Objectification, Femininity Ideology, and Sexual Assertiveness: A Model of Women’s Sexual Functioning

Numerous research studies have investigated the relationship between sexual objectification and women's experience of their bodies , while fewer have focused on the possible role of endorsement of traditional femininity ideologies in helping to explain adult sexual functioning. The studies that do exist concentrate primarily on illuminating the differences between men and women versus exploring how endorsing particular attitudes or roles may affect women's experience of their bodies. This study investigated the relationships among traditional femininity ideology, sexual victimization, and self-objectification as well as how these variables related to body monitoring , sexual assertiveness for initiation and refusal, body shame, and appearance anxiety. All eight of these variables were investigated as they related to women's reports of sexual dysfunction. The primary investigator visited a large general psychology lecture in order to inform potential participants of the research opportunity and provide them with a general understanding of the study. Only women were recruited, as evidence sugges ts a greater incidence of sexual victimization among women and greater controve rsy regarding female sexuality two elements central to this particular study. Because of the focus on sexua l experiences and sexual health, students must have reported having engaged at least once in voluntary sexual activity or intercourse to be included in the final sample. While minimal discomfort was anticipated by participating in the study, students were asked to provide personal information regarding their sexual functioning and any history of sexual abuse. All informed consent and data collection were completed online via www .survevmonkev.com. No personally ident ifying information was collected, and students participated anonymou sly. Because no signatures were required for informed consent , there was no way to connect individual students to the private information they provided. The full survey consisted of approximately 130 items from the Chi ldhood Sexual Abuse Scale, Adu lt Sexual Victimizat ion Scale, Femininity Ideology Scale, Hyperfemininity Scale, Sexual Assertiveness Scale, Objectified Body Consciousness Scale, Self-Objectification Questionnaire, Adolescent Femininity Ideology Scale, Appearance Anxiety Scale, Female Sexual Function Index, Psychosexua l Functioning Scale, and the Balanced Inventory of Desirable Responding. Latent Variable Modeling was perfonned to analyze the research data using EQS software to investigate 13 main hypot heses primarily concerning sexual victimization, femininity ideology, and sexual assertiveness. As expected, a history of sexual victimization was positively correlate d with self-objectification and negatively correlated with sexual assertiveness . Endorsement of traditional femininity ideologies was negatively correlated with sexual assertiveness for initiation, but contrary to expectations, it was not correlated with selfobjectification or sexual assertiveness for refusal. Surprising ly, sexual assertiveness for refusal was directly related to sexual dysfunction. While the majority of relat ionships posited by Objectificat ion Theory were supported , neither body shame nor appearance anxiety was found to correlate with female sexual dysfunction. On the whole, this infonnation supports the body of knowledge concerning sexual victimization and objectification but challenges the assertion that the objectification variables traditionally studied relate to female sexual functioning. TABLE OF CONTENTS ABSTRACT

Demographics indicating percentage groups for each response option Table 2 Descriptive statistics Table 3 Variable correlation matrix Table 4 Variable and indicator abbreviations Table 5 Indicator loadings from final reduced-path LVM w ith all SAS-R parameters Table 6 Reliability statist ics for each survey measure included in L VM analyses.  Objectification Theory as applied to the experience of lesbian women in Kozee and Tylka (2006).     Objectification Theory, originally proposed by Frederickso n and Roberts (1997), is now supported by a growing body of research suggesting wome n are subjected to a myriad of negative psycholog ical and behavioral consequences stemming from experiences of sexual objectification (Slater and Tiggemann, 2002;Caloge ro, 2004;Roberts and Gettman, 2004;Tylka and Hill, 2004;Green leaf and McGreer, 2006;Grabe, Hyde, and Lindberg, 2007). The theory argues that exper iences of sexual objectification lead women to self-objectify and that this internalization of objectificat ion results in an increased incidence of such negative consequences as sexua l dysfunction. Studies suggest as many as 44% of women meet criteria for sexual dysfunction as recognize d by the Ame rican Psychiatric Association (Laumann, Paik, andRosen , 1999, Bancroft, Loftus, andLong, 2003); thus, studies exploring contributors to sexual dysfunction are clearly needed. Fredrickson and Roberts argue the relationship between self-objectification and sexual dysfunction is typically mediated by women's experiences of body shame, appearance anxiety, fewer peak motivational states, and a reduced awareness of internal bodily cues. This study sought to expand this model by using latent variable modeling to explore the roles of femininity ideology and sexual assertiveness. Sexual victimization was conceptualized as the vehicle of objectification in order to explore its overall relationship to women's sexua l functioning.

Justification of the Problem
An objectifying experience occurs when a woman's body is separated from her person and valued for its use to or by others; her body is taken as entirely representative of her being and is regarded as available for the use and pleasure of others. Fredrickson and Roberts ( 1997) argue objectification can occur in three primary ways: through actual interper sonal and social encounters, in visual media that depict these encounters, and through visual media's implicit sexualizing gaze. The objectification construct readily captures both personal and vicarious experiences of constant, unrelenting sexual observation and evaluation; however, Szymanski and Henning (2007) Aubrey (2006) operationalized objectification as the frequency with which study participants watched television programs or read magazines that emphasized the body and appearance as primary components of sexual desirability. Studies such as the current survey that explored what is arguably the most severe form of objectification, sexual victimization, were clearly needed. In this study, sexual victimization and femininity ideology were hypothesized to indirectly predict women's sexua l dysfunction as mediated by: self-object ification, sexua l assertiveness for initiation and refusal, body monitoring, body shame , and appearance anxiety.

Objectification and Self-Objectification
Amidst what they consider commonplace objectification, Fredrickson and Robert s (I 997) argue the most profound effect of being regularly evaluated as sexual beings is that women learn to self-objectify at an early age. While objectification experiences remain the original source of negative consequences, self-objectificat ion is the central feature of the authors' theory. Women are habituated to look at themselves as objects, objects to be judged and appreciated by others, by external observers. The authors define self-objectification as the process of internalizing an external sexual view of one's body, noting that the many social benefits conferred upon women who prioritize their appearance (e.g., popularity, upward mobility) serve to reinforce this model of behavior and self-pe rception, even when it is not considered equally necessary for men's advancement or social acceptance. Thus, in light of how physica l appearance and sexuality influence the way women are valued, attending to these factors can be seen not as simple vanity, but as women's strategy for predicting social acceptance and treatment. By outlining the concept of self-objectification in addit ion to transient experiences of external objectification, Fredrickson and Roberts allow for both trait and state-dependent objectification. Self-objectification has long been considered the crux of the authors' theoretical model.
While Fredrickson and Robert s (1997) attribute self-objectification to the internalization of exposure to sexually objectifying images or experiences, little research examining Objectification Theory has tested this specific relation ship or, even less commonly, the relationship between sexual victimization and self-objectification. Viewing self-objectification as the core component of the theory, most research has tested models that begin with selfobjectification, ignoring how women come to adopt this perspective or whether certain objectifying images or experiences are more influential than others. There are, however, some exceptions. Moradi, Dirks , and Matteson (2005) surveyed 221 undergraduate women (M = 20.42 y/o ), asking them to report the frequency with which they were subject to a list of seven objectifying experiences (e.g., sexist comments about clothing, whistling) and to rate their degree of acceptance of various societa l standards of beauty ( e.g., "I wish I looked like a swimsuit model"), a key component of self-objectification. The author s found report s of experiencing objectification were positively correlated with reports of internalizing sociocultural standards of beauty, thus providing support for the relationship between objectifying experiences and selobjectification.

Body Monitoring
A natural consequence of being aware of the central role played by one's physical self is an increased amount of attention being paid to one's physical self, a habit referred to as "body monitoring." While Fredrickson and Roberts' (1997) original theory positions self-objectification as a precursor to body monitorin g, other researcher s have sugges ted body monitoring is itself a form of self-objectification (McKinley and Hyde, 1996). Thi s confusion has received recent attention in the literature (Tiggemann and Lynch, 2001;Tiggemann and Slater, 2001) , suggesting the two constructs constitute independent paradigms; nonetheless, when considered either synonymous with self-objectification or a behavior stemming from it, body monitoring refers to the habitual checking or imagery women engage in to monitor how their bodies appear to others.
This behavior may take the form of checking one's reflection when passing storefront windows or other reflective surfaces, visualizing one's body in a particular position or posture , adjusting one 's clothing to best concea l less socially accepted physical features, etc. Considering both the ever-present media images of idealized female bodies and also the benefits women see distributed according to approximations of that ideal, it is no wonder women are plagued by a continuous habit of monitorin g their own bodies and comparing them to the cultura l ideal. Tylka and Hill (2004) explored the relationship between a specific type of objectification, pressure for thinness, and body monitoring, conceptualizing the latter as a form of self-objectification. They surveyed 460 undergraduate women (M = 2 1 y/o), and, as predicted, latent variable modeling revealed pressure for thinness was a significant predictor of body monitoring, accounting for 26% of the variance in scores.

