- Email: [email protected]
ORIGINAL RESEARCH—OUTCOMES ASSESSMENT: Development and Validation of the Female Genital Self-Image Scale
1822
ORIGINAL RESEARCH—OUTCOMES ASSESSMENT Development and Validation of the Female Genital Self-Image Scale jsm_1728
1822..1830
Debra Herbenick, PhD, MPH, and Michael Reece, PhD, MPH Indiana University- Center for Sexual Health Promotion, Bloomington, IN, USA DOI: 10.1111/j.1743-6109.2010.01728.x
ABSTRACT
Introduction. Women’s sexual function may be influenced by various factors including medical conditions, trauma or abuse, medications, relationship dynamics, relaxation, mood, and body image. However, few studies have explored the influence of a woman’s genital self-image on her sexual function or behaviors. Aims. The purpose of this study was to establish a reliable and valid measure of female genital self-image, the Female Genital Self-Image Scale (FGSIS), and to assess the relationship between scores on the FGSIS and women’s sexual function. Methods. The FGSIS was developed in two stages. Phase One involved an analysis of cross-sectional paper-based survey data and a review of the literature. Phase Two involved a cross-sectional internet-based administration of the scale items to a total of 1,937 women. Main Outcome Measures. Psychometric properties of the scale were evaluated through the use of reliability analysis, factor analysis, and score differences based on women’s experience of orgasm from receiving cunnilingus or from self-masturbation with a vibrator. Correlation analysis was used to explore the relationship between female genital self-image and scores on the Female Sexual Function Index (FSFI). Results. The scale was found to have sufficient reliability (Cronbach’s alpha = 0.88) and one factor that explained 59.23% of the variance. Women who had ever experienced orgasm as a result of cunnilingus or self-masturbation with a vibrator and women who reported having had a gynecological exam in the previous 12 months had significantly higher FGSIS scores than those who had not (P < 0.001). Scores on the FGSIS were positively and significantly correlated with scores on all FSFI domains (P < 0.001), including the Total score, with the exception of the Desire domain. Conclusion. The FGSIS was initially found to be a reliable and valid measure though further research is needed to understand its properties in diverse populations. In addition, female genital self-image was found to be positively related to women’s sexual function. Herbenick D, and Reece M. Development and validation of the female genital self image scale. J Sex Med 2010;7:1822–1830. Key Words. Genital Self-Image; Sexual Function; Orgasm; Vulva; Female External Genitalia; Vagina
Introduction
W
omen’s sexual function may be influenced by a range of factors including medical conditions, trauma or abuse, medications, relationship dynamics, relaxation, mood, and body image [1–7]. However, few studies have explored the influence of a woman’s genital self-image on her sexual function or behaviors. Waltner, in 1986, J Sex Med 2010;7:1822–1830
proposed the concept of genital identity, which he defined as “those self-definitions, self-attitudes and subsequent feelings which arise from specific interactions and experiences which either indirectly or directly involve the genitals” [8]. He suggested that genital identity could both arise from and influence sexual interactions. It was Winter, however, in 1989, that was the first to quantitatively measure genital self-image, © 2010 International Society for Sexual Medicine
Female Genital Self-Image albeit among men, a concept that focused on men’s perceptions of the appearance and size of their genital parts [9]. In 1995, Reinholtz and Muehlenhard explored college students’ perceptions of their own and their partners’ genitals, using items created specifically for their study [10]. They found that positive genital perceptions were related to higher engagement in, and enjoyment of, sexual activities. Specifically, ratings of liking one’s own genitals were significantly correlated with enjoyment of penile-vaginal intercourse as well as receiving oral sex. The factor termed “liking one’s own genitals” consisted of three items related to comfort with, liking of and feeling proud of one’s genitals. The small size of the factor, and the fact that no items used to measure genital perceptions in this study were related to genital appearance or function (though 5 of 11 items related to genital smell or taste), suggested the need for expanding our understanding of this construct. In a 2003 study, Berman and colleagues devised a scale to measure women’s genital self-image and documented a relationship between genital selfimage and sexual function [11]. Although their findings advanced knowledge pertaining to genital self-image and sexual function, some of the scale items used in this study are problematic. The structure of some items conflicts with established recommendations of measurement by being double-barreled (assessing two or more issues or varying perspectives on an issue in a single question), which presents the possibility of an imprecise or invalid measure of the construct [12]. Another scale item, “I look at my genitals”, posed additional measurement problems as women may look at their genitals as a result of positive or negative self-image (e.g., because they find their genitals attractive or arousing or because they find them unattractive and consequently look at them with a desire to alter their appearance) or for health reasons [13–15]. Rather than being a means of measuring female genital self-image, then, such an item may be better used to understand group differences. For example, do women who look at their genitals have, on average, more positive or more negative female genital self-image than women who do not look at their genitals? As per this point, one study found that women who had looked at their genitals three or more times during the previous year had more positive perceptions of women’s genitals [16]. Two additional items in the scale by Berman and colleagues assessed the types of messages that respondents were given about their genitals while growing up. Although these may be useful measures on
1823 which to group individuals and compare their genital self-image, it is unclear why messages received as a child would be valid indicators of one’s current genital self-image. If they are, then it may be worth considering whether genital selfimage—at least as measured by their scale—is better considered as a trait rather than a state. In a 2005 study, Morrison and colleagues adapted Winter’s male genital self image scale for use among female college students [17]. However, in one administration of the scale 26.9% to 53.7% of female respondents selected the neutral response option on each item, raising issues about the validity of the overall scale because of item nonresponse. Because people frequently choose neutral response options when responding to items of a sensitive nature, it has been suggested that having neutral response choices should be avoided in order to improve validity [12]. Women’s feelings and beliefs related to their genitals may be a particularly sensitive topic for some given that women’s genitals have traditionally been shrouded in secrecy and taboo [18]. Women may enter adulthood having had little information, education or experience viewing their genitals [19]. In addition, terminology related to women’s genitals tends to be vague, some contemporary medical texts still do not feature the clitoris in diagrams, and medical dictionaries may define the clitoris in terms of the penis but not vice versa [20–23]. Any of these factors may reflect or influence how a woman feels about or experiences her genitals. In spite of researchers’ efforts to quantitatively assess female genital self-image, there has yet to be a widely used measure that assesses a range of relevant concepts and that utilizes items that are judged to be valid measures of the construct. Given the importance of genital appearance and function to many women [13,24] it is likely that these are important aspects of the construct of female genital self-image to consider in any scale of this nature. Both appearance and function were assessed in the items developed by Berman and colleagues [11] and detailed items related to appearance (but not function) were assessed in the scale modified by Morrison and colleagues [17]. In summary, studies have explored issues related to female genital self-image (or women’s “genital perceptions”) and have provided important, although preliminary, insights into this construct and its relationship to women’s sexuality. However, there remains a need for a scale that fully operationalizes the construct of genital image within a contemporary context and that has underJ Sex Med 2010;7:1822–1830
1824 gone comprehensive psychometric analyses. Being able to measure female genital self-image in a reliable and valid manner may be particularly helpful to our understanding of how procedures such as genital cosmetic surgery or treatments such as vulvectomy may affect a woman’s perceptions of the way that her genitals look or function. Aims
The purpose of this study was to establish a reliable and valid measure of female genital selfimage, the Female Genital Self Image Scale (FGSIS), and to assess the relationship between scores on the FGSIS and women’s sexual function. The FGSIS was developed in two stages, which will be reported on separately within this paper. Phase One involved the development of an initial item pool based on elicitation surveys administered to an undergraduate sample; Phase Two involved the administration of the scale and the assessment of its psychometric properties. Phase One: Development of an Item Pool
The purpose of Phase One was to develop an item pool for the FGSIS based on feelings and beliefs about women’s genitals elicited from paper-based surveys and a review of scientific and popular literature. The Institutional Review Board at the author’s institution reviewed and approved all study protocols for each phase of the research.
