Relationships Between Men's and Women's Body Image and Their Psychological, Social, and Sexual Functioning
Published in Sex Roles: A Journal of Research
The term body image is typically used to refer to perceptions and attitudes individuals hold about their bodies, although some authors argue that body image is a broader term, which encompasses behavioral aspects, such as weight loss attempts, and other indicators of investment in appearance (Banfield & McCabe, 2002). Women are generally considered to hold a more negative body image than men (Feingold & Mazzella, 1998). As a result, body dissatisfaction among women has been labeled a "normative discontent" (Rodin, Silberstein, & Striegel-Moore, 1985). However, through the use of gender-sensitive instruments that conceptualize body image concerns in terms of a desire to gain muscle, as well as to lose weight, previous beliefs that men are largely resilient to concerns about their appearance have been challenged, and there is now considerable evidence to suggest that young men are also dissatisfied with their bodies (Abell & Richards, 1996; Drewnowski & Yee, 1987).
A broad conceptualization of body image may prove important in understanding the nature of the construct among men, who appear to be less inclined than women to report holding negative attitudes toward their bodies, but do report a strong motivation to improve the appearance of their bodies (Davison, 2002). It may also be helpful to consider body image broadly when investigating its role throughout adulthood. Although the majority of research is limited to college samples, body image concerns appear to extend into later life (Montepare, 1996), and different age-related changes have been found among both men and women (Halliwell & Dittmar, 2003; Harmatz, Gronendyke, & Thomas, 1985). However, few researchers have systematically explored the development of different aspects of body image throughout the period of adulthood.
Although there has been a large body of research on the prevalence of body image concerns and potential factors associated with the development of body image, few researchers have systematically investigated the role body image plays in the day-to-day lives of individuals, beyond disturbed eating behaviors. In the present study we addressed this gap by exploring the association between body image and psychological, social, and sexual functioning among adult men and women. An innovative aspect of this study is the conceptualization of body image from a number of different aspects, making use of multiple gender-sensitive instruments, in order to understand the differential roles played by various aspects of body image. In addition, this study extends our understanding of the role of body image for adult men and women throughout the community, rather than focusing only on college students.
The associations between a disturbance in body image and psychological, social, and sexual dysfunction for different populations are currently not well understood. Previous researchers have demonstrated a relationship between body image and self-esteem among women in early adulthood (Abell & Richards, 1996; Monteath & McCabe, 1997) and in later years (Paxton & Phythian, 1999). This has led some authors to conceptualize women's body image as a component of a multidimensional global self-esteem (Marsh, 1997; O'Brien & Epstein, 1988). There are also preliminary indications that young women who report dissatisfaction with their physiques are at a greater risk of experiencing symptoms of depression or anxiety (Koenig & Wasserman, 1995; Mintz & Betz, 1986), although this relationship is less well understood among older women. There are inconsistencies in the literature, however, and it appears that results may be dependent on the particular aspect of body image measured. For example, self-esteem has been found to be unrelated to weight concerns among young women (Silberstein, Striegel-Moore, Timko, & Rodin, 1986), but strongly related to overall physical appearance (Harter, 1999). Researchers have not previously attempted to determine systematically which body image measures are most closely associated with different facets of psychological functioning. The importance of body image for the psychological functioning of men is particularly unclear, as inconsistent findings among young men stem in part from the use of different instruments, which vary in their sensitivity to measure aspects of body image most relevant to the lives of men. Of particular concern is the absence of research on the relationship between body image and self-esteem, depression, and anxiety among men from the general population.
