427-442.
Lowe, M. R. (1998). Women of steel. New York: New York University Press.
Luck, P.N., & Heiss, J. (1972). Social determinants of self-esteem in adult males. Sociology and Social Research, 57, 69-84.
Maccoby, E. E., & Jacklin, D.N. (1974). The psychology of sex differences. Stanford, California: Stanford University Press.
Mekolichick, J. (1999). Amateur bodybuilders eating and on stage: Processes of multiple identity invocation in situations of structural overlap and isolation. (Doctoral dissertation, Kent State University, 1999). Dissertation Abstracts International, 60, 2253-A
Monteach, S. A., & McCabe, M.P. (1997). The influence of societal factors on female body image. Journal of Social Psychology,137, 708-727.
Offer, D., & Howard, K.I. (1972). An empirical analysis of the Offer self-image questionnaire for adolescents. Archives of General Psychiatry, 27, 529-537.
Pope, H. G., Katz, D.L., & Hudson, J.I. (1993). Anorexia nervosa and ‘reverse anorexia’ among 108 bodybuilders. Comprehensive Psychiatry, 34, 406-409.
Pope, H.G., & Katz, D.L. (1994). Psychiatric and medical effects of anabolic-androgenic steroid use. Archives in General Psychiatry, 51, 375-382.
Reid, S., Epstein, J.S., & Benson, D.E. (1993). Living on a lighted stage: Identity salience, psychological centrality, authenticity, and role behavior of semi-professional rock musicians. In J.S. Epstein (Ed.), Adolescents and their music: If it’s too loud, you’re too old. (pp. 301-328) New York: Garland Press.
Reid, S., Epstein, J.S., & Benson, D.E. (1994). Role identity in a devalued occupation: The case of female exotic dancers. Sociological Focus, 27, 1-15.
Rosenberg, M. (1975). Sex differences in the self-concept in adolescence. Sex Roles, 1,147-159.
Rosenberg, M. (1979). Conceiving the self. New York: Basic Books.
Rosenberg, M., & Pearlin. L.I. (1978). Social class and self-esteem among children and adults. American Journal of Sociology, 84, 53-78.
Rosenberg, M., & Rosenberg, F.R. (1989). Old myths die hard: The case of Black self-esteem. Revue Internatinale de Psychologie Sociale, 2, 355-365.
Rosenberg, M., & Simmons, R.G. (1972). Black and White self-esteem: The urban school child. Arnold M. And Caroline Rose Monograph Series. Washington, D.C.: American Sociological Association.
Roussel, P., & Griffet, J. (2000). The path chosen by female bodybuilders: A tentative interpretation. Sociology of Sport Journal, 17, 130-150.
Rowley, S.J., Sellers, R.M., Chavous, T.M., & Smith, M.A. (1998). Relationship between racial identity and self-esteem in African American college and high school students. Journal of Personality and Social Psychology, 74, 715-724.
Simmons, R.G., Blyth, D.A., Van Cleave, E.F., & Bush, D.M. (1979). Entry into early adolescence: The impact of school structure, puberty, and early dating on self-esteem. American Sociological Review, 44, 948-967.
Simmons, R.G., & Rosenberg, M. (1975). Sex, sex roles, and self-image. Journal of Youth and Adolescence, 4, 229-258.
Singelis, T.M., Bond, M.H., Sharkety, W.F., & Yiu Lai, C.S. (1999). Unpackaging culture’s influence on self-esteem and embarrassability: The role of self-construals. Journal of Cross-Cultural Psychology, 30, 315-341.
Taylor, M.L., & Walsh, J. (1979). Explanations of Black self-esteem: Some empirical tests. Social Psychology Quarterly, 42, 242-253.
Thomas, V.G. (1989). Body-image satisfaction among Black women.. Journal of Social Psychology, 129, 107- 112.
Usmiani, S., & Daniluk, J. (1997). Mothers and their adolescent daughters: Relationship between self-esteem, gender role identity, and body image. Journal of Youth and Adolescence, 26, 45-62.
Wiltfang, G. L., & Scarbecz, M. (1990). Social class and adolescents’ self-esteem: Another look. Social Psychology Quarterly, 53, 174-183.
Wylie, R. C. (1979). The self-concept. Vol. 2. W. Lincoln, NE: University of Nebraska Press.
Yancey, W.L., Rigsby, L., & McCarthy, J.D. (1972). Social position and self-evaluation: The relative importance of race. American Journal of Sociology, 78, 338-359.

Authors note
The data used in this study emanate from my doctoral dissertation carried out under the direction of Dr. D.E. Benson. I thank Dr. Benson for his help to date and for his continued guidance. I also thank the anonymous reviewers of sosol for their helpful suggestions in improving the quality of this manuscript.
Correspondence concerning this article should be addressed to the author at: Department of Sociology and Anthropology, Radford University, Radford, VA 24141, U.S.A.; e-mail: jmekolic@radford.edu