Methods Item Elicitation The initial item pool was developed using data from item elicitation surveys that were collected from 370 participants (236 women and 134 men) who ranged in age from 18 to 40 years (mean 20.3 years, standard deviation [SD] = 2.1). The vast majority, 96.1% (N = 317), identified as heterosexual. Participants were recruited from undergraduate health classes and asked to complete brief and anonymous item elicitation surveys that sought the language they used to describe perceptions of genitals and attitudes about sexuality. We elicited language from both women and men in this phase in order to ensure that language was reflective of that used by both genders given that women’s perceptions of their genitals or the language they use to describe them could be influenced by male partners or influenced by shared language within couples. Respondents were randomly assigned to anonymously complete one of nine different versions of J Sex Med 2010;7:1822–1830
Herbenick and Reece the item elicitation survey. In order to elicit an expansive range of feelings, beliefs, and attitudes about women’s genitals and to do so in a manner that allowed for assessing similar concepts using a variety of question types and contexts, nine versions of the elicitation survey were used. The varying versions included open-ended questions (e.g., “What do you think your partner would like or dislike about your genitals?”), fill-in-the-blank items (e.g., “The best thing about women’s genitals is ”), as well as requests for participants’ to rank order characteristics of genitals that were important to them (e.g., smell, appearance, etc.). Some versions also presented vignettes about genitalrelated situations (e.g., performing or receiving cunnilingus, having a gynecological examination) and asked respondents to answer questions about what they would do to prepare for such events. Nine versions also enabled us to elicit a depth of detail using open-ended items but in a way that was not overly burdensome to participants.
Development of Themes and Items Content analysis was performed separately by two researchers to identify both the manifest content (e.g., content that is visible and obvious) and the latent content (e.g., the interpretation of the underlying meaning of words and text) of the participants’ responses [25,26]. The product of each content analysis was a set of meaning units (e.g., constellations of statements from the data that relate to the same central meaning). An example of a meaning unit from this data is “vaginas smell bad”. The meaning units were subsequently combined from the two researchers and each independently derived codes for each grouping of meaning units (e.g., meaning units related to smell were coded by the researchers as “smell”). After each researcher had independently coded the data, they met together for the purpose of reaching 100% consensus on the codes by reflecting and discussing on the extent to which there was agreement between the codes [27]. In an additional process to ensure 100% consensus on the codes for the meaning units, two external researchers with expertise in women’s sexuality and medicine reviewed and indicated that they were in agreement with the codes. Based on the consensus agreement of the reviewers, support existed for four major themes across the codes: smell/taste, appearance, sexual function, and shame/pride. Subsequent to the identification of the themes and based upon the language, words, and phrases elicited, the first author wrote and revised items in
1825
Female Genital Self-Image Table 1
Item pool, retention decisions based on reliability analysis and favor analysis of Female Genital Self Image Scale
Items 1. I feel positively about my genitals. 2. I am satisfied with the appearance of my genitals. 3. I would feel comfortable letting a sexual partner look at my genitals. 4. I think my genitals smell fine. 5. I think my genitals work the way they are supposed to work. 6. I feel comfortable letting a healthcare provider examine my genitals. 7. I am not embarrassed about my genitals. 8. I like how my genitals look.
9. I can see how a sexual partner would be aroused by my genitals.
10. I can see how a sexual partner would enjoy the taste of my genitals.
11. I am comfortable with how my genitals smell.
Outcome based on reliability analysis
Factor loading
Corrected item to total correlation
Item mean
Item standard deviation
Retained. Retained.
0.86 0.84
0.77 0.74
2.96 2.90
0.81 0.84
Retained.
0.81
0.72
3.07
0.79
Retained. Retained.
0.67 0.65
0.57 0.55
3.02 3.20
0.80 0.72
Retained.
0.69
0.60
3.12
0.79
Retained.
0.84
0.77
3.03
0.82
Excluded after Iteration 1 given its overlap in meaning with Item 2. Cronbach’s alpha changed from 0.93 to 0.92 with deletion. Excluded after Iteration 2 given its overlap with item 4; Cronbach’s alpha changed from 0.92 to 0.91 with deletion. Excluded after Iteration 3 given concerns that the item may measure perceptions of oral sex rather than genital self-image; Cronbach’s alpha changed from 0.91 to 0.90 with deletion. Excluded after iteration 4 given its overlap with item 6; Cronbach’s alpha changed from 0.90 to 0.88 with deletion.
collaboration with the second author and with feedback and suggestions from two other faculty members who have expertise in women’s sexuality and medicine. This resulted in a total of 11 final items written to reflect each of the four themes (see Table 1). Some items were redundant in keeping with suggestions for items writing [12]. For example, “I like how my genitals look” and “I am satisfied with the appearance of my genitals” were both included in order to test different wordings of items related to perceptions of genital appearance. A 4-point response scale (strongly disagree, disagree, agree, strongly agree) was chosen based on recommendations that a neutral response option in scales of a sensitive nature may result in respondents “fence-sitting” [12]. The initial pool of items was subsequently reviewed by two individuals with expertise in sexual health and gender studies who had reviewed the initial codes resulting from content analyses and agreed that the 11 items were reflective of the four themes underlying our conceptualization of female genital self-image, thus supporting face validity of the initial FGSIS.
Phase Two: Scale Administration and Psychometric Analysis
The purpose of Phase Two was to assess the internal consistency of the scale items and to identify whether the removal of any redundant items would help to improve the scale’s reliability. Additionally, data analyses from Phase Two included assessments of the validity and predictive capacity of the FGSIS following its administration to a convenience sample of adult women.
Methods Procedures From January through March 2008, the research team, through a collaborative research partnership with the leadership of a large U.S.-based, in-home sex toy party company, asked women who worked as party facilitators to distribute study recruitment cards to the women who attended their parties. The researchers did not have direct access to or communication with the facilitators so as to preserve their anonymity; rather, the research team drafted a recruitment email that was then emailed by company leadership to the facilitators. J Sex Med 2010;7:1822–1830
1826 Company leadership also arranged for recruitment cards to be mailed directly to party facilitators who were asked to then pass the cards along to the party attendees. This was part of a larger study about women who attend in-home sex toy parties. Recruitment cards invited attendees to visit the study website in order to learn more about the study and to decide whether they wanted to participate. The study was described as being about the characteristics of women who attend in-home sex toy parties. Individuals who agreed to participate in the study completed an anonymous, crosssectional survey and had the opportunity to enter themselves into a drawing for one of ten $50 gift cards. No other incentives were provided.
Measures The study questionnaire included items related to demographic and background variables (e.g., age, education, race/ethnicity, sexual orientation, marital status) and measures related to recent and lifetime sexual behaviors and age of first vaginal intercourse. Relevant to these analyses, respondents were asked “Have you ever had an orgasm from receiving oral sex?” (yes, no) and “Have you ever had an orgasm from self-masturbation with a vibrator?” (yes, no). A health-related item asked respondents “In the past 12 months, have you had a gynecological exam?” (yes, no). The number of items shown to each woman varied considerably depending upon the behaviors that she reported. For example, if a woman reported that she had not engaged in a certain behavior (such as masturbation) then she did not receive follow-up questions about that behavior. However, all respondents received the measures analyzed in this report. Time to complete the survey ranged from approximately 10–20 minutes. Female Sexual Function Index (FSFI). The FSFI is a 19-item measure of women’s sexual function that has established reliability, validity and temporal stability [28–32]. It measures the domains of desire, arousal, lubrication, pain, orgasm, and satisfaction, and also results in a total score. Higher scores on each domain and on the total score indicate more positive sexual function. FGSIS Test Items. The 11 FGSIS items were preceded by the following statement: “The following items are about how you feel about your own genitals (the vulva and the vagina). The word vulva refers to a woman’s external genitals (the parts that you can see from the outside such as the clitoris, pubic mound, and vaginal lips). The word vagina J Sex Med 2010;7:1822–1830
Herbenick and Reece refers to the inside part, also sometimes called the “birth canal” (this is also the part where a penis may enter or where a tampon is inserted). Please indicate how strongly you agree or disagree with each statement” (see Appendix A).
Statistical Analysis SPSS version 16.0 (Chicago, IL, USA) was used for statistical analysis. Descriptive statistics were used to report respondents’ demographic and other background data. Specific analyses conducted for psychometric evaluation of the FGSIS are described in their respective section of the results. Results
A total of 2,049 women completed the online survey and 1,937 respondents had complete data on the FGSIS (94.5%). Because the total number of recruitment cards distributed is unknown, it is not possible to calculate an overall survey response rate.