A gap also exists in our knowledge of whether a disturbance in body image is relevant to interpersonal functioning. In the 1960s and 1970s, social psychologists demonstrated the positive impact of being considered physically attractive by others on desirability as a potential dating or romantic partner (Berscheid, Dion, Walster, & Walster, 1971; Walster, Aronson, & Abrahams, 1966). Less commonly researched, however, are the social implications of an individual's own rating of his or her attractiveness or other aspects of body image. There are preliminary indications in research with college students of an association between being concerned about one's appearance and impaired social functioning. College students who perceive themselves as unattractive have been shown to be more likely to avoid cross-sex interactions (Mitchell & Orr, 1976), to engage in less intimate social interactions with members of the same and other sex (Nezlek, 1988), and to experience higher levels of social anxiety (Feingold, 1992). Negative body image may also be related to problematic sexual functioning. Researchers have found that college students with poor views of their bodies are more likely than others to avoid sexual activities (Faith & Schare, 1993), to perceive themselves as unskilled sexual partners (Holmes, Chamberlin, & Young, 1994), and to report dissatisfaction with their sex lives (Hoyt & Kogan, 2001). However, other researchers have failed to find a relationship between body image and sexual functioning; Wiederman and Hurst (1997), for example, suggested that sexuality was related to objective attractiveness among women, but not to self-ratings of their appearance.
Remarkably few researchers have made explicit reference to the social context when investigating body image, which has resulted in the impression that body image evaluations and behaviors occur in social isolation. Recently, however, there is a growing awareness of the social nature of body image among female college students through their engagement in comparisons of their own appearance with that of others; such comparisons appear to be associated with negative evaluations of their bodies (Stormer & Thompson, 1996; Thompson, Heinberg, & Tantleff, 1991). In addition, researchers have found that a concern about others evaluating one's body negatively, a variable termed social physique anxiety, is related to low levels of body satisfaction (Hart, Leary, & Rejeski, 1989). This suggests that evaluations individuals make of their bodies are related to the evaluations that they expect others may make. However, the relative importance of social aspects of body image compared to individual aspects of body image evaluations and related behaviors has not been examined. It is currently unclear whether being dissatisfied with one's physique, considering oneself unattractive, rating one's appearance as important, applying effort to improve or conceal one's body, appearance comparisons, or social physique anxiety are of greatest relevance to people's psychological, social, and sexual functioning.
There are a number of other limitations in the literature. Few researchers have examined a range of body image constructs in order to understand which aspects of body image are most relevant to particular psychological, social, and sexual functioning variables. The diversity of different evaluative and behavioral body image constructs may account for some of the inconsistent research findings. Past research has also primarily focused on college students, typically women; very few studies have included participants from the general community. As a consequence, conclusions about the role of body image in the lives of men and women cannot be made. The relevance of body image may vary with age and gender, although researchers have previously failed to address this question.
The present study was designed to investigate systematically the role of body image in the lives of men and women throughout adulthood. A cross-sectional design was employed, due to the practicalities of obtaining a sample large enough to consider body image separately among men and women of different age groups. The lack of previous research in this area supports the contribution made by exploratory designs of this kind. Multiple measures of body image, including evaluative, investment, and social aspects, were compared, in order to determine which aspects of body image were most strongly predictive of psychological (i.e., self-esteem, depression, anxiety), social (i.e., relations with members of the same and other sex, social anxiety), and sexual (i.e., sexual optimism, sexual self-efficacy, sexual satisfaction) functioning. It was hypothesized that negative body image would be associated with poor functioning in these areas. Stronger relationships between body image and psychological, social, and sexual functioning were expected for women, and for younger participants, given the emphasis in the literature on the importance of body image for these groups.
The participants were 211 men and 226 women, who ranged in age from 18 to 86 years (M = 42.26 years, SD = 17.11). This age range was divided into three groups, and each participant was assigned to one of the following age groups: young adulthood, 18-29 years (n = 129), middle adulthood, 30-49 years (n = 153), and late adulthood, 50-86 years (n = 145). This division was carried out to create equal groups to meet the requirements of parametric statistical analyses. Reported occupations and postal addresses suggest that participants represented a wide range of socioeconomic backgrounds from metropolitan and rural areas. Over 80% of participants indicated they were originally from Australia; the remainder were predominantly from Western European countries. Nearly all (95.78%) participants identified themselves as heterosexual, and over 70% were in current relationships. The weight and height of the sample corresponded well with national Australian data for men and women (Australian Bureau of Statistics, 1998). These data are documented for men and women, and each age group separately in Table I.