Body Shame
Since so few women are anywhere near capable of attaining society's idealized version of beauty despite diet, exercise, the latest fashion, beauty products, and developments in cosmetic surgery, the combination of knowing both what a male-dominated culture deems beautiful and also that that beauty is unattainable is a clear recipe for what Fredrickson and Roberts (1997) describe as body-based shame. In essence, women regularly evaluate themselves relative to an internalized cultural ideal based on phy sical appearance, and they lose every time. Given the value placed on women's physical appearance, and the inability to live in a soc ial env ironm ent without exposing one's physical self, this discrepancy translates into feelings of shame and worthlessness. Conversely, it seems possible that women who judge themselves to closely approximate beauty ideal s actually derive pleasure from view ing their bodies; they may experience a kind of body pride when aware that others would likely evaluate their bodies positively. To date, this hypothesis has not been studied in Objectification Theory research. Quinn, Kallen, and Cathey (2006) investigated the effect of self-objectification on body shame and lingering body-related thoughts among 150 men and women. Participants were asked to try on either a one-piece bathing suit or a v-neck sweater in a private dress ing room. The y then completed surv ey materials that included a self-objectifi cat ion manipulation check that asked participants to think about how wearing the clothing item made them feel before filling in 20 selfstatements regarding those feelings. Additionally, participants completed 19 items measuring experiences of body shame. As expecte d, results revealed only women reported increased body shame after the self-obje ctifying experience and that this shame med iated the extent to which self-obje ctification led to continued thoughts about the body during a subsequent free-response activity .

5
Appearance Anxiety Fredrickson and Roberts (1997) propose appearance and safety anxieties result from women not being able to anticipate when their bodies will be looked at or how they will be evaluated . Firstly , when the message is internalized that women are represented by their physical bodies and that their physica l beauty determines their overall value, it becomes clear how important women may find monitoring their bodies and how anxiety-producing it may be to constantly anticipate being looked at or evaluated. Secondly, given the relationship between physical attractiveness and perceived responsibility for sexual assault (Beneke, 1982;Jacobson and Popovich, 1983), women are not only preoccupied with their physical appearance for purposes of being accepted by others (appearance anxiety), but they are also plagued by concerns of how their physical appearance may attract unwanted attention from other s (safety anxiety).
Society encourages women to internalize an external view of themselves as sexual beings, but this internalization results in anxiety regarding both falling short of beauty standards and the poss ibility of being victimized because of approximations of those beauty standards. In a study of 160 undergraduate men and women, Roberts and Gettman (2004) investigated the relationship between a primed state of self-objectification and participants ' reports of appearance anxiety.
Students were instructed to complete a scrambled sentence task where 15 of 25 words were related to self-objectification, (e.g., slender, desirab le) body competence, (e.g. fitness, energetic) or neutral control words (e.g ., silly, honesty); participants then completed a scale assessing individuals' feelings of appearance anxiety. MANOV A results revealed a main effect for gender on the appearance anxiety measure , indicating women reported more appearance anxiety than men across the three conditions. Furthermore, women primed with self-objectification reported higher levels of appearance anxiety than those primed with body competence words. Both results are consistent with Objectification Theory's hypothesized relationship between selfobjectification and appearance anxiety in that those women who internalized unreali stic cultural 6 standards of beauty report ed greate r anxiety concerning their physical appearance, a relationship not replicated among the men surveyed

Sexual Dysfunction
Objectification Theory culminates in an evaluation of the way experiences of objectification influence depression, eating disorders , and sexual dysfunction, problems that occur most often among American women. In discussing sexual dysfunction, Fredrickson and Roberts ( 1997) argue that the shame and anxiety many women feel in regard to their physical selves may carry over into their sexual exper iences, experiences which make women undeniably physically vulnerable . Objectification Theory's framework allows for two equally viable paths in predicting health consequences: an indirect path where an accumulation of the subjective consequences of objectificat ion contributes to physica l or psychological distress and a direct path where the overt use of a women's body as a tool for another's pleasure (e.g. , sexual victimization), leads to poor mental health. Most of the research to date focuses on the indirect paths to depression and eating disorders but neglects objectifica tion ' s relationship to sexual dysfunction. Likewise, research examining objectification via sexual victimization specifically is lacking. The current study looked to bridge the gap by examin ing indirect paths between sexual victimization and participant s' reports of sexual dysfunction. Aubrey (2007) investigated the relationships among exposure to sexually objectifying media, body self-consciousness, negative body emotions (body shame, appearance anxiety), and sexual self-perceptions in a samp le of 384 undergraduate men and women (M = 19.6 y/o).
Defining body self-consc iousness as the combinat ion of self-objectification and body monitoring, her study integrates the Fredr ickson and Roberts (1997) and McKin ley and Hyde ( 1996) frameworks. The author hypothesized exposure to sexually objectifying media images would be related to poor sexual self-perceptions as mediated by body self-consciousness. Regression equations revealed objectifyi ng media exposure predicted body monitoring but not self-7 objectification, meaning the predicted effect on body self-consc iousness was only partially supported. Body monitoring, but not self-objectification, was found to fully mediate the relationship to body shame and appearance anxiety; however, both self -objectification and body monitoring were found to predict body-related self-consc iousness during physical intimacy, indicating participants who reported greater self-objectification and body monitoring were more likely to also report concerns about feeling unattractive or fat during physical intimacy. Sanchez and Kiefer (2007) found a similar result in a survey of 320 men and women recruited online . Both men's and women's reports of body shame were significantly related to increased body-related self-consc iousness during physical intimacy. Structural equation modeling revealed a negative relationship between body shame and sexual arousability that was fully mediated by body-related self-consciousness, while sexual pleasure was predicted by bodyrelated self-consciousness only . Orgasm difficulty was assessed; however, no relationship was revealed between either body shame or body-related self-consciousness and difficulty achieving orgasm. See Appendix A for a detailed synthesis of recent research investigating the traditionally studied objectification variables.

Considerations for Theory Expansion
Since Fredrickson and Roberts (1997) introduced Objectification Theory, most investigations into its major tenets have varied little from the origi nally proposed model. A limited amount of research has begun to address means of modifying or expanding the theory to incorporate supplementary constructs (see Appendix B). Two particular fields still offer enticing untapped information: femininity ideology and sexual assertiveness.

Femininity Ideology
The focus of much gender-re lated research is shifting from studying women's persona l behaviors ( enactment of feminine gender roles) to their endorsement of more global attitudes regarding women and gender (gender ideology ). While gender roles refer to the performance of a particular pattern of difference based on distinctions between men and women (Acker , 1992), 8 gender ideology references attitudes . toward widespread norms or rules regarding a particular gender (Sinn, 1997). Gender role research most typically investigates heterosexual interact ions and involves heterosexual samples; however, Kozee and Tylka (2006) explored the appropr iateness of Fredrickson and Roberts' ( 1997) model for the experiences of lesbian women (see Appendix C). Their results suggest the model best captures the correlates of objectification for lesbian women when slightly modified, but the basic structure was nonetheless supported (see Figure 2).
Objectification Theory is centered on women's internalization of the message that the physical appearance of one's body determines her value and society decides what bodies are beautiful. While messages of this sort are pervasive, so are messages regarding how women should behave with those bodies. Longstanding gender roles promote the claim that a woman's responsibility is to respond to her partner's sexua l initiations and only use sex to confirm relationship commitment or promote relationship intimacy (Amaro, Raj, and Reed, 2001 ). A compelling area for expansion of Objectification Theory is the examination of how women's internalization of messages regarding appropriate gendered behavior for women (femininity ideology) influences their experience of their bod ies as sexual beings. Research already suggests women who endorse traditional gende r roles may be more likely to experience adult sexual victimization and less likely to assert themselves in initiating wanted sex or refusing unwanted sex (Williams, 2009), but whethe r gender ideology fits into a broader model incorporating the traditional objectification variables remains unknown.