Table 2
Phase two sample characteristics (N = 2,049)
Participant characteristics
N (%)
Age (mean, standard deviation) Race American Indian/Alaskan Native Asian or Asian American Black or African American Native Hawaiin or other Pacific Islander White or Caucasian Hispanic or Latino Multiracial Other Relationship status Single In a relationship Living together Married Separated Divorced Widowed Current college/university student Education Middle school Some high school High school graduate Some college/2-year degree College graduate Graduate school Other Sexual orientation Heterosexual/straight Homosexual/lesbian Bisexual Questioning or uncertain Asexual Other
29.55 (8.4) 21 11 77 8 1,818 65 40 9
(1.0) (0.5) (3.8) (0.4) (88.7) (3.2) (2.0) (0.4)
379 297 322 938 28 62 5 567
(18.5) (14.5) (15.7) (45.8) (1.4) (3.0) (0.2) (27.7)
2 46 397 1,014 434 133 23
(0.1) (2.2) (19.4) (49.5) (21.2) (6.5) (1.1)
1,866 12 142 17 2 10
(91.1) (0.6) (6.9) (0.8) (0.1) (0.5)
Female Genital Self-Image As shown in Table 2, the mean age of the sample was young (mean = 29.55, SD = 8.4, range = 18– 68) and most respondents were white (88.7%), heterosexual (91.1%), and partnered (45.8% were married, 15.7% were cohabitating and 14.5% were in a relationship). A total of 77.2% had at least some college education. Most women (80.9%, N = 1567) had been sexually active with male partners within the past four weeks and few (0.9%, N = 17) reported sexual interactions with other women or both men and women (1.8%, N = 35) during this time period. The remainder (15.2%, N = 294) reported no partnered sexual interactions during the four weeks preceding the study. Women reported an average age at first intercourse of 16.7 years (median = 16.0, SD = 3.0).
Internal Consistency Reliability analysis was conducted on the FGSIS with each iteration in order to determine its internal consistency and to assess whether any redundant items should be deleted. Cronbach’s coefficient alpha was used as an indicator of how reliable items were in identifying the overall construct. Although this coefficient can range from 0 to 1.0, a score above 0.7 was considered the minimum and a score at or above 0.8 preferable [12]. Corrected item-to-total correlations were also used to test for internal consistency; those at or above 0.30 were generally considered sufficient [30]. As shown in Table 1, the Cronbach’s alpha for all 11 FGSIS items was very high (0.93), allowing for the possibility of deleting items for inclusion in the final scale for rational or empirical reasons [12]. Using four iterations of reliability analysis and rational decision making, items 1, 5, 7, and 9 were removed. The resulting 7-item FGSIS had a Cronbach’s alpha coefficient of 0.88 and a mean score of 21.30 (SD = 3.9; N = 1937). The scores on each item were summed for a total possible FGSIS score ranging from 7 to 28 with higher scores indicating more positive female genital self-image. Construct Validity Factor analysis was performed using principal component extraction with initial communalities of 1.0. A priori criteria for determining the number of factors were that factors had to have eigenvalues greater than 1.0; the scree plot was also examined. The scree plot indicated one factor that explained 59.23% of the variance. A large decrease
1827 was seen between the first and second eigenvalues, with small decreases thereafter (eigenvalues: 4.15, 0.88, 0.61, 0.56, 0.38, 0.31, 0.12). Factor loadings ranged from 0.65 to 0.86 (i.e., above the 0.40 cutoff) [30]. Corrected item-to-total correlations ranged from 0.55 to 0.77 (i.e., exceeded 0.30; see Table 1). These findings provided further support for the construct validity of the FGSIS. The item “I think my genitals work the way they are supposed to work” was the most highly endorsed (mean = 3.20, SD = 0.70).
Predictive Capacity Using analysis of variance, the predictive capacity of the FGSIS was assessed in relation to three dichotomously measured experiences or behaviors, including partnered sexual behaviors (having had an orgasm from cunnilingus), masturbation (solo masturbation with a vibrator), and having had a gynecological exam within the past year. Similar behaviors have been shown to be related to women’s attitudes toward women’s genitals more globally [16] and previous research has documented a relationship between genital perceptions and enjoyment of sex or ease of orgasm [10,16] particularly resulting from sexual behaviors that involve high degrees of intimacy with one’s genitals, such as cunnilingus. A higher FGSIS score among participants who reported a history of these was considered to be an indicator of the predictive capacity of the scale with regard to each. Women who had ever experienced orgasm as a result of cunnilingus, those who had experienced orgasm as a result of self-masturbation with a vibrator and women who reported having had a gynecological exam in the previous 12 months had significantly higher FGSIS scores than those who had not (P < 0.001, see Table 3), supporting the predictive capacity of the scale. Relationship to Sexual Function Bivariate correlations were used to assess the relationship between women’s scores on the FGSIS and the FSFI domains and total score. Women’s scores on the FGSIS were positively and significantly correlated with their scores on the FSFI domains of Arousal (r = 0.18, P < 0.001), Lubrication (r = 0.14, P < 0.001), Orgasm (r = 0.17, P < 0.001), Satisfaction (r = 0.19, P < 0.001), Pain (r = 0.13, P < 0.001) and the Total Score (r = 0.20, P < 0.001). There was not a significant relationship between women’s scores on the FGSIS and the FSFI Desire domain (r = 0.03, P = 0.18). J Sex Med 2010;7:1822–1830
1828 Table 3
Herbenick and Reece Predictive capacity of the female genital self image scale (N = 1,937)
Behavior
FGSIS score (mean, SD)
Ever experienced orgasm during cunnilingus Yes (N = 1,461) 21.52 (3.7) No (N = 476) 20.61 (4.1) Ever experienced orgasm during self-masturbation with a vibrator Yes (N = 1,470) 21.51 (3.9) No (N = 467) 20.62 (3.8) Had a gynecological exam in the previous 12 months Yes (N = 859) 21.50 (3.8) No (N = 1,156) 20.53 (4.1)
Discussion
The current study found sufficient support for the reliability and validity of a scale to measure female genital self-image. In this sample, female genital self-image was found to be significantly correlated with female sexual function, with the exception of the FSFI Desire domain. This is in contrast with results from the 2003 study in which Berman and colleagues found a correlation between their measure of female genital self image and the FSFI Desire domain, but no significant relationship between scores on their genital self image scale and the FSFI measures of arousal, lubrication, orgasm, pain or satisfaction [11]. In their study, however, the authors did find a relationship between their measure of female genital self image and sexual distress. Given that 7 of the 17 items in Part 1 of their two-part scale were items that measured respondents’ feelings of embarrassment, shame, anxiety or worry in relation to their genitals, it may be that their measure assesses distress or anxiety in relation to one’s genitals rather than genital self-image per se. Further research is recommended in order to better understand the nuances of these different measures. It may be useful, for example, for future research to explore the boundaries of what is meant by genital self-image vs. what may be better described as “genital distress.” FGSIS items were developed as a result of elicitation surveys that probed respondents to answer questions about a wide range of feelings and beliefs related to female genitals. It may be that FGSIS items are more expansive in representing a range of feelings about their genitals (e.g., embarrassment, comfort, cognitions, affect, and satisfaction rather than an emphasis on distress or worry). The finding that scores on the FGSIS are significantly and positively correlated with most aspects of sexual function as measured by the FSFI J Sex Med 2010;7:1822–1830
F
p
omega squared
20.37
<0.001
0.01
19.06
<0.001
0.01
7.53
<0.001
0.01
are important. Women’s feelings about their genitals have long been described as relevant to their ease of orgasm and overall sexual experience and these data provide evidence of such a relationship. A strength of this study is that a careful, rational-empirical approach to scale development was used in order to enhance the reliability and validity of the scale. Although a limitation of the study is that a convenience sample was used, a strength is that the large sample size in Phase Two was sufficient for the use of factor analysis [12]. However, caution should be exercised before generalizing these results as women who attend in-home sex toy parties may differ from the general population on key sexuality-related variables. Further research should test the FGSIS in more diverse populations within and outside of the United States to better understand the validity of the measure in other populations. The mean and standard deviation of the FGSIS in this sample should not be accepted as scale norms until further validated in a larger population-based sample. Finally, the current study was not designed to allow testing of the temporal stability of the scale; as such, future research should apply the FGSIS to the scrutiny of test-retest reliability. Findings from this study have several implications. First, given some of the limitations of previous scales that intended to assess female genital self-image as a construct, researchers should carefully consider what it is that they want to measure prior to selecting a scale. Available scales vary considerably in terms of their item content and what it is that each measures. The FGSIS developed in this study accounts for a wide range of characteristics related to female genital self-image and provides a new option for researchers and clinicians who are interested in this topic; however, further research is needed to assess its psychometric properties among a population-based sample of women.