Body Image Measures
Participants completed two subscales from the Body Image and Body Change Questionnaire (Ricciardelli & McCabe, 2001) that are related to Body Image Satisfaction and Body Image Importance. Each scale contained 10 items. An example item of body image satisfaction is "How satisfied are you with your weight?," and an example item of body image importance is "How important to you is the shape of your body, compared to other things in your life?" Responses were on a 5-point Likert scale from 1 = extremely dissatisfied/unimportant to 5 = extremely satisfied/important. Scores on each scale ranged from 10 to 50; a high score represents a high level of satisfaction with the body or a rating of appearance as highly important. These scales emerged from both exploratory and confirmatory factor analysis, and they have demonstrated high levels of internal consistency, satisfactory test-retest reliability, and concurrent and discriminant validity in previous studies with adolescents (Ricciardelli & McCabe, 2001). In the present sample, internal reliability (Cronbach's alpha) for each scale was high among both women and men ([alpha] > .90).
Participants rated their physical attractiveness using a scale specifically designed for this study, the Physical Attractiveness Scale, which measures how attractive they perceived themselves, for example, in terms of general appearance, facial attractiveness, and sexual attractiveness. This scale contains six items, an example of which is "Compared to other men, I am ..." Participants responded on a 5-point Likert scale from 1 = extremely unattractive to 5 = extremely attractive. Scores ranged from 6 to 30; a high score indicates a high self-rating of attractiveness. Internal reliability was high among both men and women ([alpha] > .90).
Two body image behaviors, body concealment (the tendency to conceal one's body from the gaze of others and to avoid discussion about body size and shape) and body improvement (engagement in attempts to improve one's body), were assessed using an instrument constructed for this study, the Body Image Behavior Scales. Items were derived in part from two extant instruments, the Body Image Avoidance Questionnaire (Rosen, Srebnik, Saltzberg, & Wendt, 1991) and the Attention to Body Shape Scale (Beebe, 1995), which were selected through exploratory and confirmatory factor analysis. The Body Concealment Scale consists of five items, an example item of which is "I avoid wearing 'revealing' clothes, like shorts or bathing suits." The Body Improvement Scale consists of three items, an example of which is "I exercise in order to get a better body." Participants responded on a 6-point Likert scale from 1 = never to 6 = always. Scores on the body concealment scale ranged from 5 to 30; a high score indicates a high engagement in attempts to conceal the body. Scores on the body improvement scale ranged from 3 to 18; a high score indicates a high engagement in attempts to improve the body. Internal reliability for each scale was high among both men and women ([alpha] > .80).
Concern about others evaluating one's body was assessed using the Social Physique Anxiety Scale (Hart et al., 1989). This scale contains 12 items, an example of which is "In the presence of others, I feel apprehensive about my physique/figure." Following the recommendation of Eklund, Kelley, and Wilson (1997), item 2 was modified (to improve performance) to "I worry about wearing clothes that might make me look too thin or overweight." Participants rated how true each of the items were using a 5-point Likert scale, from 1 = not at all true to 5 = extremely true. Scores ranged from 12 to 60; a high score indicates a high level of concern about others evaluating one's body (the responses to some items were reverse scored). Internal and test-retest reliability have been found to be adequate with a number of adult samples (Hart et al., 1989; Martin, Rejeski, Leary, McAuley, & Bane, 1997; Motl & Conroy, 2000; Petrie, Diehl, Rogers, & Johnson, 1996). Internal reliability was high among both men and women in the present sample ([alpha] > .80).
Participants indicated their level of appearance comparison by completing the Physical Appearance Comparison Scale (Thompson et al., 1991). This scale contains five items, an example of which is "At parties or other social events, I compare my physical appearance to the physical appearance of others." Responses were made on a 5-point Likert scale, from 1 = never to 5 = always. Scores ranged from 5 to 25; a high score indicates a strong tendency to compare one's own appearance with that of others. Although psychometric characteristics were found to be adequate with a university sample (Thompson et al., 1991), item 4 correlated with others at a low level in the present community sample (squared multiple correlation .70) and women ([alpha] > .80).
