<%BANNER%>

Lesbian Women and Eating Disorder Symptomatology: A Test and Extension of Objectification Theory


PAGE 1

LESBIAN WOMEN AND EATI NG DISORDER SYMPTOMATOLOGY: A TEST AND EXTENSION OF OBJECTIFICATION THEORY By TIFFANY L. GRAHAM A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2005

PAGE 2

Copyright 2005 by Tiffany L. Graham

PAGE 3

This thesis is dedicated to my best friend and mother, Donna Graham.

PAGE 4

iv ACKNOWLEDGMENTS First of all, I would like to thank Dr. B onnie Moradi for her mentorship, patience, and guidance throughout this project. I would also like to thank Dr. Mark Fondacaro and Dr. Deidre Pereira for their valuable fee dback and positive support as members of my master’s committee. I also offer a huge tha nks to Dr. Mary Fukuyama for serving as a proxy at my thesis defense, and for her feedb ack regarding my survey items and study as a whole. I would also like to thank Jackie Davis for her f eedback regarding my survey and Jim Yousse for his valuable help with creating and uploading my survey to the internet. I would also like to thank the many le sbian/gay listserves a nd organizations that helped me in promotion of the study. With th eir aid, I was able to connect with lesbian women at the international level. Thanks go to Stacey Garner for her help in promoting and recruiting participants for this study. A nd of course, I would lik e to thank my friends (particularly my cohort) and family for al l of their help and support throughout the duration of this project. I would like to thank my Dad for showing me that it really can be done. Finally, I would like to thank Bradley Jason Daniels, my partner in life and in crime, whose constant belief in me helped me to believe in myself, and that is why I was able to complete and defe nd this project. Ya! Ya! Ya!

PAGE 5

v TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iv ABSTRACT......................................................................................................................v ii CHAPTER 1 INTRODUCTION........................................................................................................1 Empirical Support for Objectification Theory Applied to Eating Disorder Symptomatology.......................................................................................................2 Application of Objectification Th eory with Lesbian Individuals.................................6 Eating Disorder Symptomatology among Lesbian Women.........................................7 The Role of Connection with the Le sbian Community in Eating Disorder Symptomatology.......................................................................................................8 The Role of Feminist Ideology in Eating Disorder Symptomatology........................10 Significance of the Study............................................................................................13 2 REVIEW OF THE LITERATURE............................................................................14 Self-Objectification and Its Link to Body Shame and Eating Disorder Symptomatology.....................................................................................................16 Sexual Objectification Experi ences and Internal ization of Socioc ultural Standards of Beauty.................................................................................................................19 Objectification Theory as it A pplies to Lesbian Individuals......................................23 Lesbian Women and Eating Disorder Symptomatology............................................25 Connection with the Lesbian Community..................................................................29 Feminist Ideology.......................................................................................................31 Hypotheses..................................................................................................................39 3 METHODS.................................................................................................................41 Participants.................................................................................................................41 Instruments.................................................................................................................42 Reported Sexual Objectification Experiences............................................................43 Connection/Disconnection with Lesbian Community................................................46 Feminist Ideology.......................................................................................................47 Internalization of Sociocu ltural Standards of Beauty.................................................48 Body Surveillance as an Indicato r of Sexual Objectification.....................................49 Body Shame................................................................................................................50

PAGE 6

vi Eating Disorder Symptomatology..............................................................................51 Demographics.............................................................................................................52 Procedure....................................................................................................................53 Statistical Analyses.....................................................................................................54 4 RESULTS...................................................................................................................56 Descriptive Statistics..................................................................................................56 Intercorrelations among Variables of Interest............................................................57 Mediations..................................................................................................................58 5 DISCUSSION.............................................................................................................63 Limitations..................................................................................................................67 Directions for Future Research...................................................................................69 Implications for Practice.............................................................................................71 Summary.....................................................................................................................73 APPENDIX A THE SEXUAL OBJECTIFICATION SUBSCALE...................................................75 B OBJECTIFICATION EXPERIENCES QUESTIONNAIRE.....................................77 C THE SOCIOCULTURAL ATTITUDES TOWARD APPEARANCE INTERNALIZATION SUBSCALE (SATAQ)..........................................................78 D CONNECTION WITH THE LESBIAN COMMUNITY SUBSCALE (CLC).........79 E ATTITUDES TOWARD FEMINISM AND THE WOMEN’S MOVEMENT (FWM) SCALE..........................................................................................................80 F BODY SURVEILLANCE SUBSCALE OF THE OBJECTIFIED BODY CONSCIOUSNESS SCALE (OBC)..........................................................................81 G BODY SHAME SUBSCALE OF THE OBJECTIFIED BODY CONSCIOUSNESS SCALE (OBC)..........................................................................82 H THE EATING ATTITUDES TEST – 26 (EAT-26)..................................................83 I DEMOGRAPHIC QUESTIONNAIRE......................................................................84 J PERTINENT TABLES AND FIGURES...................................................................87 LIST OF REFERENCES...................................................................................................90 BIOGRAPHICAL SKETCH.............................................................................................97

PAGE 7

vii Abstract of Thesis Presen ted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science LESBIAN WOMEN AND EATI NG DISORDER SYMPTOMATOLOGY: A TEST AND EXTENSION OF OBJECTIFICATION THEORY By Tiffany L. Graham December, 2005 Chair: Banafsheh Moradi Major Department: Psychology Objectification theory posits that wome n in Western societ y experience sexually objectifying events that lead to the adoption of an outside r’s perspective upon one’s own body. Such self-objectification produces increas ed body shame in response to perceived failure to live up to cultural beauty standa rds. Body shame, in turn, results in eating disorder symptomatology. Although empirical support for the tenets of objectification theory is accumulating, to date, little research exists that addresse s the applicability of objectification theory to lesbian women. In addition, two factors, connection with the lesbian community and feminist ideology, have emerged from the literature as variables that may play a role in the development of eating disorder-related attitudes and behaviors among lesbian persons. The current study tested the objectification theory framework as it applies to eating disorder symptomatol ogy in a sample of 531 lesbian women, and additionally explored the role s of connection with the lesb ian community and feminist ideology. A theoretically based path analysis was conducted to investigate relationships

PAGE 8

viii among all variables in the model. Results indicated that the obj ectification theory framework was applicable for lesbian wo men, and disconnection from the lesbian community was a stronger pred ictor of eating disorder rela ted attitudes and behaviors than feminist ideology. Implications of findi ngs and directions for future research are discussed.

PAGE 9

1 CHAPTER 1 INTRODUCTION The prevalence of eating diso rder-related atti tudes and behaviors among women in Western society is astounding. Nearly two pe rcent of women develop anorexia nervosa (Walters & Kendler, 1995), and approximately three percent battle with bulimia nervosa (Romano & Quinn, 2001). In addition, the rates of eating disorders have doubled since the 1960’s, and disordered eating behavior ca n manifest as early as elementary school (Steiner & Lock, 1998). Even among wome n who are not diagnosed with eating disorders, concern with physical appearance and body weight is so pervasive that it has been deemed “normative discontent” (Rodi n, Silberstein, & Stri egel-Moore, 1984), and scholars have advocated the c onceptualization of eating probl ems as a continuum instead of discrete diagnostic categorie s (Scarano & Kalodner-Martin, 1994). Fredrickson and Roberts (1997) proposed obj ectification theory as a framework for understanding the development of eating disord er-related attitudes and behaviors along with other health concerns among women. Objectification theory posits that women within American culture are exposed to sexua lly objectifying events that include staring, sexual ogling, catcalls, a nd blatant sexual harassment and degradation. Often, emphasis is placed on women’s individual body parts, in stead of the whole person, and strong importance is placed on a woman’s body and not her mind, skills, or abilities. Girls and women are taught at an early age that physic al appearance is important, as they are frequently objectified through various medi a images including magazines, television shows, commercials, movies, and pornographi c materials. Through su ch experiences of

PAGE 10

2 sexual objectification, women ar e socialized to become “objects to be looked at and evaluated by others” (Fredric kson & Roberts, 1997, p. 177) and they learn to treat other women in the same way. According to objectification theory, cultural experiences of sexual objectification promote self-objectification, or taking on th e perspective of an observer upon one’s own body. More specifically, as a result of pervas ive experiences of sexual objectification, girls and women come to internalize an observer’s perspectiv e upon their bodies and “view themselves as objects or sights to be appreciated by others” (Fredrickson & Roberts, 1997, p. 180). Objectific ation theory posits that chronic self-objectification, in turn, increases body shame, promotes anxiety, reduces experiences of peak motivational states or flow experiences (i.e., periods of intense concentration on a challenging and rewarding task), and decreases awareness of in ternal bodily states (e.g., heart rate, sexual arousal). These experiences in turn result in depression, sexual dysfunction, and eating disorders among women. Within this larger framework of objectif ication theory (Fredrickson & Roberts, 1997), the links among sexual obj ectification experiences, se lf-objectification, and body shame have been posited as the most cr itical to understanding eating disorder symptomatology and empirical evidence has supported these proposed links. Empirical Support for Obj ectification Theory Applied to Eating Disorder Symptomatology To date, the relationships among self -objectification, body shame, and eating disorder symptomatology have received much empirical support. One such example exists in an experiment conducted by Fred rickson, Roberts, Noll, Quinn, and Twenge (1998). According to objectifica tion theory, self-objectificati on is likely to occur when a

PAGE 11

3 woman is made more aware of her body’s appe arance. Fredrickson et al. (1998) elicited such a situation by manipulating the type of clothing worn by participants. In this study, 114 women were assigned randomly to a sw imsuit or sweater c ondition. After trying on the designated clothing, participan ts were asked to look in a full-length mirror to evaluate the clothing, and then sample a cookie for a mock taste test. Results indicated that women who were wearing a swimsuit re ported significantly higher leve ls of self-objectification and body shame, and ate fewer cookies (i.e., re strained eating) than did women in the sweater condition. Thus, a situation that he ightened self-objectif ication resulted in restrained eating and higher levels of re ported self-objectifica tion and body shame. Contributing further support for the th eory, Noll and Fredrickson (1998) and Moradi, Dirks, and Matteson (2005) also f ound that self-objectification was related positively to eating disorder symptomatol ogy. Furthermore, both studies found evidence that this link was mediated, partially, by body shame. In other words, in addition to the direct positive link between self-objectific ation and eating disorder symptoms, selfobjectification was related to greater levels of body shame, which in turn were related to higher levels of eating disorder symptomato logy. These findings were consistent with objectification theory’s proposition that body shame is a key mechanism through which self-objectification is translated into eating disorder symptoms. In contrast to the numerous studies that have examined and found support for the proposed roles of self-objectif ication and body shame in eating disorder-relat ed attitudes and behaviors, sexual objectif ication experiences are only re cently beginning to receive empirical attention in the literature on object ification theory. Hill ( 2002) began to address this gap by examining the relationship betw een sexual objectification experiences and

PAGE 12

4 self-objectification with a sample of 502, mostly White women (307 heterosexual, 33 bisexual, 155 lesbian, and 7 who did not repor t sexual orientation). Hill assessed sexual objectification experiences usi ng a combination of existing m easures that assess sexual harassment, other degrading experiences, and th e extent that women are treated as sexual objects. Results indicated that repor ts of sexual objectificati on experiences were related positively to self-objectification. However, th is relationship was moderated by age, such that the magnitude of the relationship be tween sexual objectification experiences and self-objectification was strong and positive for women between the ages of 50 and 79, but non-significant for those between the ages of 18 and 49 years old. This interaction effect might be explained by the poten tial larger accumulation of obj ectification experiences for older women than for younger women. Additional evidence suggests that internal ization of sociocultural standards of beauty, a construct not explicitly included in the objectification theory framework or Hill’s (2002) study, might be an important mechanism that translates sexual objectification experien ces into self-objectification a nd other eating disorder-related variables. For example, Morry and Stas ka (2001) found that exposure to beauty magazines, a specific type of sexual objectification experi ence, was related to higher levels of disordered eating and body shape diss atisfaction, but these links were mediated by internalization of sociocu ltural standards of beauty. Th at is, exposure to beauty magazines was related to greater acceptance of society’s beauty mandates, and such internalization in turn was related to hi gher levels of selfobjectification, body shape dissatisfaction, and eating di sorder symptomatology.

PAGE 13

5 Moradi et al. (2005) conducted the mo st comprehensive assessment of the objectification theory framework as applied to eating disorder symptoms to date by including the proposed links among sexua l objectification experiences, selfobjectification, body shame, and eating diso rder symptoms. Based on literature highlighting the importance of internalization of sociocultural standards of beauty and Morry and Staska’s (2001) findings indicating the importance of including this variable in tests of objectification theory, Moradi et al also included internalization in the model. Moradi et al.’s path analytic findings w ith over 200 undergraduate women demonstrated that reported experiences of sexual objectification were si gnificantly related to higher levels of internalization, which in tu rn was significantly linked with greater body surveillance (an indicator of self-objectification), body sh ame, and eating disorder symptoms. In addition, body surveillance was linked significantly with body shame, which was correlated significan tly with disordered eating. Moradi et al. (2005) examined the signifi cance of mediator effects and found that internalization of sociocultural standards of beauty emerged as a mediator of the link of reported experiences of sexual objectificat ion to body surveillance, body shame, and eating disorder symptoms. Furthermore, as described in objectification theory, body surveillance also mediated the link of reporte d sexual objectification experiences to body shame. Finally, consistent with previous research (e.g., Noll & Fredrickson, 1998), body shame partially mediated the link of body surveillance to eating disorder symptomatology. The overall model provided a very good fit to the data and accounted for 50% of the variance in eating disorder symptomatology.

PAGE 14

6 Taken together, the studies conducted by Hill (2002), Morry and Staska (2001), and Moradi et al. (2005) supported the role of sexual objectification experiences in the objectification theory framework. In additi on, they demonstrated the importance of attending to the role of internalization of sociocultural standards of beauty in the objectification theory framework. Application of Objectification Theory with Lesbian Individuals The tenets of objectification theory have been examined with college women and men (Fredrickson et al., 1998; Huebner & Fredrickson, 1999; Morry & Staska, 2001; Muehlenkamp & Saris-Baglama, 2002; Noll & Fredrickson, 1998; Tiggemann, & Slater, 2001), adolescent dancers and non-dancers (Slater & Tiggemann, 2002), and women who exercise (Strelan, Mchaffey, & Tiggemann, 2003). However, to date, only one study has examined the aspects of the theory as applied to lesbian persons (Hill 2002). In Hill’s (2002) previously describe d study of objectificat ion theory, sexual orientation and age were examined as potenti al moderators of the relationship between sexual objectification ex periences and self-objectification. Consistent with objectification theory, in the entire sample (i.e., heterosexua l and lesbian women), experiences of sexual objectification were related positively to self-objectific ation. Furthermore, sexual orientation did not moderate the link be tween sexual objectification and selfobjectification, suggesting that the relationshi p was similar for both the heterosexual and lesbian women samples. Although this study be gan to address the gap in data on the applicability of objectification theory to le sbian persons, it repres ented only a partial investigation of objectification theory because it did not examine key constructs such as internalization, body shame, and eating disord er symptomatology. In addition, the sample

PAGE 15

7 only consisted of White heterosexual women and White lesbian women, thereby limiting the generalizability of findings. Thus, the generalizability of the propos itions of objectification theory to understanding eating disorder symptomato logy among lesbian women is not known and research is needed to examine the applicabil ity of objectification th eory to understanding eating disorder symptoms among lesbian pers ons. Such research must attend to the scholarship on eating disorder -related attitudes and behaviors among lesbian women and include any unique factors, not included in objectificati on theory, that might shape lesbian women’s experiences of eating diso rder-related attitude s and behaviors. Eating Disorder Symptomatology among Lesbian Women Findings are mixed as to whether or not lesbian women experience different levels of eating disorder symptomatology and body satisfaction when compared with heterosexual women (Beren, Hayden, Wilfle y, & Striegel-Moore, 1997). Some studies have revealed no significant differences between lesbian and heterosexual women (Beren, Hayden, Wilfley, & Grilo, 1996; Br and, Rothblum, & Solo mon, 1992; StriegelMoore, Tucker, & Hsu, 1990), whereas other studies indicate that lesbian women experience higher levels of body satisfacti on and lower levels of eating disorder symptoms than do heterosexual women. For example, Striegel-Moore and colleagues (1990) found no significant differences in ea ting behavior and body dissatisfaction in a sample of 52 heterosexual and 30 lesbian undergraduate students. Other studies, however, have shown that lesbian women report lowe r levels of concern with body weight and physical appearance, internalization of socioc ultural standards of beauty, and disordered eating symptomatology, but higher levels of ideal body weights and body satisfaction

PAGE 16

8 than do heterosexual women (Herzog, Newma n, Yeh, & Warshaw, 1992; Share & Mintz, 2002; Siever, 1994). Myers, Taub, Morris, and Rothblum (1998) used a qualitative research design to search for an explanation of findings of di fference versus no difference between lesbian and heterosexual women’s eating disorder-rel ated attitudes and behaviors. In their telephone interviews of 18 le sbian and 2 bisexual women, respondents reported either “feeling freedom from society’s norms after coming out” or still “feeling the pressure to be thin.” This finding highli ghts the importance of attending to individual differences among lesbian women that shape their experi ences of sexual objec tification and eating disorder-related attitudes and behaviors. Ex tant literature on eating disorder-related attitudes and behaviors among lesbian women points to c onnection with the lesbian community and feminist ideology as two cri tical individual difference variables that should be examined. The Role of Connection with the Lesbian Community in Eating Disorder Symptomatology Extant findings suggest that despite expe riencing similar cultural messages as heterosexual women, lesbian women might be less likely to internalize sociocultural beauty standards, which subsequently contri butes to higher body esteem and lower levels of disordered eating behavior (Bergeron & Senn, 1998; Share & Mintz, 2002). Lesbian women’s lower level of internal ization of sociocultural standards of beauty may be due to multiple causes. For example, the lesbian subculture may be more accepting of a variety of body types and shapes (Siever, 1994). I ndeed, one participant from the telephone interviews conducted by Myers et al. (1998) reported that the lesbian community is generally more accepting of larger women. Consistent with this participant’s

PAGE 17

9 observation, empirical findings suggest that lesbian women who interact with other lesbian women are likely to experience positive health effects. For example, in their study of 188 lesbian and bisexual women, Ludwig and Brownell (1999) found that lesbian persons with friends who also identified as lesbian reported a more positive body image than le sbian persons who reported having mostly heterosexual friends. Unfortunately, sample si zes for their study were rather unbalanced in that over 80 women reported having mostly fr iends who were also lesbian or bisexual, and this group was compared with 27 women reporting having mostly heterosexual female friends, 15 women reporting having mos tly gay or bisexual male friends, and 11 women reporting having mostly heterosexual ma le friends. Nevertheless, these findings suggest that interaction w ith other lesbian women might be related to positive body image for lesbian women. Heffernan (1996) examined directly the relationship between connection with the lesbian and gay community and eating di sorder symptomatol ogy. In this study, 203 lesbian women completed questionnaires assessing lifestyle, self-esteem, attitudes about attractiveness, body esteem, and eating be havior. Involvement with lesbian/gay community was assessed with the questi on “How involved are you in lesbian/gay activities?” Results indicated that the lesbian women were not significa ntly different from heterosexual women in terms of attitudes rega rding weight and appearance, and the two groups reported similar rates of bulimia ner vosa. Only one difference emerged suggesting that the lesbian participants reported binge eating disorder more frequently than the heterosexual participants.

PAGE 18

10 Thus, in general, between group analyses suggested overall similarity between lesbian and heterosexual women’ s eating disorder-related attit udes and behaviors. Within group analyses, however, revealed that among the lesbian women in the sample, involvement in the lesbian/gay community was related to better health outcomes such as lower weight and shape concern. Heffernan (1996) concluded that connection and involvement with the lesbian/gay culture might reduce internalizati on of society’s thin beauty ideal. Given that Heffernan used only a single item to assess connection with the lesbian/gay community, however, more compre hensive assessment of this important construct is needed in future research. The Role of Feminist Ideology in Eating Disorder Symptomatology Feminist ideology is a second important in dividual difference va riable that might shape lesbian women’s eating disorder-related attitudes and behaviors. Previous research has demonstrated that women’s endorsement of feminist ideology is related to lower levels of eating disorder-related attitu des and behaviors (Dionne, Davis, Fox, & Gurevich, 1995; Snyder & Hasbrouck, 1996) For example, Snyder and Hasbrouck (1996) examined this possibility in a study of the relationshi p between feminist identity development attitudes and symptoms of disturbe d eating in a sample of 71 female college students. Their results indicated that passive acceptance of traditional gender-role stereotypes (Feminist Identity Developmen t Passive Acceptance scores) was related positively to drive for thinness and body dissa tisfaction, whereas, active commitment to feminist ideology (Feminist Ident ity Development Active Commitment scores) was related negatively to those same outcome measures. Although this study provided some support for the role of fe minist ideology in eating disorder attitudes and behaviors, the generalizability of these findings to lesbian women was not examined.

PAGE 19

11 Nevertheless, given that lesb ian women are more likely to identify with feminism than are heterosexual women (Guill e & Chrisler, 1999), this variable may also play a significant role in the development of eati ng disorder-related at titudes and behaviors among lesbian women. Bergeron and Senn (1998) addressed this possibility in their examination of attitudes regarding the body, aw areness and internalization of sociocultural standards of beauty, and feminist self-identificati on among a sample of 108 lesbian and 115 heterosexual women between the ages of 18 a nd 58. Feminist identification was assessed using the question “Would you describe your self as a feminist?” A MANOVA revealed that lesbian women reported significantly higher ideal we ights, and reported feeling stronger and more fit than th eir heterosexual counterparts. A standard multiple regression revealed that in the entire sample, internaliz ation of sociocultural standards of beauty and feminist identification both were unique predictors of body attitudes, above and beyond sexual orientation. No differences were f ound between heterosexual and lesbian women on awareness of sociocultural st andards of beauty, and this variable did no t predict body attitudes. Bergeron and Senn’s (1998) findings sugge st the importance of examining both feminist identification and internalization when examin ing body attitudes. A notable strength of this study is the large sample size obtained by snowball sampling. However, over 95% of the participants identified as White. Furthermore, feminist identity was assessed using only a single item and the author s did not examine directly eating disorder symptomatology. Nevertheless, these findings can be taken as additional support for the

PAGE 20

12 importance of examining feminist ideology and its relationship with eating disorderrelated attitudes and behaviors. Further evidence of the relationship betw een feminist ideology and disordered eating attitudes and behaviors exists in Cogan’s (1999) study of 181 lesbian and bisexual women between the ages of 17 and 58. Part icipants completed measures assessing reasons for exercise, fitness activity freque ncy, type, and duration, dieting behavior, body satisfaction, eating disorder symptoms, physi cal appearance before and after coming out, feminist self-identification, and feminist ideo logy. To assess feminist self-identification, participants responded to the question “How much do you cons ider yourself a feminist?” and to assess for feminist ideology, participants completed the 10-item Attitudes Toward Feminism and the Women’s Movement scale (FWM; Fassinger, 1994). Controlling for Body Mass Index (BMI) a nd age as covariates, it was found that those who labeled themselves as a feminist (feminist self-identif ication) and endorsed feminist ideology were overall more satisfied with their bodies than those who did not. More specifically, higher levels of femini st self-identification and endorsement of feminist ideology were related to higher body satisfaction, lower rates of bulimia, drive for thinness, and weight discre pancy, and tendency to exercise for health versus aesthetic reasons. The authors concluded that: “femin ism may be a useful tool for unlearning internalized negative body image” (p. 85). Unfortunately, th is study did not assess for the internalization of such ideals. Thus, extant research suggests that fe minist ideology should be included in examination of disordered eating sympto matology among lesbian persons. However, since it is still unknown whether or not it is connection with the lesbian community,

PAGE 21

13 feminist ideology, or both that play important roles in the in ternalization of sociocultural standards of beauty and the manifestation of disordered eating symptoms in lesbian women, both were included in the current research. Significance of the Study The present research addressed a number of gaps in the literature by examining objectification theory as it ap plied to understandi ng eating disorder symptomatology with a lesbian sample. More specifi cally, the present study examined the previously supported framework of objectification theory that includes links among se xual objectification, internalization of sociocultural beauty sta ndards, self-objectific ation, body shame, and eating disorder symptomatology (Moradi et al ., 2005). Furthermore, extant research has supported the inclusion of two additional variables, connection with the lesbian community and feminist ideology, as key predictor variables in examining the applicability of the objectification theory fr amework to lesbian women. Thus, the present study included these variables in examining objectification theory’s applicability to understanding eating disorder symptomatol ogy with lesbian wome n. The online survey method provided for the recruitment of a large sample of lesbian wo men; therefore, the results obtained were more generalizable.

PAGE 22

14 CHAPTER 2 REVIEW OF THE LITERATURE In the United States, the rate of eatin g disorders is quite high among women, and these rates continue to cl imb. According to the APA Wo rk Group on Eating Disorders (2000), approximately .5 to 3.7% of women su ffer from anorexia nervosa, and 1.1 to 4.2% suffer from bulimia nervosa. In addition, the rates of eating disorders have doubled since the 1960’s, and disordered eating behavi or can manifest as early as elementary school (Steiner & Lock, 1998). Only 1 in 40,000 women fit the size and shape of a typical supermodel (Wolf, 1991), yet women pers istently engage in behaviors to achieve the near impossible through diet, exercise, various beauty products, wardrobe, surgery, and engaging in eating disorder-related beha viors. Chronic dieting has become a way of life for some women, and up to 60% engage in these behaviors by the time they are in high school (Steiner & Lock, 1998) This pattern of chronic re striction of food intake can carry with it serious and sometimes deadly consequences. Accordi ng to Fredrickson and Roberts (1997), eating disorders are “the ex treme end of a continuum of this normative discontent” (p. 191). In sum, eating disorder symptomatology has been identified as a serious mental health concern among women. Attempts to understand, prevent, and treat eating disorder symptomatology have highlight ed intrapersonal and contextual variables that could shape the development of such symptoms. Objectification theory (Fredrickson & R oberts, 1997) provides a framework that integrates both intrapersonal and contextual factors that play a role in mental health problems, such as eating disorder sympto matology, that have higher prevalence rates

PAGE 23

15 among women than men. More specifically, ob jectification theory posits that women within American culture are exposed to sexua lly objectifying events that include staring, sexual ogling, catcalls, and blatant sexual harassment and degradation. Women often encounter situations in which emphasis is placed on women’s individual body parts, instead of the whole person, and strong impor tance is placed on a woman’s body and not her mind, skills, or abilities. Girls and women are taught at an early age that physical appearance is important, as they are frequent ly objectified through various media images including magazines, television shows, commercials, movies, and pornographic materials. Through such cultural experiences, women are socialized to become “objects to be looked at and evaluated by others” (F redrickson & Roberts, 1997, p. 177) and they learn to treat other women in the same way. According to objectification theory, this milieu of cultural experiences of sexual objectification leads to selfobjectification, a crucial part of the overall model. Selfobjectification involves taking on the perspective of an observer upon one’s own body. In other words, girls and women come to “vie w themselves as objects or sights to be appreciated by others” (Fredrickson & Robe rts, 1997, p. 180). Objectification theory posits that chronic self-objectification, in tu rn, promotes anxiety, re duces experiences of peak motivational states or flow experiences (i.e., periods of inte nse concentration on a challenging and rewarding task), decreases awar eness of internal bodily states (e.g., heart rate, sexual arousal), and increases body shame. These experiences in turn result in depression, sexual dysfunction, and eating disorders among women. Within this larger framework of objectif ication theory (Fredrickson & Roberts, 1997), the links among sexual obj ectification experiences, se lf-objectification, and body

PAGE 24

16 shame have been posited as the most cr itical to understanding eating disorder symptomatology. Empirical resear ch supporting each of the rele vant relationships in the model is described in depth below. Self-Objectification and Its Link to Body Shame and Eating Disorder Symptomatology To date, the relationships proposed in objectification theory among selfobjectification, body shame, and eating disorder symptoms have received much empirical support. One such example exists in an e xperiment conducted by Fredrickson, Roberts, Noll, Quinn, and Twenge (1998). According to objectification theor y, self-objectification is likely to occur in situations in which a woman’s se nse of her body is accentuated and she is made more aware of her body’s appearance Fredrickson et al. (1 998) elicited such a situation by manipulating the type of clothi ng worn by participants. Within two separate experiments, 114 college women were assi gned randomly to a swimsuit or sweater condition. In the first experiment, 75% of pa rticipants were Caucasian, 10% Asian, 7% Hispanic, and 7% identified with other ( unspecified) ethnicities. In the second experiment, 83% of participants identified as Caucasian, 6% African American, 5% Asian, 2% Hispanic, and 4% identified with ot her (unspecified) ethnicities. Information regarding sexual orientation was not collected from these women. After trying on the designated clothing, part icipants were asked to look in a fulllength mirror to evaluate the clothing, and th en sample a cookie for a mock taste test. Body Mass Index (BMI) was calculated and contro lled as a covariate in the analyses, in order to account for any potential confounding e ffects of obesity. Resu lts indicated that women who were wearing a swimsuit reporte d significantly higher levels of selfobjectification and body shame, and ate fewer co okies (i.e., restrained eating) than did

PAGE 25

17 women in the sweater condition. In other wo rds, a situation that heightened selfobjectification resulted in restrained ea ting and higher levels of reported selfobjectification and body shame. Noll and Fredrickson (1998) provided furt her support for the relationships among self-objectification, body sham e, and disordered eating symptomatology. Two samples of undergraduate women attending Duke Univer sity (n= 93 and 111, respectively) were administered questionnaires including the Revised Bulimia Test (BULIT-R; Thelen, Farmer, Wonderlich, & Smith, 1991), the Ea ting Attitudes Test (EAT; Garner & Garfinkel, 1979), the Revised Restraint Sc ale (Polivy, Herman, & Howard, 1988) and the Self-Objectification and Body Shame Questionn aires (designed by the researchers). The data was analyzed using multiple regre ssion, with BMI held as a covariate. Results indicated that se lf-objectification related positively with body shame and symptoms of both bulimia and anorexia, a nd body shame related positively with both anorexia and bulimia symptoms. Using Bar on and Kenny’s (1986) procedure to test for mediation, the authors found that body shame acted as a mediator of the relationship between self-objectification and disorder ed eating symptomatology. The mediational model accounted for 35% of the variance in bul imic symptoms (p < .01) and 27% of the variance in anorexia symptoms (p < .01). It is also important to note that a direct relationship was found between self-objectification and symp toms of disordered eating. Thus, body shame was a partial mediator of the link between self-objectification and eating disorder symptomatology. Further support for body shame as a mediator variable, linking self-objectification to eating disorder symptomatology, was found in Tiggemann and Slater’s (2001) study.

PAGE 26

18 This study examined the tenets of objectifica tion theory by administering, to two samples of women, measures of gene ral self-objectification, body su rveillance (the specific manifestation of self-objectif ication as persistent body m onitoring), appearance anxiety, flow, awareness of internal bodily states body shame, and disordered eating. One group of women consisted of 50 former dancers and the other of 51 undergraduate psychology students. The former dancers were Caucasian women between the ages of 17 and 25 who had studied classical ballet for at least tw o years but no longer e ngaged in dance. The undergraduate students attended The Flinders University of South Australia, and most (over 95%) identified as Caucasian. The ages of this group ranged from 17 to 24, and none of the undergraduate participan ts had studied formal dance. As predicted by the researchers, former da ncers scored higher on the measures of general self-objectification, body surveillan ce, and disordered eating symptomatology than did the group of undergradu ate students. In both samples, the relationships of selfobjectification and body surveillance to disord ered eating symptomatology was mediated by body shame, but not by anxiety, flow, or awar eness of internal states. Finally, results indicated that body surveillance emerged as the manifestation of general selfobjectification that accounted for unique variance in the reported symptoms of disordered eating. That is, in this study of former dancers and non-dancers, body surveillance was a significant unique predictor of eating di sorder symptoms, F(1,90) = 31.30, p < .001, but general self-objectification was not, F (1,90) = 0.43, p > .05. This finding suggests that it is important to include assessment of body surv eillance as the critical manifestation of self-objectification in the context of the objectification theory model.

PAGE 27

19 In sum, the studies described above all were consistent with obj ectification theory’s proposition that body shame is a key mechan ism through which self-objectification is translated into eating disorder symptoms. Furthermore, their fi ndings indicate that a direct relationship also exists betwee n self-objectification and eating disorder symptomatology. Therefore, in congruence w ith extant research, the current study will examine body shame as a partial mediator of the link of body surveillance (the critical manifestation of self-objectification) to eating disorder symptomatology. Sexual Objectification Experiences and In ternalization of Sociocultural Standards of Beauty Fredrickson and Roberts (1997) identified sexual objectificati on experiences as a key precursor to self-object ification, body shame, and eat ing disorder symptoms in objectification theory. However, few studies ha ve included this proposed role of sexual objectification experiences when examining th e model. Hill (2002) began to address this gap in her examination of the relationship be tween sexual objectific ation experiences and self-objectification. In Hill’s (2002) study, sexual objectifi cation experiences were defined as experiences in which women are treated as se xual objects, and as a result, “become their bodies” and are evaluated as such (Hill, 2002, p. 5). Experiences of sexual objectification were assessed with a questionnaire desi gned by Hill, containing 40 items measuring sexualized gaze (with and without verbal comm ents), instances of sexual harassment, and sexual assault. The CSOS (Cultural Sexual Obje ctification Scale) was derived from other measures such as the Sexual Victimizati on Measure (SWV; Belknap, Fischer, & Cullen, 1999) and the Sexual Experiences Questionnaire (SEQ; Fitz gerald, Shullman, Bailey, Richards, Swecker, Gold, Ormerod, & Weitzma n, 1988). Self-objectification was defined

PAGE 28

20 as “the extent to which indi viduals view their bodies in observable, appearance-based (objectified) terms versus non-observable, competence-based (non-objectified) terms” (Noll & Fredrickson, 1998, p. 628). This va riable was assessed using the SelfObjectification Questionnaire designed by Noll and Fredrickson (1998), and the Objectified Body Consciousness Scal e (OBC; McKinley & Hyde, 1996). Over 500 mostly White women (307 heterosexual, 33 bisexual, 155 lesbian, and 7 who did not report sexual orienta tion) participated in this web-based survey. Participants were recruited from introductory psychol ogy classes at a large public Midwestern university (n = 101), and techni ques such as snowball sampling and email listserves also were used (n = 340). In addition, over 300 letters were mailed to randomly selected university faculty and staff, and of t hose women, approximately 12% accessed the website to complete the survey (n = 61). Upon accessing the website, participants were asked to complete questionnaires measur ing reported experiences of sexual objectification as well as self-objectification. Results indicated that reported experiences of sexual objectification were related positively to self-obj ectification. However, this relationship was moderated by age, such that the magnitude of th e relationship between sexual objectification experiences and self -objectification was strong and positive for women between the ages of 50 and 79, but nonsignificant for those between the ages of 18 and 49 years old. This interaction effect might be explained by the potential larger accumulation of objectification experiences for older women than for younger women. This study adds further support for the incl usion of sexual objectif ication experiences when examining the overall obj ectification theory model.