Sexual Violence
Fredrickson and Roberts ( 1997) describe multiple forms of objectification, one of which is sexual violence, but research has focused on media images and interpersonal encounters like male gaze to the exc lusion of more severe ly objectifying experiences such as sexual victimization. An extensive body of literature exists highlighting the negative effects of both childhood and adult sexual victimization, suggesting chi ldhood sexual abuse is related to poor 9 sexual self-esteem (Van Bruggen, Runtz , and Kadlec , 2006), less sexual assertiveness for refusing unwanted sex (Morokoff, et al., 1997;Williams , 2009), adult sexual victimization, and an increased risk for HIV and other STDs (Polusny and Follette, 1995

Sexual Assertiveness
Recent studies have begun connecting experiences of victimization to women 's sexual assertiveness, showing adult sexual victimization is negatively related to women ' s tendency to refuse unwanted sex (Testa, VanZile-Tamsen, and Livingston, 2007;Williams , 2009). Given the research supporting Objectification Theory , and the notion that sex ual victimization is an extreme form of objectification , an interesting area for new research is to investigate whether any of the originally proposed variables are simila rly associated with women 's report s of sexual assertiveness. It seems reasonable to hypothesize that wome n who hav e internalized a view of themselves as sexual objects would be less likely to assert themselves when presented with unwanted sexual advances. Similar ly, those who believe a woman's value comes from providing pleasure to her partner may be less likely to assert their own sexual interests since this behavior deviates from the notion that women's bodies are tools to be used by and for others .

Overview of Model and Major Hypotheses
This study investigated the predictors that discriminate among women's reports of sexual dysfunction . Both indirect and direct paths were investigated as well as some hypothesized correlations among the independent variables . Sexual victimization and femininity ideology were evaluated as independent variables. Mediating variables included self-objectification, body monitoring , sexual assertiveness for initiation, sexual assertiveness for refusal, body shame, and appearance anxiety. Subscales or item parcels served as multiple indicators for independent as well as mediating variables. No specific hypotheses were established for the relationship between sexual victimization and femininity ideology or between sexual assertiveness and body shame or appearance anxiety, but these relationships were investigated as potential expansions to Fredrickson and Roberts ( 1997) original model. The following hypotheses were tested :

1.
Sexual victimization will be a positive predictor of self-objectification.

2.
Sexual victimization will be a negative predictor of sexual assertiveness for initiation.

3.
Sexual victimization will be a negative predictor of sexual assertiveness for refusal.

4.
Femininity ideology will be a positive predictor of self-objectification .

5.
Femininity ideology will be a negative predictor of sexual assertiveness for initiation .

6.
Femininity ideology will be a negative predictor of sexual assertiveness for refusal.

7.
Self-objectification will be a mediating positive predictor of body monitoring .

8.
Self-objectification will be a mediating negative predictor of sexual assertiveness for initiation.

9.
Self-objectification will be a mediating negative predictor of sexual assertiveness for refusal.

10.
Body monitoring will be a mediating positive predictor of body shame.

11.
Body monitoring will be a mediating positive predictor of appearance anxiety.

12.
Body shame will be a mediating pos_ itive predictor of female sexual dysfunction.

13.
Appearance anxiety will be a med iating positive predictor of fema le sexua l dysfu nction.

Participants
Because Latent Variable Modeling is one form of Structural Equation Modeli ng (SEM), a minimum of 300 participants was needed to complete the process in order to argue statist ical reliability and accuracy. The present research sample was composed of volunteer participants from a general psychology course at the University of Rhode Island . Individuals received course credit for volunteering in the study and were recruited by the researc her with a short description of the study during their class. While the investigation of femininity ideology could involve both men and women, because of the higher prevalence of sexual victimization among girls (Zier I er et al.,199 1) and the more socially controversial idea of female sexuality (Muehlenhard and McCoy,199 1), this study only recruited female participants. Add itiona lly, whi le the study was expected to draw a primarily heterosexual sample, Kozee and Tylka 's (2006) work recaps the importance of recruiting women identifying with diver se sexual orientations. Criter ia for inclusion of participant data in the research analyses included female status , being 18 years of age or older, and having engaged at least once in sexual activity or intercourse . All female psychology students 18 or older were recruited to participate regard less of age, ethnicity, sexual orientation, or stage of education; however, tho se students who reported they had never been sexually active were not included in the final samp le. See Appen dix D for a description of the racial diversity that was expected based on the demographics of the University student body at the time of recruitmen t. See Appe ndix E for a discussion of factors that influence sample size requirements.

Mate rials
See Appendix F for a copy of the comple te survey.
Demograph ics. Participant s we re asked a number of questions regardi ng age, ethnicity , year in school, religion and history of sexual experience .
Sexual . Victim ization. Five items adapted from Wyatt ( 1985) and used in Whitmire et al. (1999) were used to assess participants' history of childhood sexual abuse (e.g., "Before you were 14, did anyone older ever try to make you touch his or her genitals?"). Abuse experience items were scored on a 4-point scale from I = "no" to 4 = "many times." Cronbach's alpha reported for the full scale was .95. Twelve items adapted from the Sexual Victimization Scale (Koss and Oros, 1982) and used in Whitmire et al. ( 1999) were used to assess adult sexual victimization (e.g., "Since the age of 14, have you ever had a partner mistake how far you wanted to go with sex?"). Items were scored on a 4-point scale from I = "definitely no" to 4 = "definitely yes." Cronbach's alpha reported was .91. These 17 items provided 4 composite scores evaluating touch and penetrative childhood sexual abuse as well as adult sexual coercion and use of force.
Femininity Ideology. Ten items from two subscales of the Femininity Ideology Scale (FIS: Lehman, 2000) and five items adapted from the Hyperfemininity Scale (Mumen and Byrne, 1991) were used to assess traditional feminine gender ideology. Selected items evaluated participants' attitudes concerning women's appropriate sexual behaviors as well as behaviors appropriate for heterosexual interactions (e.g., "Women should not initiate sex" and "Women should act sexy to get what they want from men"). Items were scored on a 5-point scale from I = "strongly disagree " to 5 = "strong ly agree." Cronbach's alphas reported for complete Dependence/Deference and Purity FIS subscales were .76 and .85 respectively. Cronbach's alpha reported for the full Hyperfemininity Scale was .89. See Appendix G for a discussion of the item selection process for the femininity ideology construct.
Self-Objectification. The Self-Objectification Questionnaire (Noll and Fredrickson, 1998) was used to assess participants' physical self-concept. Participant s were asked to rank ten body attributes in order of importance to their phys ical self-concept. Five attributes were appearance-based (e.g., weight) and five were competence-based (e.g., muscular strength).
Higher scores, as determined by the difference between appearance-and competence-based subsca le scores, reflected participants' attention to observable attributes over non-ob servable attributes. Additionally, five items from the Objectified Relationship with Body (ORB) subscale of the Adolescent Femininity Ide ology Scale (AFIS: Tolman and Por sche, 2000) were used to 13 assess participants' internalization ofan objectified view of their bodies (e.g ., "I think that a girl bas to be thin to feel beautiful"). Items were scored on a 6-point scale from I = "strongly disagree" to 6 = "strongly agree ." Cronbach's alpha reported for the complete ORB subscale was .81.
Body Monitoring. The Body Surveillance subscale of the Objectified Body Consciousness Scale (McKinley and Hyde, 2006) was used to assess participants' degree of body monitoring (e.g., "During the day, I think about how I look many times"). Items were scored on a 7-point scale from 1 = "strongly disagree" to 7 = "s trongly agree." Cronbacb's alpha reported was .89.
Sexual Assertiveness. Twelve items from the Sexual Assertiveness Scale (Morokoff et al., 1997) were used to assess initiation and refusal assertiveness (e.g., "I Jet my partner know ifl want my partner to touch my genitals" and "I refuse to have sex ifl don't want to, even ifmy partner insists"). Items were scored on a 5-point scale from 1 = "never" to 5 = "always." Cronbach's alphas reported were .82 and .79 respectively.
Body Shame. The Body Shame subscale of the Objectified Body Consciousness Scale (McKinley and Hyde, 2006) was used to assess participants' degree of body shame (e.g., "I would be ashamed for people to know what I really weigh"). Items were scored on a 7-point scale from 1 = "strongly disagree" to 7 = "strongly agree." Cronbach's alpha reported was .75.
Appearance Anxiety. The Appearance Anxiety scale (Dion, Dion, and Keelan, 1990) was used to assess participants' experie nce of anxiety regarding their physical appearance (e.g., "I feel nervous about aspects of my physical appearance"). Items were scored on a 4-point scale from 1 = "sometimes" to 4 = "almost always." Cronbach's alpha reported was .86.
Female Sexual Dysfunction. Three subscales of the Female Sexual Function Index (Rosen, et al., 2000) were used to assess participants' overall sexual dysfunction according to phases of the sexual response cycle (e.g. "Over the past four weeks, how often did you feel sexual desire or interest? "). Items were scored on 5-point scales assessing frequency, intensity, confidence, difficulty, and satisfaction regarding sexual desire , arousa l, and orgasm. Cronbach's alphas reported were .89, .90, and .91 respectivel y. Three items from the Psychosexual Functioning Scale (Harlow et al., 1993) were also used to assess participants ' overall sexual satisfaction and attitudes about their personal sex lives (e.g., "I do not like some parts ofmy sex life"). Items were scored on a 5-point frequency scale from O = "never" to 4 = "always." Cronbach ' s alpha reported for the full scale was .85. Lastly , two items were included to assess participants' feelings of sexual distress and shame (e.g., "Over the past four weeks, how discouraged have you felt by your sexual response?") . These items were scored on a 5-point scale from I = "not at all" to 5 = "extremely." Response Bias. Ten items from the Impression Management subscale of the Balanced Inventory of Desirable Responding (Paulhus, 1991) were included to measure participants' tendency to respond with socially desirable answers (e.g., "I sometimes tell lies ifI have to").
Items were scored on a 7-point sca le from 1 = "not true" to 7 = "very true ." Cronbach's alpha for the full subscale has been reported between . 75 and .86.