Psychological Functioning Measures
Participants completed the Rosenberg Self-Esteem Scale (Rosenberg, 1965). This scale contains 10 items, an example of which is "I feel that I have a number of good qualities." Responses were made on a 4-point Likert scale, from 1 = strongly disagree to 4 = strongly agree. Scores ranged from 4 to 40; a high score indicates high self-esteem (the responses to some items were reverse scored). This instrument has been widely used in research, and has demonstrated good psychometric properties (Rosenberg, 1979). Internal reliability was high among both men and women in the present sample ([alpha] > .80).
Participants also completed two subscales from the Depression Anxiety Stress Sub Scales (Lovibond & Lovibond, 1995). The Depression Scale contains 14 items related to symptoms of depression, an example of which is "I felt downhearted and blue." The Anxiety Scale contains 14 items related to symptoms of anxiety, an example of which is "I felt I was close to panic." Participants were asked to indicate the extent to which they had experienced each symptom over the previous week. Responses were made on a 4-point Likert scale from 0 = did not apply to me to 3 = applied to me very much or most of the time. Scores on each scale ranged from 0 to 42; a high score indicates a high level of depression or anxiety. These subscales are reliable measures of negative affective states among nonclinical college populations (Lovibond & Lovibond, 1995). Minor modifications were made to four items to improve comprehension in a community sample, with the aim of retaining the original meaning of items. To illustrate, the item "I found it difficult to work up the initiative to do things" was modified to "I found it difficult to work up the energy to do things." Internal reliability for each scale was high among both men and women ([alpha] > .90) in the present study.
Social Functioning Measures
Participants completed the social anxiety factor of the revised Self-Consciousness Scale (Scheier & Carver, 1985). This subscale contains six items, an example of which is "It takes me time to get over my shyness in new situations." Responses were made on a 4-point Likert scale, from 1 = not at all like me to 4 = a lot like me. Scores ranged from 6 to 24; a high score represents a high level of social anxiety (the responses to one item were reverse scored). The revised Self-Consciousness Scale has demonstrated good psychometric properties with samples from the general population (Scheier & Carver, 1985). Internal reliability was moderate among men ([alpha] > .70) and high among women ([alpha] > .80) in the present study.
Social functioning was also assessed by the Same-Sex Relations and Opposite-Sex Relations subscales of the Self-Description Questionnaire III (Marsh, 1989). Each subscale contains 10 items. An example of same-sex relations is "I have few friends of the same sex that I can really count on," and an example of opposite-sex relations is "I make friends easily with members of the opposite sex." Responses to each subscale were made on an 8-point Likert scale, from 1 = definitely false to 8 = definitely true. Scores ranged from 10 to 80; a high score indicates positive same-sex or opposite-sex relations (the responses to some items were reverse scored). These subscales have been found to have adequate internal consistency and reliability in previous studies (Marsh, 1989), and internal reliability for each scale was high among both men and women in the present study ([alpha] > .80).
Sexual Functioning Measures
Sexual functioning was measured with three subscales from the Multidimensional Sexual Self-Concept Questionnaire (Snell, 1995). The Sexual Self-Efficacy Scale contains five items, an example of which is "I have the ability to take care of any sexual needs and desires that I may have." The Sexual Optimism Scale contains five items, an example of which is "I expect that the sexual aspects of my life will be positive and rewarding in the future." The Sexual Satisfaction Scale contains five items, an example of which is "I am satisfied with the way my sexual needs are currently being met." Responses to items on each scale were made on a 5-point Likert scale from 1 = not at all true to 5 = very true. Scores on each scale ranged from 5 to 25; a high score represents a high level of the construct--high sexual self-efficacy, high sexual optimism, and high sexual satisfaction (the responses to some items were reverse scored). Internal consistency of the scales has previously been found to be high, and research has produced reasonable evidence for their validity (Snell, 2001). Internal reliability for each scale was high among both men and women ([alpha] > .80) in the present study.
Participants were recruited from the general community; they were selected at random from the White Pages telephone directory of metropolitan Melbourne and a variety of rural areas in Victoria, Australia. Questionnaires were distributed by mail to individuals who agreed to participate, and were completed at home and returned via mail to the researchers. A total of 157 individuals indicated they did not want to participate in the study and received no further contact from the researchers. Of the 720 questionnaires distributed, 437 were returned, which resulted in a response rate of 60.69% among those who agreed to receive a questionnaire, and an overall response rate of 49.83% among those contacted. There was no incentive provided for individuals to participate in the study, and responses were anonymous. Completion of the questionnaire took approximately 20-30 min.