PAGE 29

21 Additional evidence suggests that internal ization of sociocultural standards of beauty, a construct not explicitly included in the objectification theory framework and not assessed in Hill’s (2002) study, might also be an important factor in understanding the role for sexual objectification experiences in eating disorder-related attitudes and behaviors. That is, internaliz ation of sociocultural standard s of beauty might be a key mechanism that translates sexual objectificat ion experiences into se lf-objectification and other eating disorder-related variables. S upport for this relationship can be found in a study conducted by Morry and Staska (2001). In this study, 89 female introductory psychology students completed me asures designed to assess exposure to fitness and beauty magazines (a specific type of se xual objectification experience), awareness and internalization of sociocultural attitudes and standards regarding appearance, selfobjectification, eating disorder sympto matology, and body shape satisfaction. The Sociocultural Attitudes Toward App earance Questionnaire (SATAQ; Heinberg, Thompson, & Stormer, 1995) was used to assess awareness (i.e., recognition of societal standards) and internalizati on (i.e., acceptance of sociocu ltural standards of beauty). Statistical analyses revealed that sexual objectification experi ences, operationalized as exposure to beauty magazines, were rela ted to higher levels of reported disordered eating symptoms and body shape dissatisfa ction, but these links were mediated by internalization of sociocultural standards of beauty. That is, exposure to beauty magazines (but not fitness magazines) was related to greater accep tance of society’s beauty mandates, and such internalization in turn was related to hi gher levels of selfobjectification, body shape dissatisfaction, a nd eating disorder symptomatology. Morry and Staska’s (2001) study was an important contribution to obj ectification theory

PAGE 30

22 research. A limitation of this study, howe ver, is that body shame was not assessed directly. In addition, Morry and Staska’s study focused on only one manifestation of sexual objectification experience (i.e., exposure to beauty magazines) and did not assess broadly the range of sexual objectification ex periences highlighted in objectification theory. In one of the most comprehensive studies of the objectification theory framework, Moradi et al. (2005) found additional support fo r the overall model, and further examined the roles of both sexua l objectification experi ences and internalization of sociocultural standards of beauty. In this study, over 200 undergraduate women in a large southeastern university completed surveys assessing th eir reported experiences of sexual objectification, internalization of sociocultural beauty sta ndards, self-objectification (manifested through body surveillance) body shame, and eating disorder symptomatology. A path analysis was c onducted using AMOS 4.01 (Arbuckle, 1999), and BMI was controlled as a covariate. Resu lts indicated that re ported experiences of sexual objectification related positively to body surveillance, body shame, and eating disorder symptoms. In addition, reported sexual objectification experiences related positively to internalization, which in turn was linked positively to body surveillance, body shame, and eating disorder symptoms. Mo radi et al. (2005) found evidence for the mediational role of internalization in the objectification theory framework. More specifically, they found that inte rnalization of sociocultural st andards of beauty acted as a partial mediator, linking reported experi ences of sexual obje ctification to body surveillance, and also mediated the link of reported experiences of sexual objectification to body shame and eating disorder symptoms. Moradi et al. (2005) also found support for

PAGE 31

23 the previously described medi ating role of body shame in the relationship between body surveillance and reported sympto ms of disordered eating. Th e overall model provided a very good fit to the data and accounted for 50% of the variance in eating disorder symptomatology. Taken together, the studies conducted by Hill (2002), Morry and Staska (2001), and Moradi et al. (2005) supported the role of sexual objectification experiences in the objectification theory framework. In additi on, they demonstrated the importance of attending to the role of internalization of sociocultural standards of beauty. However, only Moradi et al. (2005) included all variables rela ted to disordered eating symptomatology as described in objectifica tion theory. The current study will build on Moradi et al.’s study by examining their m odel, which includes relationships among reported experiences of sexual objectification, internalization of sociocultural standards of beauty, body surveillance, body shame, a nd eating disorder-related attitudes and behaviors, with a sample of lesbian women. Objectification Theory as it Applies to Lesbian Individuals The tenets of objectification theory have been examined with populations such as college women and men (Fredrickson et al ., 1998; Huebner & Fredrickson, 1999; Morry & Staska, 2001; Muehlenkamp & SarisBaglama, 2002; Noll & Fredrickson, 1998; Tiggemann, & Slater, 2001), older women (Tiggemann & Lynch, 2001), adolescent dancers and non-dancers (Slater & Tiggemann, 2002), and women who exercise (Strelan, Mehaffey, & Tiggemann, 2003). However, to date, only one study has examined the theory as it applies to le sbian persons (Hill, 2002). In Hill’s (2002) study of objectification theory, sexual orientation and age were examined as potential moderators of the relationship between se xual objectification

PAGE 32

24 experiences and self-objectificat ion. Consistent with objectifi cation theory, in the entire sample (i.e., heterosexual and lesbian women) experiences of sexual objectification were related positively to self-obj ectification. In addition, sexual orientation did not moderate the link between sexual objectification expe riences and self-obj ectification, as the relationship was similar for lesbian a nd heterosexual women. Although Hill’s study began to address the gap in data on the appli cability of objectification theory to lesbian persons, it represented only a partial investig ation of objectification theory because it did not examine key constructs such as intern alization, body shame, and disordered eating symptomatology. In addition, the sample onl y consisted of White heterosexual women and White lesbian women; thereby limiti ng the generalizability of findings. Thus, the generalizability of the propos itions of objectification theory to understanding eating disorder symptomato logy among lesbian women is not known and research is needed to examine the applicabil ity of objectification th eory to understanding eating disorder symptoms among a diverse samp le of lesbian persons. Furthermore, such research must attend to any unique factors, not included in objectification theory, that might shape lesbian women’s experiences of eating disorder-related attitudes and behaviors. Specifically, extant scholarship ha s identified two variables, connection with the lesbian community and feminist ideology, as key potential fact ors that might be related to eating disorder-re lated attitudes and behaviors among lesbian persons. The following sections provide an overview of available data on eating disorder symptomatology among lesbian women and revi ew extant literature on the links of connection with the lesbian community and femi nist ideology to eating disorder-related attitudes and behaviors.

PAGE 33

25 Lesbian Women and Eating Disorder Symptomatology Findings are mixed as to whether or not lesbian women experience different levels of eating disorder symptoms and body satisfa ction when compared with heterosexual women (Beren, Hayden, Wilfley, & Striegel-M oore, 1997). Some studi es have revealed no significant differences between lesbian a nd heterosexual women on measures of body dissatisfaction and eating disorder sympto ms (e.g., Brand, Rothblum, & Solomon, 1992; Heffernan, 1996; Striegel-Moore, Tucker, & Hs u, 1990). Other studies, such as Share and Mintz (2002), found that lesbian women expe rience lower levels of body dissatisfaction, but report similar levels of eating disorder symptoms when compared with heterosexual women. Conflicting results obtai ned by Siever (1994), s howed that lesbian women exhibit lower rates of eati ng disorder symptoms but re port similar levels of body dissatisfaction in comparison with heterose xual women. Finally, some research studies have demonstrated that lesbian women expe rience higher levels of body satisfaction and lower levels of disordered eating symp toms than heterosexual women (e.g., Herzog, Newman, Yeh, & Warshaw, 1992, Lakkis, Riccia rdelli, & Williams, 1999). Examples of research supporting each of these three c onflicting findings are described below. Striegel-Moore et al. (1990) compared 30 lesbian undergraduate women with 52 heterosexual graduate women using questionn aires measuring self-esteem, body esteem, and disordered eating attitudes and behavi ors, and very few group differences were found. Using MANOVA, no significant differen ces were found on the measures of body image satisfaction and symptoms of disorder ed eating. The differences that were found between the two groups were related to self -esteem, with the lesbian women reporting lower self-esteem than the heterosexual wo men. The researchers concluded that lesbian and heterosexual college students do not di ffer in terms of body esteem and eating

PAGE 34

26 disorder symptoms, and that perhaps the lesb ian ideology of rejecting culture’s narrowly defined beauty ideals is not enough to overcom e the socialized, intern alized beliefs about female beauty that women encounter. Beren, Hayden, Wilfley, and Grilo (1996) al so found no significant differences in body esteem and body dissatisfaction between heterosexual women and lesbian women. In this study, 257 participants (69 lesbia n women, 72 heterosexual women, 58 gay men, and 58 heterosexual men) completed measures of body dissatisfaction, self-esteem, selfconsciousness, affiliation with the lesbia n/gay community, and sexual orientation. A MANOVA revealed that the lesbian and hete rosexual women in this sample scored similarly on each of the body dissatisfaction m easures, suggesting that identifying as a lesbian person may not be enough to overcome so ciety’s pressure to conform to the ideal body type. Unfortunately, this study failed to include any type of measure of eating behavior, and therefore conclusions regard ing eating disorder sy mptomatology between lesbian and heterosexual women could not be ascertained. Share and Mintz (2002) addressed this gap in their examination of the differences between lesbian and heterosexual women on body esteem, awareness and internalization of cultural attitudes concerni ng thinness, disordered eating symptoms, physical condition, and sexual attractiveness. This study was ba sed on a sample of 173 women between the ages of 24 and 52. A total of 102 (59%) particip ants identified as exclusively or primarily heterosexual, 63 (36%) described their sexual orientation as exclus ively or primarily homosexual, and 8 (5%) reported identifying as bisexual. The participants completed the Eating Attitudes Test (EAT-26; Garner, Olms ted, Bohr, & Garfinkel, 1982) to assess for disturbed eating patterns, the Body Esteem Scale (BES; Branzoi & Shields, 1984), and

PAGE 35

27 the Sociocultural Attitudes Toward Appear ance Questionnaire (SATAQ; Heinberg et al., 1995). Body mass index was also obtained and co ntrolled as a cova riate in MANCOVAs used to compare the two groups. Statistical analyses revealed that lesb ian women reported higher levels of body esteem and lower levels of internalization of cultural beauty standards, but no differences were found on levels of awareness of cu ltural standards, disordered eating symptomatology, or body esteem. The author s concluded that although lesbian women are equally aware of sociocultu ral standards of beauty, they are less likely to internalize the cultural attitudes. However, the nonsignificant differences in eating disorder symptomatology may be an indication that identification as a lesbian woman may not offer enough of a buffer from disturbed eating. Siever (1994) also examined eating disord er-related attitudes and behavior in his study of 250 students from the University of Washington and Seattle Central Community College. The sample included 53 lesbian women, 59 gay men, 62 heterosexual women, and 63 heterosexual men. Participation involve d responding to three versions of the Body Esteem Scale (Franzoi & Herzog, 1986, 1987; Franzoi & Shields, 1984), the Body Shape Questionnaire (BSQ; Cooper, Taylor, Coope r, & Fairburn, 1987), Body size drawings (Stunkard, Sorensen, & Schulsinger, 1980), the Eating Disorder Inventory (EDI; Garner, Olmstead, & Polivy, 1983), the Eating Attitudes Test (EAT-26; Garner et al., 1982), and a demographic questionnaire designed by th e researchers. Results indicated that heterosexual and lesbian women reported si milar levels of body dissatisfaction (with lesbian women slightly, but not significantly less dissatisfied), but lesbian women scored lower on the measures of eating disorder symptomatology. The author concluded that

PAGE 36

28 “sociocultural factors can have an immunizing effect lesbians, because of a decreased emphasis on physical appearance in their commun ity, appear to be le ss vulnerable to the attitudes and behaviors that typi fy eating disorders” (p. 257). Other studies have shown that lesbian wo men report lower levels of concern with weight and physical appearance, internalizati on of sociocultural sta ndards of beauty, and disordered eating symptomatology, but hi gher levels of ideal weights and body satisfaction, than do heterosexual wo men (Abraham & Beumont, 1982; Strong, Williamson, Netemeyer, & Geer, 2000). An example of such findings can be found in Herzog et al.’s 1992 study. In this study, 109 unmarried women between the ages of 18 and 35 completed a demographic questionnai re, the Eating Disorders Inventory (EDI; Garner et al., 1983), and a set of 12 female figure drawings deve loped by Furnham and Alibhai (1983). In this task, the participants were instru cted to select their current body type and their ideal body type. The resu lts showed that lesbian women weighed significantly more than heterosexual women, de sired a significantly heavier ideal weight, were less concerned with physical appearance and weight, and repor ted a lower drive for thinness. The authors concl uded that lesbian women’s higher rates of body satisfaction and lower weight concern may be a factor in their lower rates of reported symptoms of disordered eating. Myers, Taub, Morris, and Rothblum (1998) used a qualitative research design to search for an explanation of findings of di fference versus no difference between lesbian and heterosexual women. In their telephone interviews of 18 lesbian and 2 bisexual women between the ages of 17 and 60 (mean age = 32), respondents reported either “feeling freedom from society’s norms after coming out” or still “feeling the pressure to

PAGE 37

29 be thin.” This finding highli ghts the importance of attending to individual differences among lesbian women that shape their experi ences of sexual objec tification and eating disorder-related attitudes and behaviors. Ex tant literature on eating disorder-related attitudes and behaviors among le sbian persons has identified two factors, connection with the lesbian community and feminist ideology, as critical individua l difference variables that should be examined. Connection with the Lesbian Community One explanation for the conflicting pattern of findings regard ing lesbian women and eating disorder symptoms is that even though lesbian persons are exposed to the same cultural messages as all women, sexual relations with other women may encourage body acceptance and lower concern regarding app earance and weight (Beren, et al., 1997; Siever, 1994). It may be that exposure to other lesbian women decreases the opportunity to internalize sociocultural standards of b eauty, subsequently contributing to higher body esteem and lower rates of disordered eati ng symptomatology (Share & Mintz, 2002). This lower level of internalization of sociocultural standards of beauty may be due to multiple causes. For example, the lesbian subculture may be more accepting of a variety of body types and shapes (Siever, 1994). An example of this can be found from one respondent in the Myers et al. (1998) study who stated that in general, the lesbian community is more accepting of larger women. In their interviews with 26 lesbian, libera l arts college students, Beren et al. (1997) elicited opinions rega rding lesbian beauty ideals and their sources, the experience of conflict regarding beau ty in society, the need to overcome negative stereotypes as a lesbian woman, and concerns about feminist identity that may potentially influence feelings about one’s body. Re sults indicated that lesbian women reported a conflict

PAGE 38

30 between lesbian ideology and cultural valu es regarding beauty, but that intimate involvement with other lesbian women positiv ely influenced feelings about their bodies and decreased the importance of appearance. Thus, connection with others who identify as lesbian may lead to positive health outcomes such as higher body esteem and reduced symptoms of disordered eating. Ludwig and Brownell (1999) further exam ined this possibility by studying the relationship between gender roles, group aff iliation, and body satisfaction in a sample of 188 lesbian and bisexual women. Participants in this study were recruited through the Internet and email, and thus a wide range of ages was represen ted among participants. Information regarding race and ethnicity, how ever, was not assessed. Results indicated that lesbian persons with friends who also identified as lesbian reported a more positive body image than lesbian persons who repor ted having mostly heterosexual friends. Unfortunately, Ludwig and Brownell’s sample sizes were rather unbalanced in that over 80 women reported having mostly friends who we re also lesbian or bisexual, and this group was compared with 27 women reporti ng having mostly heterosexual female friends, 15 women reporting havi ng mostly gay or bisexual male friends, and 11 women reporting having mostly heterosexual male friends. Furthermore, body image was assessed using a single measure of body satisfa ction, and information pertaining to other eating disorder-related attitudes or symp toms was not obtained. Nevertheless, these findings suggest that interaction with wome n of the same sexual orientation might be related to positive body image for lesbian women. Heffernan (1996) examined directly the relationship between connection with the lesbian and gay community and eating di sorder symptomatology. In this study,

PAGE 39

31 questionnaires were used to examine lifes tyle, disordered eating symptomatology, body esteem, attitudes about attractiveness, and self-esteem in 203 lesbian women between the ages of 17 and 65 (mean age = 34). Involve ment with lesbian/gay community was assessed with the question “How involved are yo u in lesbian/gay activ ities?” Participants were asked to rate their response using a Like rt-type scale. Findings indicated that the rate of bulimia nervosa among lesbian women was comparable to that of heterosexual women, but binge eating disorder was more frequent in the lesbian women sample. Lesbian women were not significantly different from heterosexual women regarding their attitudes about weight and appearance, or dieting. But among the lesbian women, higher levels of participation in le sbian and gay activities and orga nizations was related to lower weight concern. Thus, between group analyses suggested ove rall similarity between lesbian and heterosexual women’s eating disorder-relate d attitudes and behaviors. Within group analyses, however, revealed that among the lesbian women in the sample, involvement in the lesbian/gay community was related to better health outcomes such as lower weight and shape concern. Heffernan (1996) conclude d that connection a nd involvement with the lesbian/gay culture might reduce internalization of society’s thin beauty ideal. Given that Heffernan used only a single item to assess involvement in the lesbian/gay community, however, more comprehensive asse ssment of this important construct is needed in future research and the present study addressed this need. Feminist Ideology Feminist ideology is a second important in dividual difference va riable that might shape lesbian women’s eating disorder-related attitudes and behavior s. Indeed, previous research has demonstrated that women’s endorse ment of feminist ideology is related to

PAGE 40

32 lower levels of eating disord er-related attitudes and beha viors (e.g., Dionne, Davis, Fox, & Gurevich, 1995; Snyder & Hasbrouck, 1996) For example, Dionne et al. (1995) studied the relationship between feminist atti tudes and body satisfac tion in a sample of 200 primarily White women between the ages of 17 and 48. Participants were volunteers solicited from the student, staff, and faculty population at a large Canadian university. The authors reported recruiti ng a diverse sample, but di d not include descriptive information about the ethnic and racial bac kground of the particip ants. Participation involved completing the Body Cathexis Scale (Secord & Jourard, 1953) to assess general body dissatisfaction, the EDI (Garner & Olmsted, 1984) to assess specific body dissatisfaction, and the Composite Feminist Ideology Scale (CFIS; Dionne, 1992) to measure the degree of support for the tenets of the women’s movement. Results revealed that women’s feminist attitudes regarding phys ical attractiveness (i.e., the rejection of traditional societal beauty standards) wa s related significantly and positively to body satisfaction. Leavy and Adams (1986) also examined fe minism and its potential link to positive health outcomes such as social support, se lf-esteem, and self-accep tance. Questionnaires were used to examine the relationship between feminism, self-esteem, self-acceptance, and social support in a sample of 123 women who identified as either predominantly or exclusively homosexual. The ag es of participants ranged from 15 to 52 (mean age = 26), and 98% of the sample identified as White In their study, Leavy and Adams defined feminism as having two components: “a se t of beliefs about women’s rights” and “involvement in feminist activities” (p. 322). Thus, feminism was assessed through a two-component questionnaire adapted by the authors. The first component contained

PAGE 41

33 seven items measuring strength of agreement with feminist beliefs concerning sex roles and political action, and one se lf-perception item that required participants to rate their perception of themselves as feminists. The second component of the questionnaire included an inventory of femini st activities, and respondents we re required to report their level of participation on a 5-point Likert scale (from never to many times ). Statistical analysis revealed that lesb ian women who reported being active in feminist organizations possesse d better social support systems, higher self-esteem, and greater self-acceptance than those who did not. Feminist beliefs, however, were not correlated significantly with self-esteem, so cial support, or self-acceptance. These findings must be interpreted in light of the fact that participants’ endorsement of feminist beliefs was extremely positively skewed and such severe range restriction attenuates the potential observed relationship between femini st beliefs and the health related outcomes examined in this study. Nevertheless, when feminist activity wa s examined, lesbian women who reported being active in lesbian or feminist organizati ons reported having better social support systems, higher self-est eem, and greater selfacceptance than those who did not. The findings of this study suggest that involvement in feminist activities might be related to positive health outcomes in lesbian women. Thus, it is possible that this link might also generalize to eating disorder symptomatology. Snyder and Hasbrouck (1996) examined this possibility in a study of the relationship between feminist identity devel opment attitudes and symptoms of disturbed eating in a sample of 71 female college stude nts between the ages of 17 and 22. In this study, the relationships between feminist iden tity development attitudes, gender traits, and eating disorder symptomatology were ex amined through the use of questionnaires.

PAGE 42

34 The sample was predominantly White (95 %), and the study was conducted at a middle class, liberal arts college in which re search participation credit was given as compensation. Results indicated that th ose women who endorsed feminist values reported less dissatisfaction with their body weight and overa ll size, less concern for thinness, fewer bulimic symptoms, and fewer feelings of ineffectiveness. More specifically, passive acceptance of traditional gender -role stereotypes (Feminist Identity Development Passive Acceptance scores) related positively to drive for th inness and body dissatisfaction, whereas, active commitment to a feminist ideology (Feminist Identity Development Active Commitment scores) related negativ ely to those same outcome measures. Although this study provided some support for th e potential role of feminist ideology in reducing eating disorder attitude s and behaviors, the generaliz ability of these findings to lesbian women was not examined. Given that lesbian women are more likely to identify with feminism than are heterosexual women (Guille & Chrisler, 1999), however, feminist ideology may also play a significant role when examining eating disorder-related attitudes and behavior s among lesbian women. This possibility was addressed in Guille a nd Chrisler’s (1999) research. In this study, 217 women were recruited from colle ge campuses, community groups, bookstores, and lesbian support and activis t groups from the Connecticut, Boston, and San Francisco areas. This sample consisted of 52 adult le sbians between the ages of 25 and 70 (mean age = 38), 56 adult heterosexual women between the ages of 25 and 84 (mean age = 37), 51 young adult lesbian women between the ages of 16 and 24 (mean age = 20) and 58

PAGE 43

35 young adult heterosexual women between the ages of 15 and 24 (mean age =19) who participated in a study of “women and body image.” Participation involved the completion of que stionnaires measuring feminist identity and eating disorder symptoms. Feminist ident ity was assessed using the Feminist Identity Scale (FIS; Worell & Remer, 1992), which cons ists of four subscales: acceptance (of traditional gender roles), revelation (realization of sexism), embeddedness (immersion in female culture), and commitment (active work to improve the status of women). Eating disorder symptomatology was measured usi ng the Eating Attitudes Test (EAT; Garner & Garfinkel, 1979), the Compulsive Eating Scale (CES; Dunn & Ondercin, 1981), and the Three-Factor Eating Questi onnaire (TFEQ; Stunkard & Me ssick, 1985). Taken together, these last three questionnaires we re used to assess attitudes and behaviors associated with bulimia and anorexia nervosa, food obsessions, beliefs regarding the ab ility to resist the urge to eat, dietary restraint, disinhibition (i.e., difficulty in ceasing once one has started eating), and general hunger. The researchers used both univariate anal yses and standard multiple regression to analyze the obtained data. Results indicated th at lesbian women were significantly more likely to endorse feminist attitudes than were heterosexual women. That is, lesbian women scored significantly lower on the accep tance subscale and significantly higher on the commitment subscale than did the heterosexual women. In other words, lesbian women in this study were significantly more likely to report actively working to improve the status of all women than he terosexual women, and at the same time were less likely to report attitudes and behaviors associated wi th eating disorders. In addition, for all women, those with higher scores on the commitme nt subscale were le ss likely to restrict

PAGE 44

36 their food intake than those with higher acceptance scores. Based on these findings, Guille and Chrisler posited that feminist id eology might reduce eating disorder-related attitudes and behaviors. However, the generali zability of Guille and Chrisler’s findings is unclear since the authors did not report the demographic characteristics of their sample such as race, ethnicity, and socioeconomic st atus. Furthermore, BMI was obtained for the sample, but since it was not controlled as a covariate in the analys es, the potential for confounding exists. In a more comprehensive study, Bergeron and Senn (1998) examined attitudes regarding the body, awareness and internalization of sociocu ltural standards of beauty, and feminist self-identifica tion with a sample of 108 lesb ian and 115 heterosexual women between the ages of 18 and 58. Participants completed a demographic measure, the Body Attitude Questionnaire (BAQ; Ben-Tovim & Walker, 1991) and the Sociocultural Attitudes Towards Appearance Questionnaire (S ATAQ; Heinberg et al., 1995). Feminist identification was assessed using the question “Would you describe yourself as a feminist?” A MANOVA revealed that lesbian women reporte d significantly higher ideal weights, and reported feeling stronger and more fit than their heterosexual counterparts. A standard multiple regression revealed that in the entire sample, internalization and feminist identification both were unique predictors of body attitudes, above and beyond sexual orientation. No differences were f ound between heterosexual and lesbian women on awareness of sociocultural st andards of beauty, and this variable did no t predict body attitudes. For all women, in ternalization predicted body at titudes, and lesbian women were less likely to report internalization of sociocultural norms than heterosexual women.

PAGE 45

37 A notable strength of this study is th e large sample size obtained by snowball sampling. However, over 95% of the particip ants identified as White. Furthermore, feminist identity was assessed using only a si ngle item and the authors did not examine directly eating disorder sy mptomatology. Nevertheless, th ese findings of Bergeron and Senn (1998) suggest the importance of exam ining both feminist identification and internalization when examining body attitudes. Further evidence of the rela tionship between feminist id eology and eating disorder symptomatology exists in Cogan’s (1999) study of 181 lesbian and bisexual women recruited from Sacramento, California. Partic ipants were between the ages of 17 and 58 (mean age = 34), with 88% identifying as le sbian, and 12% identif ying as bisexual. In terms of ethnicity, 73% of the sample identi fied as White, 8% as Latina, 7% as Asian American, 6% as African American, 4% as Native American, and 2% as multiracial. Participants completed measures assessi ng reasons for exercise, fitness activity frequency, type, and duration, dieting be havior, body satisfaction, eating disorder symptoms, physical appearance before and af ter coming out, feminist self-identification, and feminist ideology. To assess for feminist se lf-identification, partic ipants responded to the question “How much do you c onsider yourself a feminist?” and to assess for feminist ideology, participants completed the 10-it em Attitudes Toward Feminism and the Woman’s Movement scale (FWM; Fassinger, 1994). Using ANOVAs (holding BMI constant), Cogan (1999) found that when compared with a sample of heterosexual women (Coga n, Bhalla, Sefa-Dedeh, & Rothblum, 1996), these lesbian women demonstrated similar drive for thinness and body dissatisfaction. Controlling for BMI and age as covariat es, it was found that those who labeled

PAGE 46

38 themselves as a feminist (feminist self-ide ntification) and endorsed feminist ideology were overall more satisfied with their bodies than those who did not. More specifically, higher levels of feminist self -identification and endorsement of feminist ideology were related to higher body satisfacti on, lower rates of bulimia, drive for thinness, and weight discrepancy, and tendency to exercise for h ealth versus aesthetic reasons. Cogan (1999) concluded that feminism might provide a means by which women can unlearn a negative body image that had been internalized by repeated exposure to thin body ideals. Unfortunately, this study did not assess fo r the internalization of such ideals. Thus, extant research suggests that fe minist ideology should be included in examination of disordered eating sympto matology among lesbian persons. However, since it is still unknown whether or not it is connection with the lesbian community, feminist ideology, or both that play important roles in the in ternalization of sociocultural standards of beauty and the manifestation of disordered eating symptoms in lesbian women, both were included in the current research. That is, this study examined both connection with the lesbian community and femi nist ideology as key predictor variables in the overall objectificat ion theory framework. In conclusion, this study aimed to expa nd upon the existing objectification theory literature by providing a comprehensive examina tion of the framework as it applies to a diverse sample of lesbian women. The simultaneous examination of sexual objectification experien ces, internalization of sociocultu ral standards of beauty, selfobjectification manifested through body surv eillance, body shame, and eating disorder symptomatology builds on prior work by Moradi et al. (2005) and provides a comprehensive test of the objectification theo ry framework as applied to eating disorder

PAGE 47

39 symptomatology. In addition, the present re search included examination of two key constructs, connection with th e lesbian community and femini st ideology, that have been identified in extant literature as importa nt factors that might shape lesbian women’s experiences of eating disorder -related attitudes and behavi ors. The present study included a more comprehensive assessment of conn ection with the lesbian community and feminist ideology than in prior studies of th ese variables, thus allowing for a thorough investigation of their role s in the overall objectifi cation theory framework. Hypotheses Based on the literature on objectificati on theory and extant research on eating disorder-related attitudes and behaviors with lesbian persons, the present study will test a model (See Figure 1) that in cludes the following hypotheses: 1). Reported experiences of sexual objectification will be related positively to body surv eillance, body shame, and eating disorder symptomatology. 2). A negative relationship will exist be tween feminist ideology and disconnection from the lesbian community. 3). Disconnection from the lesbian co mmunity will be related positively to internalization of sociocultural standards of beauty, body surveillance, body shame, and eating disorder symptomatology. 4). Feminist ideology will be related nega tively to internaliza tion of sociocultural standards of beauty, body surveillan ce, body shame, and eating disorder symptomatology. 5). Internalization of sociocultural standa rds of beauty will mediate the links of sexual objectification experiences to body surveillance, body shame, and eating disorder symptomatology (as found in Moradi et al., 2005).

PAGE 48

40 6). Body surveillance will mediate the link of sexual objectification experiences to body shame (as found in Moradi et al., 2005). 7). Body shame will partially mediate the relationship between body surveillance and eating disorder symptomatology as found in previous research (Moradi et al., 2005; Noll & Fredrickson, 1998).

PAGE 49

41 CHAPTER 3 METHODS Participants A total of 616 persons participated in this web-based, survey study (after three duplicate submissions were discarded). Of these 616 participants, 32 persons were excluded because they self-iden tified as a man (n = 2), hetero sexual (n = 2), or bisexual (n = 28), and therefore did not meet the incl usion criteria outlined in the invitation to participate. Additional respondents were excl uded from analyses due to substantial missing data (n = 17), or failure to correctly respond to at l east 5 of 6 validity check items (n = 36). Two women who did not indicate se xual orientation were included in the sample because their responses to questi ons regarding physical attraction, emotional attraction, and sexual behaviors i ndicated that they were mos tly or exclusively attracted to and had sex with women. Thus, the final sa mple used for analyses in the present study included 531 persons between the ages of 18 and 69 ( M = 35.39; Mdn = 34.0; SD = 11.68) who self-identified as either exclusively (73%; n = 385) or mostly (27%; n = 144) lesbian. Participants were from a variety of ge ographic locations with 94% reporting that they were currently living in the United Stat es (n = 498) and 6% re porting that they were currently living in other countries (n = 30) Three participants did not indicate their current residence. As for the 498 participants residing in the United States, reports of states of residence indicated that 27% were living in the South (n = 136), 24% in the West (n = 117), 20% in the Midwest (n = 98) 14% in the Southwest (n = 70), 11% in the

PAGE 50

42 Middle Atlantic (n = 56), and 4% in New Engl and (n = 21) regions of the United States (regional categories derived from www.infoplease.com which divided locations in the United States according to similarities in clim ates, geographies, traditions, and histories). Of the 30 participants living in countries other than the United States, 37% reported living in South Korea, 27% in Canada, 10% in Australia, 10% in India, 3% in the United Kingdom, 3% in Ecuador, 3% in New Zeal and, 3% in Norway, and 3% in Romania. These participants’ correct responses to validity check items indicated that they were able to read and understand the quest ions throughout the survey. With regard to race/ethnicity, 77% of th e sample identified as White/Caucasian, 7% as African American/Black, 5% as Hispan ic/Latina, 3% as Asian American/Pacific Islander, 1% as American Indian/Native Ameri can, and 7% as multi-racial or other. Forty percent of the sample reported being married /partnered, 30% identified as single, 22% reported being in a long-term dating relations hip, and 8% reported being in a casual dating relationship. In terms of highest educ ational degree obtaine d, 33% of the sample had obtained a professional degree (e.g., M.A ., M.S., Ph.D., M.D.), 32% had obtained a college degree (e.g., B.A., B.S.), 30% had co mpleted some college, and the remaining 5% had completed either a high school de gree, some high school, or less than a high school degree. In terms of employment, 64% re ported that they were employed full time, 21% part time, and 15% were not employed. With regard to social class, 54% identified as middle class, 27% as working class, 17% as upper middle class, 2% as lower class, and 1 % as upper class; 3 pa rticipants did not indicate social class. Instruments The web survey was prepared by combining the measures described below into one large questionnaire. The order of instruments for each of the two form s of the survey and

PAGE 51

43 the assignment of participants to each form was determined randomly. Due to the fact that not all measures were designed for lesb ian persons specificall y, two consultants who were lesbian women, and one of whom had e xpertise in multicultural research, reviewed, completed, and provided feedback about the survey. Following recommendations of the consultants, slight grammatical changes were made to some items in order to increase clarity. In addition, one item was adjusted slightly so that it was more appropriate for lesbian women (see descripti on of the SOS scale below). Reported Sexual Objectification Experiences The Sexual Objectification Subscale (SOS ) of Swim, Cohen, and Hyer’s (1998) measure of daily sexist events was used to assess reported experi ences of unwanted and sexual objectifying behaviors and comments. The 7 SOS items reflect the objectification experiences dimension of a larger pool of items developed based on daily diaries in which college men and women reported gender-b ased unfair or differential treatment that they observed or experienced. Because Hill (2002) found that reported experiences of sexual objectification ov er the past year were significan tly related to se lf-objectification, and to make the instructions applicable to pa rticipants who were not students, instructions for SOS items were adjusted to assess experi ences over the past year (instead of past semester as originally used by Swim et al.). Furthermore, to obtain a more thorou gh assessment of sexual objectification experiences, SOS items were supplemented w ith 6 of 18 items from Burnett’s (1995) Objectification Experiences Questionnaire (OEQ ). These six items were chosen because they tapped distinct experi ences of sexual objectification from those assessed by SOS items. Thus, participants reported their experi ences of a total of 13 sexual objectification

PAGE 52

44 events using a 5-point rating sc ale, ranging from 1 (never) to 5 (about two or more times a week over the past year). Because Hill (2002) found that experience s of sexual objectification perpetrated by both men and women over the course of a year were related positively to selfobjectification, SOS and OEQ item instructions were adjusted so that participants responded to each of the 13 items twice, on ce considering their experiences with men, and again considering their experiences with women. Sample items from the SOS include: “Had people shout sexist comments, whistle, or make catcalls at me” and “Experienced unwanted staring or ogling at myself or part s of my body when the person knew or should have known I was not interested or it was inappropriate for the situation or our relationship.” As recommended by the two consultants, the item “Had someone refer to me with a demeaning or degradi ng label specific to my gender (bitch, chick, bastard, faggot, etc) was adjust ed to include the word “dyk e” as additional example of a derogatory term. Sample items from the OEQ include: “Someone stared at your breasts while talking to you” and “S omeone made offensive, sexualized gestures toward you (e.g., pantomime of masturbation or intercourse)?” In order to examine the appropriatene ss of averaging across SOS and OEQ item ratings to compute an overall reported sexual objectification score, a principal components factor analysis was conducted w ith the 13 sexual objectif ication experiences perpetrated by women items (both SOS and OE Q). The scree plot and factor loadings suggested that all items loaded substantia lly on a single factor (item loadings: .75, .74, .74, .74, .68, .67, .66, .66, .65, .65, .61, .54, .44). Similarl y, a principal components factor analysis was conducted on the 13 sexual obj ectification experiences perpetrated by men

PAGE 53

45 items (both SOS and OEQ). Again, results of th e scree plot and factor loadings suggested that items loaded substantially on a si ngle factor (item loadings: .84, .83, .81, .80, .79, .77, .75, .71, .68, .67, .63, .56, .48). Furthermore, to examine whether sexual objectification experiences by men and women should be exam ined separately or could be combined, an additional principal compone nts factor analysis was conducted with all 26 sexual objectification expe riences items (i.e., SOS a nd OEQ items for both women and men). Factor loadings and a scree plot indicated that a ll items loaded substantially on a single factor (item loadings: .77, .75, .75, .74, .73, .72, .71, .69, .69, .67, .67, .66, .62, .61, .61, .61, .60, .59, .58, .56, .55, .55, .54, .50, .44, .41). Furthermore, scores on sexual objectification experiences pe rpetrated by women and sexual objectification experiences perpetrated by men were significantly correlated with each other (r = .62, p < .001) and suggested substantial overlap in these scores. Thus, SOS and OEQ items regarding sexual objectification experiences pe rpetrated by both men and wome n were combined into one scale, and an overall reported sexual objectification experiences mean score was computed for each participant. Moradi et al. (2005) reported an alpha in ternal consistency reliability of .87 for SOS scores. In terms of validity, Swim et al. (2001) found that women reported more sexual objectification ex periences than men, and that thes e along with other sexist events were related more strongly to anxiety for wo men than for men. However, these reported experiences of sexual objectif ication and other sexist ev ents were not related to neuroticism (discriminant validity). With regard to OEQ scores, Burnett (1995) reported alphas that ranged from .69 to .91 and test-retest reliability that ranged fr om .69 to .88 across a period of two to five

PAGE 54

46 weeks. By conducting a factor analysis, Bu rnett (1995) found evidence for discriminant validity in that sexual objectification experien ces were distinct from gender harassment and sexual abuse/coercion experiences. Scor es on the OEQ also demonstrated convergent validity, as they were moderately to strongl y related to depressi on (Burnett, 1995), selfobjectification, body image disturbances, and di sordered eating scores (Brownlow, 1997). When the combination of SOS and OEQ items was examined, alpha internal consistency reliabilities obtained for the current sample were .93 for sexual objectification experiences pe rpetrated by men, .90 for sexua l objectification experiences perpetrated by women, and .94 for overall repo rted sexual objectif ication experiences (used in the present study). Connection/Disconnection with Lesbian Community The Connection with the Lesbian Comm unity subscale (CLC) of the Lesbian Internalized Homophobia Scale (LIHS; Szym anski & Chung, 2001) was used to assess connection with the lesbian community. The CL C subscale is appropriate for use in the present study in that it assesses the extent to which a lesbian woman is connected/disconnected from the larger lesbian community, and responses can demonstrate isolation from the community, or social embeddedness in the community. The CLC consists of 13 items to which partic ipants responded using a 7-point Likert-type scale ranging from “strongly agree” to “strongly disagree.” Higher scor es indicate greater disconnection from the lesbian community. Samp le statements from the Connection with the Lesbian Community subscale include: “Att ending lesbian events and organizations is important to me” and “I am familiar with lesbian music festivals and conferences.” Scores are computed by reverse coding the a ppropriate items and then obtaining a mean score.