Procedure
After receiving Institutional Review Board approval, the researcher visited a large genera l psychology lecture in order to provide students with a brief descript ion of the study and what was being asked of study part icipants. Students received the incentive of class credit for participating. In order to minimize discomfort related to disclosing personal information about one's sexual functioning and victimization experiences, participant anonymity was provided through the use of online surveys; participants completed all survey tools online through www.surveymonkey.com. Confidentiality, participation information, and an informed consent document were provided at the beginning of the survey. Participants indicated they understand this information, and no signatures were required , thus preserving participant anonymity.
Participants accessed the online survey at their leisure and had the duration of one semester to participate. In the unlikely event the survey material caused personal distress , referrals to the Psychological Consultation Center and URI Counseling Center were provided to each participant.
Contact information for the researcher was also provided if questions or concerns arose regarding participation in the study.

Results
Women's experience of their bodies is complex and multifaceted, hardly explained by isolated relationships between individual independent and dependant variables . Using a model that allowed for an understanding of complex, multifaceted relationships was thus ideal. Due to its sanction . of mediating variables, Latent Variable Modeling produced a robust understanding of the relationships among established Objectification Theory variab les, femininity ideology, and sexual assertiveness. As recommended, a small global significance statistic (y:), a large Comparative Fit Index (CFI>.90), a small Root Mean Squared Error of Approximation (RMSEA<.06), and significant individual regression parameters with residuals less than .20 were used as criteria when evaluating the fit of the investigated model and whether relationships among variables we re adequately explained (Schumacker and Lomax, 2004). Lagrange multiplier (LM) and Wald tests were used to improve the fit of the model when adjustments to the originally proposed parameters were both statistically and theoretically supported.

Demographics and Descriptives
Of the 3 77 participants included in the final data set, the highest percentage identified as White (80.11 %), Catholic (59.42%), and somewhat religious (29.71 %). Table I lists the percentage groups for all surveyed demographic questions. Table 2 lists descriptive statistics for all variables included in major analyses. Table 3 provides a complete correlation matrix including both factor indicators and also full scale latent variables.

Preliminary Analvses
Confirmatory factor analysis (CFA) was used to explore the goodness-of-fit between the data and the proposed measurement model. Eight independent and one dependent variable were included as assessed by eighteen indicators. A significant i test of fit revealed the originally proposed model deviated significantly from the data (y: (314) = 900.683, n = 337, p < .001). The Confim1atory Fit Index (CFI) and Root Mean Square Error of Approx imation (RMSEA) , two additional model-fit statistics, also indicated the model was a poor fit with values of .84 and .08 respectively.
In an attempt to improve model fit, CF As were run to explore the effects of dichotomizing and combining the poorly loading childhood sexua l abuse items; however, the fit was still poor (r (314) = 898.14, n = 337, p < .001, CFI = .85, RMSEA = .07; x2 (288) = 698.87, n = 337, p < .001, CFI = .89, RMSEA = .07). Given this negligible improvement, a series of CF As was next run to assess how eliminating other va riable indicators with poor loadings (<.30) improved the model's fit. The childhood sexual abuse items were deleted from the variables indicating Sexual Victimization; items from the Hyperfemininity Scale were deleted from those indicating Femininity Ideology; Desire and Orgasm subscales were deleted from those indicating Female Sexual Dysfunction, and Physical Self-Concept (PSC) was dropped as an indicator of Self-Objectification. Because the removal of PSC left Self-Objec tification with only one indicator, the five items from this remaining indicator were serially divided into two item parcels.
Additionally, participants who had been previously deleted due to missing responses on the measures listed above were reinstated once those measures were removed from the model, increasing the final data set from 337 to 377 participants. Ultimately, the best goodness-of -fit was found once all of the poorly loaded indicators listed above had been removed (x2 ( I 95) = 427.53, n = 377, p < .001, CFI = .94, RMSEA = .06). While the x2 test of fit was still found to be significant, this was likely due to the large sample size. Overall, results of the final CF A suggested the proposed model, with minor adjustments in indicators, was a good fit for the data.
Abbreviations for all retained var iables and indicators can be found in Table 4.