In order to address the hypotheses outlined earlier, multivariate analyses of variance were conducted to determine the nature of sex and age differences in body image. Regression analyses were then conducted to determine which aspects of body image (if any) predicted the psychological, social, and sexual functioning of both men and women in each age group. Because of the number of analyses being conducted p < .01 was used to define significant results (Coakes & Steed, 1999).
Gender and Age Differences in Body Image
Differences in body image between men and women and among the different age groups were examined using a 2-way MANOVA, after controlling for the effects of Body Mass Index (BMI). Independent variables were gender and age group, and dependent variables were physical attractiveness, body image satisfaction, body image importance, body concealment, body improvement, social physique anxiety, and appearance comparison. Body image was found to be significantly different for men and women, F(7, 368) = 22.48, p < .001, and for different age groups, F(14, 738) = 6.00, p < .001. There was no significant interaction effect. The univariate F-tests for each dependent variable were examined in order to determine which body image variables contributed to the significant multivariate effects.
Women reported a lower level of body image satisfaction, F(1, 381) = 35.92, p < .001, and a higher level of social physique anxiety, F(1, 381) = 64.87, p < .001, than men did (see Table II). Women also reported concealing their bodies more frequently than men did, F(1, 381) = 130.38, p < .001, and they were more likely than men to engage in appearance comparisons, F(1, 381) = 25.61, p < .001. However, there were no differences between men and women in their ratings of physical attractiveness, body image importance, or level of engagement in efforts to improve their bodies.
After we controlled for the effects of BMI, we found significant differences between age groups in body image satisfaction, F(2, 381) = 11.74, p < .001, and body concealment, F(2, 381) = 5.52, p < .01; men and women in their 30s and 40s reported lower satisfaction with their bodies, and more frequent attempts to conceal their bodies, than did other participants (see Table II). Social physique anxiety scores also differed significantly between age groups, F(2, 381) = 18.97, p < .001; individuals in late adulthood reported a lower level of concern about others evaluating their bodies than did the younger participants. In addition, level of engagement in appearance comparison differed significantly between age groups, F(2, 381) = 12.34, p < .001; individuals in late adulthood were less likely than others to make appearance comparisons. Ratings of physical attractiveness, body image importance, and body improvement did not differ significantly between participants of different age groups.
Hierarchical multiple regression analyses were conducted in order to determine which aspects of body image most strongly predicted each psychological (i.e., self-esteem, depression, anxiety), social (i.e., same-sex relations, opposite-sex relations, social anxiety), and sexual functioning (i.e., sexual self-efficacy, sexual optimism, sexual satisfaction) variable. Separate analyses were conducted for men and women in each age group, as it was considered likely that the relationships would vary with both gender and age. In order to reduce the large number of independent body image variables for inclusion in each analysis, only those variables that significantly correlated with the dependent variable for each group were entered into the analysis. It was decided to control for the effects of self-esteem, depression, anxiety, and BMI, if they correlated significantly with the dependent variable. In addition, perceived relations with the other sex were considered as a potential control variable in analyses to predict sexual functioning. Control variables were entered as independent variables on the first step of each analysis, and body image variables were included as additional independent variables on the second step. The level of significance is typically corrected when there are a high number of contrasts. However, given the exploratory nature of these analyses, it was decided to consider effects significant at an alpha less than .05.
Results indicated that inclusion of body image variables at the second step significantly increased the prediction of self-esteem beyond that predicted by control variables among men in early adulthood, F change (5, 55) = 2.88, p < .05, middle adulthood, F change (4, 50) = 5.36, p < .001, and late adulthood, F change (4, 59) = 4.66, p < .01. The unique body image predictors of high self-esteem were positive ratings of physical attractiveness and a low rating of body image importance among men in early adulthood, a low level of body concealment among men in middle adulthood, and a low tendency to compare their appearance with others and high body image satisfaction among men in late adulthood (see Table III). Body image variables also significantly increased the prediction of self-esteem among women in early adulthood, F change (3, 50) = 4.60, p < .01, middle adulthood, F change (6, 84) = 5.41, p < .001, and late adulthood, F change (3, 56) = 4.37, p < .01. Although there were no unique body image predictors of self-esteem for women in early adulthood, low social physique anxiety and a low rating of body image importance predicted self-esteem among women in middle adulthood, and positive ratings of physical attractiveness predicted high self-esteem among women in late adulthood.