PAGE 55

47 In terms of reliability, the reported al pha for Connection/disconnection with the Lesbian Community scores was .87 (Szymanski & Chung, 2001). With regards to validity, Szymanski and Chung (2001) administered the LIHS to 303 female participants, and findings indicated that as expected, disconnection fr om the lesbian community correlated negatively with self-esteem (r = -.22, p < .01) and positively with loneliness (r = .38, p < .01). Also supporting the validity of CLC scores, responses to the present study’s demographic question “How connect ed or involved are you in the lesbian community?” were correlated significantly wi th CLC scores (r = -.67; p < .001). Alpha for CLC scores with the current sample was .85. Feminist Ideology The Attitudes Toward Feminism and the Women’s Movement (FWM) Scale (Fassinger, 1994) was used to assess femini st ideology. The FWM is a brief, 10-item questionnaire designed to measure agr eement with feminism and the women’s movement. Participants were asked to res pond to each of the 10 statements using a 5point rating scale from 1 (strongly disagree) to 5 (strongly agree). Sample items include: “Feminist principles should be adopted ev erywhere” and “The leaders of the women’s movement may be extreme, but they have th e right idea.” Scores are obtained by reverse coding appropriate items and then obtaining a mean, with higher scores indicating greater endorsement of feminist ideology. To examine reliability and validity, Fassi nger (1994) administered the FWM to 117 (76 women and 41 men) undergraduate psycho logy students at a large eastern public university. FWM score reliabilities were .90 for men, .87 for women, and .89 for the entire sample. As for validity, FWM scores demonstrated adequate convergent and discriminant validity. For example, Fassi nger (1994) found that FWM scores were

PAGE 56

48 independent from scores on meas ures of gender roles (r = .02, p > .05) but correlated strongly and positively with scores on four ot her feminism scales (correlations ranging from .68 to .79) and negatively with dogmatism (r = -.23, p < .05). Other evidence for validity of FWM scores exis ts in Enns’s (1987) study, in which convergent validity coefficients ranged from .36 (for involvement in activities associated with feminism) to .62 (for subjective identificati on with feminism). Enns also found discriminant validity coefficients of .23 (for ge nder roles) and -.24 (for dog matism). In the current study, responses to the demographic question aski ng “To what extent do you describe yourself as a feminist?” were significantly correlated with FWM scores (r = .71, p < .001), adding further evidence of validity for FWM scores. The alpha inte rnal consistency reliability estimate for FWM scores with the current sample was .85. Internalization of Sociocultural Standards of Beauty The Internalization subscale of the Soci ocultural Attitudes Toward Appearance Questionnaire (SATAQ; Heinberg, Thompson, & Stormer, 1995) is an 8-item, 5point Likert-type scale that assesses acceptance of sociocultural standards of beauty. Sample items include: “Women who appear in TV shows and movies project the type of appearance that I see as my goal” and “Photogr aphs of thin women ma ke me wish that I were thin.” Participants responded on a scal e ranging from 1 (compl etely disagree) to 5 (completely agree). Appropriate items were reverse coded and item ratings were then averaged with higher scores on the Internalization subscale demonstrating greater levels of internalization of socioc ultural beauty standards. In Heinberg et al.’s (1995) study of 194 female undergraduate students, Internalization scores yielded an alpha of .88. In Morry and Staska’s (2001) study, scores obtained from a sample of 89 women and 61 me n produced an alpha of .85. With regards

PAGE 57

49 to validity, scores on the Inte rnalization subscale have been shown to be distinct from scores generated from awareness of sociocu ltural standards of beauty, but significantly and positively related to disordered eating at titudes (Griffiths, Beum ont, Russell, Schotte, Thornton, Touyz, & Varano, 1999), restrained ea ting and body dissatisfaction (Griffiths, Mallia-Blanco, Boesenberg, Ellis, Fi scher, Taylor, & Wyndham, 2000), and preoccupation with body image (Morry & Stas ka, 2001). Alpha with the current sample was .87. Body Surveillance as an Indicato r of Sexual Objectification The Body Surveillance subscal e of the Objectified Body Consciousness Scale (OBC; McKinley & Hyde, 1996) wa s utilized to assess self-o bjectification, or concern with outward appearance as opposed to concern regarding how the body feels. Participants responded to eight items using a 7-point Likert-type scale, ranging from 1 (strongly disagree) to 7 (strongly agree), but NA could be selected if the item was not applicable. Following scoring instructions, appropriate items were reverse coded and “NA” responses were coded as missing. Non-mi ssing item ratings were then averaged in order to yield a scale score, with higher scores indicating higher levels of body surveillance. The Body Surveillance subscale in cludes items such as “I often worry about whether the clothes I am wearing make me look good” and “During the day, I think about how I look many times.” Scores on the Body Surveillance subscale ha ve demonstrated adequate reliability. In their sample of both young and middleaged women, McKinley and Hyde (1996) found alpha internal consistency reliab ility estimates ranging from .76 to .89. Furthermore, a two-week test-retest was c onducted, producing a reliability coefficient of .79. In their sample of over 200 women, Moradi et al. (2005) reported an alpha of .82. In

PAGE 58

50 terms of validity, consistent w ith objectification theory, wo men have been found to score higher than men on Body Surveillance (M cKinley, 1998). In addi tion, in a factor analysis, Body Surveillance emerged as a constr uct that was distinct from other factors such as body shame, although as expecte d, it was correlated pos itively with body shame and negatively with body esteem (McKinley, 199 8). Alpha internal c onsistency reliability estimate with the current sample was .86. Body Shame Body shame was measured using the B ody Shame Subscale of the Objectified Body Consciousness Scale (OBC; McKinley & Hyde, 1996). This 8-item subscale assesses guilt and negative feelings as a re sult of failing to live up to perceived cultural standards. For example, one item reads: “Whe n I can’t control my weight, I feel like something must be wrong with me.” Particip ants responded using a 7-point Likert-type scale, ranging from 1 (strongly disagree ) to 7 (strongly agree), but NA could be selected if the item was not applicable. Following sc oring instructions, a ppropriate items were reversed coded and “NA” responses were c oded as missing. Non-missing items were then averaged to produce a scale score in which higher scores indicate higher levels of body shame. OBC Body Shame subscale scores have dem onstrated adequate reliability. Across a sample of young and middle-aged women, alpha internal consistency estimates ranged from .70 to .84 (McKinley & Hyde, 1996; Mc Kinley, 1998). Moradi et al. (2005) found an alpha estimate of .81. In te rms of validity, as expected, Body Shame scores have been shown to be correlated with but have emerged as a distinct factor from control beliefs and body surveillance (McKinley & Hyde, 1996). McKi nley (1998) found that scores on the

PAGE 59

51 Body Shame subscale correlated negatively wi th body esteem and positively with body surveillance. Alpha with th e current sample was .87. Eating Disorder Symptomatology The Eating Attitudes Test-26 (EAT-26; Garner, 1997; Garner, Olmsted, Bohr, & Garfinkel, 1982) consists of 26 questions desi gned to assess dieting behaviors, bulimia and food preoccupation, and oral control. The EA T26 was selected because of its ability to assess the continuum of disturbed eating attitudes and behaviors. This follows the recommendation of Kashubeck-West, Mintz, and Saunders (2001) to examine the broad range of eating attitudes and behaviors. Part icipants indicated the frequency of such attitudes and behaviors using a 6-point, Likert-type scale, ranging from 1 (never) to 6 (always). As recommended by Garner (1997), ra tings of each of the items were then weighted from zero to three, with a score of three signifying the most “symptomatic” responses. Total EAT scores were then obt ained by summing all of the weighted item scores. The minimum score that can be obtai ned on the EAT-26 is 0, and the maximum is 78, with higher scores indicating greate r eating disorder symptomatology. Sample statements include: “Engage in dieting beha vior,” “Feel extremely guilty about eating,” “Find myself preoccupied with food,” “Cut my food into small pieces,” “Have the impulse to vomit after meals,” and “Avoid eating when I am hungry.” Scores on the EAT-26 have demonstrated good reliability an d validity, yielding Cronbach’s alphas ranging from .90 to .93 (Share & Mintz, 2002). Kashubeck-West et al. (2001) reported alphas ranging from .79 to .94 across samples. In terms of validity, scores on the EAT-26 are related to scores on other measures of eating disorder symptomotology and the EAT-26 has been used to distinguish between clinical and non-

PAGE 60

52 clinical groups (Kashubeck-West et al., 2001) Alpha for EAT-26 scores in the current sample was .86. Demographics Participants were asked to report several personal characteristics including age, height and weight (used to compute body mass index), race/ethnicity, relationship status, employment status, income, educational level, and social class. Sexual orientation was assessed utilizing a Kinsey-type scale rangi ng from 1 (exclusively lesbian) to 5 (exclusively heterosexual). To provide a more through assess ment of sexual orientation, participants were also asked to rate on a scale of 1 (low) to 5 (high) their physical and emotional attraction to both members of the same and opposite sex. Th ey were also asked a question regarding sexual behavior in which they indicated whether their sexual interactions were with thei r same gender only, same gender mostly, both genders equally, other gender mostly, or other gender only. They could also select “never had sex” if appropriate. The demographic questionnaire also included an item assessing degree of connection with the lesbian community. The it em read: “How connected or involved are you in the lesbian community?” Participants responded using a 5-point Likert scale ranging from 1 (not involved) to 5 (very i nvolved). A single item was included to assess feminist self-identification. Participants re sponded to the question “To what extent do you describe yourself as a femini st?” by using a 9-point Likert scale ranging from 0 (not at all) to 9 (very much a feminist), as us ed by Cogan (1999). These items were used as further validity checks for CLC and FWM scores.

PAGE 61

53 Procedure Participants were recruited through pers onal contacts, lesbian and/or women’s organizations and internet listserves. Adve rtisements for the study were sent to 116 listserves that were selected due to their ability to reach lesbian women throughout the United States. These listserves included na tional organizations such as the Gay and Lesbian National Hotline, the National Coal ition for LGBT health, and the Gay, Lesbian, Straight Education Network (GLSEN). In a ddition, advertisements were sent out through listserves of numerous lesbian and gay colle ge organizations. Part icipants also were recruited by sending email advertisements about the study to Yahoo and MSN groups for lesbian and gay persons. List serves that were not used for promotion of the study included those that were designated fo r men, heterosexual or bisexual women, nonEnglish speaking women, and lesbian women und er the age of 18. Flyers promoting the study also were posted in the Gainesville, Florida community to promote the study. These advertisements described th e purpose of the research, a nd highlighted the need for diversity in terms of age, race/ethnicity, and level of “outness” in the sample. The principal investigator attended club meetings organizations, events, and other activities targeted for the lesbian community to further advertise the study. Participation involved completing a series of questionnaires online via a website designed by the researcher. Although internet methodology can limit participation to individuals who have access to a computer and the internet, it has a number of benefits for recruiting large samples of lesbian or gay participants. More specifically, in the present study, a web-based survey was select ed because it provides greater anonymity for lesbian participants. Maximizing anonymity and reducing interpersonal threat is particularly important for recruiting indivi duals who are not “out ” about their sexual

PAGE 62

54 orientation and might not feel comfortable with participa ting in-person in a study about lesbian or gay persons. Furthermore, the convenience of completing a survey online (participants could log on to the website from any computer with an internet connection) facilitated the recruitment of a larger and more diverse sample in terms of age and geographic location. When participants connected to th e website, they were first shown an informed consent page that described the purpose of the study, discussed issues of confidentiality, informed respondents that they could stop f illing out the survey at any time without penalty, and provided contact in formation to participants who had questions or comments about the study. After reading the informed consent information, participants clicked a link stating: “Click here to proceed” to indicate that they had read and understood the informed consent document. Clicking the button at the bottom of the screen brought participants to one of two fo rms of the survey to be completed (the measures were counterbalanced into two different forms in order to control for order effects). Throughout the survey, participants encountered six validity check items that asked them to choose a certain response. For exam ple, one item read: “Please select Strongly Agree .” This procedure was used to help identif y random responding and to provide some indication that partic ipants read and unde rstood the questions. Upon completing the survey, participants received information abou t how to contact the researcher if they had any questions or concerns. They also receiv ed information for national support networks for eating disorders and lesbian women. Statistical Analyses As mentioned previously, data from thos e who identified as a man, bisexual, or heterosexual, as well as those who did not res pond to substantial por tions of the survey,

PAGE 63

55 were excluded from analyses, resulting in a final sample size of 531 for the present analyses. Body Mass Index was computed by calculating weight/height2 and was entered as a covariate in analyses. This is consiste nt with previous res earch, as BMI has been considered a potential confounding variable (Fredrickson, Roberts, Noll, Quinn, & Twenge, 1998). Reliability coefficients were calculated (and reported in the Instruments section) for all measures before testing the overall path model. Partial-correla tions, controlling for BMI, were computed to examine the interc orrelations among the variables of interest. Finally, all direct and indirect relations among variables of in terest, depicted in Figure 1, were examined by conducting a path analys is using AMOS, Version 5.0 (Arbuckle, 2003), a program designed specifically for structural equation modeling. The path analysis provided the strengths of all unique links among the variable s of interest, with BMI controlled. To test proposed mediator effects, guidelines outlined by Baron and Kenny were followed (1986). According to Baron and Ke nny, a mediator functions as such if it accounts for all or some of the relationshi p between the predictor variable and the criterion variable. More specifically, for a variable to be considered a mediator, significant relationships must exist between (a) the predictor and the mediator, (b) the mediator and the criterion, and (c) the predictor and criteri on. If these preconditions are satisfied, mediation is significant if the medi ator accounts for a significant amount of the predictor-criterion relationship. To test for the significance of mediations, Sobel’s formula (1982) was used (Baron & Ke nny, 1986; Frazier, Tix, & Barron, 2004).

PAGE 64

56 CHAPTER 4 RESULTS Descriptive Statistics The mean body mass index for the present sample was 29.04 (SD = 7.96) which was comparable to Heffernan’s (1996) obtai ned mean of 26.98 (SD = 6.82) in a sample of 203 lesbian women. Levels of sexual objectif ication experiences, feminist ideology, disconnection from the lesbian community, inte rnalization of socioc ultural standards of beauty, body surveillance, body shame, and eating disorder symptomatology for the current sample were generally close to the mi d range of possible scores for each measure (see Table 1). Furthermore, the present sample s’ scores on variable s of interest were comparable to those from studies that used the same instruments with similar samples of women. More specifically, th e current sample’s means and standard deviations for feminist ideology ( M = 3.92, SD = .57), disconnection from the lesbian community ( M = 2.45, SD = .94), internalization ( M = 2.13, SD = .90), body surveillance ( M = 4.01, SD = 1.27), body shame ( M = 3.12, SD = 1.38), and eating diso rder symptomatology ( M = 7.54, SD = 8.37) were comparable to those repo rted by Fassinger (1994) for feminist ideology scores in a sample of 117 female and male college students ( M = 3.52, SD = .66), Szymanski and Chung (2003) for disconnec tion from the lesbian community scores in sample of 210 lesbian and bisexual wo men (M = 2.36, SD = .91), Bergeron and Senn (1998) for internalization scores in a sample of 108 lesbian women ( M = 2.17, SD = .69), Hill (2001) for body surveillance scores in a sample of 134 lesbian women ( M = 4.12, SD = 1.19), McKinley and Hyde (1996) for body shame scores in a sample of 108

PAGE 65

57 undergraduate women ( M = 3.25, SD = 1.04), and Strong et al. (2001) for eating disorder symptoms in a sample of 82 lesbian women ( M = 7.74 SD = 9.59). Because the sexual objectification experiences measure used in the present study was a combination of previously used scales, the present samples’ scores on sexual objectification experiences ( M = 1.72, SD = .63) could not be compared to that of prior samples. To test for order effects across the tw o orders of the survey, a MANOVA was conducted with survey order as the independent variable and th e variables of interest (i.e., BMI, sexual objectification experiences, fe minist ideology, disconnection from the lesbian community, internalization of sociocultural standards of beauty, body surveillance, body shame, a nd eating disorder symptoma tology) as the dependent variables. A small, but signifi cant overall effect was found ( F [1, 530] = 1.99, p < .05, p 2 = .03) but the only significan t univariate effect was a difference between the two groups on reported experiences of sexual objectification ( F [1, 530] = 4.10, p < .05, p 2 = .008). Given that survey order accounted fo r less than 1% of variance in sexual objectification scores and did not result in a significant difference in any of the other seven variables of interest, survey or der was not deemed to be problematic. Intercorrelations among Variables of Interest Partial correlations, controlling for BMI, we re computed to test the relations among variables of interest and ev aluate whether pre-conditions for mediation were met (see Table 1). Consistent with H ypothesis 1, after controlling fo r BMI, reported experiences of sexual objectification we re correlated positively with body surveillance (r = .22, p < .001), body shame (r = .19, p < .001), and eating disorder symptomatology (r = .17, p < .001). In addition, consistent with Hypothesis 2, after controlling for BMI, a negative

PAGE 66

58 correlation existed between feminist ideo logy and disconnection from the lesbian community (r = -.33, p < .001). After controlling for BMI, Hypothesis 3 also was supported, as indicated by significant positive correlations between disconnection with the lesbian community and internalization of sociocultural standard s of beauty (r = .14, p = .001), body surveillance (r = .14, p = .001), body shame (r = .17, p < .001), and eating disorder symptoms (r = .09, p < .05). Hypothesis 4 was only partially supported; with BMI controlled, feminist ideology was not si gnificantly correlated to internalization, body shame, and eating disorder symptoms but was significantly and negatively correlated to body su rveillance (r = -.10, p < .05). Mediations To test the mediations proposed in Hypotheses 5, 6, and 7, Baron and Kenny’s (1986) procedures were followed. According to these authors, for a variable to be considered as a mediator, si gnificant relations must exist between (a) the predictor and the mediator, (b) the mediator and the criter ion, and (c) the predic tor and criterion. These preconditions were satisfied for Hypothe ses 5, 6, and 7 (see Table 1 for partialcorrelations). That is, for Hypothesis 5, re ported experiences of sexual objectification (predictor) were correlated si gnificantly with internalization (potential mediator), which in turn was correlated significantly with body surveillance, body shame, and eating disorder symptomatology (crite rion variables). In addition, reported sexual objectification experiences (predictor) were correlated significantly with body surveillance, body shame, and eating disorder symptomatology (criterion variables). With rega rd to Hypothesis 6, reported sexual objectification experiences (predictor) were significantly related to body surveillance (potential mediator), which in turn was significantly related to body shame (criterion). Reported sexual objec tification experiences (predict or) were also significantly

PAGE 67

59 related to body shame (crite rion). For Hypothesis 7, body su rveillance (predictor) was correlated significantly with body shame ( potential mediator), and body shame was correlated significantly with eating diso rder symptomatology (criterion). Body surveillance (predictor) was also signif icantly correlated with eating disorder symptomatology (criterion). According to Baron and Kenny (1986), if th ese conditions are satisfied, a variable acts as a mediator to the exte nt that it accounts for the rela tionship between the predictor and criterion variable(s). In order to test the signifi cance of mediations, Amos 5.0 (Arbuckle, 2003) was used to conduct a path analysis of a fu lly saturated model in which all direct and indirect paths were estimated (see the model presented in Figure 1). Again, body mass index was entered as a covari ate in the model. Maximum likelihood estimation was utilized with the covariance matr ix of the variables of interest as input. Given that the model tested was fully satu rated, values for the Goodness of Fit Index (GFI), Incremental Fit Index (IFI), Compara tive Fit Index (CFI), and the Normed Fit Index (NFI) were all 1.0. The overall mode l account for 45% of the variance in body shame, 38% of the variance in eating disord er symptoms, 32% of the variance in body surveillance, and 5% of the variance in inte rnalization of socioc ultural standards of beauty. As indicated in Figure 2, most standa rdized path coefficients were significant, indicating significant unique di rect links. Significant unique direct links did not emerge, however, from feminist ideology to interna lization, body surveillance, and eating disorder symptomatology; from disconnection from the lesbian community to body surveillance and eating disorder symptomatology; a nd from reported sexual objectification experiences to body shame and eating disord er symtomatology. The only unique link that

PAGE 68

60 was in the unexpected direction wa s the significant, albeit small (.07, p < .05) positive unique link between feminist ideology and body shame (with BMI and other exogenous variables controlled). To test the significance of mediations through internal ization of sociocultural beauty standards, body surveillance, a nd body shame (the proposed mediators), appropriate standardized path coefficients we re multiplied to compute indirect effects, a procedure recommended by Cohen and Cohen (1983). Next, Sobel’s formula (1982) was used to determine whether or not the indirect effects were significantly different from zero. Hypothesis 5 proposed that internaliza tion of sociocultural standards of beauty would mediate the links of sexual objectification experien ces to body surveillance, body shame, and eating disorder symptomatology (a s found in Moradi et al., 2005). Consistent with this hypothesis, through in ternalization of soci ocultural standards of beauty, reported experiences of sexual objectif ication had a significant in direct link of .08 (.15 x .52; z = 3.41; p < .001) to body surveillance, .05 (.15 x .31; z = 3.21, p < .01) to body shame, and .02 (.15 x .14; z = 2.31, p < .05) to eating disorder sympto matology. In addition to these significant indirect relations, reported experiences of sexual objectification also had a significant direct link to body surveillance, but not to body shame or eating disorder symptoms. Therefore, consistent with Hypot hesis 5, internalization of sociocultural beauty standards partially mediated the li nk between reported se xual objectification experiences and body surveillance, and fully mediated the links of reported sexual objectification experiences to body shame and eating disorder symptomatology. Hypothesis 6 proposed that body surveill ance would mediate the link of sexual objectification experiences to body shame (as found in Moradi et al., 2005). Consistent

PAGE 69

61 with this hypothesis, through body surveillance, a significant indirect link of .05 (.13 x .38; z = 3.31, p < .001) was obtained between reported sexual objectificat ion experiences and body shame. Because no significant dir ect link existed between reported sexual objectification experiences and body shame, body surveillance act ed as a full mediator of this link, supporting Hypothesis 6. Hypothesis 7 proposed that body shame woul d partially mediate the relationship between body surveillance and eating disorder symptomatology as found in previous research (Moradi et al., 2005; Noll & Fredrick son, 1998). Consistent with this hypothesis, through body shame, a significant indirect link of .13 (.38 x .34; z = 5.84, p < .001) was found between body surveillance and eating disorder symptomatology. Because there was also a direct relationship betw een body surveillance and eating disorder symptomatology, body shame acted as a par tial mediator. Thus, Hypothesis 7 was supported. Finally, the fully saturated model was comp ared to an alternative trimmed model that eliminated the non-significant direct paths (a) from feminist ideology to internalization, body surveillance, and eating disorder symptoms, (b) from disconnection from the lesbian community to body survei llance and eating disorder symptomatology, and (c) from reported sexual objectificati on experiences to body shame and eating disorder symptoms. The goodness of fit indices for this mode l were above the acceptable cut offs and nearly identical to those obtaine d from the original model (GFI = .99; IFI = .99; CFI = .99; NFI = .99). The trimmed model explained 45% of the variance in body shame, 38% of the variance in eating disord er symptomatology, 31% of the variance in body surveillance, and 5% of the variance in in ternalization of socioc ultural standards of

PAGE 70

62 beauty. The variance accounted for by the trimme d model was the same as that accounted for in the fully saturated model, except for variance accounted fo r in body surveillance, which dropped from 32% to 31%. In addition, the magnitude of path coefficients was comparable across the two models. Thus, comp ared with the fully saturated model, the trimmed model appears to be more parsimoni ous but equally appropr iate in explaining the relationships among the variables of interest.

PAGE 71

63 CHAPTER 5 DISCUSSION Objectification theory (Fredrickson & Robe rts, 1997) and empirical investigations of its tenets point to the importance of sexual objectificati on experiences, selfobjectification (manifested as body surveilla nce), and body shame as predictors of women’s eating disorder symp tomatology. In addition, Moradi et al.’s (2005) findings highlighted the role of internalization of sociocultural standards of beauty in the objectification theory framework. The pres ent study was the firs t to examine the objectification theory framework and the additional role of internalization of sociocultural standards of beauty in eating disorder symptomatology with a large sample of lesbian women. Furthermore, in the c ontext of objectification theory, this study examined the roles of feminist ideology and connection/disconnection from the lesbian community, each of which has been identified as a key predictor of eating disorder symptoms among lesbian persons. Overall, findings of the present study suggested that relations outlined in objectification theory and s upported in prior tests of obj ectification theory with heterosexual women are also supported w ith lesbian women. More specifically, consistent with objectificati on theory’s conceptualizati on of the role of sexual objectification experiences in eating disorder symptoms and their precursors, in the present study, partial co rrelations (with BMI controlled) indicated that reported sexual objectification experiences were significantl y and positively related to body surveillance, body shame, and eating disorder symptomato logy. Furthermore, results of the path

PAGE 72

64 analysis suggested that sexual objectification experiences were related positively and uniquely to internalization of sociocultural beauty standa rds and body surveillance when feminist ideology, disconnection from the lesbian community, and BMI were accounted for. In addition, internalization of sociocultura l standards of beauty partially mediated the links of sexual objectification experiences to body surveillance, body shame, and eating disorder symptomatology, suggesting that th rough internalization, reported experiences of sexual objectification might be translat ed into body surveillance, body shame, and eating disorder symptoms. These findings are c onsistent with previous research (Moradi et al., 2005; Morry & Staska, 2001) with hete rosexual women and further highlight the importance of including both se xual objectification experiences and internalization of sociocultural standards of beauty when inves tigating the tenets of objectification theory. Also consistent with prior tests of objec tification theory, (e.g., Moradi et al., 2005), the current findings supported body surveilla nce as a full mediator of the relation between sexual objectification experiences and body shame. Thus, it appears that body surveillance might be a key mechanism thr ough which sexually objectifying experiences are translated into body shame. In addition, body shame acted as a partial mediator in the link of body surveillance to eating disorder symptomatology, supporting the findings of extant literature (Greenle af, 2005; Moradi et al., 2005, Noll & Fredrickson, 1998) and suggesting that chronic body monitoring might be translated into eating disorder symptomatology by promoting body shame. Ta ken together, these findings support the generalizeability and potential utility of the objectification theory framework as applied to understanding eating disorder sympto matology among lesbian women. Indeed, the

PAGE 73

65 overall path model examined in the present study explained 38% of the variance of eating disorder symptoms in this sample of 531 lesbian women. In addition to extending research on object ification theory to lesbian women, the present study integrated the pr eviously highlighted potential roles of feminist ideology and connection/disconnection from the lesb ian community in its examination of objectification theory. Consistent with previous findings (Syzmanski & Chung, 2003), present results indicated that feminist ideology and disconnection from the lesbian community were significantly and negativel y correlated with one another (with BMI controlled). Furthermore, as hypothesized, di sconnection from the lesbian community was significantly and positively related to internalization of sociocultural standards of beauty, body surveillance, body shame, and eating disorder symp tomatology. In other words, disconnection from the lesbian commun ity was correlated with greater levels of eating disorder symptoms and their correlate s for lesbian women. This finding supports extant literature in which interaction and i nvolvement with other lesbian women has been found to be related to positive health out comes such as higher body esteem, higher body image, lower weight concern, a nd lower levels of ea ting disorder symptoms (Beren et al., 1997; Heffernan, 1996; Ludwig & Brownell, 1999). In terms of feminist ideology as a predic tor variable, partial correlations (with BMI controlled) revealed only one significant, but negative relationship between feminist ideology and body surveillance. Thus, it seems that endorsi ng feminist ideology may be related to lower levels of body monitoring. Interestingly, a partial correlation (with BMI controlled) indicated no re lationship between feminist ideology and body shame; however, when feminist ideology was considered in the context of disconnection from

PAGE 74

66 the lesbian community via the path analysis, feminist ideology was significantly and positively related to body shame. Although this relationship was very small (.07; p < .05), it may be that the positive health outcomes a ssociated with feminist ideology as found in previous literature (e.g., Leavy & Adams, 1986) are accounted for largely by overlap with connection with the lesbian community. Indeed, the present re sults indicated that compared with feminist ideology, disconn ection from the lesbian community was a stronger and more consistent correlate of eating disorder constructs included in the model, as evidenced by the patt ern of partial correlations (w ith BMI controlled) and path coefficients between disconnection with the lesbian community versus feminist ideology and the four outcome variables. The small but positive relation that em erged between feminist ideology and body shame, when connection/disconnection from the lesbian community, sexual objectification experiences, and BM I were accounted for, might be interpreted in light of a study conducted by Eliason and Morgan (1998), in which lesbian women were asked to define what it meant to be a lesbian. Those who gave political definitions (e.g., “woman identification,” “affiliation with other oppres sed groups”) were more likely to identify themselves as feminist and get involved in political activities promoting lesbian women, but at the same time, were si gnificantly more likely to have a history of eating disorders than those who gave non-politic al answers (e.g., “sex/love wi th women,” “lesbianism as one small aspect of a person”). Interpreted in light of the present findings and prior research on feminist ideology and connectio n/disconnection from the lesbian community, Eliason and Morgan’s findings suggest an interesting possibility. When considered separately, connection with th e lesbian community and femini st ideology each are related

PAGE 75

67 to lower levels of eating disorder-related variables. However, lesbian women who have an integrated lesbian and feminist political identity may have experienced greater levels of eating disorder symptoms in the past a nd possibly also experi ence greater body shame. It may be that under some conditions, eating disorder symptoms become so detrimental that some lesbian women turn to the reje ction of society’s beauty standards through increasing their endorsement of feminist principles and by becoming more connected with the lesbian community. This causal hypothesis is sp eculative, and longitudinal research is needed to ascertain the direction of relations between connection/disconnection with the lesbian community, femi nist ideology, and eating disorder symtomatology. Limitations The present findings must be interpreted in light of a number of limitations. First, the present data were collected using an onlin e, self-report survey in which participation was voluntary. As such, the stigma associated with eating disorder-r elated attitudes and behaviors may have limited participation fr om those experiencing high levels of such symptoms. Although scores from this sample are similar to those obtained in other nonclinical samples of lesbian women, the ge neralizability of the present findings are limited to nonclinical lesbian populations. To ad dress this limitation, future research is needed to test the tenets of objectification theory with samples of lesbian women who are experiencing clinically si gnificant levels of eating disorder symptomatology. Another potential limitation of the present study (that is shared by any study that explicitly recruits lesbian and gay participants) is that the majority of participants might be “out” about their sexual orientation a nd so at least somewhat comfortable with participating in a study about lesbian persons. Because “ outness” is an individual