1829
Female Genital Self-Image By taking a careful rational-empirical approach to scale development, scale items were written that may be useful to clinicians individually as well as through the use of the total 7-item scale. Specifically, although the FGSIS may be useful to clinicians and researchers as they attempt to learn more about the influence of women’s genital self-image on their sexual function, responses to individual scale items may provide helpful insights or encourage conversations between clinicians and clients or patients. For example, although a woman may score highly on the scale as a whole (indicating a generally positive genital self-image), clinicians should take note of any negative endorsements of items that indicate a woman may feel negative about certain aspects of her genitals such as their appearance, function, or smell, or her ability to feel comfortable having her genitals seen by a sexual partner or healthcare provider. Given that patients may often find it difficult to approach their healthcare providers with questions or concerns related to their sexual health and well-being, the ability to gather such information via a questionnaire may be a useful starting point in discussing sexuality and genital health with patients or clients. In the future, researchers might use the FGSIS in clinical samples in order to understand the extent to which it provides useful insights into women’s genital self-image before and after treatment for vulvar intraepithelial neoplasia, vulvar cancer, vulvar dermatoses, pelvic organ prolapse or other gynecological conditions that may affect the appearance, function or smell of a woman’s genitals, or women’s comfort letting others view their genitals [33–36]. The FGSIS may prove to be an important tool in our understanding of the female genital self-image of women who seek laser vaginal “rejuvenation”, labiaplasty or other elective genital procedures. Its use may also provide helpful insights into which aspects of genital self-image (e.g., concerns about appearance vs. smell) are most relevant to specific domains of sexual function such as orgasm or arousal. Clinicians may find the FGSIS to be an effective measure in therapeutic situations, too, as it may provide useful information about the different aspects of women’s feelings about their genitals. For example, a woman may feel comfortable letting a sexual partner view her genitals but highly uncomfortable allowing a healthcare provider to view her genitals (or vice versa) and being able to look at these item responses individually may be beneficial.
Conclusions
In summary, the FGSIS is a reliable and valid measure and future research should examine its psychometric properties in diverse samples to better understand its possible uses as well as the relationship between genital self-image and sexual function. Corresponding Author: Debra Herbenick, PhD, MPH, Center for Sexual Health Promotion, Indiana University, HPER 116, Bloomington, IN 47405, USA. Tel: 812-322-3777; Fax: 812-855-3936; E-mail: debby@ indiana.edu Conflict of Interest: None. Statement of Authorship
Category 1 (a) Conception and Design Debra Herbenick; Michael Reece (b) Acquisition of Data Debra Herbenick; Michael Reece (c) Analysis and Interpretation of Data Debra Herbenick; Michael Reece
Category 2 (a) Drafting the Article Debra Herbenick (b) Revising It for Intellectual Content Debra Herbenick; Michael Reece
Category 3 (a) Final Approval of the Completed Article Debra Herbenick; Michael Reece References 1 Nurnberg HG, Hensley PL, Heiman JR, Croft HA, DeBattista C, Paine S. Sildenafil treatment of women with antidepressant-associated sexual dysfunction: A randomized controlled trial. JAMA 2008;300:395–404. 2 Herbenick D, Reece M, Sanders SA, Dodge B, Ghassemi A, Fortenberry JD. Prevalence and characteristics of vibrator use by women in the United States: Results from a nationally representative study. J Sex Med 2009;6:1857–66. 3 Kelly MP, Strassberg DS, Kircher JR. Attitudinal and experiential correlates of anorgasmia. Arch Sexual Behav 1990;19: 165–77. 4 Herbenick D, Reece M, Hollub A, Satinsky S, Dodge B. Young female breast cancer survivors: Their sexual function and interest in sexual enhancement products and services. Cancer Nurs 2008;31:417–25. 5 Metz M, Epstein N. Assessing the role of relationship conflict in sexual dysfunction. J Sex Marital Ther 2002;28:139–64. 6 Laumann EO, Paik A, Rosen R. Sexual dysfunction in the United States: Prevalence and predictors. JAMA 1999;281: 537–44. 7 Meston C, Levin R, Sipski M, Hull E, Heiman J. Women’s orgasm. Annu Rev Sex Res 2004;1:173–257.
J Sex Med 2010;7:1822–1830
1830 8 Waltner R. Genital identity: A core component of sexual- and self-identity. J Sex Res 1986;22:399–402. 9 Winter H. An examination of the relationships between penis size and body image, genital image, and perception of sexual competency in the male. Dissertation. New York, 1999. 10 Reinholtz RK, Muehlenhard CL. Genital perceptions and sexual activity in a college population. J Sex Res 1995;32:155– 65. 11 Berman L, Berman J, Miles M, Pollets D, Powell JA. Genital self-image as a component of sexual health: Relationship between genital self-image, female sexual function, and quality of life measures. Jsex Marital Ther 2003;29:11–21. 12 DeVellis RF. Scale development: Theory and applications. 2nd edition. Sage University Applied Social Research Methods Series. Thousand Oaks, CA: Sage Publications; 2003. 13 Koning M, Ingeborg AZ, Bouman TK, van der Lei B. Female attitudes regarding labia minora appearance and reduction with consideration of media influence. Aesthetic Surg J 2009; 29:65–71. 14 Ensler E. The vagina monologues. New York: Villard; 1998. 15 Lawhead RA, Majmudar B. Early diagnosis of vulvar neoplasia as a result of vulvar self-examination. J Reprod Med 1990;35: 1134–7. 16 Herbenick D. The development and validation of a scale to measure attitudes toward women’s genitals. Int J Sex Health 2010;21:153–66. 17 Morrison TG, Bearden A, Ellis SR, Harriman R. Correlates of genital perceptions among Canadian post-secondary students. Electronic J Human Sexuality 2005. Available at: http://www. ejhs.org/volume8/GenitalPerceptions.htm (accessed November 5, 2009). 18 Braun V, Wilkinson S. Socio-cultural representations of the vagina. J Reproduct Infant Psychol 2001;19:17–32. 19 Waskul DD, Vannini P, Wiesen D. Women and their clitoris: Personal discovery, signification, and use. Symbol Interactism 2007;30:151–74. 20 Braun V, Kitzinger C. “Snatch”, “hole” or “honey pot”? Semantic categories and the problem of nonspecificity in female genital slang. J Sex Res 2001;38:146–58. 21 Braun V, Kitzinger C. Telling it straight? Dictionary definitions of women’s genitals. J Sociolinguistics 2002;5:214–32. 22 Snell RS. Clinical anatomy for medical students. 3rd edition. London: Little, Brown and Co.; 1986. 23 Williams PL. Grays anatomy: The anatomical basis of medicine and surgery. 38th edition. Edinburgh: Churchill Livingstone; 1996.