Latent Variable Modeling
Once the measurement model was successfully adjusted for fit, the data was subjected to Latent Variable Modeling (LVM). The original LVM analysis ( Figure 3) revealed adequate goodness-of-fit (r (209) = 535.12, n = 377, p < .001, CFI = .91, RMSEA = .06). Lagrange multiplier test results sugges ted two theoretically ju stified additional parameters: SO predicting BS and SAS-R predictin g FSD. A modified model ( Figure 4) was thus analyzed with a second L VM. Eight latent variables were hypothesized to predict one outcome measure through both direct and mediat ed paths. Each latent variable was indicated by at least two manifest variables, and most indicator s were item parcels devised from large mea surement tools. · Once again, the y: test offit was significant (x2 (207) = 468.04 , n = 377, p < .00 I) ; however , additional fit statistics suggeste d the modified model was a goo d fit for the data (CFI = .93, RMSEA = .06).
Of the 19 parameters investi gated in the modified model, 7 involved only variables traditionally studied in Objectification research. Of tho se seven par ameters, five were found to be significant. SY (.3 1) was found to be significant ly associated with SO (R 2 = . I 0), indicating women who experienced sexua l victimization were more likely to rep ort greater selfobjectificatio n than those without vict imization histories. SO was also found to be significant ly related to BM (R 2 = .46) and BS (R 2 = .43) with corre lation coeffici ents of .68 and .46 respectively , indicating great er experience of self-objectificati on was related to more body monit oring and body shame. BM was also found to be significantly associated with AA (R 2 = .40) and BS with corre lation coefficients of .60 and .23 respecti vely, indicating body monitoring was associated with greater experienc e of appearance anxiety and body shame. While these results all support the established Objectification Theory model , neither BS nor AA was found to be significantly associated with FSD, meaning participant s who reported body shame or appearance anxiety were no more likely to report sexual dysfunct ion than those who did not. BS and AA were, however, sign ificantly related to each other.
FGI , SAS-I, and SAS-R were investigated as potential additions to the well-established Objectification Theory model. They contributed twelve parameters to the analyzed model, and results revealed five of these parameters were significant. SAS-I (R 2 = .08) was found to have a significant negat ive relationship with SY (-. I 8) and FGI (-. I 8) but not SO. The relationship between SAS-I and FGI was the only relationship involving FGI found to be significant. Results revealed women who reported traditional femininity ideologies we re not likely to report more self-objectification or less sexual assertiveness for refusal as originally hypothesized.
Furthermore , SAS-I was not significantly related to BS or AA as originally expected. These results suggest women who reported experience of sexual victimization or who endorsed traditional femininity ideologies were less likely to report sexua l assertiveness for initiation. Less sexual assertiveness for initiation did not, however, relate to participants reports of body shame or appearance anxiety.
While only two of the parameter s involving SAS-I were significant, three of the relationships involving SAS-R found support in the data. A signi ficant negative relation ship was found between SV (-.33) and SAS-R (R 2 = .14), indicating women who reported experiences of sexual victimization were less likely to report sexual assertiveness for refusal. The hypothe sized relationship between SO and SAS-R was not found to be significant. A significant negative relation ship was also found between SAS-R (-.12) and AA, meaning women who reported greater sexual assertiveness for refusal were less likely to report feelings of appearance anxiety. The relationships in the modified L VM mimicked those in the original L VM with one exception related to SAS-R. In the original model, the direct relationship between SO and BS was not included , and results reveal ed a sma ll but significant negative association between SAS-R (-. 17) and BS (R 2 = .35). With the addition of the parameter from SO to BS in the modified model, the relationship between SAS-Rand BS lost sign ificance, but the variance explained among body shame scores increased to 43%.
Of the three variables ultimately investigated as predicto rs ofFSD, only the parameter added in the modified model was found to be significant. Results revealed FSD (R 2 = .19) was significantly associated with SAS-R (-.34) but not the two variab les originally hypothesized to predict it, BS and AA. These results suggest women who experienced sexual dysfunction were 20 less likely to be sexually assertive in refusing unwanted sex. Cont rary to study hypotheses, however, women with sexual dysfunction were not more likely to report body shame or appearance anxiety. FSD exhibited a medium effect size indicating approximately 19% of the variance in reports of female sexual dysfunction was explained by model variables.
A series ofLVMs was next conducted to explore bow min imizing or eliminating the contributions ofFGI, SAS-I, and SAS-R would affect the overall model fit. SAS-R was found to be significantly related to SV and AA, and it also had a strong relationship to the model's only outcome variable, FSD . While SAS-I was significantly related to SV and FGI, its contribution to the model beyond that was negligible. Neither FGI nor SAS-I was found to be significantly related to the variables traditionally studied in Objectification research , thus reducing or eliminating their contributions to the model was expected to improve parsimony as well as goodness-of-fit.
When all prediction parameters from FGI, SAS-I, and SAS-R to other model variab les were eliminated ( Figure 5), BS C.20) was found to be significant ly related to FSD, but only 5% of the variance in sexual dysfunction was explained, a drop from 19%. Likewise, goodness -of-fit was found to be significantly worse Ci! C216) = 507.242, n = 377, p < .00 1, CFI = .92, RMSEA = .06), indicating the statist ically significant paths from SAS-R to AA and FSD did play an important role in the fit of the investigated model. When the stronger of these parameters CSAS-R to FSD) was reinstated, L VM results revealed a x2 test of fit still worse than that of the modified model Ci! C215) = 4 77.66, n = 377, p < .00 I); however, additiona l fit statistics suggested equal fit CCFI = .93, RMSEA = .06). This change was stat istically no different than reinstat ing both parameter s Ci! C214) = 474.79 , n = 377, p < .001, CFI = .93, RMSEA = .06), thus the relationship between SAS-R and FSD was shown to be a particularly important addition to the model. When SAS -R was included as a predictor of AA, FSD, and BS as originally propo sed, L VM results revealed a x2 goodness -of-fit statistically no different than the modified model Ci! 21 (213) = 471.28, n = 377, p < .001, CFI = .93, RMSEA = .06). Because the variance in sexual dysfunction scores improved slight ly to 20% and fewer parameters were included than in the modified version, it could arguably be considered .the more pars imonious of the two. The correlation coefficients and effect sizes for each statistically significant parameter in this last reduced-path model can be found in Figure 6.
The L VM results described above illustrate mixed support for the investigated hypotheses. Hypotheses 1, 2, and 3 proposed SV would be positively correlated with SO and negatively correlated with SAS-I and SAS-R. In support of these hypotheses, women's reports of sexual victimization were found to be associated with greater self-objectification, less sexual assertiveness for initiation, and less sexual assertiveness for refusal. Hypothesis 5 proposed FGI would be negatively associated with SAS-I. In support of this hypothesis, results revealed women who endorsed more traditional femininity ideologies were less likely to report sexual assertiveness for initiation. Conversely, results did not support Hypothesis 4 or 6 which argued FGI would be positively correlated with SO and negatively corre lated with SAS-R . Women who reported traditional femininity ideologies were not more likely to report self-objectification or less likely to report sexual assertiveness for refusal. Hypothesis 7 proposed SO would be positively associated with BM, and results supported this claim. Women who reported greater self-objectification were more likely to also report body monitoring. Conversely, results did not support Hypothesis 8 or 9 which proposed SO would be negative ly correlated with SAS-I and SAS-R. Women who reported engaging in more self-objectification were not found to report less sexual assertiveness for either initiation or refusal. Hypotheses 10 and 11 proposed BM would be positively correlated with BS and AA. Results supported both hypotheses as women who reported frequent body monitoring were more likely to also report feelings of body shame and appearance anxiety. Hypothesis 12 proposed BS wou ld be posit ively related to FSD. Results revealed support for this claim, but only when SAS-R was not included in mode l analyses. The significant statistical relationship between women's reports of body sham e and sexual 22 dysfunction disappeared when sexual assertiveness for refusal was investigated simultaneously.
Contrary to Hypothesis 13, results did not support a relationship between AA and FSD. Women who reported appearance anxiety were not more likely to report sexua l dysfunction.
In addition to the hypothesized relationships investigated , four exploratory parameters were also analyzed to evaluate whether a relationship existed among sexual assertiveness, body shame, and appearance anxiety. No significant relation ship was found between SAS-I and either BS or AA; and relation ships involving SAS-R were inconsisten t. Results from the origina l L VM revealed SAS-R was negatively correlated with BS and AA, suggest ing women who reported more sexual assertiveness for refusal were less likely to experience either body shame or appearance anxiety. When the relation ship between SO and BS was added to the model, however, the relationship between SAS-R and BS lost significance. Furthermore, the relationship between SAS-Rand AA was only found to be significant when parameters connecting SAS-R to BS, AA, and FSD were all analyzed simultaneously. Clearly the relationsh ip between women's sexual assertiveness for refusal and feelings of body shame or appearance anxiety needs further exploration.
Standardized residuals, LM and Wald tests revealed no meaningful way to improve the model , but a final L VM was run to compare the fit of the reduced-path models to a model in which FGI and SAS-I, two poorly performing variables, had been removed completely ( Figure 7).
This condensed model had a better i goodness-of-fit than any of the reduced-path models (x2 (122) = 299.3 7, n = 377, p < .001, CFI = .95, RMSEA = .06); however, it is not a nested model and cannot be directly compared. Overall, it could arguably be considered the most parsimonious given it was revealed to be a good fit to the data with fewer variab les than the modified or reduced-path models; however, the explained variance in sexual dysfunct ion was slightly lower at 18%. The correlation coefficie nts and effec t sizes for each parameter in the condensed model can be found in Figure 7. Table 5 provides a list of all factor ind icators included in final analyses and 23 their individual loadings as revealed in the last reduced-path model. Table 6 lists Cronbach 's alpha reliability statistics for each factor measure included in L VM analyses .