PAGE 76

68 difference variable that may inhibit some individuals from participating in a study advertised for lesbian women, the present sa mple may not capture the experiences of lesbian women who are less “out” about their se xual orientation. As such, the participants in the present study may have been more connected to the lesbian community. Because current findings indicated that disconnection from the lesbian community was related to internalization of sociocultural standards of beauty, body surveillance, body shame, and eating disorder symptomatology, it may be that lesbian women who are less out and more disconnected from the lesbian community might be at greater risk for eating disorder attitudes and behaviors. Th e current sample obtained sc ores that were around the midrange for connection/disconnection with the lesbian community. Therefore, future studies should attend to the experiences of lesbian women who are less “out” and more disconnected from the lesbian community than those in the current sample. A third potential limitation is that even though the survey incl uded validity check items to ensure that participants understood th e survey instructions and items, there was no way to ensure that the respondents actually met the participation requirements. This limitation is inherent in all volunteer, self-repo rt studies (online or in person) and for the present study; however, the bene fits of the online survey (e.g. facilitating participation of less out participants, larger sample size) outweighed the costs. With such a large sample size, it is unlikely that any submissions fr om those who failed to meet participation criteria would substantially skew results, but findings should s till be interpre ted with this limitation in mind. A fourth potential limitation is that al though the present sample was diverse in terms of age and geographical location, res pondents were predominantly Caucasian and

PAGE 77

69 well-educated. This imbalance, which has o ccurred within other lesbian samples (e.g., Bergeron & Senn, 1998; Hill, 2002), limits the ge neralizability of findings to Caucasian lesbian women with at least a college edu cation. To address this limitation, future research is needed to develop and evaluate recruitment strategies designed to increase representation of racial/ethnic minority le sbian women as well as lower and working class lesbian women. A final potential limitation, which is shared in much of the eating disorder research, is that there was no way to ensure that self-r eports of weight and height were accurate. However, prior evidence exists that self-report information for BMI are comparable to actually measured information (e.g., Koslow sky, Scheinberg, & Bleich, 1994; Tienboon, Wahlqvist, & Rutishauser, 1992). In addition, as one participan t addressed in a feedback email, medical conditions that may affect BMI (e.g., diabetes, hypoglycemia) were not assessed. In general, participan ts’ health backgrounds are not assessed in eating disorder research, and future studies could address this potential confo und by including a brief medical history measure in the questionnair e. Despite these lim itations, the present findings can inform research and prac tice in a number of important ways. Directions for Future Research Findings from the current study can be used to inform future research conducted on objectification theory with le sbian persons. More specificall y, findings indicate that it is important to include sexual objectification experiences, inte rnalization of sociocultural standards of beauty, body su rveillance, and body shame as key predictors of eating disorder symptomatology with this population. Furthermore, the present findings suggest that connection/disconnection from the lesbia n community is an important variable to consider in applications of objectification theory to lesbian women. Indeed,

PAGE 78

70 connection/disconnection with the lesbian community might subsume the previously observed links of feminist ideology to eating disorder-related cons tructs. On the other hand, the unique links of feminist ideology to eating disorder-related constructs, when connection/disconnection from the lesbian co mmunity is considered, remains unclear. A critical direction for future research is to further examine the temporal links between lesbian feminist ideology and eating disorder symptomatology. More specifically, longitudinal research is needed to expand upon Eliason and Morgan’s (1998) findings and identify the exact role of le sbian feminist ideology as it relates to involvement in political activ ities and both past and presen t eating disorder symptoms. Furthermore, future research is needed to understand the nature of the relations between feminist ideology and eating disorder-related attitudes and behavior s in the context of connection/disconnection with the lesbian community. The present study focused on feminist ideology as a potential predictor va riable in the model and found that links of feminist ideology to eating disorder constr ucts were generally nonsignificant when BMI and connection/disconnection from the lesb ian community were accounted for. An additional possibility that is wo rth exploring in future resear ch is that feminist ideology might act as a moderator variable, affecting the relations of connection with the lesbian community with internalization of sociocul tural standards of beauty and body shame. That is, for lesbian women who are disconnect ed from the lesbian community, feminist ideology may be related to lower levels of eat ing disorder-related constructs. For lesbian women who are highly connected to the le sbian community, feminist ideology may be related to greater levels of eating disorder -related constructs. Eliason and Morgan’s findings are consistent with such a possibili ty, if past eating disorder symptoms are

PAGE 79

71 considered. Future research is needed that will test the potential interaction effect between connection with the le sbian community and feminist ideology in relation to past and present eating disorder sy mptomatology for lesbian women. Another important area for further inves tigation is to explor e what aspects of lesbian identity and community might be rela ted to lower eating diso rder-related attitudes and behaviors. For example, Kaminski ( 2000) conducted 19 interviews with lesbian women about their identity and health, and found that, compared to those who experienced their environments to be hostile, homophobic, or conservative, participants who perceived their environment to be suppor tive were more likely to adopt feminist principles and were more likely to experience positive health outcomes such as greater self-acceptance and reduced a nxiety, depression, and substance abuse. Thus, it seems that perceived social support, which may be obt ained through connection with the lesbian community, could be an important factor to consider when examining predictors of health-related concerns such as eating disorders among lesbian women. Clearly, future research is needed to tease apart aspects of connection/disconnection from the lesbian community such as feminist self-identifica tion, feminist ideology, a nd social support that might play a role in the promotion of hea lth and reduction of eati ng disorder and other symptomatology for lesbian (and other) women. Implications for Practice In addition to informing future research, findings from the current study can also be used to inform counselors’ and clinicians ’ efforts to reduce eat ing disorder-related attitudes and behaviors among lesbian wo men. Findings from the current study and previous research with heterosexual wome n (e.g., Moradi, 2005) support tenets of the objectification theory framework, and it w ould be beneficial for mental health

PAGE 80

72 professionals to provide appropr iate education regarding obje ctification theory so that both heterosexual women and lesbian wome n can recognize sexua l objectification experiences as they occur. More specifically, findings from th e current study indicate that sexual objectification experiences are related to internalization of sociocultural beauty standards and body surveillan ce, both of which are related to eating disorder symptomatology. These correlational findings provide the groundwork for testing the directions of causality implie d in objectification theory am ong these variables. If such causality is supported, learning to identify sexual objectificat ion experiences when they occur may be beneficial for women because they can then actively work against internalizing societal beauty standards, b ecome aware of daily chronic body monitoring, work towards reducing feelings of shame re garding their bodies, and thus, lower their risk of developing eating disorder symptomatology. Strategies specific to lesbian women might include the development of a workshop in which counselors and clinicians can e ducate lesbian women regarding the potential benefits of connection with the lesbian community, reduction of the internalization of sociocultural standards of beauty, and pr evention of body shame. Together, lesbian women can discuss their perceptions of the lesbian community, identify cultural beauty standards, develop strategies to actively work against endorsi ng such standards, identify goals to reduce body monitoring, and promote positive attitudes toward their bodies. Based on extant literature (e.g., Kaminski, 2000) as well as findings from the present study, such association and inte raction with other lesbian women may lead to positive health outcomes. The present cross-sectio nal findings provide the groundwork for

PAGE 81

73 exploring whether such interven tions would result in reducing the risk for eating disorder symptomatology among lesbian women. As for individual therapy, it may be benefi cial for mental health professionals to adopt treatment strategies for their lesbia n clients that include providing education regarding society’s current stan dards of beauty (specifically those associated with lesbian women and beauty), promoting acceptan ce of one’s physical appearance, and encouraging increased connection with the le sbian community. Future research regarding the exact role of feminist id eology is needed to ascertain whether or not therapy should include the provision of edu cation and support regarding th e development of feminist ideology. It is important to note that these therapeutic strategies have the potential to help heterosexual women as well. Future research is needed that will address the potential benefits of connection with the lesbian community for heterosexual women of all ages and backgrounds. Summary Overall, findings of the present study suppor ted the applicability of the tenets of objectification theory to unde rstanding eating disorder symptomatology of lesbian women. The present findings also pointed to the importance of cons idering the role of connection with the lesbian community when investigating eating disorder symptomatology with lesbian women, but also raised questions about the unique role of feminist ideology beyond connection with th e lesbian community. Finally, the present findings lay the groundwork for examining th e objectification theory framework with more diverse samples of lesbian women, e xploring longitudinal links between sexual objectification experiences, c onnection with the lesbian co mmunity, feminist ideology,

PAGE 82

74 and the predictors of eating di sorder symptoms, and using ex perimental designs to begin to elucidate potential causal relations between variables in the model.

PAGE 83

75 APPENDIX A THE SEXUAL OBJECTIFICATION SUBSCALE Please use the following scale to indicate how often during the pa st year you have experienced each of the events below. For each item, respond once considering your experiences with men and respond again c onsidering your experiences with women. 1. Never 2. About once during the past year 3. About once a month during the past year 4. About once a week during the past year 5. About two or more times per week during past year 1. Had people shout sexist comments, whistle, or make catcalls at me. By men 1 2 3 4 5 By women 1 2 3 4 5 2. Had someone refer to me with a demeaning or degrading label specific to my gender (bitch, chick, dyke, bastard, faggot, etc). By men 1 2 3 4 5 By women 1 2 3 4 5 3 Had sexist comments ma de about parts of my body or clothing. By men 1 2 3 4 5 By women 1 2 3 4 5 4. Had sexist comments ma de about parts of my body or clothing. By men 1 2 3 4 5 By women 1 2 3 4 5 5. Heard someone make comments about sexual behavior I might do or thi ngs they would want to do with me. By men 1 2 3 4 5 By women 1 2 3 4 5 6. Someone did or said something that made me feel threatened sexually. By men 1 2 3 4 5 By women 1 2 3 4 5 7. Experienced unwanted staring or ogling at myself or parts of my body when the person knew or should have known I was not interested or it was inappropriate fo r the situation or our relationship. By men 1 2 3 4 5 By women 1 2 3 4 5

PAGE 84

76 8. Experienced unwanted flirting when the person knew or should have known I was not interested or it was inappropriate fo r the situation or our relationship. By men 1 2 3 4 5 By women 1 2 3 4 5

PAGE 85

77 APPENDIX B OBJECTIFICATION EXPERI ENCES QUESTIONNAIRE Please use the following scale to indicate how often during the pa st year you have experienced each of the events below. For each item, respond once considering your experiences with men and respond again c onsidering your experiences with women. 1 Never 2 About once during the past year 3 About once a month during the past year 4 About once a week during the past year 5 About two or more times per week during past year 1. Been "checked out" (i.e., had your body stared at in an intrusive way) by a person in public. By men 1 2 3 4 5 By women 1 2 3 4 5 2. Your appearance/body commented on in a way that you felt was inappropriate. By men 1 2 3 4 5 By women 1 2 3 4 5 3 Your romantic partner (current or former) "checked out" other women in your presence. By men 1 2 3 4 5 By women 1 2 3 4 5 4. Someone stared at your br easts while talking to you. By men 1 2 3 4 5 By women 1 2 3 4 5 5. Someone made offensive, sexualized gestures toward you (e.g., pantomime of masturbation or intercourse)? By men 1 2 3 4 5 By women 1 2 3 4 5 6. Felt that a date was more interested in your body (and gaining access to it) than in you as a person. By men 1 2 3 4 5 By women 1 2 3 4 5

PAGE 86

78 APPENDIX C THE SOCIOCULTURAL ATTITUDES TOWARD APPEARANCE INTERNALIZATION SUBSCALE (SATAQ) Please read each of the following items and select the number that best reflects your agreement with the statement. 1. = completely disagree 2. = somewhat disagree 3. = neither agree nor disagree 4. = somewhat agree 5. = completely agree 1. Women who appear in TV shows and movies project the type of appearance that I see as my goal. 1 2 3 4 5 2. I believe that clothes look better on thin models. 1 2 3 4 5 3. Music videos that show thin wo men make me wish that I were thin. 1 2 3 4 5 4. I do not wish to look like the models in the magazines. 1 2 3 4 5 5. I tend to compare my body to people in magazines and on TV. 1 2 3 4 5 6. Photographs of thin women make me wish that I were thin. 1 2 3 4 5 7. I wish I looked like a swimsuit model. 1 2 3 4 5 8. I often read magazines like Cosmopolitan Vogue and Glamour and compare my appearance to the models. 1 2 3 4 5

PAGE 87

79 APPENDIX D CONNECTION WITH THE LESBIAN COMMUNITY SUBSCALE (CLC) Please read each of the following items and select the number that best reflects your agreement with the statement. 6. Strongly Disagree 7. Moderately Disagree 8. Slightly Disagree 9. Neither Agree nor Disagree 10. Slightly Agree 11. Moderately Agree 12. Strongly Agree 1. Many of my friends are lesbians. 1 2 3 4 5 6 7 2. Attending lesbian even ts and organizations is important to me. 1 2 3 4 5 6 7 3. I feel isolated and separate from other lesbians. 1 2 3 4 5 6 7 4. When interacting with members of the lesbian community, I often feel diffe rent and alone, like I don’t fit in. 1 2 3 4 5 6 7 5. Having lesbian friends is important to me. 1 2 3 4 5 6 7 6. I am familiar with lesbian books and/or magazines. 1 2 3 4 5 6 7 7. Being a part of the lesbian community is important to me. 1 2 3 4 5 6 7 8. I feel comfortable joining a lesbian social group, lesbian sports team, or lesbian organization. 1 2 3 4 5 6 7 9. Social situations with othe r lesbians make me feel uncomfortable. 1 2 3 4 5 6 7 10. I am familiar with lesbian movies and/or music. 1 2 3 4 5 6 7 11. I am aware of the history concerning the development of lesbian communities and/or the lesbian/gay rights movement. 1 2 3 4 5 6 7 12. I am familiar with lesbian music festivals and conferences. 1 2 3 4 5 6 7 13. I am familiar with community resources for lesbians (i.e., bookstores, support groups, bars, etc.). 1 2 3 4 5 6 7

PAGE 88

80 APPENDIX E ATTITUDES TOWARD FEMINISM AND THE WOMEN’S MOVEMENT (FWM) SCALE Please read each of the following items and select the number that best reflects your agreement with the statement. 13. Strongly Disagree 14. Somewhat Disagree 15. Neither Agree nor Disagree 16. Somewhat Agree 17. Strongly Agree 1. The leaders of the women’s m ovement have the right idea. 1 2 3 4 5 2. There are better ways for women to fight for equality than through the women’s movement. 1 2 3 4 5 3. Feminists are too visionary for a practical world. 1 2 3 4 5 4. People would favor women’s libera tion more if they knew more about it. 1 2 3 4 5 5. The women’s movement has positiv ely influenced relationships between men and women. 1 2 3 4 5 6. The women’s movement is too radi cal and extreme in its views. 1 2 3 4 5 7. Feminist principles should be adopted everywhere. 1 2 3 4 5 8. I would be overjoyed if wome n’s liberation gained more strength in this country. 1 2 3 4 5 9. The women’s movement has made important gains in equal rights and political power for women. 1 2 3 4 5 10. Feminists are a menace to this nation and the world. 1 2 3 4 5

PAGE 89

81 APPENDIX F BODY SURVEILLANCE SUBSCALE OF THE OBJECTIFIED BODY CONSCIOUSNESS SCALE (OBC) Please read each of the following items and select the number that best reflects your agreement with the statement. Circle NA onl y if the statement does not apply to you. Do not circle NA if you don't agree with the st atement. For example if the statement says "When I am happy, I feel like singing" a nd you don't feel like singing when you are happy, then you would circle one of the disa gree choices. You woul d only circle NA if you were never happy. 18. Strongly Disagree 19. Moderately Disagree 20. Slightly Disagree 21. Neither Disagree nor Agree 22. Slightly Agree 23. Moderately Agree 24. Strongly Agree 25. Item does not apply 1. I rarely think about how I look. 1 2 3 4 5 6 7 NA 2. I think it is more importa nt that my clothes are comfortable than whethe r they look good on me. 1 2 3 4 5 6 7 NA 3. I think more about how my body feels than how my body looks. 1 2 3 4 5 6 7 NA 4. I rarely compare how I look with how other people look. 1 2 3 4 5 6 7 NA 5. During the day, I think about how I look many times. 1 2 3 4 5 6 7 NA 6. I often worry about whether the clothes I am wearing make me look good. 1 2 3 4 5 6 7 NA 7. I rarely worry about how I look to other people. 1 2 3 4 5 6 7 NA 8. I am more concerned with what my body can do than how it looks. 1 2 3 4 5 6 7 NA

PAGE 90

82 APPENDIX G BODY SHAME SUBSCALE OF THE OBJECTIFIED BODY CONSCIOUSNESS SCALE (OBC) Please read each of the following items and select the number that best reflects your agreement with the statement. Circle NA onl y if the statement does not apply to you. Do not circle NA if you don't agree with the st atement. For example if the statement says "When I am happy, I feel like singing" a nd you don't feel like singing when you are happy, then you would circle one of the disa gree choices. You woul d only circle NA if you were never happy. 26. Strongly Disagree 27. Moderately Disagree 28. Slightly Disagree 29. Neither Disagree nor Agree 30. Slightly Agree 31. Moderately Agree 32. Strongly Agree 33. Item does not apply 1. When I can’t control my weight, I feel like something must be wrong with me. 1 2 3 4 5 6 7 NA 2. I feel ashamed of myself when I haven’t made the effort to look my best. 1 2 3 4 5 6 7 NA 3. I feel like I must be a bad person when I don’t look as good as I could. 1 2 3 4 5 6 7 NA 4. I would be ashamed for people to know what I really weigh. 1 2 3 4 5 6 7 NA 5. Even when I can’t control my weight, I think I’m an okay person. 1 2 3 4 5 6 7 NA 6. I never worry that something is wrong with me when I am not exercising as much as I should. 1 2 3 4 5 6 7 NA 7. When I’m not exercising enough, I question whether I am a good enough person. 1 2 3 4 5 6 7 NA 8. When I’m not the size I th ink I should be, I feel ashamed. 1 2 3 4 5 6 7 NA

PAGE 91

83 APPENDIX H THE EATING ATTITUDES TEST – 26 (EAT-26) For each of the following questions, please select the response that best describes you. 34. Never 35. Rarely 36. Sometimes 37. Often 38. Usually 39. Always 1. Am terrified about being overweight. 1 2 3 4 5 6 2. Avoid eating when I am hungry. 1 2 3 4 5 6 3. Find myself preoccupied with food. 1 2 3 4 5 6 4. Have gone on eating binges wher e I feel that I may not be able to stop. 1 2 3 4 5 6 5. Cut my food into small pieces. 1 2 3 4 5 6 6. Aware of the calorie content of foods that I eat. 1 2 3 4 5 6 7. Particularly avoid food w ith a high carbohydrate content (i.e., bread, rice, potatoes, etc.) 1 2 3 4 5 6 8. Feel that others would prefer if I ate more. 1 2 3 4 5 6 9. Vomit after I have eaten. 1 2 3 4 5 6 10. Feel extremely guilty after eating. 1 2 3 4 5 6 11. Am preoccupied with a desire to be thinner. 1 2 3 4 5 6 12. Think about burning up calories when I exercise. 1 2 3 4 5 6 13. Other people think that I am too thin. 1 2 3 4 5 6 14. Am preoccupied with the t hought of having fat on my body. 1 2 3 4 5 6 15. Take longer than others to eat my meals. 1 2 3 4 5 6 16. Avoid foods with sugar in them. 1 2 3 4 5 6 17. Eat diet foods. 1 2 3 4 5 6 18. Feel that food controls my life. 1 2 3 4 5 6 19. Display self-control around food. 1 2 3 4 5 6 20. Feel that others pressure me to eat. 1 2 3 4 5 6 21. Give too much time and thought to food. 1 2 3 4 5 6 22. Feel uncomfortable after eating sweets. 1 2 3 4 5 6 23. Engage in dieting behavior. 1 2 3 4 5 6 24. Like my stomach to be empty. 1 2 3 4 5 6 25. Enjoy trying new rich foods. 1 2 3 4 5 6 26. Have the impulse to vomit after meals. 1 2 3 4 5 6

PAGE 92

84 APPENDIX I DEMOGRAPHIC QUESTIONNAIRE Please tell us a little about yourself. This information will be used only to describe the sample as a group. 1. Age: _______ 2. Gender _____Male _____Female ____Transgender 3. Your current relationship stat us (please select the best descriptor): 4. ____Single ____Married/Partnered ____Da ting, long term ____Dating, casual 5. Completed Education (please select one): _____ Less than High School _____ Some High School _____ High School Graduate _____ Some College _____ College Degree (e.g. B. A., B.S.) _____ Professional Degree (e.g., MBA, MS, Ph.D, M. D.) 6. Current Employment status (p lease select the one best descriptor): _____ Employed Full Time _____Employed Part Time _____Not employed 7. Yearly household income (income of those on whom you rely financially): ______Below $10,000 ______$60,001 to $70,000 ______$10,001 to $20,000 ______$70,001 to $80,000 ______$20,001 to $30,000 ______$80,001 to $90,000 ______$30,001 to $40,000 ______$90,001 to $100,000 ______$40,001 to $50,000 ______$100,001 to $110,000

PAGE 93

85 ______$50,001 to $60,000 ______Above $110,001 8. Your current social class (p lease select the one best descriptor): _____ lower class _____working class _____ middle class _____ upper middle class _____upper class 9. Race/ethnicity (Please check one) _____ African American/Black _____ Asian American/Pacific Islander _____ American Indian/Native American _____ Hispanic/Latino/a – White _____ Hispanic/Latino/a – Black _____ Multi-racial, please specify: ___________________________ _____ White/Caucasian _____ Other, please specify: ___________________________ 10. Current height ______feet _______inches 11. Current weight in pounds__________________ 12. Your sexual orientation (p lease check the one best descriptor): _____ Exclusively lesbian _____ Mostly lesbian _____ Bisexual _____ Mostly Heterosexual _____ Exclusively Heterosexual 13. How much are you physically attracted to members of your own sex? __________ _____ _____ low moderate high

PAGE 94

86 14. How much are you physically attracted to members of the other sex? _________ _____ _____ _____ low moderate high 15. How much are you emotionally attracted to members of your own sex? _____ _____ _____ low moderate high 16. How much are you emotionally attracted to members of the other sex? _____ _____ _____ low moderate high 17. Sexual behavior: Have you had sex with persons of your own gender, the other gender, or both genders? ____ ____ ____ ____ ____ ____ Never had sex My own gender only My own gender mostly Both genders equally Other gender mostly Other gender only 18. How connected or involved are you in the lesbian community? Please select one. ______ ______ ______ ______ ______ _______ very slightly a little moderately quite a bit extremely not at all 19. To what extent do you describe yourself as a feminist? _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ 0 9 not at all very much a feminist 20. Finally, we would like to obtain informa tion regarding the ge ographic location of our sample. This information will remain conf idential. Please fill in the city, state, and country in which you cu rrently reside down below: City:_____________________ State:_____________________ Country:__________________

PAGE 95

87 APPENDIX J PERTINENT TABLES AND FIGURES

PAGE 96

88Summary Statistics and Partial Correl ations Among Variables of Interest with Body Mass Index Controlled Variables 1 2 3 4 5 6 Possible Range Sample Range M SD 1. Reported Sexual Objectification Experiences 1.00-5.00 1.00-4.69 1.72 .63 .94 2. Feminist Ideology -.16** 1.00-5.00 1.30-5.00 3.92 .57 .85 3. Disconnection from Lesbian Community .10* .33** 1.00-7.00 1.00-5.92 2.45 .94 .85 4. Internalization .15** .01 .14** 1.00-5.00 1.00-5.00 2.13 .90 .87 5. Body Surveillance .22** -.10* .14* .54** 1.00-7.00 1.00-7.00 4.01 1.27 .86 6. Body Shame .19** .00 .17**.55**.58** 1.00-7.00 1.00-7.00 3.12 1.38 .87 7. Eating disorder symptoms .17** .01 .09* .45**.51**.55**0.00-78.00 0.00-51.00 7.54 8.37 .86 Note. *p < .05. **p <.001. Higher scores indicate higher levels of the construct assessed.

PAGE 97

89 Hypothesized Path Model Trimmed Model Depicting Relationships Among Variables of Interest with Body Mass Index Controlled

PAGE 98

90 LIST OF REFERENCES Abraham, S. F., & Beumont, P. J. (1982) Varieties of psychosexual experience in patients with anorexia nervosa. International Journal of Eating Disorders, 1 1019. American Psychiatric Association Work Group on Eating Disorders (2000). Practice guidelines for the treatment of patient s with eating disorders (revision). American Journal of Psychiatry, 157 1-39. Arbuckle, J. L. (2003). Amos (Version 5.0) [Computer softwa re]. Chicago: SmallWaters Corporation. Baron, R. M., & Kenny, D. A. (1986). The mode rator-mediator variab le distinction in social psychological research: Con ceptual, strategic, and statistical consideration. Journal of Personality and Social Psychology 51 1173-1182. Belknap, J., Fischer, B. S., & Cullen, F. T. (1999). The development of a comprehensive measure of the sexual victimization of college women. Violence Against Women, 5 185–24. Ben-Tovim, D. I., & Walker, M. K. (1991). The development of the Ben-Tovim-Walker Body Attitudes Questionnaire (BAQ), a new measure of women’s attitudes towards their own bodies. Psychological Medicine, 21 775-784. Bergeron, S. M., & Senn, C. Y. (1998). B ody image and sociocultural norms: A comparison of heterosexual and lesbian women. Psychology of Women Quarterly 22 385-401. Beren, S. E., Hayden, H. A., Wilfley, D. E ., & Grilo, C. M. (1996). The influence of sexual orientation on body dissatisf action in adult men and women. International Journal of Eating Disorders, 20 (2), 135-141. Beren, S. E., Hayden, H.A., Wilfley, D.E ., & Striegel-Moore, R. H. (1997). Body dissatisfaction among lesbian college st udents: The conflict of straddling mainstream and lesbian cultures. Psychology of Women Quarterly 21 431-445. Brand, P. A., Rothblum, E. D., & Solomon, L. J. (1992). A comparison of lesbians, gay men, and heterosexuals on wei ght and restrained eating. International Journal of Eating Disorders, 11 (3), 253-259.

PAGE 99

91 Brownlow, B. (1997). The relationship between objectification, body image disturbances, and disordered eating: Investigating race, socioeconomic status, acculturation, and self-objectifica tion as mediators Unpublished doctoral dissertation, Duke University. Burnett, R. E. (1995). Gendered objectification experie nces: Construct validity, implications for dysphoria and depression, and phenomenology Unpublished doctoral dissertation, Duke University. Cogan, J. C. (1999). Lesbians walk the tightrope of beauty: Thin is in but femme is out. Journal of Lesbian Studies, 3 (4), 77-89. Cogan, J. C., Bhalla, S. K., Sefa-Dedeh, A., & Rothblum, E. D. (1996). A comparison study of United States and African student s on perceptions of obesity and thinness. Journal of Cross-Cultural Psychology, 27 98-113. Cohen, J., & Cohen, P. (1983). Applied multiple regression/correlation analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Erlbaum. Cooper, P. J., Taylor, M. J., Cooper, Z., & Fairburn, C. G. (1987). The development and validation of the Body Shape Questionnaire. International Journal of Eating Disorders, 6 485-494. Dionne, M. (1992). Feminist ideology as a predicto r of body dissatisfaction in women Unpublished master’s thesis, York Univ ersity, North York, Ontario, Canada. Dionne, M., Davis, C., Fox, J., & Gurevich, M. (1995). Feminist ideo logy as a predictor of body dissatisfaction in women. Sex Roles, 33 (3/4), 277-287. Dunn, P. K., & Ondercin, P. (1981). Personality variables related to compulsive eating in college women. Journal of Clinical Psychology, 37 43-49. Eliason, M. J., & Morgan, K. S. (1998). Lesbians define themselves: Diversity in lesbian identification. Journal of Gay, Lesbian, & Bisexual Identity, 3 (1), 47-63. Enns, C. Z. (1987). A comparison of non-feminist and pr ofeminist women’s reactions to nonsexist, liberal feminist, and radical feminist therapy Unpublished doctoral dissertation, University of California/Santa Barbara. Fassinger, R. E. (1994). Development and tes ting of the Attitudes Toward Feminism and the Women’s Movement (FWM) scale. Psychology of Women Quarterly, 18 389402. Fitzgerald, L. F., Shullman, S. L., Baile y, N., Richards, M., Swecker, J., Gold, Y., Ormerod, M., & Weitzman, L. (1988). The incidence and dimensions of sexual harassment in academia and the workplace. Journal of Vocational Behavior, 32 152–175.

PAGE 100

92 Frazier, P.A., Tix, A. P., & Barron, K. E. ( 2004). Testing moderator and mediator effects in counseling psychology research. Journal of Counseling Psychology, 51 (1), 115134. Franzoi, S. E., & Herzog, M. E. (1986). The Body Esteem Scale: A convergent and discriminant validity study. Journal of Personality Assessment, 50 24-31. Franzoi, S. E., & Herzog, M. E. (1987). Judging physical attractiveness: What body aspects do we use? Personality and Social Psychology Bulletin, 13 19-33. Franzoi, S. L., & Shields, S. A. (1984). The Body Esteem Scale: Multidimensional structure and sex differen ces in a college population Journal of Personality Assessment, 48 (2), 173-178. Fredrickson, B. L., & Roberts, T. A. (1997). Objectificati on theory: Toward understanding women’s lived experien ces and mental health risks. Psychology of Women Quarterly 21 173-206. Fredrickson, B. L., Roberts, T. A, Noll, S. M., Quinn, D. M, & Twenge, J. M. (1998). That swimsuit becomes you: Sex differen ces in self-objectification, restrained eating, and math performance. Journal of Personality & Social Psychology 75 (1), 269-284. Furnham, A., & Alibhai, N. ( 1983). Cross-cultural differences in the perception of female body shapes. Psychological Medicine, 13 (4), 829-837. Garner, D. M. (1997). Psychoeducational principles in treatment. In: D.M. Garner & P.E. Garfinkel (Eds.) Handbook of Treatment for Eating Disorders New York: Guilford Press. Garner, D. M., & Garfinkel, P. E. (1979). The Eating Attitudes Test: An index of the symptoms of anorexia nervosa. Psychological Medicine, 9 (2), 273–279. Garner, D. M., & Olmsted, M. P. (1984). Eating Disorder Inventory Manual Odessa, FL: Psychological Assessment Resources. Garner, D. M., Olmsted, M. P., Bohr, Y. & Garfinkel, P. E. (1982). The Eating Attitudes Test: Psychometric features and clinical correlates. Psychological Medicine 12 871-878. Garner, D. M., Olmsted, M. P., & Polivy, J. (1983). Development and validation of multidimensional eating disorder inventory for anorexia nervosa and bulimia. International Journal of Eating Disorders, 2 15-35. Greenleaf, C. (2005). Self-objectific ation among physically active women. Sex Roles, 52 (1-2), 51-62.

PAGE 101

93 Griffiths, R. A., Beumont, P. J. V., Russe ll, J., Schotte, D., Thornton, C., Touyz, S., & Varano, P. (1999). Sociocultural attitudes to wards appearance in dieting disordered and nondieting disordered subjects. European Eating Disorders Review, 7 193203. Griffiths, R. A., Mallia-Blanco, R., Boesenber g, E., Ellis, C., Fischer, K., Taylor, M., & Wyndham, J. (2000). Restrained eating and so ciocultural attitudes to appearance and general dissatisfaction. European Eating Disorders Review, 8 394-402. Guille, C., & Chrisler, J. C. (1999). Does fe minism serve a protective function against eating disorders? Journal of Lesbian Studies, 3 (4), 141-148. Heffernan, K. (1996). Eating disorders and weight concern among lesbians. International Journal of Eating Disorders 19 (2), 127-138. Heinberg, L. J., Thompson, J. K., & Stormer, S. (1995). Development and validation of the Sociocultural Attitudes To ward Appearance Questionnaire. International Journal of Eating Disorders 17 81-89. Herzog, D. B., Newman, K. L., Yeh, C. J., & Warshaw, M. (1992). Body image satisfaction in homosexual and heterosexual women. International Journal of Eating Disorders, 11 (4), 391-396. Hill, M. S. (2002). Examining objectification theory: Sexual objectif ication’s link with self-objectification and moderation by sexua l orientation and age in White women. Unpublished doctoral dissertati on, The University of Akron. Huebner, D. M, & Fredrickson, B. L. (1999). Gender differences in memory perspectives: Evidence for self-objectification in women. Sex Roles, 41 (5/6), 459467. InfoPlease (2005). State Regions Retrieved June 17, 2005, from Infoplease, Pearson Education Web site: http://www.infoplease.com/ipa/A0770177.html Kaminski (2000). Lesbian health: Social context, sexual identity and well-being. Journal of Lesbian Studies, 4 (3), 87-101. Kashubeck-West, S., Mintz, L. B., & Sa unders, K. J. (2001). Assessment of eating disorders in women. The Counseling Psychologist, 29 627-634. Koslowsky, M., Scheinberg, Z., Bleich, A., & Mark, M. (1994). Pred icting actual weight from self-report data. Educational & Psychological Measurement, 54 (1), 168-173. Lakkis, J., Ricciardelli, L. A ., & Williams, R. J. (1999). Ro le of sexual orientation and gender-related traits in disordered eating. Sex Roles, 41 (1-2), 1-16.