Inclusion of body image variables failed significantly to increase the prediction of depression or anxiety beyond the effect of control variables among most groups. However, body image variables entered at the second step significantly increased the prediction of depression among women in late adulthood, F change (4, 46) = 4.57, p < .01; high social physique anxiety acted as a unique body image predictor (see Table III). Body image variables entered at the second step significantly increased the prediction of anxiety among men in late adulthood, F change (2, 62) = 6.65, p < .01; a high level of appearance comparison acted as a unique body image predictor. For the predictor of anxiety among women in late adulthood, F change (4, 56) = 4.16, p < .01, although no specific body image predictor was found to explain unique variance.
Body image variables significantly increased the prediction of social anxiety at the second step, beyond the effect of control variables, among men in middle adulthood, F change (2, 52) = 4.54, p < .05; the unique body image predictor was a high level of appearance comparison (see Table IV). Inclusion of body image variables did not significantly increase the prediction of social anxiety among men in early or late adulthood, beyond the effect of control variables. Among women, inclusion of body image variables significantly increased the prediction of social anxiety during late adulthood, F change (6, 51) = 3.63, p < .01, but not at other ages. The unique body image predictors of social anxiety among women in late adulthood were high social physique anxiety and a high level of body improvement.
Inclusion of body image variables, entered as a group at the second step, did not significantly increase the prediction of same-sex relations among men in early or late adulthood, or among women of any age group, beyond the effect of control variables. However, a significant increase in the prediction of same-sex relations was found among men in middle adulthood, F change (5, 49) = 2.61, p < .05. Positive same-sex relations were uniquely predicted by positive ratings of physical attractiveness among this group (see Table IV). Inclusion of body image variables at this step significantly increased the prediction of positive cross-sex relations among men in young adulthood, F change (2, 57) = 4.17, p < .05; a low level of body concealment acted as a unique body image predictor, but did not increase the prediction of cross-sex relations beyond the effect of control variables among any other group.
Inclusion of body image variables, entered as a group at the second step, did not significantly increase the prediction of sexual self-efficacy or sexual satisfaction among women in any age group, or among men in early or late adulthood, beyond the effect of control variables. Among men in middle adulthood, however, inclusion of body image variables significantly increased the prediction of sexual self-efficacy, F change (5, 46) = 3.69, p < .01, and sexual satisfaction, F change (4, 49) = 6.27, p < .001; high body image satisfaction acted as the unique body image variable in both instances (see Table IV). A low tendency to compare their appearance to that of others and a low level of body concealment also predicted sexual satisfaction.
The group of body image variables, entered at the second step, did not significantly increase the prediction of sexual optimism among men or women in early or late adulthood beyond the effect of control variables. Inclusion of body image variables significantly increased the prediction of sexual optimism among men in middle adulthood, however, F change (4, 48) = 6.69, p < .001; low social physique anxiety acted as a unique body image predictor (see Table IV). Although body image variables increased the prediction of sexual optimism as a group among women in middle adulthood, F change (6, 81) = 2.72, p < .05, there were no unique body image predictors.
In the present study we considered a number of aspects of body image among men and women across different stages of adulthood. Body image concerns were generally found to be more prevalent among women than men; women reported lower satisfaction with their bodies and a greater tendency to conceal their bodies. Women appeared to be more focused on the social aspects of body image; they compared their appearance to that of others more frequently than men did, and they reported higher levels of social physique anxiety, which indicates that they were more concerned about others evaluating their appearance negatively. However, there were no gender differences in ratings of physical attractiveness or the perceived importance of appearance in the lives of men and women, and men were just as likely as women to report engaging in efforts to improve their bodies.