Herbenick and Reece 24 Bancroft JH, Loftus J, Long JS. Distress about sex: A national survey of women in heterosexual relationships. Arch Sex Behav 2003;32:193–208. 25 Downe-Wamboldt B. Content analysis: Method, applications, and issues. Health Care Women Int 1992;13:313–21. 26 Kondracki NL, Wellma NS, Amundson DR. Content analysis: Review of methods and their applications in nutrition education. J Nutr Educ Behav 2002;34:224–30. 27 Graneheim UH, Lundman B. Qualitative content analysis in nursing research: Concepts, procedures and measures to achieve trustworthiness. Nurs Educ Today 2004;24:105–12. 28 Rosen R, Brown C, Heiman J, Leiblum S, Meston C, Shabsigh R, Ferguson D, D’agostino R. The Female Sexual Function Index (FSFI): A multidimensional self-report instrument for the assessment of female sexual function. J Sex Marital Ther 2000;26:191–208. 29 Ter Kuile MM, Brauer M, Laan E. The Female Sexual Function Index (FSFI) and the Female Sexual Distress Scale (FSDS): Psychometric properties within a Dutch population. J Sex Marital Ther 2006;32:289–304. 30 Verit FF, Verit A. Validation of the female sexual function index in women with chronic pelvic pain. J Sex Med 2007;4: 1635–41. 31 Wiegel M, Meston C, Rosen R. The Female Sexual Function Index (FSFI): Cross-validation and development of clinical cutoff scores. J Sex Marital Ther 2005;31:1–20. 32 Witting K, Santtila P, Jern P, Varionen M, Wager I, Höglund M, Johansoon A, Vikström N, Sandnabba NK. Evaluation of the Female Sexual Function Index in a population based sample from Finland. Arch Sex Behav 2008;37:912–24. 33 Lowenstein L, Gamble T, Sanses TV, van Raalte H, Carberry C, Jakus S, Pham T, Nguyen A, Hoskey K, Kenton K. Changes in sexual function after treatment for prolapse are related to the improvement in body image perception. J Sex Med 2009;6:2286–91. 34 Miklos JR, Moore RD. Labiaplasty of the labia minora: Patients’ indications for pursuing surgery. J Sex Med 2008; 5:1492–5. 35 Burrows LJ, Shaw HA, Goldstein AT. The vulvar dermatoses. Review. J Sex Med 2008;5:276–83. 36 Goodman MP, Placik OJ, Benson RH, Miklos JR, Moore RD, Jason RA, Matlock DL, Simopoulos AF, Stern BH, Stanton RA, Kolb SE, Gonzalez F. A Large Multicenter Outcome Study of Female Genital Plastic Surgery. J Sex Med 2009; Nov 12 [Epub ahead of print] doi: 10.1111/j.1743-6109.2009. 01573.x.
Appendix A Female Genital Self Image Scale
The following items are about how you feel about your own genitals (the vulva and the vagina). The word vulva refers to a woman’s external genitals (the parts that you can see from the outside such as the clitoris, pubic mound, and vaginal lips). The word vagina refers to the inside part, also sometimes called the “birth canal” (this is also the part where a penis may enter or where a tampon is inserted). Please indicate how strongly you agree or disagree with each statement. Please mark an “X” in the box to indicate how strongly you agree or disagree with each statement. Strongly Disagree I I I I I I I
feel positively about my genitals. am satisfied with the appearance of my genitals. would feel comfortable letting a sexual partner look at my genitals. think my genitals smell fine. think my genitals work the way they are supposed to work. feel comfortable letting a healthcare provider examine my genitals. am not embarrassed about my genitals.
J Sex Med 2010;7:1822–1830
Disagree
Agree
Strongly Agree
ORIGINAL RESEARCH—OUTCOMES ASSESSMENT Development and Validation of the Female Genital Self-Image Scale jsm_1728
1822..1830
Debra Herbenick, PhD, MPH, and Michael Reece, PhD, MPH Indiana University- Center for Sexual Health Promotion, Bloomington, IN, USA DOI: 10.1111/j.1743-6109.2010.01728.x
ABSTRACT
Introduction. Women’s sexual function may be influenced by various factors including medical conditions, trauma or abuse, medications, relationship dynamics, relaxation, mood, and body image. However, few studies have explored the influence of a woman’s genital self-image on her sexual function or behaviors. Aims. The purpose of this study was to establish a reliable and valid measure of female genital self-image, the Female Genital Self-Image Scale (FGSIS), and to assess the relationship between scores on the FGSIS and women’s sexual function. Methods. The FGSIS was developed in two stages. Phase One involved an analysis of cross-sectional paper-based survey data and a review of the literature. Phase Two involved a cross-sectional internet-based administration of the scale items to a total of 1,937 women. Main Outcome Measures. Psychometric properties of the scale were evaluated through the use of reliability analysis, factor analysis, and score differences based on women’s experience of orgasm from receiving cunnilingus or from self-masturbation with a vibrator. Correlation analysis was used to explore the relationship between female genital self-image and scores on the Female Sexual Function Index (FSFI). Results. The scale was found to have sufficient reliability (Cronbach’s alpha = 0.88) and one factor that explained 59.23% of the variance. Women who had ever experienced orgasm as a result of cunnilingus or self-masturbation with a vibrator and women who reported having had a gynecological exam in the previous 12 months had significantly higher FGSIS scores than those who had not (P < 0.001). Scores on the FGSIS were positively and significantly correlated with scores on all FSFI domains (P < 0.001), including the Total score, with the exception of the Desire domain. Conclusion. The FGSIS was initially found to be a reliable and valid measure though further research is needed to understand its properties in diverse populations. In addition, female genital self-image was found to be positively related to women’s sexual function. Herbenick D, and Reece M. Development and validation of the female genital self image scale. J Sex Med 2010;7:1822–1830. Key Words. Genital Self-Image; Sexual Function; Orgasm; Vulva; Female External Genitalia; Vagina
Introduction
W
omen’s sexual function may be influenced by a range of factors including medical conditions, trauma or abuse, medications, relationship dynamics, relaxation, mood, and body image [1–7]. However, few studies have explored the influence of a woman’s genital self-image on her sexual function or behaviors. Waltner, in 1986, J Sex Med 2010;7:1822–1830
proposed the concept of genital identity, which he defined as “those self-definitions, self-attitudes and subsequent feelings which arise from specific interactions and experiences which either indirectly or directly involve the genitals” [8]. He suggested that genital identity could both arise from and influence sexual interactions. It was Winter, however, in 1989, that was the first to quantitatively measure genital self-image, © 2010 International Society for Sexual Medicine
Female Genital Self-Image albeit among men, a concept that focused on men’s perceptions of the appearance and size of their genital parts [9]. In 1995, Reinholtz and Muehlenhard explored college students’ perceptions of their own and their partners’ genitals, using items created specifically for their study [10]. They found that positive genital perceptions were related to higher engagement in, and enjoyment of, sexual activities. Specifically, ratings of liking one’s own genitals were significantly correlated with enjoyment of penile-vaginal intercourse as well as receiving oral sex. The factor termed “liking one’s own genitals” consisted of three items related to comfort with, liking of and feeling proud of one’s genitals. The small size of the factor, and the fact that no items used to measure genital perceptions in this study were related to genital appearance or function (though 5 of 11 items related to genital smell or taste), suggested the need for expanding our understanding of this construct. In a 2003 study, Berman and colleagues devised a scale to measure women’s genital self-image and documented a relationship between genital selfimage and sexual function [11]. Although their findings advanced knowledge pertaining to genital self-image and sexual function, some of the scale items used in this study are problematic. The structure of some items conflicts with established recommendations of measurement by being double-barreled (assessing two or more issues or varying perspectives on an issue in a single question), which presents the possibility of an imprecise or invalid measure of the construct [12]. Another scale item, “I look at my genitals”, posed additional measurement problems as women may look at their genitals as a result of positive or negative self-image (e.g., because they find their genitals attractive or arousing or because they find them unattractive and consequently look at them with a desire to alter their appearance) or for health reasons [13–15]. Rather than being a means of measuring female genital self-image, then, such an item may be better used to understand group differences. For example, do women who look at their genitals have, on average, more positive or more negative female genital self-image than women who do not look at their genitals? As per this point, one study found that women who had looked at their genitals three or more times during the previous year had more positive perceptions of women’s genitals [16]. Two additional items in the scale by Berman and colleagues assessed the types of messages that respondents were given about their genitals while growing up. Although these may be useful measures on
1823 which to group individuals and compare their genital self-image, it is unclear why messages received as a child would be valid indicators of one’s current genital self-image. If they are, then it may be worth considering whether genital selfimage—at least as measured by their scale—is better considered as a trait rather than a state. In a 2005 study, Morrison and colleagues adapted Winter’s male genital self image scale for use among female college students [17]. However, in one administration of the scale 26.9% to 53.7% of female respondents selected the neutral response option on each item, raising issues about the validity of the overall scale because of item nonresponse. Because people frequently choose neutral response options when responding to items of a sensitive nature, it has been suggested that having neutral response choices should be avoided in order to improve validity [12]. Women’s feelings and beliefs related to their genitals may be a particularly sensitive topic for some given that women’s genitals have traditionally been shrouded in secrecy and taboo [18]. Women may enter adulthood having had little information, education or experience viewing their genitals [19]. In addition, terminology related to women’s genitals tends to be vague, some contemporary medical texts still do not feature the clitoris in diagrams, and medical dictionaries may define the clitoris in terms of the penis but not vice versa [20–23]. Any of these factors may reflect or influence how a woman feels about or experiences her genitals. In spite of researchers’ efforts to quantitatively assess female genital self-image, there has yet to be a widely used measure that assesses a range of relevant concepts and that utilizes items that are judged to be valid measures of the construct. Given the importance of genital appearance and function to many women [13,24] it is likely that these are important aspects of the construct of female genital self-image to consider in any scale of this nature. Both appearance and function were assessed in the items developed by Berman and colleagues [11] and detailed items related to appearance (but not function) were assessed in the scale modified by Morrison and colleagues [17]. In summary, studies have explored issues related to female genital self-image (or women’s “genital perceptions”) and have provided important, although preliminary, insights into this construct and its relationship to women’s sexuality. However, there remains a need for a scale that fully operationalizes the construct of genital image within a contemporary context and that has underJ Sex Med 2010;7:1822–1830
1824 gone comprehensive psychometric analyses. Being able to measure female genital self-image in a reliable and valid manner may be particularly helpful to our understanding of how procedures such as genital cosmetic surgery or treatments such as vulvectomy may affect a woman’s perceptions of the way that her genitals look or function. Aims
The purpose of this study was to establish a reliable and valid measure of female genital selfimage, the Female Genital Self Image Scale (FGSIS), and to assess the relationship between scores on the FGSIS and women’s sexual function. The FGSIS was developed in two stages, which will be reported on separately within this paper. Phase One involved the development of an initial item pool based on elicitation surveys administered to an undergraduate sample; Phase Two involved the administration of the scale and the assessment of its psychometric properties. Phase One: Development of an Item Pool
The purpose of Phase One was to develop an item pool for the FGSIS based on feelings and beliefs about women’s genitals elicited from paper-based surveys and a review of scientific and popular literature. The Institutional Review Board at the author’s institution reviewed and approved all study protocols for each phase of the research.