PAGE 102

94 Leavy, R. L., & Adams, E. M. (1986). Femini sm as a correlate of self-esteem, selfacceptance, and social support among lesbians. Psychology of Women Quarterly, 10 321-326. Ludwig, M. R., & Brownell, K. D. (1999). Lesbians, bisexual women, and body image: An investigation of gender ro les and social group affiliation. International Journal of Eating Disorders 25 (1), 89-97. McKinley, N. M. (1998). Gender differen ces in undergraduates’ body esteem: The mediating effect of objectified body c onsciousness and actual/ideal weight discrepancy. Sex Roles, 39 (1-2), 113-123. McKinley, N. M, & Hyde, J. S. (1996). The Objectified Body Consciousness Scale: Development and validation. Psychology of Women Quarterly 20 181-215. Moradi, B., Dirks, D., & Matteson, A. V. (2005). Roles of sexual objectification experiences and internalization of sta ndards of beauty in eating disorder symptomatology: A test and exte nsion of objectif ication theory. Journal of Counseling Psychology, 52 (3), 420-428. Morry, M. M., & Staska, S. L. (2001). Maga zine exposure: In ternalization, selfobjectification, eating attitudes, and body satisfaction in male and female university students. Canadian Journal of Behavioural Science 33 (4), 269-279. Muehlenkamp, J. J., & Saris-Baglama, R. N. (2002). Self-obj ectification and its psychological outcomes for college women. Psychology of Women Quarterly 26 (4), 371-379. Myers, A., Taub, J., Morris, J. F., & Rothbl um, E. D. (1999). Beauty mandates and the appearance obsession: Are lesbian and bisexual women better off? Journal of Lesbian Studies, 3 (4), 15-26. Noll, S. M, & Fredrickson, B. L. ( 1998). A mediational model linking selfobjectification, body shame, and disordered eating. Psychology of Women Quarterly 22 (4), 623-636. Polivy, J., Herman, C. P., & Howard, K. ( 1988). The Restraint Scale: Assessment of dieting. In M. Hersen & A. S. Bellack (Eds.), Dictionary of behavioral assessment techniques (pp. 377–380. New York: Pergamon. Rodin, J., Silberstein, L., & Striegel-Moor e, R. H. (1984). Women and weight: A normative discontent. In T. B. Sonderegger (Ed.), Nebraska Symposium on Motivation 1984: Psychology and Gender (Vol. 32, pp. 267-307). Lincoln, NE: University of Nebraska Press. Romano, S. J. & Quinn, L. (2001). Evaluati on and treatment of bulimia nervosa. Primary Psychiatry, 8 (2), 57-62.

PAGE 103

95 Scarano, G. M. & Kalodner-Martin, C. R. (1994). A description of the continuum of eating disorders: Implications for intervention and research. Journal of Counseling & Development, 72 (4), 356-361. Share, T. L, & Mintz, L. B. (2002). Diffe rences between lesbians and heterosexual women in disordered eati ng and related attitudes. Journal of Homosexuality 42 (4), 89-106. Siever, M. D. (1994). Sexual orientation and ge nder as factors in socioculturally acquired vulnerability to body dissatisfa ction and eating disorders. Journal of Consulting Clinical Psychology 62 (2), 252-260. Slater, A., & Tiggemann, M. ( 2002). A test of objectification theory in adolescent girls. SexRoles 46 (9-10), 343-349. Snyder, R., & Hasbrouck, L. (1996). Feminist identity, gender trai ts, and symptoms of disturbed eating among college women Psychology of Women Quarterly, 20 593598. Sobel, M. E. (1982). Asymptotic confidence in tervals for indirect e ffects in structural equations models. In S. Leinhart (Ed.), Sociological methodology 1982 (pp. 290312). San Francisco: Jossey-Bass. Steiner, H., & Lock, J. (1998). Anorexia ne rvosa and bulimia nervosa in children and adolescents: A review of the past 10 years. Journal of the American Academy of Child & Adolescent Psychiatry, 37 (4), 352-359. Strelan, P., Mehaffey, S. J., & Tiggemann, M. (2003). Self-objectifica tion and esteem in young women: The mediating role of reasons for exercise. Sex Roles 48 (1-2), 8995. Striegel-Moore, R. H., Tucker, N., & Hs u, J. (1990). Body image dissatisfaction and disordered eating in lesbian college students. International Journal of Eating Disorders, 9 (5), 493-500. Strong, S. M., Williamson, D. A., Netemeyer, R. G., & Geer, J. H. (2000). Eating disorder symptoms and concerns about body differ as a function of gender and sexual orientation. Journal of Social and Clinical Psychology, 19 (2), 240-255. Stunkard, A., Sorensen, T., & Schulsinger, F. (1980). Use of the Danish Adoption Register for the study of obesity and thinness. In S. Kety (Ed.), The genetics of neurological and psychiatric disorders (pp. 115-120). New York: Raven Press. Stunkard, A. J., & Messick, S. (1985). The Thre e-Factor Eating Questionnaire to measure dietary restraint, di sinhibition, and hunger. Journal of Psychosomatic Research, 29 71-83.

PAGE 104

96 Swim, J. K., Cohen, L. L., & Hyers, L. L. (1998). Experiencing everyday prejudice and discrimination. In J. K. Swim & C. Stangor (Eds.), Prejudice: The target’s perspective (pp. 37-60). New York: Academic. Szymanski, D. M, & Chung, Y. B. (2001). Th e Lesbian Internalized Homophobia Scale: A rational/theore tical approach. Journal of Homosexuality 41 (2), 37-52. Szymanski, D. M., & Chung, Y. B. (2003). Fe minist attitudes and coping resources as correlates of lesbian internalized heterosexism. Feminism & Psychology, 13 (3), 369-389. Thelen, M. H., Farmer, J., & Wonderlich, S. (1991). A revision of the Bulimia Test: The BULIT-R. Psychological Assessment, 3 (1), 119-124. Tienboon, P., Wahlqvist, M. L., & Rutishauser I. H. (1992). Self-reported weight and height in adolescents and their parents. Journal of Adolescent Health, 13 (6), 528532. Tiggemann, M., & Lynch, J. E. (2001). Body imag e across the life span in adult women: The role of self-objectification. Developmental Psychology, 37 (2), 243-253. Tiggemann, M., & Slater, A. (2001). A test of objectification theory in former dancers and non-dancers. Psychology of Women Quarterly 25 (1), 57-64. Walters, E. E., & Kendler, K. S. (1995). Anor exia nervosa and anorexic-like syndromes in a population-based female twin sample American Journal of Psychiatry, 152 64-71. Wolf, N. (1991). The beauty myth: How images of beauty are used against women. New York: Anchor Books. Worell, J., & Remer, P. (1992). Feminist perspectives in therapy: An empowerment model for women New York: Wiley.

PAGE 105

97 BIOGRAPHICAL SKETCH Tiffany Leigh Graham was born in Fort La uderdale, Florida, and graduated summa cum laude from Ball State University in May 2003 with a B.S. in psychology. She immediately entered graduate school in the Department of Psychology at the University of Florida, where she is pursui ng a Ph.D. in counseling psychology.


Permanent Link: http://ufdc.ufl.edu/UFE0013334/00001

Material Information

Title: Lesbian Women and Eating Disorder Symptomatology: A Test and Extension of Objectification Theory
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0013334:00001

Permanent Link: http://ufdc.ufl.edu/UFE0013334/00001

Material Information

Title: Lesbian Women and Eating Disorder Symptomatology: A Test and Extension of Objectification Theory
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0013334:00001


This item has the following downloads:


Full Text












LESBIAN WOMEN AND EATING DISORDER SYMPTOMATOLOGY: A TEST
AND EXTENSION OF OBJECTIFICATION THEORY















By

TIFFANY L. GRAHAM


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE

UNIVERSITY OF FLORIDA


2005

































Copyright 2005

by

Tiffany L. Graham
































This thesis is dedicated to my best friend and mother, Donna Graham.















ACKNOWLEDGMENTS

First of all, I would like to thank Dr. Bonnie Moradi for her mentorship, patience,

and guidance throughout this project. I would also like to thank Dr. Mark Fondacaro and

Dr. Deidre Pereira for their valuable feedback and positive support as members of my

master's committee. I also offer a huge thanks to Dr. Mary Fukuyama for serving as a

proxy at my thesis defense, and for her feedback regarding my survey items and study as

a whole. I would also like to thank Jackie Davis for her feedback regarding my survey

and Jim Yousse for his valuable help with creating and uploading my survey to the

internet. I would also like to thank the many lesbian/gay listserves and organizations that

helped me in promotion of the study. With their aid, I was able to connect with lesbian

women at the international level. Thanks go to Stacey Garner for her help in promoting

and recruiting participants for this study. And of course, I would like to thank my friends

(particularly my cohort) and family for all of their help and support throughout the

duration of this project. I would like to thank my Dad for showing me that it really can be

done. Finally, I would like to thank Bradley Jason Daniels, my partner in life and in

crime, whose constant belief in me helped me to believe in myself, and that is why I was

able to complete and defend this project. Ya! Ya! Ya!
















TABLE OF CONTENTS
page

A C K N O W L E D G M E N T S ................................................................................................. iv

ABSTRACT ............... ................... ........ .............. vii

CHAPTER

1 IN T R O D U C T IO N ............................................................................. .....................

Empirical Support for Objectification Theory Applied to Eating Disorder
Sym ptom atology ................ .... .................................. .. .. ..... ...... .. ... ... ... .. 2
Application of Objectification Theory with Lesbian Individuals.............................6
Eating Disorder Symptomatology among Lesbian Women.................................7
The Role of Connection with the Lesbian Community in Eating Disorder
Sym ptom atology ................................ .. .............. .. ...... .. .. .......... .. 8
The Role of Feminist Ideology in Eating Disorder Symptomatology ................... 10
Significan ce of th e Stu dy ...................... .. .. ............. ............................................. 13

2 REVIEW OF THE LITERATURE ......................................... ...............14

Self-Objectification and Its Link to Body Shame and Eating Disorder
Sym ptom atology ........................ ... .... .............. ....... ...................... .... .. 16
Sexual Objectification Experiences and Internalization of Sociocultural Standards
of Beauty .............. . .... .................... ........ ...................... 19
Objectification Theory as it Applies to Lesbian Individuals...............................23
Lesbian Women and Eating Disorder Symptomatology ........................................25
Connection with the Lesbian Community ...................................... ............... 29
F em insist Ideology .............. ................ ........... ...... ........ ...... ............. 31
H ypotheses ................................................. 39

3 M E T H O D S ........................................................................................................... 4 1

P a rtic ip a n ts ........................................................................................................... 4 1
In strum ents ........................ .................................... ................4 2
Reported Sexual Objectification Experiences ................. ................. ............43
Connection/Disconnection with Lesbian Community ............... ..... ...........46
Fem inist Ideology ........................................47
Internalization of Sociocultural Standards of Beauty .................................... 48
Body Surveillance as an Indicator of Sexual Objectification ...............................49
B o d y S h a m e ...................... .. ............. .. ..................................................5 0









Eating D disorder Sym ptom atology ........................................ ......................... 51
D em graphics .........................................................................................................52
P ro c e d u re .......................................................................................5 3
Statistical A analyses ......................................................... ...... .... ...... .. 54

4 R E S U L T S .............................................................................5 6

D descriptive Statistics .................... .. .......................................... ... 56
Intercorrelations among Variables of Interest ...... ................. .. ........... ........ 57
M e d iatio n s ................................................................5 8

5 DISCUSSION ............ ......... ........ .. ...............63

Limitations ............... ...... ... ... .............. ... .................... 67
Directions for Future Research.......... .......................... ................ ... 69
Im plications for P ractice......... ..................... .................................... ............... 7 1
Summary ................................ ............................73

APPENDIX

A THE SEXUAL OBJECTIFICATION SUBSCALE ............................................75

B OBJECTIFICATION EXPERIENCES QUESTIONNAIRE...............................77

C THE SOCIOCULTURAL ATTITUDES TOWARD APPEARANCE
INTERNALIZATION SUBSCALE (SATAQ)............................... ...............78

D CONNECTION WITH THE LESBIAN COMMUNITY SUBSCALE (CLC) .........79

E ATTITUDES TOWARD FEMINISM AND THE WOMEN'S MOVEMENT
(FW M ) SC A L E .......................... ...................... .. .. .... ........ ........ 80

F BODY SURVEILLANCE SUB SCALE OF THE OBJECTIFIED BODY
CON SCIOU SNESS SCALE (OBC) ........................................ ...................... 81

G BODY SHAME SUBSCALE OF THE OBJECTIFIED BODY
CON SCIOUSNESS SCALE (OBC) ........................................ ...................... 82

H THE EATING ATTITUDES TEST 26 (EAT-26) ..............................................83

I DEMOGRAPHIC QUESTIONNAIRE...................... ... ......................... 84

J PERTINENT TABLES AND FIGURES ........................................ .....................87

LIST OF REFEREN CES ............................................................................. 90

BIOGRAPH ICAL SKETCH ...................................................... 97















Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science

LESBIAN WOMEN AND EATING DISORDER SYMPTOMATOLOGY: A TEST
AND EXTENSION OF OBJECTIFICATION THEORY

By

Tiffany L. Graham

December, 2005

Chair: Banafsheh Moradi
Major Department: Psychology

Objectification theory posits that women in Western society experience sexually

objectifying events that lead to the adoption of an outsider's perspective upon one's own

body. Such self-objectification produces increased body shame in response to perceived

failure to live up to cultural beauty standards. Body shame, in turn, results in eating

disorder symptomatology. Although empirical support for the tenets of objectification

theory is accumulating, to date, little research exists that addresses the applicability of

objectification theory to lesbian women. In addition, two factors, connection with the

lesbian community and feminist ideology, have emerged from the literature as variables

that may play a role in the development of eating disorder-related attitudes and behaviors

among lesbian persons. The current study tested the objectification theory framework as

it applies to eating disorder symptomatology in a sample of 531 lesbian women, and

additionally explored the roles of connection with the lesbian community and feminist

ideology. A theoretically based path analysis was conducted to investigate relationships









among all variables in the model. Results indicated that the objectification theory

framework was applicable for lesbian women, and disconnection from the lesbian

community was a stronger predictor of eating disorder related attitudes and behaviors

than feminist ideology. Implications of findings and directions for future research are

discussed.














CHAPTER 1
INTRODUCTION

The prevalence of eating disorder-related attitudes and behaviors among women in

Western society is astounding. Nearly two percent of women develop anorexia nervosa

(Walters & Kendler, 1995), and approximately three percent battle with bulimia nervosa

(Romano & Quinn, 2001). In addition, the rates of eating disorders have doubled since

the 1960's, and disordered eating behavior can manifest as early as elementary school

(Steiner & Lock, 1998). Even among women who are not diagnosed with eating

disorders, concern with physical appearance and body weight is so pervasive that it has

been deemed "normative discontent" (Rodin, Silberstein, & Striegel-Moore, 1984), and

scholars have advocated the conceptualization of eating problems as a continuum instead

of discrete diagnostic categories (Scarano & Kalodner-Martin, 1994).

Fredrickson and Roberts (1997) proposed objectification theory as a framework for

understanding the development of eating disorder-related attitudes and behaviors along

with other health concerns among women. Objectification theory posits that women

within American culture are exposed to sexually objectifying events that include staring,

sexual ogling, catcalls, and blatant sexual harassment and degradation. Often, emphasis is

placed on women's individual body parts, instead of the whole person, and strong

importance is placed on a woman's body and not her mind, skills, or abilities. Girls and

women are taught at an early age that physical appearance is important, as they are

frequently objectified through various media images including magazines, television

shows, commercials, movies, and pornographic materials. Through such experiences of









sexual objectification, women are socialized to become "objects to be looked at and

evaluated by others" (Fredrickson & Roberts, 1997, p. 177) and they learn to treat other

women in the same way.

According to objectification theory, cultural experiences of sexual objectification

promote self-objectification, or taking on the perspective of an observer upon one's own

body. More specifically, as a result of pervasive experiences of sexual objectification,

girls and women come to internalize an observer's perspective upon their bodies and

"view themselves as objects or sights to be appreciated by others" (Fredrickson &

Roberts, 1997, p. 180). Objectification theory posits that chronic self-objectification, in

turn, increases body shame, promotes anxiety, reduces experiences of peak motivational

states or flow experiences (i.e., periods of intense concentration on a challenging and

rewarding task), and decreases awareness of internal bodily states (e.g., heart rate, sexual

arousal). These experiences in turn result in depression, sexual dysfunction, and eating

disorders among women.

Within this larger framework of objectification theory (Fredrickson & Roberts,

1997), the links among sexual objectification experiences, self-objectification, and body

shame have been posited as the most critical to understanding eating disorder

symptomatology and empirical evidence has supported these proposed links.

Empirical Support for Objectification Theory Applied to Eating Disorder
Symptomatology

To date, the relationships among self-objectification, body shame, and eating

disorder symptomatology have received much empirical support. One such example

exists in an experiment conducted by Fredrickson, Roberts, Noll, Quinn, and Twenge

(1998). According to objectification theory, self-objectification is likely to occur when a









woman is made more aware of her body's appearance. Fredrickson et al. (1998) elicited

such a situation by manipulating the type of clothing worn by participants. In this study,

114 women were assigned randomly to a swimsuit or sweater condition. After trying on

the designated clothing, participants were asked to look in a full-length mirror to evaluate

the clothing, and then sample a cookie for a mock taste test. Results indicated that women

who were wearing a swimsuit reported significantly higher levels of self-objectification

and body shame, and ate fewer cookies (i.e., restrained eating) than did women in the

sweater condition. Thus, a situation that heightened self-objectification resulted in

restrained eating and higher levels of reported self-objectification and body shame.

Contributing further support for the theory, Noll and Fredrickson (1998) and

Moradi, Dirks, and Matteson (2005) also found that self-objectification was related

positively to eating disorder symptomatology. Furthermore, both studies found evidence

that this link was mediated, partially, by body shame. In other words, in addition to the

direct positive link between self-objectification and eating disorder symptoms, self-

objectification was related to greater levels of body shame, which in turn were related to

higher levels of eating disorder symptomatology. These findings were consistent with

objectification theory's proposition that body shame is a key mechanism through which

self-objectification is translated into eating disorder symptoms.

In contrast to the numerous studies that have examined and found support for the

proposed roles of self-objectification and body shame in eating disorder-related attitudes

and behaviors, sexual objectification experiences are only recently beginning to receive

empirical attention in the literature on objectification theory. Hill (2002) began to address

this gap by examining the relationship between sexual objectification experiences and









self-objectification with a sample of 502, mostly White women (307 heterosexual, 33

bisexual, 155 lesbian, and 7 who did not report sexual orientation). Hill assessed sexual

objectification experiences using a combination of existing measures that assess sexual

harassment, other degrading experiences, and the extent that women are treated as sexual

obj ects.

Results indicated that reports of sexual objectification experiences were related

positively to self-objectification. However, this relationship was moderated by age, such

that the magnitude of the relationship between sexual objectification experiences and

self-objectification was strong and positive for women between the ages of 50 and 79, but

non-significant for those between the ages of 18 and 49 years old. This interaction effect

might be explained by the potential larger accumulation of objectification experiences for

older women than for younger women.

Additional evidence suggests that internalization of sociocultural standards of

beauty, a construct not explicitly included in the objectification theory framework or

Hill's (2002) study, might be an important mechanism that translates sexual

objectification experiences into self-objectification and other eating disorder-related

variables. For example, Morry and Staska (2001) found that exposure to beauty

magazines, a specific type of sexual objectification experience, was related to higher

levels of disordered eating and body shape dissatisfaction, but these links were mediated

by internalization of sociocultural standards of beauty. That is, exposure to beauty

magazines was related to greater acceptance of society's beauty mandates, and such

internalization in turn was related to higher levels of self-objectification, body shape

dissatisfaction, and eating disorder symptomatology.









Moradi et al. (2005) conducted the most comprehensive assessment of the

objectification theory framework as applied to eating disorder symptoms to date by

including the proposed links among sexual objectification experiences, self-

objectification, body shame, and eating disorder symptoms. Based on literature

highlighting the importance of internalization of sociocultural standards of beauty and

Morry and Staska's (2001) findings indicating the importance of including this variable

in tests of objectification theory, Moradi et al. also included internalization in the model.

Moradi et al.'s path analytic findings with over 200 undergraduate women demonstrated

that reported experiences of sexual objectification were significantly related to higher

levels of internalization, which in turn was significantly linked with greater body

surveillance (an indicator of self-objectification), body shame, and eating disorder

symptoms. In addition, body surveillance was linked significantly with body shame,

which was correlated significantly with disordered eating.

Moradi et al. (2005) examined the significance of mediator effects and found that

internalization of sociocultural standards of beauty emerged as a mediator of the link of

reported experiences of sexual objectification to body surveillance, body shame, and

eating disorder symptoms. Furthermore, as described in objectification theory, body

surveillance also mediated the link of reported sexual objectification experiences to body

shame. Finally, consistent with previous research (e.g., Noll & Fredrickson, 1998), body

shame partially mediated the link of body surveillance to eating disorder

symptomatology. The overall model provided a very good fit to the data and accounted

for 50% of the variance in eating disorder symptomatology.









Taken together, the studies conducted by Hill (2002), Morry and Staska (2001), and

Moradi et al. (2005) supported the role of sexual objectification experiences in the

objectification theory framework. In addition, they demonstrated the importance of

attending to the role of internalization of sociocultural standards of beauty in the

objectification theory framework.

Application of Objectification Theory with Lesbian Individuals

The tenets of objectification theory have been examined with college women and

men (Fredrickson et al., 1998; Huebner & Fredrickson, 1999; Morry & Staska, 2001;

Muehlenkamp & Saris-Baglama, 2002; Noll & Fredrickson, 1998; Tiggemann, & Slater,

2001), adolescent dancers and non-dancers (Slater & Tiggemann, 2002), and women who

exercise (Strelan, Mchaffey, & Tiggemann, 2003). However, to date, only one study has

examined the aspects of the theory as applied to lesbian persons (Hill 2002).

In Hill's (2002) previously described study of objectification theory, sexual

orientation and age were examined as potential moderators of the relationship between

sexual objectification experiences and self-objectification. Consistent with objectification

theory, in the entire sample (i.e., heterosexual and lesbian women), experiences of sexual

objectification were related positively to self-objectification. Furthermore, sexual

orientation did not moderate the link between sexual objectification and self-

objectification, suggesting that the relationship was similar for both the heterosexual and

lesbian women samples. Although this study began to address the gap in data on the

applicability of objectification theory to lesbian persons, it represented only a partial

investigation of objectification theory because it did not examine key constructs such as

internalization, body shame, and eating disorder symptomatology. In addition, the sample









only consisted of White heterosexual women and White lesbian women, thereby limiting

the generalizability of findings.

Thus, the generalizability of the propositions of objectification theory to

understanding eating disorder symptomatology among lesbian women is not known and

research is needed to examine the applicability of objectification theory to understanding

eating disorder symptoms among lesbian persons. Such research must attend to the

scholarship on eating disorder-related attitudes and behaviors among lesbian women and

include any unique factors, not included in objectification theory, that might shape

lesbian women's experiences of eating disorder-related attitudes and behaviors.

Eating Disorder Symptomatology among Lesbian Women

Findings are mixed as to whether or not lesbian women experience different levels

of eating disorder symptomatology and body satisfaction when compared with

heterosexual women (Beren, Hayden, Wilfley, & Striegel-Moore, 1997). Some studies

have revealed no significant differences between lesbian and heterosexual women

(Beren, Hayden, Wilfley, & Grilo, 1996; Brand, Rothblum, & Solomon, 1992; Striegel-

Moore, Tucker, & Hsu, 1990), whereas other studies indicate that lesbian women

experience higher levels of body satisfaction and lower levels of eating disorder

symptoms than do heterosexual women. For example, Striegel-Moore and colleagues

(1990) found no significant differences in eating behavior and body dissatisfaction in a

sample of 52 heterosexual and 30 lesbian undergraduate students. Other studies, however,

have shown that lesbian women report lower levels of concern with body weight and

physical appearance, internalization of sociocultural standards of beauty, and disordered

eating symptomatology, but higher levels of ideal body weights and body satisfaction









than do heterosexual women (Herzog, Newman, Yeh, & Warshaw, 1992; Share & Mintz,

2002; Siever, 1994).

Myers, Taub, Morris, and Rothblum (1998) used a qualitative research design to

search for an explanation of findings of difference versus no difference between lesbian

and heterosexual women's eating disorder-related attitudes and behaviors. In their

telephone interviews of 18 lesbian and 2 bisexual women, respondents reported either

"feeling freedom from society's norms after coming out" or still "feeling the pressure to

be thin." This finding highlights the importance of attending to individual differences

among lesbian women that shape their experiences of sexual objectification and eating

disorder-related attitudes and behaviors. Extant literature on eating disorder-related

attitudes and behaviors among lesbian women points to connection with the lesbian

community and feminist ideology as two critical individual difference variables that

should be examined.

The Role of Connection with the Lesbian Community in Eating Disorder
Symptomatology


Extant findings suggest that despite experiencing similar cultural messages as

heterosexual women, lesbian women might be less likely to internalize sociocultural

beauty standards, which subsequently contributes to higher body esteem and lower levels

of disordered eating behavior (Bergeron & Senn, 1998; Share & Mintz, 2002). Lesbian

women's lower level of internalization of sociocultural standards of beauty may be due to

multiple causes. For example, the lesbian subculture may be more accepting of a variety

of body types and shapes (Siever, 1994). Indeed, one participant from the telephone

interviews conducted by Myers et al. (1998) reported that the lesbian community is

generally more accepting of larger women. Consistent with this participant's









observation, empirical findings suggest that lesbian women who interact with other

lesbian women are likely to experience positive health effects.

For example, in their study of 188 lesbian and bisexual women, Ludwig and

Brownell (1999) found that lesbian persons with friends who also identified as lesbian

reported a more positive body image than lesbian persons who reported having mostly

heterosexual friends. Unfortunately, sample sizes for their study were rather unbalanced

in that over 80 women reported having mostly friends who were also lesbian or bisexual,

and this group was compared with 27 women reporting having mostly heterosexual

female friends, 15 women reporting having mostly gay or bisexual male friends, and 11

women reporting having mostly heterosexual male friends. Nevertheless, these findings

suggest that interaction with other lesbian women might be related to positive body

image for lesbian women.

Heffernan (1996) examined directly the relationship between connection with the

lesbian and gay community and eating disorder symptomatology. In this study, 203

lesbian women completed questionnaires assessing lifestyle, self-esteem, attitudes about

attractiveness, body esteem, and eating behavior. Involvement with lesbian/gay

community was assessed with the question "How involved are you in lesbian/gay

activities?" Results indicated that the lesbian women were not significantly different from

heterosexual women in terms of attitudes regarding weight and appearance, and the two

groups reported similar rates of bulimia nervosa. Only one difference emerged suggesting

that the lesbian participants reported binge eating disorder more frequently than the

heterosexual participants.









Thus, in general, between group analyses suggested overall similarity between

lesbian and heterosexual women's eating disorder-related attitudes and behaviors. Within

group analyses, however, revealed that among the lesbian women in the sample,

involvement in the lesbian/gay community was related to better health outcomes such as

lower weight and shape concern. Heffernan (1996) concluded that connection and

involvement with the lesbian/gay culture might reduce internalization of society's thin

beauty ideal. Given that Heffernan used only a single item to assess connection with the

lesbian/gay community, however, more comprehensive assessment of this important

construct is needed in future research.

The Role of Feminist Ideology in Eating Disorder Symptomatology

Feminist ideology is a second important individual difference variable that might

shape lesbian women's eating disorder-related attitudes and behaviors. Previous research

has demonstrated that women's endorsement of feminist ideology is related to lower

levels of eating disorder-related attitudes and behaviors (Dionne, Davis, Fox, &

Gurevich, 1995; Snyder & Hasbrouck, 1996). For example, Snyder and Hasbrouck

(1996) examined this possibility in a study of the relationship between feminist identity

development attitudes and symptoms of disturbed eating in a sample of 71 female college

students. Their results indicated that passive acceptance of traditional gender-role

stereotypes (Feminist Identity Development Passive Acceptance scores) was related

positively to drive for thinness and body dissatisfaction, whereas, active commitment to

feminist ideology (Feminist Identity Development Active Commitment

scores) was related negatively to those same outcome measures. Although this

study provided some support for the role of feminist ideology in eating disorder attitudes

and behaviors, the generalizability of these findings to lesbian women was not examined.









Nevertheless, given that lesbian women are more likely to identify with feminism than

are heterosexual women (Guille & Chrisler, 1999), this variable may also play a

significant role in the development of eating disorder-related attitudes and behaviors

among lesbian women.

Bergeron and Senn (1998) addressed this possibility in their examination of

attitudes regarding the body, awareness and internalization of sociocultural standards of

beauty, and feminist self-identification among a sample of 108 lesbian and 115

heterosexual women between the ages of 18 and 58. Feminist identification was assessed

using the question "Would you describe yourself as a feminist?" A MANOVA revealed

that lesbian women reported significantly higher ideal weights, and reported feeling

stronger and more fit than their heterosexual counterparts. A standard multiple regression

revealed that in the entire sample, internalization of sociocultural standards of beauty and

feminist identification both were unique predictors of body attitudes, above and beyond

sexual orientation. No differences were found between heterosexual and lesbian women

on awareness of sociocultural standards of beauty, and this variable did not predict body

attitudes.

Bergeron and Senn's (1998) findings suggest the importance of examining both

feminist identification and internalization when examining body attitudes. A notable

strength of this study is the large sample size obtained by snowball sampling. However,

over 95% of the participants identified as White. Furthermore, feminist identity was

assessed using only a single item and the authors did not examine directly eating disorder

symptomatology. Nevertheless, these findings can be taken as additional support for the









importance of examining feminist ideology and its relationship with eating disorder-

related attitudes and behaviors.

Further evidence of the relationship between feminist ideology and disordered

eating attitudes and behaviors exists in Cogan's (1999) study of 181 lesbian and bisexual

women between the ages of 17 and 58. Participants completed measures assessing

reasons for exercise, fitness activity frequency, type, and duration, dieting behavior, body

satisfaction, eating disorder symptoms, physical appearance before and after coming out,

feminist self-identification, and feminist ideology. To assess feminist self-identification,

participants responded to the question "How much do you consider yourself a feminist?"

and to assess for feminist ideology, participants completed the 10-item Attitudes Toward

Feminism and the Women's Movement scale (FWM; Fassinger, 1994).

Controlling for Body Mass Index (BMI) and age as covariates, it was found that

those who labeled themselves as a feminist (feminist self-identification) and endorsed

feminist ideology were overall more satisfied with their bodies than those who did not.

More specifically, higher levels of feminist self-identification and endorsement of

feminist ideology were related to higher body satisfaction, lower rates of bulimia, drive

for thinness, and weight discrepancy, and tendency to exercise for health versus aesthetic

reasons. The authors concluded that: "feminism may be a useful tool for unlearning

internalized negative body image" (p. 85). Unfortunately, this study did not assess for the

internalization of such ideals.

Thus, extant research suggests that feminist ideology should be included in

examination of disordered eating symptomatology among lesbian persons. However,

since it is still unknown whether or not it is connection with the lesbian community,









feminist ideology, or both that play important roles in the internalization of sociocultural

standards of beauty and the manifestation of disordered eating symptoms in lesbian

women, both were included in the current research.

Significance of the Study

The present research addressed a number of gaps in the literature by examining

objectification theory as it applied to understanding eating disorder symptomatology with

a lesbian sample. More specifically, the present study examined the previously supported

framework of objectification theory that includes links among sexual objectification,

internalization of sociocultural beauty standards, self-objectification, body shame, and

eating disorder symptomatology (Moradi et al., 2005). Furthermore, extant research has

supported the inclusion of two additional variables, connection with the lesbian

community and feminist ideology, as key predictor variables in examining the

applicability of the objectification theory framework to lesbian women. Thus, the present

study included these variables in examining objectification theory's applicability to

understanding eating disorder symptomatology with lesbian women. The online survey

method provided for the recruitment of a large sample of lesbian women; therefore, the

results obtained were more generalizable.














CHAPTER 2
REVIEW OF THE LITERATURE

In the United States, the rate of eating disorders is quite high among women, and

these rates continue to climb. According to the APA Work Group on Eating Disorders

(2000), approximately .5 to 3.7% of women suffer from anorexia nervosa, and 1.1 to

4.2% suffer from bulimia nervosa. In addition, the rates of eating disorders have doubled

since the 1960's, and disordered eating behavior can manifest as early as elementary

school (Steiner & Lock, 1998). Only 1 in 40,000 women fit the size and shape of a

typical supermodel (Wolf, 1991), yet women persistently engage in behaviors to achieve

the near impossible through diet, exercise, various beauty products, wardrobe, surgery,

and engaging in eating disorder-related behaviors. Chronic dieting has become a way of

life for some women, and up to 60% engage in these behaviors by the time they are in

high school (Steiner & Lock, 1998). This pattern of chronic restriction of food intake can

carry with it serious and sometimes deadly consequences. According to Fredrickson and

Roberts (1997), eating disorders are "the extreme end of a continuum of this normative

discontent" (p. 191). In sum, eating disorder symptomatology has been identified as a

serious mental health concern among women. Attempts to understand, prevent, and treat

eating disorder symptomatology have highlighted intrapersonal and contextual variables

that could shape the development of such symptoms.

Objectification theory (Fredrickson & Roberts, 1997) provides a framework that

integrates both intrapersonal and contextual factors that play a role in mental health

problems, such as eating disorder symptomatology, that have higher prevalence rates









among women than men. More specifically, objectification theory posits that women

within American culture are exposed to sexually objectifying events that include staring,

sexual ogling, catcalls, and blatant sexual harassment and degradation. Women often

encounter situations in which emphasis is placed on women's individual body parts,

instead of the whole person, and strong importance is placed on a woman's body and not

her mind, skills, or abilities. Girls and women are taught at an early age that physical

appearance is important, as they are frequently objectified through various media images

including magazines, television shows, commercials, movies, and pornographic

materials. Through such cultural experiences, women are socialized to become "objects

to be looked at and evaluated by others" (Fredrickson & Roberts, 1997, p. 177) and they

learn to treat other women in the same way.

According to objectification theory, this milieu of cultural experiences of sexual

objectification leads to self-objectification, a crucial part of the overall model. Self-

objectification involves taking on the perspective of an observer upon one's own body. In

other words, girls and women come to "view themselves as objects or sights to be

appreciated by others" (Fredrickson & Roberts, 1997, p. 180). Objectification theory

posits that chronic self-objectification, in turn, promotes anxiety, reduces experiences of

peak motivational states or flow experiences (i.e., periods of intense concentration on a

challenging and rewarding task), decreases awareness of internal bodily states (e.g., heart

rate, sexual arousal), and increases body shame. These experiences in turn result in

depression, sexual dysfunction, and eating disorders among women.

Within this larger framework of objectification theory (Fredrickson & Roberts,

1997), the links among sexual objectification experiences, self-objectification, and body









shame have been posited as the most critical to understanding eating disorder

symptomatology. Empirical research supporting each of the relevant relationships in the

model is described in depth below.