Discussion
The principal aim of this study was to explore whether Objectification Theory could be meaningfully expanded to include femininity ideology and sexual assertiveness. Secondly, unlike most Objectification research, this study incorporated both external and internal experiences of objectification, operationalizing external objectification as the experience of sexual victimization instead of exposure to sexualized media or whistling, etc. as commonly captured. Lastly, sexual dysfunction has long been hypothesized to be a consequence of objectification and the related experiences of body shame and appearance anxiety; however, most research in the field has targeted disordered eating or depression as the outcome of interest. It became clear while evaluating the L VM results that sexual assertiveness for refusal adds to the understanding captured by the Objectification model. Furthermore, sexual victimization in particular was found to relate to women's experience of self-objectification just as past research has shown this relationship when investigating less severe forms of objectification. While femininity ideology and sexua l assertiveness for initiation were not found to contribute to expanding the Objectification model, the lack of relationship between female sexual dysfunction and either body shame or appearance anxiety challenges past assertions that objectification ultimately leads to problems with sexual functioning.
Strong support was found for the Objectification model as traditionally studied. With the exception of the lack of relationship between female sexual dysfunction and either body shame or appearance anxiety, there were no surprises in how the Objectification Theory variables performed. While not included as a direct relationship in the model as originally proposed, the revelation that women who reported more self-objectification also reported more body shame does have support in prior research (e.g., Quinn, Kallen, and Cathey, 2006). This finding conflicts with results of Aubrey (2007) which also revealed a fully mediated relationship and served as the basis for how these variables were incorporated into the original model. Aubrey's study may also help clarify the lack of association with female sexual dysfunction. As discussed previously, in Aubrey's research, reports of body self-consciousn ess during physical intimacy functioned as indicators of sexual dysfunction. In the current study, female sexual dysfunction was operationalized as difficulty with physiological arousal as well as feelings of distress or dissatisfaction with one's sex life. It could be argued that the discrepancy in result s is due to the discrepancy in operationalization of dysfunction. Additional research may find that Objectification Theory explains female sexual dysfunction · as relate s to women's percept ions of their bodies during sex but not as relates to physiological measures or general satisfaction with sexual experiences. Alternatively, because a relationship between body shame and female sexual dysfunction was revealed when a condensed version of the model was analyzed, the role of sexual assertiveness for refusal also be responsible. As discussed previously, Sanchez and Kiefer (2007) found body-related self-consciousness to be a mediator of the relation ship between body shame and sexual dysfunction as measured by difficulty with arousal. This particular variable was not included in the present study, but the results argue for the same type of relationship to female sexual dysfunction as presented in the current study, a mediated one. While additional research may reveal sexual assertiveness for refusal to be a reliable correlate of female sexual dysfunction, its role in the current model remain s somewhat unclear.
As discussed, the primary objective of the current study was to exp lore possible expansions of Objectification Theory. Result s did not reveal support for broadening the Objectification model to include femininity ideology or sexual assertiveness for initiation; however, sexual assertiveness for refusal was exposed as an important, if somewhat inconsistent, additional factor. The relationship revealed between sexual victimization and sexual assertiveness for refusal was in line with previous research, show ing that women who experience sexual coercion or vict imization by force are less likely to assert themselves in refusing unwanted sex. The unique contribution of sexual assertiveness for refusal, however, is in its relationship to the traditionally studied Objectification Theory variables. Negat ive relationships were found between sexual assertiveness for refusal and both body shame and appearance anxiety as well as 26 between sexual assertiveness for refusal and female sexual dysfunction. This findin g suggests women who are less likely to assert themselves in unwanted sexual situations are more likely to experience shame, anxiety , and dysfunction as relates to their bodies and sexual experiences. As mentioned, the role of sexual assertiveness for refusal is somewhat unclear, as it was no longer revealed as a significant correlate to body shame when the relationship between selfobjectification and body shame was added to the model. What never changed, however, was the strong association to female sexua l dysfunction.
In reviewing the se relationships, a pattern emerges whereby one may hypothesize objectification leads to body monitoring which leads to appearance anxiety. This anxiety relates to less assertiveness for refusing unwanted sex which then results in problems wit h sexual functioning and sat isfactio n. It must be kept in mind, however, that the nature of cross-sectional data precludes drawing causal conclusions. Thus , hypothesizing that female sexual dysfunction leads to appearance anxiety, reduced sexual assertiveness for refusal and more frequent body monitoring is equally likely based on this data. Neverthe less, given the longitudinal evidence that supports Objectification Theory (Aubrey, 2006;Grabe, Hyde, and Lindberg (2007), justification exists for further researching directional paths that result in fema le sexual dysfunction.
While the variables traditionally studied in Objectification research performed as expected, the weaknesses in the model were the exploratory relationships investigated in an attempt to link Objectification Theory to femini~ity ideology and sexua l assertiveness for initiation. Prior research has not focused on the relationship between femininity ideology and self-objectification nor tried to evaluate how sexual assertiveness is associated with body shame or appearance anxiety within the context of Objectification Theory. · One may hypothesize that an association between sexua l assertiveness for initiation and body shame or appearance anxiety was masked in the present study by the significant relationship found between sexual assertiveness for initiation and femininity ideology: women who do not assert themselves in initiating sex may not feel shame or anxiety about their bodies because they believe it is unfeminine for women to assert 27 themselves. Likewise, women who are more likely to assert themselves in initiating sex may not experience body shame or appearance anxiety because they hold less traditional views regarding women's roles and capabilities in sexual situations. As for the lack ofrelationship between femininity ideology and self-objectification, given the relative newness of the femininity ideology construct, it may be additional efforts are needed to develop its measurement before ruling out a relationship to self -objectification. Similarly, a relationship between self -objec tification and sexual assertiveness for initiation may have shown itself had a direct approach been taken to trigger participants' sel f-objectificat ion, such as previous researchers' method of having participants try on swimsuits and view themselves in a mirror before answering survey questions about self-objectification. Clearly additional research is needed to test any of these explanations.
In previous research, Williams, et al. (2009) found women's identification with traditional feminine gender roles to be associated with both experience of adult sexual victimization and also sexua l assertiveness for refusing unwanted sex. Neither relation ship was significantly rep licated in the current study; however, FGI did show small correlations with SV and SAS -R (.07 and -.10 respectively). One explanation for the discrepanc y in results may be that the previous study explored women's personal endorsement of traditional feminine role behaviors while the present study aimed to capture participants' reports of femininity ideology. It may be that traditionally feminine characteristics and behaviors are associated with victimization and less sexual assertiveness for refusal while global attitudes regard ing what is feminine are not.
Additionally, toward this aim of measuring a slightly different construct, diffe rent measures were used in the current study than those that resulted in the sign ificant relationships found by Williams, et al. No conclusions can be drawn regarding the way feminine traits and femininity ideologies relate differently until researc h is conducted to compare them directly.
Aside from the nonsignificant relationships discussed above, resu lts from the current study support the relationships revealed by previous research. Most of the study's significant relationships involve the variables traditionally studied in Objectification research and sexua l 28 assertiveness. As reported by Moradi, Dirks , and Matteson (2005), experience of objectification was found to be significantly associated with self-objectification. Like Quinn, Kallen, and Cathey (2006), results revealed a significant relationship between self-objectification and body shame, a direct relationship not originally included in the model. The meditational relationship through body monitoring, however, supports results by Aubrey (2007). While Roberts and Gettman (2004)

29
Unique to this Objectification Theory research was the inclusion of sexual assertiveness variables. While sexual assertiveness for initiation was not found to relate to the primary Objectification variab les, sexual assertiveness for refusal had an undeniable impact. Given the inconsistency in results, the association to body shame and appearance anxiety remain s unclear; however, sexual assertiveness for refusal was revealed to be the strongest correlate of female sexual dysfunction, statist ically eliminating any significant relationship between dysfunction and other Objectification variables. Similarly, the significant association with sexual victimization suggests sexual assertiveness for refusal should be incorporated when attempts to bridge the fields of victimization and objectification occur.
Lastly , the failure of this study to support previous research regarding predictors of female sexual dysfunction suggests more clarification is needed in identifying what aspects of sexual functioning are affected by Objectification variables. As discus sed previously, it may be found that psychological and physiologica l aspects of sexua l functioning relate differently to the variables traditionally studied. More specifically, it may be that female sexual dysfunct ion is found in Objectification research when operationalized as negative perceptions of one's body during sex rather than as organic problems in functioning.

Limitations
As a statistical method, L VM is comprehensive but not devoid of limitat ions. Crosssectional data, when well-specified, can contribute to hypothesizing causal relationships, but true causality require s not only advanced research des igns, but designs that include temporal ordering in addition to evidence of association and sufficient isolation of potentia l confounds (Bullock , Harlow , and Mulaik, 1994). Whi le cross-sectional data helped clarify the relationships among the independent and dependent variables investigated, it did not allow causal conclusions to be drawn. In the current study, for example, it was expected that latent variable modeling wou ld identify self -objectification as a predictor of body mon itoring; however, this relationship, even though significant, would have been revealed regard less of the directionali ty of the variables. It 30 is only with longitudinal data that one has the ability to argue the direction of influence between the variables tested.
Additionally, due to problems with their statistical fit, items related to participants ' experience of childhood sexual abuse were not included in analyses as originally planned. Thus , results only reflect relationships to adult sexual victimization and cannot be generalized to participants' abuse experiences during childhood. Extensive research exists highlighting the negative effects of childhood sexual abuse, and it would be expected that these experiences relate to the broader Objectification model as clear ly as experiences in adulthood do, if not more so.
Future research would benefit from investigating the role of childhood sexual abuse and whether it mimics or diverges from results found when incorporating adult victimization experiences alone.
Lastly, use of university students was essentia l in providing immediate availability of the greatest number of research participants; however, it must be kept in mind that a relatively ethnically and socioeconomically homogenous sample was the result of using this source. The fact that most participants were white and college educated should automatically challenge any temptation to generalize study findings. Before broader generalizations are considered, studies incorporating nationally representative samples must be conducted. Currently, findings only apply to individuals matching the demographics of this participant pool; further application of these findings must be limited to hypotheses regarding a similar samp le.

Conclusions
In final review, this study offers an additiona l variable for consideration when conducting Objectification research while questioning the link to female sexual dysfunction and offering a bridge to the field of sexual victimization. The central model proposed by Fred rickson and Roberts (1997) remains intact and gains more support; however, additional research is needed to refine its relationship to female sexual dysfunction.