Methods Item Elicitation The initial item pool was developed using data from item elicitation surveys that were collected from 370 participants (236 women and 134 men) who ranged in age from 18 to 40 years (mean 20.3 years, standard deviation [SD] = 2.1). The vast majority, 96.1% (N = 317), identified as heterosexual. Participants were recruited from undergraduate health classes and asked to complete brief and anonymous item elicitation surveys that sought the language they used to describe perceptions of genitals and attitudes about sexuality. We elicited language from both women and men in this phase in order to ensure that language was reflective of that used by both genders given that women’s perceptions of their genitals or the language they use to describe them could be influenced by male partners or influenced by shared language within couples. Respondents were randomly assigned to anonymously complete one of nine different versions of J Sex Med 2010;7:1822–1830
Herbenick and Reece the item elicitation survey. In order to elicit an expansive range of feelings, beliefs, and attitudes about women’s genitals and to do so in a manner that allowed for assessing similar concepts using a variety of question types and contexts, nine versions of the elicitation survey were used. The varying versions included open-ended questions (e.g., “What do you think your partner would like or dislike about your genitals?”), fill-in-the-blank items (e.g., “The best thing about women’s genitals is ”), as well as requests for participants’ to rank order characteristics of genitals that were important to them (e.g., smell, appearance, etc.). Some versions also presented vignettes about genitalrelated situations (e.g., performing or receiving cunnilingus, having a gynecological examination) and asked respondents to answer questions about what they would do to prepare for such events. Nine versions also enabled us to elicit a depth of detail using open-ended items but in a way that was not overly burdensome to participants.
Development of Themes and Items Content analysis was performed separately by two researchers to identify both the manifest content (e.g., content that is visible and obvious) and the latent content (e.g., the interpretation of the underlying meaning of words and text) of the participants’ responses [25,26]. The product of each content analysis was a set of meaning units (e.g., constellations of statements from the data that relate to the same central meaning). An example of a meaning unit from this data is “vaginas smell bad”. The meaning units were subsequently combined from the two researchers and each independently derived codes for each grouping of meaning units (e.g., meaning units related to smell were coded by the researchers as “smell”). After each researcher had independently coded the data, they met together for the purpose of reaching 100% consensus on the codes by reflecting and discussing on the extent to which there was agreement between the codes [27]. In an additional process to ensure 100% consensus on the codes for the meaning units, two external researchers with expertise in women’s sexuality and medicine reviewed and indicated that they were in agreement with the codes. Based on the consensus agreement of the reviewers, support existed for four major themes across the codes: smell/taste, appearance, sexual function, and shame/pride. Subsequent to the identification of the themes and based upon the language, words, and phrases elicited, the first author wrote and revised items in
1825
Female Genital Self-Image Table 1
Item pool, retention decisions based on reliability analysis and favor analysis of Female Genital Self Image Scale
Items 1. I feel positively about my genitals. 2. I am satisfied with the appearance of my genitals. 3. I would feel comfortable letting a sexual partner look at my genitals. 4. I think my genitals smell fine. 5. I think my genitals work the way they are supposed to work. 6. I feel comfortable letting a healthcare provider examine my genitals. 7. I am not embarrassed about my genitals. 8. I like how my genitals look.
9. I can see how a sexual partner would be aroused by my genitals.
10. I can see how a sexual partner would enjoy the taste of my genitals.
11. I am comfortable with how my genitals smell.
Outcome based on reliability analysis
Factor loading
Corrected item to total correlation
Item mean
Item standard deviation
Retained. Retained.
0.86 0.84
0.77 0.74
2.96 2.90
0.81 0.84
Retained.
0.81
0.72
3.07
0.79
Retained. Retained.
0.67 0.65
0.57 0.55
3.02 3.20
0.80 0.72
Retained.
0.69
0.60
3.12
0.79
Retained.
0.84
0.77
3.03
0.82
Excluded after Iteration 1 given its overlap in meaning with Item 2. Cronbach’s alpha changed from 0.93 to 0.92 with deletion. Excluded after Iteration 2 given its overlap with item 4; Cronbach’s alpha changed from 0.92 to 0.91 with deletion. Excluded after Iteration 3 given concerns that the item may measure perceptions of oral sex rather than genital self-image; Cronbach’s alpha changed from 0.91 to 0.90 with deletion. Excluded after iteration 4 given its overlap with item 6; Cronbach’s alpha changed from 0.90 to 0.88 with deletion.
collaboration with the second author and with feedback and suggestions from two other faculty members who have expertise in women’s sexuality and medicine. This resulted in a total of 11 final items written to reflect each of the four themes (see Table 1). Some items were redundant in keeping with suggestions for items writing [12]. For example, “I like how my genitals look” and “I am satisfied with the appearance of my genitals” were both included in order to test different wordings of items related to perceptions of genital appearance. A 4-point response scale (strongly disagree, disagree, agree, strongly agree) was chosen based on recommendations that a neutral response option in scales of a sensitive nature may result in respondents “fence-sitting” [12]. The initial pool of items was subsequently reviewed by two individuals with expertise in sexual health and gender studies who had reviewed the initial codes resulting from content analyses and agreed that the 11 items were reflective of the four themes underlying our conceptualization of female genital self-image, thus supporting face validity of the initial FGSIS.
Phase Two: Scale Administration and Psychometric Analysis
The purpose of Phase Two was to assess the internal consistency of the scale items and to identify whether the removal of any redundant items would help to improve the scale’s reliability. Additionally, data analyses from Phase Two included assessments of the validity and predictive capacity of the FGSIS following its administration to a convenience sample of adult women.
Methods Procedures From January through March 2008, the research team, through a collaborative research partnership with the leadership of a large U.S.-based, in-home sex toy party company, asked women who worked as party facilitators to distribute study recruitment cards to the women who attended their parties. The researchers did not have direct access to or communication with the facilitators so as to preserve their anonymity; rather, the research team drafted a recruitment email that was then emailed by company leadership to the facilitators. J Sex Med 2010;7:1822–1830
1826 Company leadership also arranged for recruitment cards to be mailed directly to party facilitators who were asked to then pass the cards along to the party attendees. This was part of a larger study about women who attend in-home sex toy parties. Recruitment cards invited attendees to visit the study website in order to learn more about the study and to decide whether they wanted to participate. The study was described as being about the characteristics of women who attend in-home sex toy parties. Individuals who agreed to participate in the study completed an anonymous, crosssectional survey and had the opportunity to enter themselves into a drawing for one of ten $50 gift cards. No other incentives were provided.