Self-Objectification and Its Link to Body Shame and Eating Disorder
Symptomatology

To date, the relationships proposed in objectification theory among self-

objectification, body shame, and eating disorder symptoms have received much empirical

support. One such example exists in an experiment conducted by Fredrickson, Roberts,

Noll, Quinn, and Twenge (1998). According to objectification theory, self-objectification

is likely to occur in situations in which a woman's sense of her body is accentuated and

she is made more aware of her body's appearance. Fredrickson et al. (1998) elicited such

a situation by manipulating the type of clothing worn by participants. Within two separate

experiments, 114 college women were assigned randomly to a swimsuit or sweater

condition. In the first experiment, 75% of participants were Caucasian, 10% Asian, 7%

Hispanic, and 7% identified with other (unspecified) ethnicities. In the second

experiment, 83% of participants identified as Caucasian, 6% African American, 5%

Asian, 2% Hispanic, and 4% identified with other (unspecified) ethnicities. Information

regarding sexual orientation was not collected from these women.

After trying on the designated clothing, participants were asked to look in a full-

length mirror to evaluate the clothing, and then sample a cookie for a mock taste test.

Body Mass Index (BMI) was calculated and controlled as a covariate in the analyses, in

order to account for any potential confounding effects of obesity. Results indicated that

women who were wearing a swimsuit reported significantly higher levels of self-

objectification and body shame, and ate fewer cookies (i.e., restrained eating) than did









women in the sweater condition. In other words, a situation that heightened self-

objectification resulted in restrained eating and higher levels of reported self-

objectification and body shame.

Noll and Fredrickson (1998) provided further support for the relationships among

self-objectification, body shame, and disordered eating symptomatology. Two samples of

undergraduate women attending Duke University (n= 93 and 111, respectively) were

administered questionnaires including the Revised Bulimia Test (BULIT-R; Thelen,

Farmer, Wonderlich, & Smith, 1991), the Eating Attitudes Test (EAT; Garner &

Garfinkel, 1979), the Revised Restraint Scale (Polivy, Herman, & Howard, 1988) and the

Self-Objectification and Body Shame Questionnaires (designed by the researchers). The

data was analyzed using multiple regression, with BMI held as a covariate.

Results indicated that self-objectification related positively with body shame and

symptoms of both bulimia and anorexia, and body shame related positively with both

anorexia and bulimia symptoms. Using Baron and Kenny's (1986) procedure to test for

mediation, the authors found that body shame acted as a mediator of the relationship

between self-objectification and disordered eating symptomatology. The mediational

model accounted for 35% of the variance in bulimic symptoms (p < .01) and 27% of the

variance in anorexia symptoms (p < .01). It is also important to note that a direct

relationship was found between self-objectification and symptoms of disordered eating.

Thus, body shame was a partial mediator of the link between self-objectification and

eating disorder symptomatology.

Further support for body shame as a mediator variable, linking self-objectification

to eating disorder symptomatology, was found in Tiggemann and Slater's (2001) study.









This study examined the tenets of objectification theory by administering, to two samples

of women, measures of general self-objectification, body surveillance (the specific

manifestation of self-objectification as persistent body monitoring), appearance anxiety,

flow, awareness of internal bodily states, body shame, and disordered eating. One group

of women consisted of 50 former dancers and the other of 51 undergraduate psychology

students. The former dancers were Caucasian women between the ages of 17 and 25 who

had studied classical ballet for at least two years but no longer engaged in dance. The

undergraduate students attended The Flinders University of South Australia, and most

(over 95%) identified as Caucasian. The ages of this group ranged from 17 to 24, and

none of the undergraduate participants had studied formal dance.

As predicted by the researchers, former dancers scored higher on the measures of

general self-objectification, body surveillance, and disordered eating symptomatology

than did the group of undergraduate students. In both samples, the relationships of self-

objectification and body surveillance to disordered eating symptomatology was mediated

by body shame, but not by anxiety, flow, or awareness of internal states. Finally, results

indicated that body surveillance emerged as the manifestation of general self-

objectification that accounted for unique variance in the reported symptoms of disordered

eating. That is, in this study of former dancers and non-dancers, body surveillance was a

significant unique predictor of eating disorder symptoms, F(1,90) = 31.30, p < .001, but

general self-objectification was not, F (1,90) = 0.43, p > .05. This finding suggests that it

is important to include assessment of body surveillance as the critical manifestation of

self-objectification in the context of the objectification theory model.









In sum, the studies described above all were consistent with objectification theory's

proposition that body shame is a key mechanism through which self-objectification is

translated into eating disorder symptoms. Furthermore, their findings indicate that a

direct relationship also exists between self-objectification and eating disorder

symptomatology. Therefore, in congruence with extant research, the current study will

examine body shame as a partial mediator of the link of body surveillance (the critical

manifestation of self-objectification) to eating disorder symptomatology.

Sexual Objectification Experiences and Internalization of Sociocultural Standards
of Beauty

Fredrickson and Roberts (1997) identified sexual objectification experiences as a

key precursor to self-objectification, body shame, and eating disorder symptoms in

objectification theory. However, few studies have included this proposed role of sexual

objectification experiences when examining the model. Hill (2002) began to address this

gap in her examination of the relationship between sexual objectification experiences and

self-objectification.

In Hill's (2002) study, sexual objectification experiences were defined as

experiences in which women are treated as sexual objects, and as a result, "become their

bodies" and are evaluated as such (Hill, 2002, p. 5). Experiences of sexual objectification

were assessed with a questionnaire designed by Hill, containing 40 items measuring

sexualized gaze (with and without verbal comments), instances of sexual harassment, and

sexual assault. The CSOS (Cultural Sexual Objectification Scale) was derived from other

measures such as the Sexual Victimization Measure (SWV; Belknap, Fischer, & Cullen,

1999) and the Sexual Experiences Questionnaire (SEQ; Fitzgerald, Shullman, Bailey,

Richards, Swecker, Gold, Ormerod, & Weitzman, 1988). Self-objectification was defined









as "the extent to which individuals view their bodies in observable, appearance-based

(objectified) terms versus non-observable, competence-based (non-objectified) terms"

(Noll & Fredrickson, 1998, p. 628). This variable was assessed using the Self-

Objectification Questionnaire designed by Noll and Fredrickson (1998), and the

Objectified Body Consciousness Scale (OBC; McKinley & Hyde, 1996).

Over 500 mostly White women (307 heterosexual, 33 bisexual, 155 lesbian, and 7

who did not report sexual orientation) participated in this web-based survey. Participants

were recruited from introductory psychology classes at a large public Midwestern

university (n = 101), and techniques such as snowball sampling and email listserves also

were used (n = 340). In addition, over 300 letters were mailed to randomly selected

university faculty and staff, and of those women, approximately 12% accessed the

website to complete the survey (n = 61). Upon accessing the website, participants were

asked to complete questionnaires measuring reported experiences of sexual

objectification as well as self-objectification. Results indicated that reported experiences

of sexual objectification were related positively to self-objectification. However, this

relationship was moderated by age, such that the magnitude of the relationship between

sexual objectification experiences and self-objectification was strong and positive for

women between the ages of 50 and 79, but non-significant for those between the ages of

18 and 49 years old. This interaction effect might be explained by the potential larger

accumulation of objectification experiences for older women than for younger women.

This study adds further support for the inclusion of sexual objectification experiences

when examining the overall objectification theory model.









Additional evidence suggests that internalization of sociocultural standards of

beauty, a construct not explicitly included in the objectification theory framework and not

assessed in Hill's (2002) study, might also be an important factor in understanding the

role for sexual objectification experiences in eating disorder-related attitudes and

behaviors. That is, internalization of sociocultural standards of beauty might be a key

mechanism that translates sexual objectification experiences into self-objectification and

other eating disorder-related variables. Support for this relationship can be found in a

study conducted by Morry and Staska (2001). In this study, 89 female introductory

psychology students completed measures designed to assess exposure to fitness and

beauty magazines (a specific type of sexual objectification experience), awareness and

internalization of sociocultural attitudes and standards regarding appearance, self-

objectification, eating disorder symptomatology, and body shape satisfaction. The

Sociocultural Attitudes Toward Appearance Questionnaire (SATAQ; Heinberg,

Thompson, & Stormer, 1995) was used to assess awareness (i.e., recognition of societal

standards) and internalization (i.e., acceptance of sociocultural standards of beauty).

Statistical analyses revealed that sexual objectification experiences, operationalized

as exposure to beauty magazines, were related to higher levels of reported disordered

eating symptoms and body shape dissatisfaction, but these links were mediated by

internalization of sociocultural standards of beauty. That is, exposure to beauty

magazines (but not fitness magazines) was related to greater acceptance of society's

beauty mandates, and such internalization in turn was related to higher levels of self-

objectification, body shape dissatisfaction, and eating disorder symptomatology. Morry

and Staska's (2001) study was an important contribution to objectification theory









research. A limitation of this study, however, is that body shame was not assessed

directly. In addition, Morry and Staska's study focused on only one manifestation of

sexual objectification experience (i.e., exposure to beauty magazines) and did not assess

broadly the range of sexual objectification experiences highlighted in objectification

theory.

In one of the most comprehensive studies of the objectification theory framework,

Moradi et al. (2005) found additional support for the overall model, and further examined

the roles of both sexual objectification experiences and internalization of sociocultural

standards of beauty. In this study, over 200 undergraduate women in a large southeastern

university completed surveys assessing their reported experiences of sexual

objectification, internalization of sociocultural beauty standards, self-objectification

(manifested through body surveillance), body shame, and eating disorder

symptomatology. A path analysis was conducted using AMOS 4.01 (Arbuckle, 1999),

and BMI was controlled as a covariate. Results indicated that reported experiences of

sexual objectification related positively to body surveillance, body shame, and eating

disorder symptoms. In addition, reported sexual objectification experiences related

positively to internalization, which in turn was linked positively to body surveillance,

body shame, and eating disorder symptoms. Moradi et al. (2005) found evidence for the

mediational role of internalization in the objectification theory framework. More

specifically, they found that internalization of sociocultural standards of beauty acted as a

partial mediator, linking reported experiences of sexual objectification to body

surveillance, and also mediated the link of reported experiences of sexual objectification

to body shame and eating disorder symptoms. Moradi et al. (2005) also found support for









the previously described mediating role of body shame in the relationship between body

surveillance and reported symptoms of disordered eating. The overall model provided a

very good fit to the data and accounted for 50% of the variance in eating disorder

symptomatology.

Taken together, the studies conducted by Hill (2002), Morry and Staska (2001), and

Moradi et al. (2005) supported the role of sexual objectification experiences in the

objectification theory framework. In addition, they demonstrated the importance of

attending to the role of internalization of sociocultural standards of beauty. However,

only Moradi et al. (2005) included all variables related to disordered eating

symptomatology as described in objectification theory. The current study will build on

Moradi et al.'s study by examining their model, which includes relationships among

reported experiences of sexual objectification, internalization of sociocultural standards

of beauty, body surveillance, body shame, and eating disorder-related attitudes and

behaviors, with a sample of lesbian women.

Objectification Theory as it Applies to Lesbian Individuals

The tenets of objectification theory have been examined with populations such as

college women and men (Fredrickson et al., 1998; Huebner & Fredrickson, 1999; Morry

& Staska, 2001; Muehlenkamp & Saris-Baglama, 2002; Noll & Fredrickson, 1998;

Tiggemann, & Slater, 2001), older women (Tiggemann & Lynch, 2001), adolescent

dancers and non-dancers (Slater & Tiggemann, 2002), and women who exercise (Strelan,

Mehaffey, & Tiggemann, 2003). However, to date, only one study has examined the

theory as it applies to lesbian persons (Hill, 2002).

In Hill's (2002) study of objectification theory, sexual orientation and age were

examined as potential moderators of the relationship between sexual objectification









experiences and self-objectification. Consistent with objectification theory, in the entire

sample (i.e., heterosexual and lesbian women), experiences of sexual objectification were

related positively to self-objectification. In addition, sexual orientation did not moderate

the link between sexual objectification experiences and self-objectification, as the

relationship was similar for lesbian and heterosexual women. Although Hill's study

began to address the gap in data on the applicability of objectification theory to lesbian

persons, it represented only a partial investigation of objectification theory because it did

not examine key constructs such as internalization, body shame, and disordered eating

symptomatology. In addition, the sample only consisted of White heterosexual women

and White lesbian women; thereby limiting the generalizability of findings.

Thus, the generalizability of the propositions of objectification theory to

understanding eating disorder symptomatology among lesbian women is not known and

research is needed to examine the applicability of objectification theory to understanding

eating disorder symptoms among a diverse sample of lesbian persons. Furthermore, such

research must attend to any unique factors, not included in objectification theory, that

might shape lesbian women's experiences of eating disorder-related attitudes and

behaviors. Specifically, extant scholarship has identified two variables, connection with

the lesbian community and feminist ideology, as key potential factors that might be

related to eating disorder-related attitudes and behaviors among lesbian persons. The

following sections provide an overview of available data on eating disorder

symptomatology among lesbian women and review extant literature on the links of

connection with the lesbian community and feminist ideology to eating disorder-related

attitudes and behaviors.









Lesbian Women and Eating Disorder Symptomatology

Findings are mixed as to whether or not lesbian women experience different levels

of eating disorder symptoms and body satisfaction when compared with heterosexual

women (Beren, Hayden, Wilfley, & Striegel-Moore, 1997). Some studies have revealed

no significant differences between lesbian and heterosexual women on measures of body

dissatisfaction and eating disorder symptoms (e.g., Brand, Rothblum, & Solomon, 1992;

Heffernan, 1996; Striegel-Moore, Tucker, & Hsu, 1990). Other studies, such as Share and

Mintz (2002), found that lesbian women experience lower levels of body dissatisfaction,

but report similar levels of eating disorder symptoms when compared with heterosexual

women. Conflicting results obtained by Siever (1994), showed that lesbian women

exhibit lower rates of eating disorder symptoms but report similar levels of body

dissatisfaction in comparison with heterosexual women. Finally, some research studies

have demonstrated that lesbian women experience higher levels of body satisfaction and

lower levels of disordered eating symptoms than heterosexual women (e.g., Herzog,

Newman, Yeh, & Warshaw, 1992, Lakkis, Ricciardelli, & Williams, 1999). Examples of

research supporting each of these three conflicting findings are described below.

Striegel-Moore et al. (1990) compared 30 lesbian undergraduate women with 52

heterosexual graduate women using questionnaires measuring self-esteem, body esteem,

and disordered eating attitudes and behaviors, and very few group differences were

found. Using MANOVA, no significant differences were found on the measures of body

image satisfaction and symptoms of disordered eating. The differences that were found

between the two groups were related to self-esteem, with the lesbian women reporting

lower self-esteem than the heterosexual women. The researchers concluded that lesbian

and heterosexual college students do not differ in terms of body esteem and eating









disorder symptoms, and that perhaps the lesbian ideology of rejecting culture's narrowly

defined beauty ideals is not enough to overcome the socialized, internalized beliefs about

female beauty that women encounter.

Beren, Hayden, Wilfley, and Grilo (1996) also found no significant differences in

body esteem and body dissatisfaction between heterosexual women and lesbian women.

In this study, 257 participants (69 lesbian women, 72 heterosexual women, 58 gay men,

and 58 heterosexual men) completed measures of body dissatisfaction, self-esteem, self-

consciousness, affiliation with the lesbian/gay community, and sexual orientation. A

MANOVA revealed that the lesbian and heterosexual women in this sample scored

similarly on each of the body dissatisfaction measures, suggesting that identifying as a

lesbian person may not be enough to overcome society's pressure to conform to the ideal

body type. Unfortunately, this study failed to include any type of measure of eating

behavior, and therefore conclusions regarding eating disorder symptomatology between

lesbian and heterosexual women could not be ascertained.

Share and Mintz (2002) addressed this gap in their examination of the differences

between lesbian and heterosexual women on body esteem, awareness and internalization

of cultural attitudes concerning thinness, disordered eating symptoms, physical condition,

and sexual attractiveness. This study was based on a sample of 173 women between the

ages of 24 and 52. A total of 102 (59%) participants identified as exclusively or primarily

heterosexual, 63 (36%) described their sexual orientation as exclusively or primarily

homosexual, and 8 (5%) reported identifying as bisexual. The participants completed the

Eating Attitudes Test (EAT-26; Garner, Olmsted, Bohr, & Garfinkel, 1982) to assess for

disturbed eating patterns, the Body Esteem Scale (BES; Branzoi & Shields, 1984), and









the Sociocultural Attitudes Toward Appearance Questionnaire (SATAQ; Heinberg et al.,

1995). Body mass index was also obtained and controlled as a covariate in MANCOVAs

used to compare the two groups.

Statistical analyses revealed that lesbian women reported higher levels of body

esteem and lower levels of internalization of cultural beauty standards, but no differences

were found on levels of awareness of cultural standards, disordered eating

symptomatology, or body esteem. The authors concluded that although lesbian women

are equally aware of sociocultural standards of beauty, they are less likely to internalize

the cultural attitudes. However, the non-significant differences in eating disorder

symptomatology may be an indication that identification as a lesbian woman may not

offer enough of a buffer from disturbed eating.

Siever (1994) also examined eating disorder-related attitudes and behavior in his

study of 250 students from the University of Washington and Seattle Central Community

College. The sample included 53 lesbian women, 59 gay men, 62 heterosexual women,

and 63 heterosexual men. Participation involved responding to three versions of the Body

Esteem Scale (Franzoi & Herzog, 1986, 1987; Franzoi & Shields, 1984), the Body Shape

Questionnaire (BSQ; Cooper, Taylor, Cooper, & Fairburn, 1987), Body size drawings

(Stunkard, Sorensen, & Schulsinger, 1980), the Eating Disorder Inventory (EDI; Garner,

Olmstead, & Polivy, 1983), the Eating Attitudes Test (EAT-26; Garner et al., 1982), and

a demographic questionnaire designed by the researchers. Results indicated that

heterosexual and lesbian women reported similar levels of body dissatisfaction (with

lesbian women slightly, but not significantly less dissatisfied), but lesbian women scored

lower on the measures of eating disorder symptomatology. The author concluded that









socioculturall factors can have an immunizing effect lesbians, because of a decreased

emphasis on physical appearance in their community, appear to be less vulnerable to the

attitudes and behaviors that typify eating disorders" (p. 257).

Other studies have shown that lesbian women report lower levels of concern with

weight and physical appearance, internalization of sociocultural standards of beauty, and

disordered eating symptomatology, but higher levels of ideal weights and body

satisfaction, than do heterosexual women (Abraham & Beumont, 1982; Strong,

Williamson, Netemeyer, & Geer, 2000). An example of such findings can be found in

Herzog et al.'s 1992 study. In this study, 109 unmarried women between the ages of 18

and 35 completed a demographic questionnaire, the Eating Disorders Inventory (EDI;

Garner et al., 1983), and a set of 12 female figure drawings developed by Fumham and

Alibhai (1983). In this task, the participants were instructed to select their current body

type and their ideal body type. The results showed that lesbian women weighed

significantly more than heterosexual women, desired a significantly heavier ideal weight,

were less concerned with physical appearance and weight, and reported a lower drive for

thinness. The authors concluded that lesbian women's higher rates of body satisfaction

and lower weight concern may be a factor in their lower rates of reported symptoms of

disordered eating.

Myers, Taub, Morris, and Rothblum (1998) used a qualitative research design to

search for an explanation of findings of difference versus no difference between lesbian

and heterosexual women. In their telephone interviews of 18 lesbian and 2 bisexual

women between the ages of 17 and 60 (mean age = 32), respondents reported either

"feeling freedom from society's norms after coming out" or still "feeling the pressure to









be thin." This finding highlights the importance of attending to individual differences

among lesbian women that shape their experiences of sexual objectification and eating

disorder-related attitudes and behaviors. Extant literature on eating disorder-related

attitudes and behaviors among lesbian persons has identified two factors, connection with

the lesbian community and feminist ideology, as critical individual difference variables

that should be examined.

Connection with the Lesbian Community

One explanation for the conflicting pattern of findings regarding lesbian women

and eating disorder symptoms is that even though lesbian persons are exposed to the

same cultural messages as all women, sexual relations with other women may encourage

body acceptance and lower concern regarding appearance and weight (Beren, et al., 1997;

Siever, 1994). It may be that exposure to other lesbian women decreases the opportunity

to internalize sociocultural standards of beauty, subsequently contributing to higher body

esteem and lower rates of disordered eating symptomatology (Share & Mintz, 2002). This

lower level of internalization of sociocultural standards of beauty may be due to multiple

causes. For example, the lesbian subculture may be more accepting of a variety of body

types and shapes (Siever, 1994). An example of this can be found from one respondent in

the Myers et al. (1998) study who stated that in general, the lesbian community is more

accepting of larger women.

In their interviews with 26 lesbian, liberal arts college students, Beren et al. (1997)

elicited opinions regarding lesbian beauty ideals and their sources, the experience of

conflict regarding beauty in society, the need to overcome negative stereotypes as a

lesbian woman, and concerns about feminist identity that may potentially influence

feelings about one's body. Results indicated that lesbian women reported a conflict









between lesbian ideology and cultural values regarding beauty, but that intimate

involvement with other lesbian women positively influenced feelings about their bodies

and decreased the importance of appearance. Thus, connection with others who identify

as lesbian may lead to positive health outcomes such as higher body esteem and reduced

symptoms of disordered eating.

Ludwig and Brownell (1999) further examined this possibility by studying the

relationship between gender roles, group affiliation, and body satisfaction in a sample of

188 lesbian and bisexual women. Participants in this study were recruited through the

Internet and email, and thus a wide range of ages was represented among participants.

Information regarding race and ethnicity, however, was not assessed. Results indicated

that lesbian persons with friends who also identified as lesbian reported a more positive

body image than lesbian persons who reported having mostly heterosexual friends.

Unfortunately, Ludwig and Brownell's sample sizes were rather unbalanced in that over

80 women reported having mostly friends who were also lesbian or bisexual, and this

group was compared with 27 women reporting having mostly heterosexual female

friends, 15 women reporting having mostly gay or bisexual male friends, and 11 women

reporting having mostly heterosexual male friends. Furthermore, body image was

assessed using a single measure of body satisfaction, and information pertaining to other

eating disorder-related attitudes or symptoms was not obtained. Nevertheless, these

findings suggest that interaction with women of the same sexual orientation might be

related to positive body image for lesbian women.

Heffernan (1996) examined directly the relationship between connection with the

lesbian and gay community and eating disorder symptomatology. In this study,









questionnaires were used to examine lifestyle, disordered eating symptomatology, body

esteem, attitudes about attractiveness, and self-esteem in 203 lesbian women between the

ages of 17 and 65 (mean age = 34). Involvement with lesbian/gay community was

assessed with the question "How involved are you in lesbian/gay activities?" Participants

were asked to rate their response using a Likert-type scale. Findings indicated that the

rate of bulimia nervosa among lesbian women was comparable to that of heterosexual

women, but binge eating disorder was more frequent in the lesbian women sample.

Lesbian women were not significantly different from heterosexual women regarding their

attitudes about weight and appearance, or dieting. But among the lesbian women, higher

levels of participation in lesbian and gay activities and organizations was related to lower

weight concern.

Thus, between group analyses suggested overall similarity between lesbian and

heterosexual women's eating disorder-related attitudes and behaviors. Within group

analyses, however, revealed that among the lesbian women in the sample, involvement in

the lesbian/gay community was related to better health outcomes such as lower weight

and shape concern. Heffernan (1996) concluded that connection and involvement with

the lesbian/gay culture might reduce internalization of society's thin beauty ideal. Given

that Heffernan used only a single item to assess involvement in the lesbian/gay

community, however, more comprehensive assessment of this important construct is

needed in future research and the present study addressed this need.

Feminist Ideology

Feminist ideology is a second important individual difference variable that might

shape lesbian women's eating disorder-related attitudes and behaviors. Indeed, previous

research has demonstrated that women's endorsement of feminist ideology is related to









lower levels of eating disorder-related attitudes and behaviors (e.g., Dionne, Davis, Fox,

& Gurevich, 1995; Snyder & Hasbrouck, 1996). For example, Dionne et al. (1995)

studied the relationship between feminist attitudes and body satisfaction in a sample of

200 primarily White women between the ages of 17 and 48. Participants were volunteers

solicited from the student, staff, and faculty population at a large Canadian university.

The authors reported recruiting a diverse sample, but did not include descriptive

information about the ethnic and racial background of the participants. Participation

involved completing the Body Cathexis Scale (Secord & Jourard, 1953) to assess general

body dissatisfaction, the EDI (Garner & Olmsted, 1984) to assess specific body

dissatisfaction, and the Composite Feminist Ideology Scale (CFIS; Dionne, 1992) to

measure the degree of support for the tenets of the women's movement. Results revealed

that women's feminist attitudes regarding physical attractiveness (i.e., the rejection of

traditional societal beauty standards) was related significantly and positively to body

satisfaction.

Leavy and Adams (1986) also examined feminism and its potential link to positive

health outcomes such as social support, self-esteem, and self-acceptance. Questionnaires

were used to examine the relationship between feminism, self-esteem, self-acceptance,

and social support in a sample of 123 women who identified as either predominantly or

exclusively homosexual. The ages of participants ranged from 15 to 52 (mean age = 26),

and 98% of the sample identified as White. In their study, Leavy and Adams defined

feminism as having two components: "a set of beliefs about women's rights" and

"involvement in feminist activities" (p. 322). Thus, feminism was assessed through a

two-component questionnaire adapted by the authors. The first component contained









seven items measuring strength of agreement with feminist beliefs concerning sex roles

and political action, and one self-perception item that required participants to rate their

perception of themselves as feminists. The second component of the questionnaire

included an inventory of feminist activities, and respondents were required to report their

level of participation on a 5-point Likert scale (from never to many times).

Statistical analysis revealed that lesbian women who reported being active in

feminist organizations possessed better social support systems, higher self-esteem, and

greater self-acceptance than those who did not. Feminist beliefs, however, were not

correlated significantly with self-esteem, social support, or self-acceptance. These

findings must be interpreted in light of the fact that participants' endorsement of feminist

beliefs was extremely positively skewed and such severe range restriction attenuates the

potential observed relationship between feminist beliefs and the health related outcomes

examined in this study. Nevertheless, when feminist activity was examined, lesbian

women who reported being active in lesbian or feminist organizations reported having

better social support systems, higher self-esteem, and greater self-acceptance than those

who did not. The findings of this study suggest that involvement in feminist activities

might be related to positive health outcomes in lesbian women. Thus, it is possible that

this link might also generalize to eating disorder symptomatology.

Snyder and Hasbrouck (1996) examined this possibility in a study of the

relationship between feminist identity development attitudes and symptoms of disturbed

eating in a sample of 71 female college students between the ages of 17 and 22. In this

study, the relationships between feminist identity development attitudes, gender traits,

and eating disorder symptomatology were examined through the use of questionnaires.









The sample was predominantly White (95%), and the study was conducted at a middle

class, liberal arts college in which research participation credit was given as

compensation.

Results indicated that those women who endorsed feminist values reported less

dissatisfaction with their body weight and overall size, less concern for thinness, fewer

bulimic symptoms, and fewer feelings of ineffectiveness. More specifically, passive

acceptance of traditional gender-role stereotypes (Feminist Identity Development Passive

Acceptance scores) related positively to drive for thinness and body dissatisfaction,

whereas, active commitment to a feminist ideology (Feminist Identity Development

Active Commitment scores) related negatively to those same outcome measures.

Although this study provided some support for the potential role of feminist ideology in

reducing eating disorder attitudes and behaviors, the generalizability of these findings to

lesbian women was not examined. Given that lesbian women are more likely to identify

with feminism than are heterosexual women (Guille & Chrisler, 1999), however, feminist

ideology may also play a significant role when examining eating disorder-related

attitudes and behaviors among lesbian women.

This possibility was addressed in Guille and Chrisler's (1999) research. In this

study, 217 women were recruited from college campuses, community groups, bookstores,

and lesbian support and activist groups from the Connecticut, Boston, and San Francisco

areas. This sample consisted of 52 adult lesbians between the ages of 25 and 70 (mean

age = 38), 56 adult heterosexual women between the ages of 25 and 84 (mean age = 37),

51 young adult lesbian women between the ages of 16 and 24 (mean age = 20) and 58









young adult heterosexual women between the ages of 15 and 24 (mean age =19) who

participated in a study of "women and body image."

Participation involved the completion of questionnaires measuring feminist identity

and eating disorder symptoms. Feminist identity was assessed using the Feminist Identity

Scale (FIS; Worell & Remer, 1992), which consists of four subscales: acceptance (of

traditional gender roles), revelation (realization of sexism), embeddedness (immersion in

female culture), and commitment (active work to improve the status of women). Eating

disorder symptomatology was measured using the Eating Attitudes Test (EAT; Garner &

Garfinkel, 1979), the Compulsive Eating Scale (CES; Dunn & Ondercin, 1981), and the

Three-Factor Eating Questionnaire (TFEQ; Stunkard & Messick, 1985). Taken together,

these last three questionnaires were used to assess attitudes and behaviors associated with

bulimia and anorexia nervosa, food obsessions, beliefs regarding the ability to resist the

urge to eat, dietary restraint, disinhibition (i.e., difficulty in ceasing once one has started

eating), and general hunger.

The researchers used both univariate analyses and standard multiple regression to

analyze the obtained data. Results indicated that lesbian women were significantly more

likely to endorse feminist attitudes than were heterosexual women. That is, lesbian

women scored significantly lower on the acceptance subscale and significantly higher on

the commitment subscale than did the heterosexual women. In other words, lesbian

women in this study were significantly more likely to report actively working to improve

the status of all women than heterosexual women, and at the same time were less likely to

report attitudes and behaviors associated with eating disorders. In addition, for all

women, those with higher scores on the commitment subscale were less likely to restrict









their food intake than those with higher acceptance scores. Based on these findings,

Guille and Chrisler posited that feminist ideology might reduce eating disorder-related

attitudes and behaviors. However, the generalizability of Guille and Chrisler's findings is

unclear since the authors did not report the demographic characteristics of their sample

such as race, ethnicity, and socioeconomic status. Furthermore, BMI was obtained for the

sample, but since it was not controlled as a covariate in the analyses, the potential for

confounding exists.

In a more comprehensive study, Bergeron and Senn (1998) examined attitudes

regarding the body, awareness and internalization of sociocultural standards of beauty,

and feminist self-identification with a sample of 108 lesbian and 115 heterosexual women

between the ages of 18 and 58. Participants completed a demographic measure, the Body

Attitude Questionnaire (BAQ; Ben-Tovim & Walker, 1991) and the Sociocultural

Attitudes Towards Appearance Questionnaire (SATAQ; Heinberg et al., 1995). Feminist

identification was assessed using the question "Would you describe yourself as a

feminist?" A MANOVA revealed that lesbian women reported significantly higher ideal

weights, and reported feeling stronger and more fit than their heterosexual counterparts.

A standard multiple regression revealed that in the entire sample, internalization and

feminist identification both were unique predictors of body attitudes, above and beyond

sexual orientation. No differences were found between heterosexual and lesbian women

on awareness of sociocultural standards of beauty, and this variable did not predict body

attitudes. For all women, internalization predicted body attitudes, and lesbian women

were less likely to report internalization of sociocultural norms than heterosexual women.









A notable strength of this study is the large sample size obtained by snowball

sampling. However, over 95% of the participants identified as White. Furthermore,

feminist identity was assessed using only a single item and the authors did not examine

directly eating disorder symptomatology. Nevertheless, these findings ofBergeron and

Senn (1998) suggest the importance of examining both feminist identification and

internalization when examining body attitudes.

Further evidence of the relationship between feminist ideology and eating disorder

symptomatology exists in Cogan's (1999) study of 181 lesbian and bisexual women

recruited from Sacramento, California. Participants were between the ages of 17 and 58

(mean age = 34), with 88% identifying as lesbian, and 12% identifying as bisexual. In

terms of ethnicity, 73% of the sample identified as White, 8% as Latina, 7% as Asian

American, 6% as African American, 4% as Native American, and 2% as multiracial.

Participants completed measures assessing reasons for exercise, fitness activity

frequency, type, and duration, dieting behavior, body satisfaction, eating disorder

symptoms, physical appearance before and after coming out, feminist self-identification,

and feminist ideology. To assess for feminist self-identification, participants responded to

the question "How much do you consider yourself a feminist?" and to assess for feminist

ideology, participants completed the 10-item Attitudes Toward Feminism and the

Woman's Movement scale (FWM; Fassinger, 1994).

Using ANOVAs (holding BMI constant), Cogan (1999) found that when compared

with a sample of heterosexual women (Cogan, Bhalla, Sefa-Dedeh, & Rothblum, 1996),

these lesbian women demonstrated similar drive for thinness and body dissatisfaction.

Controlling for BMI and age as covariates, it was found that those who labeled









themselves as a feminist (feminist self-identification) and endorsed feminist ideology

were overall more satisfied with their bodies than those who did not. More specifically,

higher levels of feminist self-identification and endorsement of feminist ideology were

related to higher body satisfaction, lower rates of bulimia, drive for thinness, and weight

discrepancy, and tendency to exercise for health versus aesthetic reasons. Cogan (1999)

concluded that feminism might provide a means by which women can unlearn a negative

body image that had been internalized by repeated exposure to thin body ideals.

Unfortunately, this study did not assess for the internalization of such ideals.

Thus, extant research suggests that feminist ideology should be included in

examination of disordered eating symptomatology among lesbian persons. However,

since it is still unknown whether or not it is connection with the lesbian community,

feminist ideology, or both that play important roles in the internalization of sociocultural

standards of beauty and the manifestation of disordered eating symptoms in lesbian

women, both were included in the current research. That is, this study examined both

connection with the lesbian community and feminist ideology as key predictor variables

in the overall objectification theory framework.

In conclusion, this study aimed to expand upon the existing objectification theory

literature by providing a comprehensive examination of the framework as it applies to a

diverse sample of lesbian women. The simultaneous examination of sexual

objectification experiences, internalization of sociocultural standards of beauty, self-

objectification manifested through body surveillance, body shame, and eating disorder

symptomatology builds on prior work by Moradi et al. (2005) and provides a

comprehensive test of the objectification theory framework as applied to eating disorder









symptomatology. In addition, the present research included examination of two key

constructs, connection with the lesbian community and feminist ideology, that have been

identified in extant literature as important factors that might shape lesbian women's

experiences of eating disorder-related attitudes and behaviors. The present study included

a more comprehensive assessment of connection with the lesbian community and

feminist ideology than in prior studies of these variables, thus allowing for a thorough

investigation of their roles in the overall objectification theory framework.