Body image concerns were relatively consistent throughout adulthood, which supports previous indications of the high prevalence of body image concerns among individuals beyond their college-aged years (Allaz, Bernstein, Rouget, Archinard, & Morabia, 1998; Ben-Tovim & Walker, 1994; Pliner, Chaiken, & Flett, 1990). There were some developmental trends, however, as men and women in their 30s and 40s were more vulnerable than other groups to dissatisfaction with their bodies and engaged in more attempts to conceal their bodies, for example, with nonrevealing clothing. This highlights the importance of attending to body image among adults beyond early adulthood, which is typically considered the most vulnerable period for body image disturbance. A developmental shift was also apparent in later years, most particularly in relation to the social aspects of body image. Although men and women over 50 years of age tended to make evaluations of their own appearance that were just as negative as those of younger participants, and did not perceive their appearance to be any less important than younger participants did, they reported less concern about others evaluating their bodies, and they were less likely to compare their appearance with that of others.
This exploratory study was designed to examine the relationships between different aspects of body image and psychological, social, and sexual functioning, rather than simply to document the existence or prevalence of body image concerns. Previous research, based on correlational analyses, has tended to conclude that a negative body image is associated with impaired psychological and interpersonal functioning. However, we used hierarchical regression analyses that controlled for the effects of possible moderator variables (self-esteem, depression, anxiety, BMI, and cross-sex relations), and found that body image variables did not contribute to a unique understanding of psychological, social, and sexual functioning among most groups.
An exception was found for self-esteem as a dependent variable. Self-esteem was predicted by body image variables among all groups. There were few gender differences in the overall strength of the association between body image and self-esteem, a finding that supports a number of previous studies of college students (e.g., Abell & Richards, 1996; Stowers & Durm, 1996), but is inconsistent with the conclusions of other researchers (e.g., Tiggemann, 1994) and the findings from a recent review (Powell & Hendricks, 1999). In the present study, although men at all stages of adulthood were less likely than women to hold a global negative body image, once developed, a poor body image was as strongly related to the general self-concept of men as it was of women. However, the particular aspect of body image most relevant to self-esteem differed according to age and gender. For example, physical attractiveness played an important role among men in early adulthood, but was more relevant to women's self-esteem in later years. Gender differences in the types of body image variables relevant to self-esteem may explain some of the inconsistencies in the literature, given that previous researchers exploring the relationship between body image and self-esteem have typically employed a single measure of body image.
The absence of relationships between body image and other aspects of psychological, social, and sexual functioning among most groups in this study appears to be best explained by shared relationships with self-esteem. To illustrate, although depression and body image variables were generally correlated, consistent with earlier research (Denniston, Roth, & Gilroy, 1992; Mable, Balance, & Galgan, 1986; Sarwer, Wadden, & Foster, 1998), associations were no longer present among most groups when we controlled for self-esteem. This is a surprising finding, given the attention paid by researchers to the importance of body image in understanding depression among women. In contrast to conceptualizations of body dissatisfaction as either a symptom or source of depression (Boggiano & Barrett, 1991; Koenig & Wasserman, 1995; McCarthy, 1990), it may be better understood in this context as an aspect of self-esteem (Allgood-Merten, Lewinsohn, & Hops, 1990). Thus, although men and women with a negative body image were more likely than others to report negative social and sexual functioning and to experience symptoms of depression and anxiety, this appeared to be due to the presence of a negative general self-concept.
This conclusion is made tentatively, given that it is contrary to much of the literature, and may be considered a preliminary finding. However, with the exception of depression, the relationships between body image and psychological, social, and sexual functioning have received little previous empirical investigation, even among samples of young women. In the limited research available authors failed to consider the role of self-esteem, with the exception of Allgood-Merten et al. (1990) whose conclusions support those of the present study. The current methodology does not allow for a direct evaluation of the relationships for men and women of different age groups, due to limitations in sample sizes. Replication of the findings is recommended, particularly using methods of analyses that allow for modeling of relationships, with particular attention paid to the role of self-esteem. For example, self-esteem may act as an important mediating factor between body image and day-to-day functioning.