Measures The study questionnaire included items related to demographic and background variables (e.g., age, education, race/ethnicity, sexual orientation, marital status) and measures related to recent and lifetime sexual behaviors and age of first vaginal intercourse. Relevant to these analyses, respondents were asked “Have you ever had an orgasm from receiving oral sex?” (yes, no) and “Have you ever had an orgasm from self-masturbation with a vibrator?” (yes, no). A health-related item asked respondents “In the past 12 months, have you had a gynecological exam?” (yes, no). The number of items shown to each woman varied considerably depending upon the behaviors that she reported. For example, if a woman reported that she had not engaged in a certain behavior (such as masturbation) then she did not receive follow-up questions about that behavior. However, all respondents received the measures analyzed in this report. Time to complete the survey ranged from approximately 10–20 minutes. Female Sexual Function Index (FSFI). The FSFI is a 19-item measure of women’s sexual function that has established reliability, validity and temporal stability [28–32]. It measures the domains of desire, arousal, lubrication, pain, orgasm, and satisfaction, and also results in a total score. Higher scores on each domain and on the total score indicate more positive sexual function. FGSIS Test Items. The 11 FGSIS items were preceded by the following statement: “The following items are about how you feel about your own genitals (the vulva and the vagina). The word vulva refers to a woman’s external genitals (the parts that you can see from the outside such as the clitoris, pubic mound, and vaginal lips). The word vagina J Sex Med 2010;7:1822–1830
Herbenick and Reece refers to the inside part, also sometimes called the “birth canal” (this is also the part where a penis may enter or where a tampon is inserted). Please indicate how strongly you agree or disagree with each statement” (see Appendix A).
Statistical Analysis SPSS version 16.0 (Chicago, IL, USA) was used for statistical analysis. Descriptive statistics were used to report respondents’ demographic and other background data. Specific analyses conducted for psychometric evaluation of the FGSIS are described in their respective section of the results. Results
A total of 2,049 women completed the online survey and 1,937 respondents had complete data on the FGSIS (94.5%). Because the total number of recruitment cards distributed is unknown, it is not possible to calculate an overall survey response rate.
Table 2
Phase two sample characteristics (N = 2,049)
Participant characteristics
N (%)
Age (mean, standard deviation) Race American Indian/Alaskan Native Asian or Asian American Black or African American Native Hawaiin or other Pacific Islander White or Caucasian Hispanic or Latino Multiracial Other Relationship status Single In a relationship Living together Married Separated Divorced Widowed Current college/university student Education Middle school Some high school High school graduate Some college/2-year degree College graduate Graduate school Other Sexual orientation Heterosexual/straight Homosexual/lesbian Bisexual Questioning or uncertain Asexual Other
29.55 (8.4) 21 11 77 8 1,818 65 40 9
(1.0) (0.5) (3.8) (0.4) (88.7) (3.2) (2.0) (0.4)
379 297 322 938 28 62 5 567
(18.5) (14.5) (15.7) (45.8) (1.4) (3.0) (0.2) (27.7)
2 46 397 1,014 434 133 23
(0.1) (2.2) (19.4) (49.5) (21.2) (6.5) (1.1)
1,866 12 142 17 2 10
(91.1) (0.6) (6.9) (0.8) (0.1) (0.5)
Female Genital Self-Image As shown in Table 2, the mean age of the sample was young (mean = 29.55, SD = 8.4, range = 18– 68) and most respondents were white (88.7%), heterosexual (91.1%), and partnered (45.8% were married, 15.7% were cohabitating and 14.5% were in a relationship). A total of 77.2% had at least some college education. Most women (80.9%, N = 1567) had been sexually active with male partners within the past four weeks and few (0.9%, N = 17) reported sexual interactions with other women or both men and women (1.8%, N = 35) during this time period. The remainder (15.2%, N = 294) reported no partnered sexual interactions during the four weeks preceding the study. Women reported an average age at first intercourse of 16.7 years (median = 16.0, SD = 3.0).
Internal Consistency Reliability analysis was conducted on the FGSIS with each iteration in order to determine its internal consistency and to assess whether any redundant items should be deleted. Cronbach’s coefficient alpha was used as an indicator of how reliable items were in identifying the overall construct. Although this coefficient can range from 0 to 1.0, a score above 0.7 was considered the minimum and a score at or above 0.8 preferable [12]. Corrected item-to-total correlations were also used to test for internal consistency; those at or above 0.30 were generally considered sufficient [30]. As shown in Table 1, the Cronbach’s alpha for all 11 FGSIS items was very high (0.93), allowing for the possibility of deleting items for inclusion in the final scale for rational or empirical reasons [12]. Using four iterations of reliability analysis and rational decision making, items 1, 5, 7, and 9 were removed. The resulting 7-item FGSIS had a Cronbach’s alpha coefficient of 0.88 and a mean score of 21.30 (SD = 3.9; N = 1937). The scores on each item were summed for a total possible FGSIS score ranging from 7 to 28 with higher scores indicating more positive female genital self-image. Construct Validity Factor analysis was performed using principal component extraction with initial communalities of 1.0. A priori criteria for determining the number of factors were that factors had to have eigenvalues greater than 1.0; the scree plot was also examined. The scree plot indicated one factor that explained 59.23% of the variance. A large decrease
1827 was seen between the first and second eigenvalues, with small decreases thereafter (eigenvalues: 4.15, 0.88, 0.61, 0.56, 0.38, 0.31, 0.12). Factor loadings ranged from 0.65 to 0.86 (i.e., above the 0.40 cutoff) [30]. Corrected item-to-total correlations ranged from 0.55 to 0.77 (i.e., exceeded 0.30; see Table 1). These findings provided further support for the construct validity of the FGSIS. The item “I think my genitals work the way they are supposed to work” was the most highly endorsed (mean = 3.20, SD = 0.70).