1
Future research is plainly needed to clarify the particular aspects of sexual functioning impacted by objectification. Additionally, the role of sexual assertiveness for refusal deserves further investigation and consideration as an addition to the Objectification model. Given the relationships between sexual assertiveness for refusal, sexual victimization, and female sexual dysfunction , it seems assertiveness must be taken into account when trying to explain female sexual dysfunction from the perspective of objectifying experiences.      Table 3 cont inued   SAS-I-IP !  SAS-I-IP2 SAS -I-IP3 SAS-I SAS-I-IPJ 1.0 SAS-I-IP2 . 6 Table 6 Reliability statistics for each survey measure included in L VM analyses.

51
Appendix A

Synthesis of Reviewed Research
Many components of Fredrickson and Roberts' (1997) Objectification Theory are thoroughly supported by existing research. Self-objectification, body monitoring and body shame have been revea led as integral parts of an understanding of women's experience of their bodies, and while the degree of influence of factors mediating the relationships between these constructs and health conditions such as sexual dysfunction is less abundant, the research clearly suggests Objectification Theory variables contribute to women's health.
Eighteen empirical articles dating from 2001 to 2007 were reviewed to inform the preceding discussion. Of these articles, four tested relationships involving objectifying experiences -through either direct interpersonal interactions or exposure to sexually objectifying media images. Results across the four articles found support for a positive association between participants' reports of objectifying experiences and self-objectification, body surveillan ce, body shame, appearance anxiety, and symptoms of disordered eating (Moradi , Dirks, and Matteson, 2005;Aubrey, 2006;Kozee and Tylka , 2006;Aubrey, 2007). Additionally, one article provided support for the relationship between objectifying experiences and reduced awareness of internal bodily cues (Kozee and Tylka, 2006). While the effects of objectifying experience s themselves are one of the least studied variables in tests of Objectificat ion Theory, the literature does support Fredrickson and Roberts' ( 1997) initial claims that a culture which consistently evaluates women by their physica l appearance and highlights bodie s and sexuality over competence or intelligence is bound to encou rage women's negative self-apprai sals. Tests of more varied examples of objectifying experiences (e.g., sexua l harassment) would further strengthen the support for this relationship.
These results suggest that even with the inconsistent conceptualization of self-objectification as synonymous with or independent of body monitoring, preoccupation with one's body and physica l appearance as perceived by others is a major contributor to women's experience of themselves. Additional clarification is needed in regard to which is the best way to organize and operationalize the relationship between self-objectifica tion and bod y monitoring; however, these constructs are arguably two of the three most salient, most empirica lly supported elements of Objectification Theory.
Body shame, the third strongly supported Objectification Theory construct , was tested in thirteen of the reviewed articles. Support was found for positive corre lations between body shame and both depression (Muehlenkamp, Swanson, and Brausch, 2005;Grabe, Hyde, and Lindberg, 2007;Szymanski and Henning, 2007) and disordered eat ing (Tiggemann and Lynch , 2001 ;Tiggemann and Slater, 2001;S·later and Tiggemann, 2002;Tylka and Hill, 2004;Moradi, Dirks, and Matteson, 2005;Greenleaf and McGreer , 2006;Kozee and Tylka , 2006;Sanchez and Kwang, 2007). Additionally, body shame was shown to be positive ly related to body -related selfconsciousness during physica l intimacy (Sanchez and Kiefe r, 2007) and negatively related to awareness of internal bodily states (Tylka and Hill, 2004;Kozee and Tylka, 2006). As can be seen from reviewing the relat ionships mentioned in the above paragraphs, not only has body shame been upheld as asserting its own influence on other Objectification Theory variables, but it has also clearly been established as a variable that is influenced by other Objectification Theory variables. In this sense, body shame has emerged as one of the centra l mediating factors between women's exper ience ofobjec tification, body monitoring, and health factors such as disordered eating, sexual dysfu nction, and depression.
Arguably one of the most important aims of Objectificat ion Theory is the pred iction or explanation of health conditions found to be predominant among women. Of the eighteen articles reviewed, fourteen tested predictors of depress ion, eating disorders, and/or elements of sexual dysfunction. Among those examining depression, posit ive corre lations were found between symptoms of depression and self-objec tification (Muehlenkam p and Saris -Baglama, 2002); appeara nce anxiety (Szymanski and Henning , 2007); ruminat ion, and body shame (Grabe, Hyde, and Lindbe rg, 2007) , Furthermore, depress ion was found to be negatively correlated with participan ts' peak motivat ional states (Szymanski and Henning, 2007). Elements of sexual dysfu nction were related to self-objectificat ion (Aubrey, 2007;Roberts and Gettman, 2004); body surveillance (Aubrey , 2007); and body shame (Sanchez and Kiefer, 2007). Across studies, eating disorde r symptomatology was found to be correlated with sexually objectifying experiences , self-54 objectification , body monitoring , body shame, lowered awareness of interna l bodily states , appearance anxiety, relationship contingency, and depression (Tiggemann and Lynch , 2001;Tiggemann and Slater, 200 l ;Slater and Tiggemann , 2002;Mue hlenkamp and Saris-Baglama , 2002;Tylka and Hill, 2004;Morad i, Dirks , and Matteson, 2005;Greenleaf and McGreer , 2006;Kozee and Tylka, 2006;Sanchez and Kwan g, 2007). While some health risks (e.g., eating disorders) have received more attention and support than others (e.g., sexual dysfu nction), none has received as much attention as the relationships among variables hypothesized to predict them.
In terms of testing the overall approp riateness of the Objectification Theory model, howeve r, studi es incorporating these variables in hypothesized explanations of depression , eating disorders, and sexua l dysfunction are frequently considered successful.
Clearly the research to date provides compell ing evidence in support of Objecti fication Theory , particularly in regard to relationships among self-objectification , body monitoring, and body shame. The role of variables less often studied, however, (e.g., appearance anxiety, flow, internal awareness), is not as well understood. This imbalance in research focus muddle s the clear outline of a path from experience of objectification to health factors like depression, eating disorder s, and sexual dysfunction. As with most theories, additional research is needed .

55
Appendix B Objectification Theory Expanded Sanchez and Kwang (2007) introduced a potential addition to Objectification Theory by investigating the role of relationship-based self-esteem. They argue women are more likely to experience body shame and disordered eating if they base their self-esteem on being in or maintaining romantic relationships, a pattern the authors conceptualize as relationship contingency. Just as Fredrickson and Roberts' (1997) argue that social pressures to match a culturally specified beauty ideal are internalized by women as measures of self-wo rth , Sanchez and Kwang argue the pressures women feel to find romantic partners may be internalized and similarly affect their self-worth via changes in self-esteem. When romantic relationships fail or are not momentarily present, the authors argue there is the potential for women to devalue themselves because of an internalizat ion of the message that women need romantic relationships.
Furthermore, aware of the value men attach to physical appearance in selecting a partner, the authors argue a preoccupation with one's body image may serve as a means of finding and maintaining romantic relationships.
Like Fredrickson and Roberts' (I 997) assertion that body shame contributes to disordered eating, Sanchez and Kwang (2007)  provide compelling evidence to consider broadening that conceptualization to include the messages women receive about their need for romantic relationships. While the cross-sectional 56 study design limits the inferences drawn from these resu lts, the inclusion of both college students and women recruited from the community at large adds force to these findings beyond that from more restricted samples. An unfortunately common drawback to this and most other Objectification Theory studies, however, is the homogenous sample, comprised of primarily white, heterosexual women. This fact automatically limits the degree to which these results can be generalized.