Of interest in this study is the finding that body image played a role in psychological functioning among men and women over 50 years of age, in contrast to other adults. This was the only group for whom body image contributed to a unique understanding of depression and anxiety, beyond the shared association with self-esteem. Social aspects of body image were most relevant, as men in late adulthood who engaged in a high level of appearance comparison reported higher levels of anxiety and self-esteem than did men who were not concerned about how they looked in comparison to others. In addition, women in late adulthood who were highly concerned about how others may evaluate their appearance were more likely than other women their age to report symptoms of depression and social anxiety. Thus, although in general older men and women were less concerned about the social aspect of body image than younger individuals, the minority who did hold such concerns experienced symptoms of negative psychological adjustment.
Although body image was found to play a less important role in social and sexual functioning than previously proposed, it did appear to have particular relevance to the social and sexual functioning of men during middle adulthood, that is, men between the ages of 30 and 50 years. Men undergo a number of changes at this stage of their lives, in their interpersonal relationships, their roles at work, their families, and also in their physiques. It is during this developmental period when the negative physical effects of aging tend to become particularly apparent; men continue to gain body fat up until the age of 50 years, particularly around the abdomen area (Bemben, Massey, Bemben, Boileau, & Misner, 1998). Men do not typically express concerns about these changes directly, and they report a more positive body image than similarly aged women, both in this study and in previous research (Feingold & Mazzella, 1998). However, it appears that a minority of men, who present with the type of body image disturbance more typically observed among women, such as low satisfaction with their appearance, high social physique anxiety, attempts to conceal their bodies from others, and a tendency to compare their appearance to others, are more likely to experience significant difficulties in their interpersonal functioning, most noticeably in the sexual arena. Social aspects of body image played a particularly important role in middle-aged men's interpersonal functioning. To illustrate, high social physique anxiety was a particularly strong predictor of low sexual optimism, which suggests that middle-aged men who were concerned about others evaluating their bodies were likely to expect unrewarding future sexual interactions.
In contrast to the findings with men, women who expressed dissatisfaction with their bodies, and women who worried about how they "shaped up" in comparison with others and how others may perceive their bodies, experienced relatively few problems in their psychological, social, or sexual functioning beyond poor general self-esteem. The well-established, normative nature of women's views of their bodies may result in their body image concerns having only a limited negative association with other aspects of women's lives. This point has been made previously in relation to women's views of their sexuality (Wiederman & Hurst, 1997), but can be extended to include more general psychological and social functioning.
This research has demonstrated the importance of considering multiple measures of body image, given that different measures were associated with different aspects of psychological, social, and sexual functioning. Social aspects of body image, particularly concerns about how others may evaluate one's body, are a particular area that requires further research. The results of the present research also demonstrated the importance of investigating the effects of body image separately for men and women and for different age groups. This is the first study to demonstrate that body image may play different roles in the lives of different adult populations. Replication of these findings is required, particularly in longitudinal research, in order to explore potential underlying mechanisms to explain the role of body image in the psychological, social, and sexual functioning of men and women at different stages of adult development. The current sample was divided into three broad age categories, on the basis of sample size. Future researchers exploring the development of body image in adulthood should consider theoretically developed stages of adult development when selecting appropriate age categories to investigate. For example, body image may play a different role in the lives of adults 50-65 years than for adults in later years. Smaller, more homogeneous groups may demonstrate differences in the development of body image and highlight specific associations of body image and day-to-day functioning at different ages.
This study was limited by the use of correlational data. Small sample sizes in each group precluded the use of more sophisticated techniques, such as structural equation modeling, which may be employed in future research with larger samples to model relationships between body image and psychological, social, and sexual functioning variables. An investigation of these relationships was beyond the scope of this article, and they were not accounted for in the present analysis, which was focused on understanding which specific aspects of body image were of most relevance to particular aspects of day-to-day functioning. Future researchers may gainfully model the nature of the relationships between different aspects of body image for different populations. It is hoped that increased acknowledgement of the complexity of the body image construct, particularly in relation to the varied roles it plays in the lives of adult men and women, will stimulate further theoretical and empirical development in this area.
Continue to part 2 to see the tables