Hypotheses

Based on the literature on objectification theory and extant research on eating

disorder-related attitudes and behaviors with lesbian persons, the present study will test a

model (See Figure 1) that includes the following hypotheses:1). Reported experiences of

sexual objectification will be related positively to body surveillance, body shame, and

eating disorder symptomatology.

2). A negative relationship will exist between feminist ideology and disconnection

from the lesbian community.

3). Disconnection from the lesbian community will be related positively to

internalization of sociocultural standards of beauty, body surveillance, body shame, and

eating disorder symptomatology.

4). Feminist ideology will be related negatively to internalization of sociocultural

standards of beauty, body surveillance, body shame, and eating disorder

symptomatology.

5). Internalization of sociocultural standards of beauty will mediate the links of

sexual objectification experiences to body surveillance, body shame, and eating disorder

symptomatology (as found in Moradi et al., 2005).






40


6). Body surveillance will mediate the link of sexual objectification experiences to

body shame (as found in Moradi et al., 2005).

7). Body shame will partially mediate the relationship between body surveillance

and eating disorder symptomatology as found in previous research (Moradi et al., 2005;

Noll & Fredrickson, 1998).














CHAPTER 3
METHODS

Participants

A total of 616 persons participated in this web-based, survey study (after three

duplicate submissions were discarded). Of these 616 participants, 32 persons were

excluded because they self-identified as a man (n = 2), heterosexual (n = 2), or bisexual

(n = 28), and therefore did not meet the inclusion criteria outlined in the invitation to

participate. Additional respondents were excluded from analyses due to substantial

missing data (n = 17), or failure to correctly respond to at least 5 of 6 validity check items

(n = 36). Two women who did not indicate sexual orientation were included in the

sample because their responses to questions regarding physical attraction, emotional

attraction, and sexual behaviors indicated that they were mostly or exclusively attracted

to and had sex with women. Thus, the final sample used for analyses in the present study

included 531 persons between the ages of 18 and 69 (M= 35.39; Mdn = 34.0; SD=

11.68) who self-identified as either exclusively (73%; n = 385) or mostly (27%; n = 144)

lesbian.

Participants were from a variety of geographic locations with 94% reporting that

they were currently living in the United States (n = 498) and 6% reporting that they were

currently living in other countries (n = 30). Three participants did not indicate their

current residence. As for the 498 participants residing in the United States, reports of

states of residence indicated that 27% were living in the South (n = 136), 24% in the

West (n = 117), 20% in the Midwest (n = 98), 14% in the Southwest (n = 70), 11% in the









Middle Atlantic (n = 56), and 4% in New England (n = 21) regions of the United States

(regional categories derived from www.infoplease.com which divided locations in the

United States according to similarities in climates, geographies, traditions, and histories).

Of the 30 participants living in countries other than the United States, 37% reported

living in South Korea, 27% in Canada, 10% in Australia, 10% in India, 3% in the United

Kingdom, 3% in Ecuador, 3% in New Zealand, 3% in Norway, and 3% in Romania.

These participants' correct responses to validity check items indicated that they were able

to read and understand the questions throughout the survey.

With regard to race/ethnicity, 77% of the sample identified as White/Caucasian, 7%

as African American/Black, 5% as Hispanic/Latina, 3% as Asian American/Pacific

Islander, 1% as American Indian/Native American, and 7% as multi-racial or other. Forty

percent of the sample reported being married/partnered, 30% identified as single, 22%

reported being in a long-term dating relationship, and 8% reported being in a casual

dating relationship. In terms of highest educational degree obtained, 33% of the sample

had obtained a professional degree (e.g., M.A., M.S., Ph.D., M.D.), 32% had obtained a

college degree (e.g., B.A., B.S.), 30% had completed some college, and the remaining

5% had completed either a high school degree, some high school, or less than a high

school degree. In terms of employment, 64% reported that they were employed full time,

21% part time, and 15% were not employed. With regard to social class, 54% identified

as middle class, 27% as working class, 17% as upper middle class, 2% as lower class, and

1 % as upper class; 3 participants did not indicate social class.

Instruments

The web survey was prepared by combining the measures described below into one

large questionnaire. The order of instruments for each of the two forms of the survey and









the assignment of participants to each form was determined randomly. Due to the fact

that not all measures were designed for lesbian persons specifically, two consultants who

were lesbian women, and one of whom had expertise in multicultural research, reviewed,

completed, and provided feedback about the survey. Following recommendations of the

consultants, slight grammatical changes were made to some items in order to increase

clarity. In addition, one item was adjusted slightly so that it was more appropriate for

lesbian women (see description of the SOS scale below).

Reported Sexual Objectification Experiences

The Sexual Objectification Subscale (SOS) of Swim, Cohen, and Hyer's (1998)

measure of daily sexist events was used to assess reported experiences of unwanted and

sexual objectifying behaviors and comments. The 7 SOS items reflect the objectification

experiences dimension of a larger pool of items developed based on daily diaries in

which college men and women reported gender-based unfair or differential treatment that

they observed or experienced. Because Hill (2002) found that reported experiences of

sexual objectification over the past year were significantly related to self-obj ectification,

and to make the instructions applicable to participants who were not students, instructions

for SOS items were adjusted to assess experiences over the past year (instead of past

semester as originally used by Swim et al.).

Furthermore, to obtain a more thorough assessment of sexual objectification

experiences, SOS items were supplemented with 6 of 18 items from Burnett's (1995)

Objectification Experiences Questionnaire (OEQ). These six items were chosen because

they tapped distinct experiences of sexual objectification from those assessed by SOS

items. Thus, participants reported their experiences of a total of 13 sexual objectification









events using a 5-point rating scale, ranging from 1 (never) to 5 (about two or more times

a week over the past year).

Because Hill (2002) found that experiences of sexual objectification perpetrated by

both men and women over the course of a year were related positively to self-

objectification, SOS and OEQ item instructions were adjusted so that participants

responded to each of the 13 items twice, once considering their experiences with men,

and again considering their experiences with women. Sample items from the SOS

include: "Had people shout sexist comments, whistle, or make catcalls at me" and

"Experienced unwanted staring or ogling at myself or parts of my body when the person

knew or should have known I was not interested or it was inappropriate for the situation

or our relationship." As recommended by the two consultants, the item "Had someone

refer to me with a demeaning or degrading label specific to my gender (bitch, chick,

bastard, faggot, etc) was adjusted to include the word "dyke" as additional example of a

derogatory term. Sample items from the OEQ include: "Someone stared at your breasts

while talking to you" and "Someone made offensive, sexualized gestures toward you

(e.g., pantomime of masturbation or intercourse)?"

In order to examine the appropriateness of averaging across SOS and OEQ item

ratings to compute an overall reported sexual objectification score, a principal

components factor analysis was conducted with the 13 sexual objectification experiences

perpetrated by women items (both SOS and OEQ). The scree plot and factor loadings

suggested that all items loaded substantially on a single factor (item loadings: .75, .74,

.74, .74, .68, .67, .66, .66, .65, .65, .61, .54, .44). Similarly, a principal components factor

analysis was conducted on the 13 sexual objectification experiences perpetrated by men









items (both SOS and OEQ). Again, results of the scree plot and factor loadings suggested

that items loaded substantially on a single factor (item loadings: .84, .83, .81, .80, .79,

.77, .75, .71, .68, .67, .63, .56, .48). Furthermore, to examine whether sexual

objectification experiences by men and women should be examined separately or could

be combined, an additional principal components factor analysis was conducted with all

26 sexual objectification experiences items (i.e., SOS and OEQ items for both women

and men). Factor loadings and a scree plot indicated that all items loaded substantially on

a single factor (item loadings: .77, .75, .75, .74, .73, .72, .71, .69, .69, .67, .67, .66, .62,

.61, .61, .61, .60, .59, .58, .56, .55, .55, .54, .50, .44, .41). Furthermore, scores on sexual

objectification experiences perpetrated by women and sexual objectification experiences

perpetrated by men were significantly correlated with each other (r = .62, p < .001) and

suggested substantial overlap in these scores. Thus, SOS and OEQ items regarding sexual

objectification experiences perpetrated by both men and women were combined into one

scale, and an overall reported sexual objectification experiences mean score was

computed for each participant.

Moradi et al. (2005) reported an alpha internal consistency reliability of .87 for

SOS scores. In terms of validity, Swim et al. (2001) found that women reported more

sexual objectification experiences than men, and that these along with other sexist events

were related more strongly to anxiety for women than for men. However, these reported

experiences of sexual objectification and other sexist events were not related to

neuroticism discriminantt validity).

With regard to OEQ scores, Burnett (1995) reported alphas that ranged from .69 to

.91 and test-retest reliability that ranged from .69 to .88 across a period of two to five









weeks. By conducting a factor analysis, Burnett (1995) found evidence for discriminant

validity in that sexual objectification experiences were distinct from gender harassment

and sexual abuse/coercion experiences. Scores on the OEQ also demonstrated convergent

validity, as they were moderately to strongly related to depression (Burnett, 1995), self-

objectification, body image disturbances, and disordered eating scores (Brownlow, 1997).

When the combination of SOS and OEQ items was examined, alpha internal

consistency reliabilities obtained for the current sample were .93 for sexual

objectification experiences perpetrated by men, .90 for sexual objectification experiences

perpetrated by women, and .94 for overall reported sexual objectification experiences

(used in the present study).

Connection/Disconnection with Lesbian Community

The Connection with the Lesbian Community subscale (CLC) of the Lesbian

Internalized Homophobia Scale (LIHS; Szymanski & Chung, 2001) was used to assess

connection with the lesbian community. The CLC subscale is appropriate for use in the

present study in that it assesses the extent to which a lesbian woman is

connected/disconnected from the larger lesbian community, and responses can

demonstrate isolation from the community, or social embeddedness in the community.

The CLC consists of 13 items to which participants responded using a 7-point Likert-type

scale ranging from "strongly agree" to "strongly disagree." Higher scores indicate greater

disconnection from the lesbian community. Sample statements from the Connection with

the Lesbian Community subscale include: "Attending lesbian events and organizations is

important to me" and "I am familiar with lesbian music festivals and conferences."

Scores are computed by reverse coding the appropriate items and then obtaining a mean

score.









In terms of reliability, the reported alpha for Connection/disconnection with the

Lesbian Community scores was .87 (Szymanski & Chung, 2001). With regards to

validity, Szymanski and Chung (2001) administered the LIHS to 303 female participants,

and findings indicated that as expected, disconnection from the lesbian community

correlated negatively with self-esteem (r = -.22, p < .01) and positively with loneliness (r

= .38, p < .01). Also supporting the validity of CLC scores, responses to the present

study's demographic question "How connected or involved are you in the lesbian

community?" were correlated significantly with CLC scores (r = -.67; p < .001). Alpha

for CLC scores with the current sample was .85.

Feminist Ideology

The Attitudes Toward Feminism and the Women's Movement (FWM) Scale

(Fassinger, 1994) was used to assess feminist ideology. The FWM is a brief, 10-item

questionnaire designed to measure agreement with feminism and the women's

movement. Participants were asked to respond to each of the 10 statements using a 5-

point rating scale from 1 (strongly disagree) to 5 (strongly agree). Sample items include:

"Feminist principles should be adopted everywhere" and "The leaders of the women's

movement may be extreme, but they have the right idea." Scores are obtained by reverse

coding appropriate items and then obtaining a mean, with higher scores indicating greater

endorsement of feminist ideology.

To examine reliability and validity, Fassinger (1994) administered the FWM to 117

(76 women and 41 men) undergraduate psychology students at a large eastern public

university. FWM score reliabilities were .90 for men, .87 for women, and .89 for the

entire sample. As for validity, FWM scores demonstrated adequate convergent and

discriminant validity. For example, Fassinger (1994) found that FWM scores were









independent from scores on measures of gender roles (r = .02, p > .05) but correlated

strongly and positively with scores on four other feminism scales (correlations ranging

from .68 to .79) and negatively with dogmatism (r = -.23, p < .05). Other evidence for

validity of FWM scores exists in Enns's (1987) study, in which convergent validity

coefficients ranged from .36 (for involvement in activities associated with feminism) to

.62 (for subjective identification with feminism). Enns also found discriminant validity

coefficients of .23 (for gender roles) and -.24 (for dogmatism). In the current study,

responses to the demographic question asking "To what extent do you describe yourself

as a feminist?" were significantly correlated with FWM scores (r = .71, p < .001), adding

further evidence of validity for FWM scores. The alpha internal consistency reliability

estimate for FWM scores with the current sample was .85.

Internalization of Sociocultural Standards of Beauty

The Internalization subscale of the Sociocultural Attitudes Toward Appearance

Questionnaire (SATAQ; Heinberg, Thompson, & Stormer, 1995) is an 8-item, 5- point

Likert-type scale that assesses acceptance of sociocultural standards of beauty. Sample

items include: "Women who appear in TV shows and movies project the type of

appearance that I see as my goal" and "Photographs of thin women make me wish that I

were thin." Participants responded on a scale ranging from 1 (completely disagree) to 5

(completely agree). Appropriate items were reverse coded and item ratings were then

averaged with higher scores on the Internalization subscale demonstrating greater levels

of internalization of sociocultural beauty standards.

In Heinberg et al.'s (1995) study of 194 female undergraduate students,

Internalization scores yielded an alpha of .88. In Morry and Staska's (2001) study, scores

obtained from a sample of 89 women and 61 men produced an alpha of .85. With regards









to validity, scores on the Internalization subscale have been shown to be distinct from

scores generated from awareness of sociocultural standards of beauty, but significantly

and positively related to disordered eating attitudes (Griffiths, Beumont, Russell, Schotte,

Thornton, Touyz, & Varano, 1999), restrained eating and body dissatisfaction (Griffiths,

Mallia-Blanco, Boesenberg, Ellis, Fischer, Taylor, & Wyndham, 2000), and

preoccupation with body image (Morry & Staska, 2001). Alpha with the current sample

was .87.

Body Surveillance as an Indicator of Sexual Objectification

The Body Surveillance subscale of the Objectified Body Consciousness Scale

(OBC; McKinley & Hyde, 1996) was utilized to assess self-objectification, or concern

with outward appearance as opposed to concern regarding how the body feels.

Participants responded to eight items using a 7-point Likert-type scale, ranging from 1

(strongly disagree) to 7 (strongly agree), but NA could be selected if the item was not

applicable. Following scoring instructions, appropriate items were reverse coded and

"NA" responses were coded as missing. Non-missing item ratings were then averaged in

order to yield a scale score, with higher scores indicating higher levels of body

surveillance. The Body Surveillance subscale includes items such as "I often worry about

whether the clothes I am wearing make me look good" and "During the day, I think about

how I look many times."

Scores on the Body Surveillance subscale have demonstrated adequate reliability.

In their sample of both young and middle-aged women, McKinley and Hyde (1996)

found alpha internal consistency reliability estimates ranging from .76 to .89.

Furthermore, a two-week test-retest was conducted, producing a reliability coefficient of

.79. In their sample of over 200 women, Moradi et al. (2005) reported an alpha of .82. In









terms of validity, consistent with objectification theory, women have been found to score

higher than men on Body Surveillance (McKinley, 1998). In addition, in a factor

analysis, Body Surveillance emerged as a construct that was distinct from other factors

such as body shame, although as expected, it was correlated positively with body shame

and negatively with body esteem (McKinley, 1998). Alpha internal consistency reliability

estimate with the current sample was .86.

Body Shame

Body shame was measured using the Body Shame Subscale of the Objectified

Body Consciousness Scale (OBC; McKinley & Hyde, 1996). This 8-item subscale

assesses guilt and negative feelings as a result of failing to live up to perceived cultural

standards. For example, one item reads: "When I can't control my weight, I feel like

something must be wrong with me." Participants responded using a 7-point Likert-type

scale, ranging from 1 (strongly disagree) to 7 (strongly agree), but NA could be selected

if the item was not applicable. Following scoring instructions, appropriate items were

reversed coded and "NA" responses were coded as missing. Non-missing items were then

averaged to produce a scale score in which higher scores indicate higher levels of body

shame.

OBC Body Shame subscale scores have demonstrated adequate reliability. Across a

sample of young and middle-aged women, alpha internal consistency estimates ranged

from .70 to .84 (McKinley & Hyde, 1996; McKinley, 1998). Moradi et al. (2005) found

an alpha estimate of .81. In terms of validity, as expected, Body Shame scores have been

shown to be correlated with but have emerged as a distinct factor from control beliefs and

body surveillance (McKinley & Hyde, 1996). McKinley (1998) found that scores on the









Body Shame subscale correlated negatively with body esteem and positively with body

surveillance. Alpha with the current sample was .87.

Eating Disorder Symptomatology

The Eating Attitudes Test-26 (EAT-26; Garner, 1997; Garner, Olmsted, Bohr, &

Garfinkel, 1982) consists of 26 questions designed to assess dieting behaviors, bulimia

and food preoccupation, and oral control. The EAT- 26 was selected because of its ability

to assess the continuum of disturbed eating attitudes and behaviors. This follows the

recommendation of Kashubeck-West, Mintz, and Saunders (2001) to examine the broad

range of eating attitudes and behaviors. Participants indicated the frequency of such

attitudes and behaviors using a 6-point, Likert-type scale, ranging from 1 (never) to 6

(always). As recommended by Garner (1997), ratings of each of the items were then

weighted from zero to three, with a score of three signifying the most "symptomatic"

responses. Total EAT scores were then obtained by summing all of the weighted item

scores. The minimum score that can be obtained on the EAT-26 is 0, and the maximum is

78, with higher scores indicating greater eating disorder symptomatology. Sample

statements include: "Engage in dieting behavior," "Feel extremely guilty about eating,"

"Find myself preoccupied with food," "Cut my food into small pieces," "Have the

impulse to vomit after meals," and "Avoid eating when I am hungry."

Scores on the EAT-26 have demonstrated good reliability and validity, yielding

Cronbach's alphas ranging from .90 to .93 (Share & Mintz, 2002). Kashubeck-West et al.

(2001) reported alphas ranging from .79 to .94 across samples. In terms of validity, scores

on the EAT-26 are related to scores on other measures of eating disorder

symptomotology and the EAT-26 has been used to distinguish between clinical and non-









clinical groups (Kashubeck-West et al., 2001). Alpha for EAT-26 scores in the current

sample was .86.

Demographics

Participants were asked to report several personal characteristics including age,

height and weight (used to compute body mass index), race/ethnicity, relationship status,

employment status, income, educational level, and social class. Sexual orientation was

assessed utilizing a Kinsey-type scale ranging from 1 (exclusively lesbian) to 5

(exclusively heterosexual). To provide a more through assessment of sexual orientation,

participants were also asked to rate on a scale of 1 (low) to 5 (high) their physical and

emotional attraction to both members of the same and opposite sex. They were also asked

a question regarding sexual behavior in which they indicated whether their sexual

interactions were with their same gender only, same gender mostly, both genders equally,

other gender mostly, or other gender only. They could also select "never had sex" if

appropriate.

The demographic questionnaire also included an item assessing degree of

connection with the lesbian community. The item read: "How connected or involved are

you in the lesbian community?" Participants responded using a 5-point Likert scale

ranging from 1 (not involved) to 5 (very involved). A single item was included to assess

feminist self-identification. Participants responded to the question "To what extent do

you describe yourself as a feminist?" by using a 9-point Likert scale ranging from 0 (not

at all) to 9 (very much a feminist), as used by Cogan (1999). These items were used as

further validity checks for CLC and FWM scores.









Procedure

Participants were recruited through personal contacts, lesbian and/or women's

organizations and internet listserves. Advertisements for the study were sent to 116

listserves that were selected due to their ability to reach lesbian women throughout the

United States. These listserves included national organizations such as the Gay and

Lesbian National Hotline, the National Coalition for LGBT health, and the Gay, Lesbian,

Straight Education Network (GLSEN). In addition, advertisements were sent out through

listserves of numerous lesbian and gay college organizations. Participants also were

recruited by sending email advertisements about the study to Yahoo and MSN groups for

lesbian and gay persons. Listserves that were not used for promotion of the study

included those that were designated for men, heterosexual or bisexual women, non-

English speaking women, and lesbian women under the age of 18. Flyers promoting the

study also were posted in the Gainesville, Florida community to promote the study. These

advertisements described the purpose of the research, and highlighted the need for

diversity in terms of age, race/ethnicity, and level of outnesss" in the sample. The

principal investigator attended club meetings, organizations, events, and other activities

targeted for the lesbian community to further advertise the study.

Participation involved completing a series of questionnaires online via a website

designed by the researcher. Although internet methodology can limit participation to

individuals who have access to a computer and the internet, it has a number of benefits

for recruiting large samples of lesbian or gay participants. More specifically, in the

present study, a web-based survey was selected because it provides greater anonymity for

lesbian participants. Maximizing anonymity and reducing interpersonal threat is

particularly important for recruiting individuals who are not "out" about their sexual









orientation and might not feel comfortable with participating in-person in a study about

lesbian or gay persons. Furthermore, the convenience of completing a survey online

(participants could log on to the website from any computer with an internet connection)

facilitated the recruitment of a larger and more diverse sample in terms of age and

geographic location.

When participants connected to the website, they were first shown an informed

consent page that described the purpose of the study, discussed issues of confidentiality,

informed respondents that they could stop filling out the survey at any time without

penalty, and provided contact information to participants who had questions or comments

about the study. After reading the informed consent information, participants clicked a

link stating: "Click here to proceed" to indicate that they had read and understood the

informed consent document. Clicking the button at the bottom of the screen brought

participants to one of two forms of the survey to be completed (the measures were

counterbalanced into two different forms in order to control for order effects).

Throughout the survey, participants encountered six validity check items that asked them

to choose a certain response. For example, one item read: "Please select Strongly Agree."

This procedure was used to help identify random responding and to provide some

indication that participants read and understood the questions. Upon completing the

survey, participants received information about how to contact the researcher if they had

any questions or concerns. They also received information for national support networks

for eating disorders and lesbian women.

Statistical Analyses

As mentioned previously, data from those who identified as a man, bisexual, or

heterosexual, as well as those who did not respond to substantial portions of the survey,









were excluded from analyses, resulting in a final sample size of 531 for the present

analyses. Body Mass Index was computed by calculating weight/height2 and was entered

as a covariate in analyses. This is consistent with previous research, as BMI has been

considered a potential confounding variable (Fredrickson, Roberts, Noll, Quinn, &

Twenge, 1998).

Reliability coefficients were calculated (and reported in the Instruments section) for

all measures before testing the overall path model. Partial-correlations, controlling for

BMI, were computed to examine the intercorrelations among the variables of interest.

Finally, all direct and indirect relations among variables of interest, depicted in Figure 1,

were examined by conducting a path analysis using AMOS, Version 5.0 (Arbuckle,

2003), a program designed specifically for structural equation modeling. The path

analysis provided the strengths of all unique links among the variables of interest, with

BMI controlled.

To test proposed mediator effects, guidelines outlined by Baron and Kenny were

followed (1986). According to Baron and Kenny, a mediator functions as such if it

accounts for all or some of the relationship between the predictor variable and the

criterion variable. More specifically, for a variable to be considered a mediator,

significant relationships must exist between (a) the predictor and the mediator, (b) the

mediator and the criterion, and (c) the predictor and criterion. If these preconditions are

satisfied, mediation is significant if the mediator accounts for a significant amount of the

predictor-criterion relationship. To test for the significance of mediations, Sobel's

formula (1982) was used (Baron & Kenny, 1986; Frazier, Tix, & Barron, 2004).














CHAPTER 4
RESULTS

Descriptive Statistics

The mean body mass index for the present sample was 29.04 (SD = 7.96) which

was comparable to Heffernan's (1996) obtained mean of 26.98 (SD = 6.82) in a sample

of 203 lesbian women. Levels of sexual objectification experiences, feminist ideology,

disconnection from the lesbian community, internalization of sociocultural standards of

beauty, body surveillance, body shame, and eating disorder symptomatology for the

current sample were generally close to the mid range of possible scores for each measure

(see Table 1). Furthermore, the present samples' scores on variables of interest were

comparable to those from studies that used the same instruments with similar samples of

women. More specifically, the current sample's means and standard deviations for

feminist ideology (M= 3.92, SD = .57), disconnection from the lesbian community (M=

2.45, SD = .94), internalization (M= 2.13, SD = .90), body surveillance (M= 4.01, SD=

1.27), body shame (M= 3.12, SD = 1.38), and eating disorder symptomatology (M=

7.54, SD = 8.37) were comparable to those reported by Fassinger (1994) for feminist

ideology scores in a sample of 117 female and male college students (M 3.52, SD =

.66), Szymanski and Chung (2003) for disconnection from the lesbian community scores

in sample of 210 lesbian and bisexual women (M = 2.36, SD = .91), Bergeron and Senn

(1998) for internalization scores in a sample of 108 lesbian women (M 2.17, SD = .69),

Hill (2001) for body surveillance scores in a sample of 134 lesbian women (M = 4.12, SD

= 1.19), McKinley and Hyde (1996) for body shame scores in a sample of 108









undergraduate women (M= 3.25, SD = 1.04), and Strong et al. (2001) for eating disorder

symptoms in a sample of 82 lesbian women (M 7.74, SD = 9.59). Because the sexual

objectification experiences measure used in the present study was a combination of

previously used scales, the present samples' scores on sexual objectification experiences

(M= 1.72, SD = .63) could not be compared to that of prior samples.

To test for order effects across the two orders of the survey, a MANOVA was

conducted with survey order as the independent variable and the variables of interest (i.e.,

BMI, sexual objectification experiences, feminist ideology, disconnection from the

lesbian community, internalization of sociocultural standards of beauty, body

surveillance, body shame, and eating disorder symptomatology) as the dependent

variables. A small, but significant overall effect was found (F [1, 530] = 1.99,p < .05,

p2 = .03) but the only significant univariate effect was a difference between the two

groups on reported experiences of sexual objectification (F [1, 530] = 4.10, p < .05, rp =

.008). Given that survey order accounted for less than 1% of variance in sexual

objectification scores and did not result in a significant difference in any of the other

seven variables of interest, survey order was not deemed to be problematic.

Intercorrelations among Variables of Interest

Partial correlations, controlling for BMI, were computed to test the relations among

variables of interest and evaluate whether pre-conditions for mediation were met (see

Table 1). Consistent with Hypothesis 1, after controlling for BMI, reported experiences

of sexual objectification were correlated positively with body surveillance (r = .22, p <

.001), body shame (r = .19, p < .001), and eating disorder symptomatology (r = .17, p <

.001). In addition, consistent with Hypothesis 2, after controlling for BMI, a negative









correlation existed between feminist ideology and disconnection from the lesbian

community (r = -.33, p < .001). After controlling for BMI, Hypothesis 3 also was

supported, as indicated by significant positive correlations between disconnection with

the lesbian community and internalization of sociocultural standards of beauty (r = .14, p

=.001), body surveillance (r = .14, p= .001), body shame (r .17, p < .001), and eating

disorder symptoms (r = .09, p < .05). Hypothesis 4 was only partially supported; with

BMI controlled, feminist ideology was not significantly correlated to internalization,

body shame, and eating disorder symptoms, but was significantly and negatively

correlated to body surveillance (r = -.10, p < .05).

Mediations

To test the mediations proposed in Hypotheses 5, 6, and 7, Baron and Kenny's

(1986) procedures were followed. According to these authors, for a variable to be

considered as a mediator, significant relations must exist between (a) the predictor and

the mediator, (b) the mediator and the criterion, and (c) the predictor and criterion. These

preconditions were satisfied for Hypotheses 5, 6, and 7 (see Table 1 for partial-

correlations). That is, for Hypothesis 5, reported experiences of sexual objectification

(predictor) were correlated significantly with internalization (potential mediator), which

in turn was correlated significantly with body surveillance, body shame, and eating

disorder symptomatology (criterion variables). In addition, reported sexual objectification

experiences (predictor) were correlated significantly with body surveillance, body shame,

and eating disorder symptomatology (criterion variables). With regard to Hypothesis 6,

reported sexual objectification experiences (predictor) were significantly related to body

surveillance (potential mediator), which in turn was significantly related to body shame

(criterion). Reported sexual objectification experiences (predictor) were also significantly









related to body shame (criterion). For Hypothesis 7, body surveillance (predictor) was

correlated significantly with body shame (potential mediator), and body shame was

correlated significantly with eating disorder symptomatology (criterion). Body

surveillance (predictor) was also significantly correlated with eating disorder

symptomatology (criterion).

According to Baron and Kenny (1986), if these conditions are satisfied, a variable

acts as a mediator to the extent that it accounts for the relationship between the predictor

and criterion variable(s). In order to test the significance of mediations, Amos 5.0

(Arbuckle, 2003) was used to conduct a path analysis of a fully saturated model in which

all direct and indirect paths were estimated (see the model presented in Figure 1). Again,

body mass index was entered as a covariate in the model. Maximum likelihood

estimation was utilized with the covariance matrix of the variables of interest as input.

Given that the model tested was fully saturated, values for the Goodness of Fit Index

(GFI), Incremental Fit Index (IFI), Comparative Fit Index (CFI), and the Normed Fit

Index (NFI) were all 1.0. The overall model account for 45% of the variance in body

shame, 38% of the variance in eating disorder symptoms, 32% of the variance in body

surveillance, and 5% of the variance in internalization of sociocultural standards of

beauty. As indicated in Figure 2, most standardized path coefficients were significant,

indicating significant unique direct links. Significant unique direct links did not emerge,

however, from feminist ideology to internalization, body surveillance, and eating disorder

symptomatology; from disconnection from the lesbian community to body surveillance

and eating disorder symptomatology; and from reported sexual objectification

experiences to body shame and eating disorder symtomatology. The only unique link that









was in the unexpected direction was the significant, albeit small (.07, p < .05) positive

unique link between feminist ideology and body shame (with BMI and other exogenous

variables controlled).

To test the significance of mediations through internalization of sociocultural

beauty standards, body surveillance, and body shame (the proposed mediators),

appropriate standardized path coefficients were multiplied to compute indirect effects, a

procedure recommended by Cohen and Cohen (1983). Next, Sobel's formula (1982) was

used to determine whether or not the indirect effects were significantly different from

zero. Hypothesis 5 proposed that internalization of sociocultural standards of beauty

would mediate the links of sexual objectification experiences to body surveillance, body

shame, and eating disorder symptomatology (as found in Moradi et al., 2005). Consistent

with this hypothesis, through internalization of sociocultural standards of beauty, reported

experiences of sexual objectification had a significant indirect link of.08 (.15 x .52; z =

3.41;p < .001) to body surveillance, .05 (.15 x .31; z = 3.21,p < .01) to body shame, and

.02 (.15 x .14; z = 2.31, p < .05) to eating disorder symptomatology. In addition to these

significant indirect relations, reported experiences of sexual objectification also had a

significant direct link to body surveillance, but not to body shame or eating disorder

symptoms. Therefore, consistent with Hypothesis 5, internalization of sociocultural

beauty standards partially mediated the link between reported sexual objectification

experiences and body surveillance, and fully mediated the links of reported sexual

objectification experiences to body shame and eating disorder symptomatology.

Hypothesis 6 proposed that body surveillance would mediate the link of sexual

objectification experiences to body shame (as found in Moradi et al., 2005). Consistent









with this hypothesis, through body surveillance, a significant indirect link of .05 (.13 x

.38; z = 3.31, p < .001) was obtained between reported sexual objectification experiences

and body shame. Because no significant direct link existed between reported sexual

objectification experiences and body shame, body surveillance acted as a full mediator of

this link, supporting Hypothesis 6.

Hypothesis 7 proposed that body shame would partially mediate the relationship

between body surveillance and eating disorder symptomatology as found in previous

research (Moradi et al., 2005; Noll & Fredrickson, 1998). Consistent with this hypothesis,

through body shame, a significant indirect link of .13 (.38 x .34; z = 5.84, p < .001) was

found between body surveillance and eating disorder symptomatology. Because there

was also a direct relationship between body surveillance and eating disorder

symptomatology, body shame acted as a partial mediator. Thus, Hypothesis 7 was

supported.

Finally, the fully saturated model was compared to an alternative trimmed model

that eliminated the non-significant direct paths (a) from feminist ideology to

internalization, body surveillance, and eating disorder symptoms, (b) from disconnection

from the lesbian community to body surveillance and eating disorder symptomatology,

and (c) from reported sexual objectification experiences to body shame and eating

disorder symptoms. The goodness of fit indices for this model were above the acceptable

cut offs and nearly identical to those obtained from the original model (GFI = .99; IFI=

.99; CFI = .99; NFI = .99). The trimmed model explained 45% of the variance in body

shame, 38% of the variance in eating disorder symptomatology, 31% of the variance in

body surveillance, and 5% of the variance in internalization of sociocultural standards of






62


beauty. The variance accounted for by the trimmed model was the same as that accounted

for in the fully saturated model, except for variance accounted for in body surveillance,

which dropped from 32% to 31%. In addition, the magnitude of path coefficients was

comparable across the two models. Thus, compared with the fully saturated model, the

trimmed model appears to be more parsimonious but equally appropriate in explaining

the relationships among the variables of interest.














CHAPTER 5
DISCUSSION

Objectification theory (Fredrickson & Roberts, 1997) and empirical investigations

of its tenets point to the importance of sexual objectification experiences, self-

objectification (manifested as body surveillance), and body shame as predictors of

women's eating disorder symptomatology. In addition, Moradi et al.'s (2005) findings

highlighted the role of internalization of sociocultural standards of beauty in the

objectification theory framework. The present study was the first to examine the

objectification theory framework and the additional role of internalization of

sociocultural standards of beauty in eating disorder symptomatology with a large sample

of lesbian women. Furthermore, in the context of objectification theory, this study

examined the roles of feminist ideology and connection/disconnection from the lesbian

community, each of which has been identified as a key predictor of eating disorder

symptoms among lesbian persons.

Overall, findings of the present study suggested that relations outlined in

objectification theory and supported in prior tests of objectification theory with

heterosexual women are also supported with lesbian women. More specifically,

consistent with objectification theory's conceptualization of the role of sexual

objectification experiences in eating disorder symptoms and their precursors, in the

present study, partial correlations (with BMI controlled) indicated that reported sexual

objectification experiences were significantly and positively related to body surveillance,

body shame, and eating disorder symptomatology. Furthermore, results of the path









analysis suggested that sexual objectification experiences were related positively and

uniquely to internalization of sociocultural beauty standards and body surveillance when

feminist ideology, disconnection from the lesbian community, and BMI were accounted

for. In addition, internalization of sociocultural standards of beauty partially mediated the

links of sexual objectification experiences to body surveillance, body shame, and eating

disorder symptomatology, suggesting that through internalization, reported experiences

of sexual objectification might be translated into body surveillance, body shame, and

eating disorder symptoms. These findings are consistent with previous research (Moradi

et al., 2005; Morry & Staska, 2001) with heterosexual women and further highlight the

importance of including both sexual objectification experiences and internalization of

sociocultural standards of beauty when investigating the tenets of objectification theory.