57
Appendix C Objectification and Lesbian Women Kozee and Tylka (2006) , Group comparisons revealed lesbian women reported more body monitoring and fewer symptoms of disordered eating than heterosexual women, but no other significant differences between groups were revealed. As expected, path analyses indicated the proposed mode l was an excellent fit to the data provided by the sample of heterosexual women, the only non-significant path being from body monitoring to internal awareness. The same model, however , was a poor fit to the data provided by the samp le of lesbian women. Results suggested four model modifications, and a second analysis following these changes revealed the adjus ted model provided a much better fit to the lesbian sample data. This final model suggested interper sonal experiences of sexual object ification predicted body shame and awareness of internal bodily 58 states in addition to body monitoring . Furthermore, eating disorder symptomatolo gy was directly predicted by body monitoring as well as through the mediational relationship with body shame .
Finally, body shame predicted awareness of internal bodily states , but internal awareness had no significant effect on disordered eating .
While both sets of analyses support the inclusion of the same set of variables, results suggest objectification is indirectly related to negative consequences among heterosexual women but directly related to negative consequences among lesbian women . Kozee and Tylka (2006) assert their results suggest the relationships among Objectification Theory variables may likely be more complex for lesbian women as compared to heterosexual women. of undergraduate students did not report a racial demographic. It was expected that the make-up of the study sample would reflect these underlying demographics; however, because participants were recruited from psychology courses specifically rather than the university at large, the demographic characteristics of the final sample were also expected to vary slightly. In the largest psychology lecture, the ratio of women to men was higher than the demographics for the university population; however, no additional statistics reflecting the subset of students enrolled in psychology courses were available . Because the purpose of this study was to investigate hypotheses derived from Objectification Theory which was developed to explain the consequences of women's experiences of objectification, only women were recruited to participate in this study. Additionally, as mentioned previously, sexual abuse is more prevalent among women.
Just as racial and ethnic homogeneity was expected, participants were expected to fall within a very limited age range, 18-22. Using a young adult sample was in line with the majority of research in the field of Objectification Theory, but this limitation must be kept in mind when considering research results. The current study sought to expand our understanding of variables that influence the sexual functioning of young adult women; however, additional research will be needed before any significant relationships can be judged to be independent of participants' developmental stage.
While research has yet to reveal that Objectification Theory (Fredrickson and Roberts , 1997)  another. It is the way a woman evaluates her appro ximation to a standard of beauty that matters , not the specifics of that standard. Following this logic , the current study presumes that Objectification Theory will apply to women of different ethnic /racial backgrounds . It should be noted that this study did not test these assumptions. As stated, this study was expected to draw a primarily white sample ; however , research has yet to suggest the proposed hypotheses would not be supported by data from a more racially and ethnically diverse sample. Future research should test these assumptions by investigating more diverse samples.

61
Append ix E

Sample Size Determinants
Assuming that an effect truly exists in the population, power is the probability that a statistical test will be significant. Increasing power decreases the chance of falsely concluding that no effect exists, in other words, the possibility of committing a Type II error is decreased.
For example, in this study, committing a Type II error would mean that sexual victimization and femininity ideology both affect one 's sexual functioning, but statistical analyses failed to recognize a signific ant association. Because statistical power increases as a function of sample size, if the aim is to avoid this error, a larger sample must be recruited. Large samp les facilitate finding significa nce because the effect size does not need to be as big as would be required for some small samples. When distinguishing between results that are consequential and those that are negligible, effects sizes and theory are crucial. The current L VM design investigated sexua l victimization, femininity ideology, and seven additional mediating and dependent variables.
When associations were found to be significant, large effect sizes and strong theoretical support were used to decide which relationships were central to understanding the model. Because this study recruited a University sample versus a national samp le, having too large a samp le size was not a conside rable threat. Furthermore, by implementing the common ly used p < .05 alpha value to determine statist ical significance, a significant finding was attributed to chance factors in no more than 5% of cases.
According to Kline (2005), because SEM is considered a large samp le model, no less than 200 participants should comprise any SEM design. As such, having small (N < I 00) or medium (N = 100-200) sample s could have resulted in problem s durin g analysis. In addition , because some protection must be in place in expectation of incomplete data or individuals not completing the study, the participant pool recruited initially {n=500) was even larger than what was expected to be used in the end.

62
Appendix F Demographics For the following questions, please check or provide the answer that is best for you.   As children, many women are in sexual situations with someone older than them. A sexual situation could mean someone showin g his or her genita ls to you. It could mean someone touching you in a sexua l way. It could also mean a man or boy putting his penis in your mouth, vagina, or rectum.
Think back to when you were a child up to 14 years of age, and answer the next quest ions. I understand that these may be difficult questions to answer, but please try to answer them as honestly as you can. Plea se check your best answer.
Before yo u were 14 years old: In this next set of questions, a "sex partner " is any perso n that yo u have done any of these things with: Oral sex: your mouth on your partner's genital s or yo ur partner 's mouth on yo ur genitals; Vaginal sex: a man putting hi s penis in your vagina; Appendix G As noted by Noar (2003), "To have confidence in the findings of our studies, we must first have confidence in the quality of our measures (pg. 622) ." Throughout the history of gender role studies, instruments such as the Bern Sex Ro le Inventory (BSRI: Bern, 1974 ), and Attitudes Towards Women Scale (A WS: Spence and Helmreich, 1972) have been frequently used, and while research has both critiqued and supported the validity of these scales (Holt and Ellis, 1998;Tolman and Porche, 2000;Hoffman, 2001;Hoffman and Borders, 200 1), a more recent concern has been their appropriateness for the shift toward studying gender ideology rather than gender traits or an individual's enactment ofa gender role. When it becomes clear from a review of the literature, as is the case with feminine gender ideology, that empir ically developed and validated measurement instruments fall short of meeting the researcher's current needs, the temptation to embark on addressi ng that absence with a new scale may immed iately present itse lf. The process is by no means simplistic, but with a clearly articulated theory and a serious of sound methodological steps, the result may be a much needed alternative to a bevy of poorly designed or ill-fitting scales.
Given the existing plethora of psychological studies, it is genera lly unlikely that no one has ever investigated a researcher's interest area before. With th is in mind, scale develope rs are encouraged to conduct a thorough review of the literature to verify no established scales exist that appropriately and suffic iently meet the needs of the task at hand. As explained by Meyer,Edwards,and Rossi (I 995), "the process oftest selection serves to ensure that the instrument is of good quality and is approp riate to the clinical or research question at hand (pg. 25)." Concerning the field of femininity ideology, a number of existing scales deserved review before deciding on a measu rement approach for the current study.
The Adolescent Femin inity Ideology Scale (AFIS: To lman and Porche, 2000) was developed in response to the perception that gender-research measures focused on personality traits or gender roles rather than widespread internalization of femin ine gender norms.
Additionally, noting the adult samples used for the development of most gender -related measures , the authors wished to develop a scale specific for use with adolescents. The resulting instrument was created to target adolescents' internalization of two components of femininity ideology: an "inauthentic self in intimate relationships with others and having an objectified relationship with one's body (pg. 365)." The measure was developed using a highly diverse sample and focus groups that provided qualitative in addition to quantitative guidance. The current study, however, aimed to measure women's internalization of different social messages than those assessed by the AFIS. While Tolman and Porche (2000) were interested in self-objectification and girls' inauthentic self in relationships, the present study was concerned with women's expectations of adult sexual relationships and what they considered to be appropriate sexual behavior for women .
The Femininity Ideology Scale (FIS: Levant, 2007) was recently developed for use with adult samples and more closely satisfied the needs of the current research. The FIS is presented as a broad -spectrum measure of femininity ideology and thus measures respondents' attitudes across five broad domains: 1) Stereotypic image and activities , 2) Dependence or deference, 3) Purity, 4) Caretaking, and 5) Emotionality. While the full FIS scale is a general measure of femininity ideology rather than a focused assessment of perceptions of appropriate female sexuality, two of the domains represented by FIS scores were relevant to the current study.
Dependency/Deference and Purity items such as "Women shou ld not initiate sex" and "Women should dress conservatively so they do not appear loose" are consistent with a traditional understanding of femin inity and thus served as appropriate items for the study. The Purity and Dependency/Deference subscales satisfied the theoretical needs of the study ; however, adding items to improve the scale's reliability was expected to help satisfy the study ' s psychometric needs, as initial estimates of internal consistency suggested the FIS has an adequate but low reliability (Cronbach's alpha= .70).
They Hyperfemininity Scale (Murnen and Byrne, 1991) targets women's enactment of traditional femin ine gender role behaviors; however, it falls short of measuring femininity 78 ideology by focusing on women's personal beliefs as opposed to behaviors or attitudes they consider universally appropriate for women. Nevertheless , because the scale was designed for women respondents and assesses gender role endorsements specifically in the context of sexual relationships, it provides a useful resource for supplementing the items provided by the FIS subscales. With simple modifications, (e .g., "Women should state their sexual needs clear ly and concisely" vs. "I try to state my sexual needs clearly and concisely") , Hyperfemininity Scale items were adjusted to assess participants' attitudes about appropriate sexua l behavior for women in general rather than just the attitudes that dictate individuals' own personal behaviors.
Given all of the above consideration s, using the Purity and Dependency/Deference subscales of the FIS and a selection of modified Hyperfemininity Scale items appeared to be the best approach to measuring the target ideology of the current study, feminine sexuality.