Also consistent with prior tests of objectification theory, (e.g., Moradi et al., 2005),

the current findings supported body surveillance as a full mediator of the relation

between sexual objectification experiences and body shame. Thus, it appears that body

surveillance might be a key mechanism through which sexually objectifying experiences

are translated into body shame. In addition, body shame acted as a partial mediator in the

link of body surveillance to eating disorder symptomatology, supporting the findings of

extant literature (Greenleaf, 2005; Moradi et al., 2005, Noll & Fredrickson, 1998) and

suggesting that chronic body monitoring might be translated into eating disorder

symptomatology by promoting body shame. Taken together, these findings support the

generalizeability and potential utility of the objectification theory framework as applied

to understanding eating disorder symptomatology among lesbian women. Indeed, the









overall path model examined in the present study explained 38% of the variance of eating

disorder symptoms in this sample of 531 lesbian women.

In addition to extending research on objectification theory to lesbian women, the

present study integrated the previously highlighted potential roles of feminist ideology

and connection/disconnection from the lesbian community in its examination of

objectification theory. Consistent with previous findings (Syzmanski & Chung, 2003),

present results indicated that feminist ideology and disconnection from the lesbian

community were significantly and negatively correlated with one another (with BMI

controlled). Furthermore, as hypothesized, disconnection from the lesbian community

was significantly and positively related to internalization of sociocultural standards of

beauty, body surveillance, body shame, and eating disorder symptomatology. In other

words, disconnection from the lesbian community was correlated with greater levels of

eating disorder symptoms and their correlates for lesbian women. This finding supports

extant literature in which interaction and involvement with other lesbian women has been

found to be related to positive health outcomes such as higher body esteem, higher body

image, lower weight concern, and lower levels of eating disorder symptoms (Beren et al.,

1997; Heffernan, 1996; Ludwig & Brownell, 1999).

In terms of feminist ideology as a predictor variable, partial correlations (with BMI

controlled) revealed only one significant, but negative relationship between feminist

ideology and body surveillance. Thus, it seems that endorsing feminist ideology may be

related to lower levels of body monitoring. Interestingly, a partial correlation (with BMI

controlled) indicated no relationship between feminist ideology and body shame;

however, when feminist ideology was considered in the context of disconnection from









the lesbian community via the path analysis, feminist ideology was significantly and

positively related to body shame. Although this relationship was very small (.07; p < .05),

it may be that the positive health outcomes associated with feminist ideology as found in

previous literature (e.g., Leavy & Adams, 1986) are accounted for largely by overlap

with connection with the lesbian community. Indeed, the present results indicated that

compared with feminist ideology, disconnection from the lesbian community was a

stronger and more consistent correlate of eating disorder constructs included in the

model, as evidenced by the pattern of partial correlations (with BMI controlled) and path

coefficients between disconnection with the lesbian community versus feminist ideology

and the four outcome variables.

The small but positive relation that emerged between feminist ideology and body

shame, when connection/disconnection from the lesbian community, sexual

objectification experiences, and BMI were accounted for, might be interpreted in light of

a study conducted by Eliason and Morgan (1998), in which lesbian women were asked to

define what it meant to be a lesbian. Those who gave political definitions (e.g., "woman

identification," "affiliation with other oppressed groups") were more likely to identify

themselves as feminist and get involved in political activities promoting lesbian women,

but at the same time, were significantly more likely to have a history of eating disorders

than those who gave non-political answers (e.g., "sex/love with women," "lesbianism as

one small aspect of a person"). Interpreted in light of the present findings and prior

research on feminist ideology and connection/disconnection from the lesbian community,

Eliason and Morgan's findings suggest an interesting possibility. When considered

separately, connection with the lesbian community and feminist ideology each are related









to lower levels of eating disorder-related variables. However, lesbian women who have

an integrated lesbian and feminist political identity may have experienced greater levels

of eating disorder symptoms in the past and possibly also experience greater body shame.

It may be that under some conditions, eating disorder symptoms become so detrimental

that some lesbian women turn to the rejection of society's beauty standards through

increasing their endorsement of feminist principles and by becoming more connected

with the lesbian community. This causal hypothesis is speculative, and longitudinal

research is needed to ascertain the direction of relations between

connection/disconnection with the lesbian community, feminist ideology, and eating

disorder symtomatology.

Limitations

The present findings must be interpreted in light of a number of limitations. First,

the present data were collected using an online, self-report survey in which participation

was voluntary. As such, the stigma associated with eating disorder-related attitudes and

behaviors may have limited participation from those experiencing high levels of such

symptoms. Although scores from this sample are similar to those obtained in other

nonclinical samples of lesbian women, the generalizability of the present findings are

limited to nonclinical lesbian populations. To address this limitation, future research is

needed to test the tenets of objectification theory with samples of lesbian women who are

experiencing clinically significant levels of eating disorder symptomatology.

Another potential limitation of the present study (that is shared by any study that

explicitly recruits lesbian and gay participants) is that the majority of participants might

be "out" about their sexual orientation and so at least somewhat comfortable with

participating in a study about lesbian persons. Because outnesss" is an individual









difference variable that may inhibit some individuals from participating in a study

advertised for lesbian women, the present sample may not capture the experiences of

lesbian women who are less "out" about their sexual orientation. As such, the participants

in the present study may have been more connected to the lesbian community. Because

current findings indicated that disconnection from the lesbian community was related to

internalization of sociocultural standards of beauty, body surveillance, body shame, and

eating disorder symptomatology, it may be that lesbian women who are less out and more

disconnected from the lesbian community might be at greater risk for eating disorder

attitudes and behaviors. The current sample obtained scores that were around the

midrange for connection/disconnection with the lesbian community. Therefore, future

studies should attend to the experiences of lesbian women who are less "out" and more

disconnected from the lesbian community than those in the current sample.

A third potential limitation is that even though the survey included validity check

items to ensure that participants understood the survey instructions and items, there was

no way to ensure that the respondents actually met the participation requirements. This

limitation is inherent in all volunteer, self-report studies (online or in person) and for the

present study; however, the benefits of the online survey (e.g., facilitating participation of

less out participants, larger sample size) outweighed the costs. With such a large sample

size, it is unlikely that any submissions from those who failed to meet participation

criteria would substantially skew results, but findings should still be interpreted with this

limitation in mind.

A fourth potential limitation is that although the present sample was diverse in

terms of age and geographical location, respondents were predominantly Caucasian and









well-educated. This imbalance, which has occurred within other lesbian samples (e.g.,

Bergeron & Senn, 1998; Hill, 2002), limits the generalizability of findings to Caucasian

lesbian women with at least a college education. To address this limitation, future

research is needed to develop and evaluate recruitment strategies designed to increase

representation of racial/ethnic minority lesbian women as well as lower and working

class lesbian women.

A final potential limitation, which is shared in much of the eating disorder research,

is that there was no way to ensure that self-reports of weight and height were accurate.

However, prior evidence exists that self-report information for BMI are comparable to

actually measured information (e.g., Koslowsky, Scheinberg, & Bleich, 1994; Tienboon,

Wahlqvist, & Rutishauser, 1992). In addition, as one participant addressed in a feedback

email, medical conditions that may affect BMI (e.g., diabetes, hypoglycemia) were not

assessed. In general, participants' health backgrounds are not assessed in eating disorder

research, and future studies could address this potential confound by including a brief

medical history measure in the questionnaire. Despite these limitations, the present

findings can inform research and practice in a number of important ways.

Directions for Future Research

Findings from the current study can be used to inform future research conducted on

objectification theory with lesbian persons. More specifically, findings indicate that it is

important to include sexual objectification experiences, internalization of sociocultural

standards of beauty, body surveillance, and body shame as key predictors of eating

disorder symptomatology with this population. Furthermore, the present findings suggest

that connection/disconnection from the lesbian community is an important variable to

consider in applications of objectification theory to lesbian women. Indeed,









connection/disconnection with the lesbian community might subsume the previously

observed links of feminist ideology to eating disorder-related constructs. On the other

hand, the unique links of feminist ideology to eating disorder-related constructs, when

connection/disconnection from the lesbian community is considered, remains unclear.

A critical direction for future research is to further examine the temporal links

between lesbian feminist ideology and eating disorder symptomatology. More

specifically, longitudinal research is needed to expand upon Eliason and Morgan's (1998)

findings and identify the exact role of lesbian feminist ideology as it relates to

involvement in political activities and both past and present eating disorder symptoms.

Furthermore, future research is needed to understand the nature of the relations between

feminist ideology and eating disorder-related attitudes and behaviors in the context of

connection/disconnection with the lesbian community. The present study focused on

feminist ideology as a potential predictor variable in the model and found that links of

feminist ideology to eating disorder constructs were generally nonsignificant when BMI

and connection/disconnection from the lesbian community were accounted for. An

additional possibility that is worth exploring in future research is that feminist ideology

might act as a moderator variable, affecting the relations of connection with the lesbian

community with internalization of sociocultural standards of beauty and body shame.

That is, for lesbian women who are disconnected from the lesbian community, feminist

ideology may be related to lower levels of eating disorder-related constructs. For lesbian

women who are highly connected to the lesbian community, feminist ideology may be

related to greater levels of eating disorder-related constructs. Eliason and Morgan's

findings are consistent with such a possibility, if past eating disorder symptoms are









considered. Future research is needed that will test the potential interaction effect

between connection with the lesbian community and feminist ideology in relation to past

and present eating disorder symptomatology for lesbian women.

Another important area for further investigation is to explore what aspects of

lesbian identity and community might be related to lower eating disorder-related attitudes

and behaviors. For example, Kaminski (2000) conducted 19 interviews with lesbian

women about their identity and health, and found that, compared to those who

experienced their environments to be hostile, homophobic, or conservative, participants

who perceived their environment to be supportive were more likely to adopt feminist

principles and were more likely to experience positive health outcomes such as greater

self-acceptance and reduced anxiety, depression, and substance abuse. Thus, it seems that

perceived social support, which may be obtained through connection with the lesbian

community, could be an important factor to consider when examining predictors of

health-related concerns such as eating disorders among lesbian women. Clearly, future

research is needed to tease apart aspects of connection/disconnection from the lesbian

community such as feminist self-identification, feminist ideology, and social support that

might play a role in the promotion of health and reduction of eating disorder and other

symptomatology for lesbian (and other) women.

Implications for Practice

In addition to informing future research, findings from the current study can also be

used to inform counselors' and clinicians' efforts to reduce eating disorder-related

attitudes and behaviors among lesbian women. Findings from the current study and

previous research with heterosexual women (e.g., Moradi, 2005) support tenets of the

objectification theory framework, and it would be beneficial for mental health









professionals to provide appropriate education regarding objectification theory so that

both heterosexual women and lesbian women can recognize sexual objectification

experiences as they occur. More specifically, findings from the current study indicate that

sexual objectification experiences are related to internalization of sociocultural beauty

standards and body surveillance, both of which are related to eating disorder

symptomatology. These correlational findings provide the groundwork for testing the

directions of causality implied in objectification theory among these variables. If such

causality is supported, learning to identify sexual objectification experiences when they

occur may be beneficial for women because they can then actively work against

internalizing societal beauty standards, become aware of daily chronic body monitoring,

work towards reducing feelings of shame regarding their bodies, and thus, lower their

risk of developing eating disorder symptomatology.

Strategies specific to lesbian women might include the development of a workshop

in which counselors and clinicians can educate lesbian women regarding the potential

benefits of connection with the lesbian community, reduction of the internalization of

sociocultural standards of beauty, and prevention of body shame. Together, lesbian

women can discuss their perceptions of the lesbian community, identify cultural beauty

standards, develop strategies to actively work against endorsing such standards, identify

goals to reduce body monitoring, and promote positive attitudes toward their bodies.

Based on extant literature (e.g., Kaminski, 2000) as well as findings from the present

study, such association and interaction with other lesbian women may lead to positive

health outcomes. The present cross-sectional findings provide the groundwork for









exploring whether such interventions would result in reducing the risk for eating disorder

symptomatology among lesbian women.

As for individual therapy, it may be beneficial for mental health professionals to

adopt treatment strategies for their lesbian clients that include providing education

regarding society's current standards of beauty (specifically those associated with lesbian

women and beauty), promoting acceptance of one's physical appearance, and

encouraging increased connection with the lesbian community. Future research regarding

the exact role of feminist ideology is needed to ascertain whether or not therapy should

include the provision of education and support regarding the development of feminist

ideology. It is important to note that these therapeutic strategies have the potential to help

heterosexual women as well. Future research is needed that will address the potential

benefits of connection with the lesbian community for heterosexual women of all ages

and backgrounds.

Summary

Overall, findings of the present study supported the applicability of the tenets of

objectification theory to understanding eating disorder symptomatology of lesbian

women. The present findings also pointed to the importance of considering the role of

connection with the lesbian community when investigating eating disorder

symptomatology with lesbian women, but also raised questions about the unique role of

feminist ideology beyond connection with the lesbian community. Finally, the present

findings lay the groundwork for examining the objectification theory framework with

more diverse samples of lesbian women, exploring longitudinal links between sexual

objectification experiences, connection with the lesbian community, feminist ideology,






74


and the predictors of eating disorder symptoms, and using experimental designs to begin

to elucidate potential causal relations between variables in the model.















APPENDIX A
THE SEXUAL OBJECTIFICATION SUB SCALE

Please use the following scale to indicate how often during the past year you have

experienced each of the events below. For each item, respond once considering your

experiences with men and respond again considering your experiences with women.

1. Never
2. About once during the past year
3. About once a month during the past year
4. About once a week during the past year
5. About two or more times per week during past year

1. Had people shout sexist comments, whistle, or By men 1 2 3 4 5
make catcalls at me.
By women 1 2 3 4 5
2. Had someone refer to me with a demeaning or By men 1 2 3 4 5
degrading label specific to my gender (bitch,
chick, dyke, bastard, faggot, etc).
By women 1 2 3 4 5
3 Had sexist comments made about parts of my By men 1 2 3 4 5
body or clothing.
By women 1 2 3 4 5
4. Had sexist comments made about parts of my By men 1 2 3 4 5
body or clothing.
By women 1 2 3 4 5
5. Heard someone make comments about sexual By men 1 2 3 4 5
behavior I might do or things they would want to
do with me.
By women 1 2 3 4 5
6. Someone did or said something that made me By men 1 2 3 4 5
feel threatened sexually.
By women 1 2 3 4 5
7. Experienced unwanted staring or ogling at By men 1 2 3 4 5
myself or parts of my body when the person
knew or should have known I was not interested
or it was inappropriate for the situation or our
relationship.
By women 1 2 3 4 5









8. Experienced unwanted flirting when the person
knew or should have known I was not interested
or it was inappropriate for the situation or our
relationship.


By men



By women


1 2 3 4 5


1 2 3 4















APPENDIX B
OBJECTIFICATION EXPERIENCES QUESTIONNAIRE

Please use the following scale to indicate how often during the past year you have

experienced each of the events below. For each item, respond once considering your

experiences with men and respond again considering your experiences with women.

1 Never
2 About once during the past year
3 About once a month during the past year
4 About once a week during the past year
5 About two or more times per week during past year

1. Been "checked out" (i.e., had your body stared at By men 1 2 3 4 5
in an intrusive way) by a person in public.
By women 1 2 3 4 5
2. Your appearance/body commented on in a way By men 1 2 3 4 5
that you felt was inappropriate.
By women 1 2 3 4 5
3 Your romantic partner (current or former) By men 1 2 3 4 5
"checked out" other women in your presence.
By women 1 2 3 4 5
4. Someone stared at your breasts while talking to By men 1 2 3 4 5
you.
By women 1 2 3 4 5
5. Someone made offensive, sexualized gestures By men 1 2 3 4 5
toward you (e.g., pantomime of masturbation or
intercourse)?
By women 1 2 3 4 5
6. Felt that a date was more interested in your body By men 1 2 3 4 5
(and gaining access to it) than in you as a person.
By women 1 2 3 4 5















APPENDIX C
THE SOCIOCULTURAL ATTITUDES TOWARD APPEARANCE
INTERNALIZATION SUB SCALE (SATAQ)

Please read each of the following items and select the number that best reflects your

agreement with the statement.

1. = completely disagree
2. = somewhat disagree
3. = neither agree nor disagree
4. = somewhat agree
5. = completely agree


1. Women who appear in TV shows and movies project the type
of appearance that I see as my goal.
2. I believe that clothes look better on thin models.
3. Music videos that show thin women make me wish that I were
thin.
4. I do not wish to look like the models in the magazines.
5. I tend to compare my body to people in magazines and on TV.
6. Photographs of thin women make me wish that I were thin.
7. I wish I looked like a swimsuit model.
8. I often read magazines like Cosmopolitan, Vogue, and Glamour
and compare my appearance to the models.


1 2 3 4 5















APPENDIX D
CONNECTION WITH THE LESBIAN COMMUNITY SUB SCALE (CLC)

Please read each of the following items and select the number that best reflects your

agreement with the statement.


Strongly Disagree
Moderately Disagree
Slightly Disagree
Neither Agree nor Disagree
Slightly Agree
Moderately Agree
Strongly Agree


1. Many of my friends are lesbians.
2. Attending lesbian events and organizations is important to
me.
3. I feel isolated and separate from other lesbians.
4. When interacting with members of the lesbian
community, I often feel different and alone, like I don't
fit in.
5. Having lesbian friends is important to me.
6. I am familiar with lesbian books and/or magazines.
7. Being a part of the lesbian community is important to me.
8. I feel comfortable joining a lesbian social group, lesbian
sports team, or lesbian organization.
9. Social situations with other lesbians make me feel
uncomfortable.
10. I am familiar with lesbian movies and/or music.
11. I am aware of the history concerning the development of
lesbian communities and/or the lesbian/gay rights
movement.
12. I am familiar with lesbian music festivals and
conferences.
13. I am familiar with community resources for lesbians (i.e.,
bookstores, support groups, bars, etc.).


1 2 3 4 5 6 7


1 2 3 4 5 6 7

1 2 3 4 5 6 7
1234567















APPENDIX E
ATTITUDES TOWARD FEMINISM AND THE WOMEN'S MOVEMENT (FWM)
SCALE

Please read each of the following items and select the number that best reflects your

agreement with the statement.


Strongly Disagree
Somewhat Disagree
Neither Agree nor Disagree
Somewhat Agree
Strongly Agree


1. The leaders of the women's movement have the right idea.
2. There are better ways for women to fight for equality than
through the women's movement.
3. Feminists are too visionary for a practical world.
4. People would favor women's liberation more if they knew more
about it.
5. The women's movement has positively influenced relationships
between men and women.
6. The women's movement is too radical and extreme in its views.
7. Feminist principles should be adopted everywhere.
8. I would be overjoyed if women's liberation gained more
strength in this country.
9. The women's movement has made important gains in equal
rights and political power for women.
10. Feminists are a menace to this nation and the world.


2 3 4 5


1 2 3 4 5

1 2 3 4 5
12345















APPENDIX F
BODY SURVEILLANCE SUB SCALE OF THE OBJECTIFIED BODY
CONSCIOUSNESS SCALE (OBC)

Please read each of the following items and select the number that best reflects your

agreement with the statement. Circle NA only if the statement does not apply to you. Do

not circle NA if you don't agree with the statement. For example if the statement says

"When I am happy, I feel like singing" and you don't feel like singing when you are

happy, then you would circle one of the disagree choices. You would only circle NA if

you were never happy.

18. Strongly Disagree
19. Moderately Disagree
20. Slightly Disagree
21. Neither Disagree nor Agree
22. Slightly Agree
23. Moderately Agree
24. Strongly Agree
25. Item does not apply

1. I rarely think about how I look. 1 2 3 4 5 6 7 NA
2. I think it is more important that my clothes are 1 2 3 4 5 6 7 NA
comfortable than whether they look good on me.
3. I think more about how my body feels than how 1 2 3 4 5 6 7 NA
my body looks.
4. I rarely compare how I look with how other people 1 2 3 4 5 6 7 NA
look.
5. During the day, I think about how I look many 1 2 3 4 5 6 7 NA
times.
6. I often worry about whether the clothes I am 1 2 3 4 5 6 7 NA
wearing make me look good.
7. I rarely worry about how I look to other people. 1 2 3 4 5 6 7 NA
8. I am more concerned with what my body can do 1 2 3 4 5 6 7 NA
than how it looks.















APPENDIX G
BODY SHAME SUB SCALE OF THE OBJECTIFIED BODY CONSCIOUSNESS
SCALE (OBC)

Please read each of the following items and select the number that best reflects your

agreement with the statement. Circle NA only if the statement does not apply to you. Do

not circle NA if you don't agree with the statement. For example if the statement says

"When I am happy, I feel like singing" and you don't feel like singing when you are

happy, then you would circle one of the disagree choices. You would only circle NA if

you were never happy.

26. Strongly Disagree
27. Moderately Disagree
28. Slightly Disagree
29. Neither Disagree nor Agree
30. Slightly Agree
31. Moderately Agree
32. Strongly Agree
33. Item does not apply
1. When I can't control my weight, I feel like 1 2 3 4 5 6 7 NA
something must be wrong with me.
2. I feel ashamed of myself when I haven't made the 1 2 3 4 5 6 7 NA
effort to look my best.
3. I feel like I must be a bad person when I don't look 1 2 3 4 5 6 7 NA
as good as I could.
4. I would be ashamed for people to know what I 1 2 3 4 5 6 7 NA
really weigh.
5. Even when I can't control my weight, I think I'm 1 2 3 4 5 6 7 NA
an okay person.
6. I never worry that something is wrong with me 1 2 3 4 5 6 7 NA
when I am not exercising as much as I should.
7. When I'm not exercising enough, I question 1 2 3 4 5 6 7 NA
whether I am a good enough person.
8. When I'm not the size I think I should be, I feel 1 2 3 4 5 6 7 NA
ashamed.















APPENDIX H
THE EATING ATTITUDES TEST 26 (EAT-26)

For each of the following questions, please select the response that best describes

you.

34. Never
35. Rarely
36. Sometimes
37. Often
38. Usually
39. Always
1. Am terrified about being overweight. 1 2 3 4 5 6
2. Avoid eating when I am hungry. 1 2 3 4 5 6
3. Find myself preoccupied with food. 1 2 3 4 5 6
4. Have gone on eating binges where I feel that I may not be 1 2 3 4 5 6
able to stop.
5. Cut my food into small pieces. 1 2 3 4 5 6
6. Aware of the calorie content of foods that I eat. 1 2 3 4 5 6
7. Particularly avoid food with a high carbohydrate content 1 2 3 4 5 6
(i.e., bread, rice, potatoes, etc.)
8. Feel that others would prefer if I ate more. 1 2 3 4 5 6
9. Vomit after I have eaten. 1 2 3 4 5 6
10. Feel extremely guilty after eating. 1 2 3 4 5 6
11. Am preoccupied with a desire to be thinner. 1 2 3 4 5 6
12. Think about burning up calories when I exercise. 1 2 3 4 5 6
13. Other people think that I am too thin. 1 2 3 4 5 6
14. Am preoccupied with the thought of having fat on my body. 1 2 3 4 5 6
15. Take longer than others to eat my meals. 1 2 3 4 5 6
16. Avoid foods with sugar in them. 1 2 3 4 5 6
17. Eat diet foods. 1 2 3 4 5 6
18. Feel that food controls my life. 1 2 3 4 5 6
19. Display self-control around food. 1 2 3 4 5 6
20. Feel that others pressure me to eat. 1 2 3 4 5 6
21. Give too much time and thought to food. 1 2 3 4 5 6
22. Feel uncomfortable after eating sweets. 1 2 3 4 5 6
23. Engage in dieting behavior. 1 2 3 4 5 6
24. Like my stomach to be empty. 1 2 3 4 5 6
25. Enjoy trying new rich foods. 1 2 3 4 5 6
26. Have the impulse to vomit after meals. 1 2 3 4 5 6














APPENDIX I
DEMOGRAPHIC QUESTIONNAIRE

Please tell us a little about yourself. This information will be used only to describe

the sample as a group.

1. Age:

2. Gender Male Female Transgender

3. Your current relationship status (please select the best descriptor):

4. Single Married/Partnered Dating, long term Dating, casual

5. Completed Education (please select one):

Less than High School

Some High School

High School Graduate

Some College

College Degree (e.g. B. A., B.S.)

Professional Degree (e.g., MBA, MS, Ph.D, M. D.)

6. Current Employment status (please select the one best descriptor):

Employed Full Time Employed Part Time Not employed

7. Yearly household income (income of those on whom you rely financially):

Below $10,000 $60,001 to $70,000

$10,001 to $20,000 $70,001 to $80,000

$20,001 to $30,000 $80,001 to $90,000

$30,001 to $40,000 $90,001 to $100,000

$40,001 to $50,000 $100,001 to $110,000









$50,001 to $60,000 Above $110,001

8. Your current social class (please select the one best descriptor):

lower class working class middle class

upper middle class upper class

9. Race/ethnicity (Please check one)

African American/Black

Asian American/Pacific Islander

American Indian/Native American

Hispanic/Latino/a White

Hispanic/Latino/a Black

Multi-racial, please specify:

White/Caucasian

Other, please specify:

10. Current height feet inches

11. Current weight in pounds

12. Your sexual orientation (please check the one best descriptor):

Exclusively lesbian

Mostly lesbian

Bisexual

Mostly Heterosexual

Exclusively Heterosexual

13. How much are you physically attracted to members of your own sex?


moderate


low


high






86


14. How much are you physically attracted to members of the other sex?


low


moderate


high


15. How much are you emotionally attracted to members of your own sex?


low


moderate


high


16. How much are you emotionally attracted to members of the other sex?


moderate


high


17. Sexual behavior: Have you had sex with persons of your own gender, the other
gender, or both genders?


Never had My own My own gender Both Other Other
sex gender mostly genders gender gender
only equally mostly only
18. How connected or involved are you in the lesbian community? Please select one.


very slightly a little


moderately


quite a bit extremely


not at all


19. To what extent do you describe yourself as a feminist?


9

very much a feminist


not at all


20. Finally, we would like to obtain information regarding the geographic location of
our sample. This information will remain confidential. Please fill in the city, state,
and country in which you currently reside down below:

City:

State:


Country:















APPENDIX J
PERTINENT TABLES AND FIGURES













Summary Statistics and Partial Correlations Among Variables of Interest with Body Mass Index Controlled

Variables 1 2 3 4 5 6 Possible Sample M SD oc
Range Range
1. Reported Sexual 1.00-5.00 1.00-4.69 1.72 .63 .94
Objectification
Experiences
2. Feminist Ideology -.16** 1.00-5.00 1.30-5.00 3.92 .57 .85
3. Disconnection .10* -1.00-7.00 1.00-5.92 2.45 .94 .85
from Lesbian .33**
Community
4. Internalization .15** .01 .14** 1.00-5.00 1.00-5.00 2.13 .90 .87
5. Body Surveillance .22** -.10* .14* .54** 1.00-7.00 1.00-7.00 4.01 1.27 .86
6. Body Shame .19** .00 .17** .55** .58** 1.00-7.00 1.00-7.00 3.12 1.38 .87
7. Eating disorder .17** .01 .09* .45** .51** .55** 0.00-78.00 0.00-51.00 7.54 8.37 .86
symptoms_______
Note. *p < .05. **p <.001. Higher scores indicate higher levels of the construct assessed.

































Hypothesized Path Model


Objectification .15
Experiences

Trimmed Model Depicting Relationships Among Variables of Interest with Body Mass

Index Controlled















LIST OF REFERENCES


Abraham, S. F., & Beumont, P. J. (1982). Varieties of psychosexual experience in
patients with anorexia nervosa. International Journal ofEating Disorders, 1, 10-
19.

American Psychiatric Association Work Group on Eating Disorders (2000). Practice
guidelines for the treatment of patients with eating disorders (revision). American
Journal ofPsychiatry, 157, 1-39.

Arbuckle, J. L. (2003). Amos (Version 5.0) [Computer software]. Chicago: SmallWaters
Corporation.

Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in
social psychological research: Conceptual, strategic, and statistical
consideration. Journal ofPersonality and Social Psychology, 51, 1173-1182.

Belknap, J., Fischer, B. S., & Cullen, F. T. (1999). The development of a comprehensive
measure of the sexual victimization of college women. Violence Against Women, 5,
185-24.

Ben-Tovim, D. I., & Walker, M. K. (1991). The development of the Ben-Tovim-Walker
Body Attitudes Questionnaire (BAQ), a new measure of women's attitudes towards
their own bodies. Psychological Medicine, 21, 775-784.

Bergeron, S. M., & Senn, C. Y. (1998). Body image and sociocultural norms: A
comparison of heterosexual and lesbian women. Psychology of Women Quarterly,
22, 385-401.

Beren, S. E., Hayden, H. A., Wilfley, D. E., & Grilo, C. M. (1996). The influence of
sexual orientation on body dissatisfaction in adult men and women. International
Journal ofEating Disorders, 20(2), 135-141.

Beren, S. E., Hayden, H.A., Wilfley, D.E., & Striegel-Moore, R. H. (1997). Body
dissatisfaction among lesbian college students: The conflict of straddling
mainstream and lesbian cultures. Psychology of Women Quarterly, 21, 431-445.

Brand, P. A., Rothblum, E. D., & Solomon, L. J. (1992). A comparison of lesbians, gay
men, and heterosexuals on weight and restrained eating. International Journal of
Eating Disorders, 11(3), 253-259.









Brownlow, B. (1997). The relationship between objectification, body image disturbances,
and disordered eating: Investigating race, socioeconomic status, acculturation,
and self-objectification as mediators. Unpublished doctoral dissertation, Duke
University.

Burnett, R. E. (1995). Gendered objectification experiences: Construct validity,
implicationsfor dysphoria and depression, and phenomenology. Unpublished
doctoral dissertation, Duke University.

Cogan, J. C. (1999). Lesbians walk the tightrope of beauty: Thin is in but femme is out.
Journal ofLesbian Studies, 3(4), 77-89.

Cogan, J. C., Bhalla, S. K., Sefa-Dedeh, A., & Rothblum, E. D. (1996). A comparison
study of United States and African students on perceptions of obesity and thinness.
Journal of Cross-Cultural Psychology, 27, 98-113.

Cohen, J., & Cohen, P. (1983). Applied multiple regression/correlation analysis for the
behavioral sciences (2nd ed.). Hillsdale, NJ: Erlbaum.

Cooper, P. J., Taylor, M. J., Cooper, Z., & Fairburn, C. G. (1987). The development and
validation of the Body Shape Questionnaire. International Journal ofEating
Disorders, 6, 485-494.

Dionne, M. (1992). Feminist ideology as a predictor of body dissatisfaction in women.
Unpublished master's thesis, York University, North York, Ontario, Canada.

Dionne, M., Davis, C., Fox, J., & Gurevich, M. (1995). Feminist ideology as a predictor
of body dissatisfaction in women. Sex Roles, 33 (3/4), 277-287.

Dunn, P. K., & Ondercin, P. (1981). Personality variables related to compulsive eating in
college women. Journal of Clinical Psychology, 37, 43-49.

Eliason, M. J., & Morgan, K. S. (1998). Lesbians define themselves: Diversity in lesbian
identification. Journal of Gay, Lesbian, & Bisexual Identity, 3(1), 47-63.

Enns, C. Z. (1987). A comparison of non-feminist andprofeminist women's reactions to
nonsexist, liberal feminist, and radical feminist therapy. Unpublished doctoral
dissertation, University of California/Santa Barbara.

Fassinger, R. E. (1994). Development and testing of the Attitudes Toward Feminism and
the Women's Movement (FWM) scale. Psychology of Women Quarterly, 18, 389-
402.

Fitzgerald, L. F., Shullman, S. L., Bailey, N., Richards, M., Swecker, J., Gold, Y.,
Ormerod, M., & Weitzman, L. (1988). The incidence and dimensions of sexual
harassment in academia and the workplace. Journal of Vocational Behavior, 32,
152-175.









Frazier, P.A., Tix, A. P., & Barron, K. E. (2004). Testing moderator and mediator effects
in counseling psychology research. Journal of Counseling Psychology, 51(1), 115-
134.

Franzoi, S. E., & Herzog, M. E. (1986). The Body Esteem Scale: A convergent and
discriminant validity study. Journal ofPersonality Assessment, 50, 24-31.

Franzoi, S. E., & Herzog, M. E. (1987). Judging physical attractiveness: What body
aspects do we use? Personality and Social Psychology Bulletin, 13, 19-33.

Franzoi, S. L., & Shields, S. A. (1984). The Body Esteem Scale: Multidimensional
structure and sex differences in a college population. Journal of Personality
Assessment, 48(2), 173-178.

Fredrickson, B. L., & Roberts, T. A. (1997). Objectification theory: Toward
understanding women's lived experiences and mental health risks. Psychology of
Women Quarterly, 21, 173-206.

Fredrickson, B. L., Roberts, T. A, Noll, S. M., Quinn, D. M, & Twenge, J. M. (1998).
That swimsuit becomes you: Sex differences in self-objectification, restrained
eating, and math performance. Journal ofPersonality & Social Psychology, 75(1),
269-284.

Furnham, A., & Alibhai, N. (1983). Cross-cultural differences in the perception of female
body shapes. Psychological Medicine, 13(4), 829-837.

Garner, D. M. (1997). Psychoeducational principles in treatment. In: D.M. Garner & P.E.

Garfinkel (Eds.) Handbook of Treatment for Eating Disorders, New York: Guilford
Press.

Garner, D. M., & Garfinkel, P. E. (1979). The Eating Attitudes Test: An index of the
symptoms of anorexia nervosa. Psychological Medicine, 9(2), 273-279.

Garner, D. M., & Olmsted, M. P. (1984). Eating Disorder Inventory Manual. Odessa, FL:
Psychological Assessment Resources.

Garner, D. M., Olmsted, M. P., Bohr, Y. & Garfinkel, P. E. (1982). The Eating Attitudes
Test: Psychometric features and clinical correlates. Psychological Medicine, 12,
871-878.

Garner, D. M., Olmsted, M. P., & Polivy, J. (1983). Development and validation of
multidimensional eating disorder inventory for anorexia nervosa and bulimia.
International Journal ofEating Disorders, 2, 15-35.

Greenleaf, C. (2005). Self-objectification among physically active women. Sex Roles,
52(1-2), 51-62.