Predictive Capacity Using analysis of variance, the predictive capacity of the FGSIS was assessed in relation to three dichotomously measured experiences or behaviors, including partnered sexual behaviors (having had an orgasm from cunnilingus), masturbation (solo masturbation with a vibrator), and having had a gynecological exam within the past year. Similar behaviors have been shown to be related to women’s attitudes toward women’s genitals more globally [16] and previous research has documented a relationship between genital perceptions and enjoyment of sex or ease of orgasm [10,16] particularly resulting from sexual behaviors that involve high degrees of intimacy with one’s genitals, such as cunnilingus. A higher FGSIS score among participants who reported a history of these was considered to be an indicator of the predictive capacity of the scale with regard to each. Women who had ever experienced orgasm as a result of cunnilingus, those who had experienced orgasm as a result of self-masturbation with a vibrator and women who reported having had a gynecological exam in the previous 12 months had significantly higher FGSIS scores than those who had not (P < 0.001, see Table 3), supporting the predictive capacity of the scale. Relationship to Sexual Function Bivariate correlations were used to assess the relationship between women’s scores on the FGSIS and the FSFI domains and total score. Women’s scores on the FGSIS were positively and significantly correlated with their scores on the FSFI domains of Arousal (r = 0.18, P < 0.001), Lubrication (r = 0.14, P < 0.001), Orgasm (r = 0.17, P < 0.001), Satisfaction (r = 0.19, P < 0.001), Pain (r = 0.13, P < 0.001) and the Total Score (r = 0.20, P < 0.001). There was not a significant relationship between women’s scores on the FGSIS and the FSFI Desire domain (r = 0.03, P = 0.18). J Sex Med 2010;7:1822–1830
1828 Table 3
Herbenick and Reece Predictive capacity of the female genital self image scale (N = 1,937)
Behavior
FGSIS score (mean, SD)
Ever experienced orgasm during cunnilingus Yes (N = 1,461) 21.52 (3.7) No (N = 476) 20.61 (4.1) Ever experienced orgasm during self-masturbation with a vibrator Yes (N = 1,470) 21.51 (3.9) No (N = 467) 20.62 (3.8) Had a gynecological exam in the previous 12 months Yes (N = 859) 21.50 (3.8) No (N = 1,156) 20.53 (4.1)
Discussion
The current study found sufficient support for the reliability and validity of a scale to measure female genital self-image. In this sample, female genital self-image was found to be significantly correlated with female sexual function, with the exception of the FSFI Desire domain. This is in contrast with results from the 2003 study in which Berman and colleagues found a correlation between their measure of female genital self image and the FSFI Desire domain, but no significant relationship between scores on their genital self image scale and the FSFI measures of arousal, lubrication, orgasm, pain or satisfaction [11]. In their study, however, the authors did find a relationship between their measure of female genital self image and sexual distress. Given that 7 of the 17 items in Part 1 of their two-part scale were items that measured respondents’ feelings of embarrassment, shame, anxiety or worry in relation to their genitals, it may be that their measure assesses distress or anxiety in relation to one’s genitals rather than genital self-image per se. Further research is recommended in order to better understand the nuances of these different measures. It may be useful, for example, for future research to explore the boundaries of what is meant by genital self-image vs. what may be better described as “genital distress.” FGSIS items were developed as a result of elicitation surveys that probed respondents to answer questions about a wide range of feelings and beliefs related to female genitals. It may be that FGSIS items are more expansive in representing a range of feelings about their genitals (e.g., embarrassment, comfort, cognitions, affect, and satisfaction rather than an emphasis on distress or worry). The finding that scores on the FGSIS are significantly and positively correlated with most aspects of sexual function as measured by the FSFI J Sex Med 2010;7:1822–1830
F
p
omega squared
20.37
<0.001
0.01
19.06
<0.001
0.01
7.53
<0.001
0.01
are important. Women’s feelings about their genitals have long been described as relevant to their ease of orgasm and overall sexual experience and these data provide evidence of such a relationship. A strength of this study is that a careful, rational-empirical approach to scale development was used in order to enhance the reliability and validity of the scale. Although a limitation of the study is that a convenience sample was used, a strength is that the large sample size in Phase Two was sufficient for the use of factor analysis [12]. However, caution should be exercised before generalizing these results as women who attend in-home sex toy parties may differ from the general population on key sexuality-related variables. Further research should test the FGSIS in more diverse populations within and outside of the United States to better understand the validity of the measure in other populations. The mean and standard deviation of the FGSIS in this sample should not be accepted as scale norms until further validated in a larger population-based sample. Finally, the current study was not designed to allow testing of the temporal stability of the scale; as such, future research should apply the FGSIS to the scrutiny of test-retest reliability. Findings from this study have several implications. First, given some of the limitations of previous scales that intended to assess female genital self-image as a construct, researchers should carefully consider what it is that they want to measure prior to selecting a scale. Available scales vary considerably in terms of their item content and what it is that each measures. The FGSIS developed in this study accounts for a wide range of characteristics related to female genital self-image and provides a new option for researchers and clinicians who are interested in this topic; however, further research is needed to assess its psychometric properties among a population-based sample of women.
1829
Female Genital Self-Image By taking a careful rational-empirical approach to scale development, scale items were written that may be useful to clinicians individually as well as through the use of the total 7-item scale. Specifically, although the FGSIS may be useful to clinicians and researchers as they attempt to learn more about the influence of women’s genital self-image on their sexual function, responses to individual scale items may provide helpful insights or encourage conversations between clinicians and clients or patients. For example, although a woman may score highly on the scale as a whole (indicating a generally positive genital self-image), clinicians should take note of any negative endorsements of items that indicate a woman may feel negative about certain aspects of her genitals such as their appearance, function, or smell, or her ability to feel comfortable having her genitals seen by a sexual partner or healthcare provider. Given that patients may often find it difficult to approach their healthcare providers with questions or concerns related to their sexual health and well-being, the ability to gather such information via a questionnaire may be a useful starting point in discussing sexuality and genital health with patients or clients. In the future, researchers might use the FGSIS in clinical samples in order to understand the extent to which it provides useful insights into women’s genital self-image before and after treatment for vulvar intraepithelial neoplasia, vulvar cancer, vulvar dermatoses, pelvic organ prolapse or other gynecological conditions that may affect the appearance, function or smell of a woman’s genitals, or women’s comfort letting others view their genitals [33–36]. The FGSIS may prove to be an important tool in our understanding of the female genital self-image of women who seek laser vaginal “rejuvenation”, labiaplasty or other elective genital procedures. Its use may also provide helpful insights into which aspects of genital self-image (e.g., concerns about appearance vs. smell) are most relevant to specific domains of sexual function such as orgasm or arousal. Clinicians may find the FGSIS to be an effective measure in therapeutic situations, too, as it may provide useful information about the different aspects of women’s feelings about their genitals. For example, a woman may feel comfortable letting a sexual partner view her genitals but highly uncomfortable allowing a healthcare provider to view her genitals (or vice versa) and being able to look at these item responses individually may be beneficial.
Conclusions
In summary, the FGSIS is a reliable and valid measure and future research should examine its psychometric properties in diverse samples to better understand its possible uses as well as the relationship between genital self-image and sexual function. Corresponding Author: Debra Herbenick, PhD, MPH, Center for Sexual Health Promotion, Indiana University, HPER 116, Bloomington, IN 47405, USA. Tel: 812-322-3777; Fax: 812-855-3936; E-mail: debby@ indiana.edu Conflict of Interest: None. Statement of Authorship
Category 1 (a) Conception and Design Debra Herbenick; Michael Reece (b) Acquisition of Data Debra Herbenick; Michael Reece (c) Analysis and Interpretation of Data Debra Herbenick; Michael Reece
Category 2 (a) Drafting the Article Debra Herbenick (b) Revising It for Intellectual Content Debra Herbenick; Michael Reece
Category 3 (a) Final Approval of the Completed Article Debra Herbenick; Michael Reece References 1 Nurnberg HG, Hensley PL, Heiman JR, Croft HA, DeBattista C, Paine S. Sildenafil treatment of women with antidepressant-associated sexual dysfunction: A randomized controlled trial. JAMA 2008;300:395–404. 2 Herbenick D, Reece M, Sanders SA, Dodge B, Ghassemi A, Fortenberry JD. Prevalence and characteristics of vibrator use by women in the United States: Results from a nationally representative study. J Sex Med 2009;6:1857–66. 3 Kelly MP, Strassberg DS, Kircher JR. Attitudinal and experiential correlates of anorgasmia. Arch Sexual Behav 1990;19: 165–77. 4 Herbenick D, Reece M, Hollub A, Satinsky S, Dodge B. Young female breast cancer survivors: Their sexual function and interest in sexual enhancement products and services. Cancer Nurs 2008;31:417–25. 5 Metz M, Epstein N. Assessing the role of relationship conflict in sexual dysfunction. J Sex Marital Ther 2002;28:139–64. 6 Laumann EO, Paik A, Rosen R. Sexual dysfunction in the United States: Prevalence and predictors. JAMA 1999;281: 537–44. 7 Meston C, Levin R, Sipski M, Hull E, Heiman J. Women’s orgasm. Annu Rev Sex Res 2004;1:173–257.
J Sex Med 2010;7:1822–1830
1830 8 Waltner R. Genital identity: A core component of sexual- and self-identity. J Sex Res 1986;22:399–402. 9 Winter H. An examination of the relationships between penis size and body image, genital image, and perception of sexual competency in the male. Dissertation. New York, 1999. 10 Reinholtz RK, Muehlenhard CL. Genital perceptions and sexual activity in a college population. J Sex Res 1995;32:155– 65. 11 Berman L, Berman J, Miles M, Pollets D, Powell JA. Genital self-image as a component of sexual health: Relationship between genital self-image, female sexual function, and quality of life measures. Jsex Marital Ther 2003;29:11–21. 12 DeVellis RF. Scale development: Theory and applications. 2nd edition. Sage University Applied Social Research Methods Series. Thousand Oaks, CA: Sage Publications; 2003. 13 Koning M, Ingeborg AZ, Bouman TK, van der Lei B. Female attitudes regarding labia minora appearance and reduction with consideration of media influence. Aesthetic Surg J 2009; 29:65–71. 14 Ensler E. The vagina monologues. New York: Villard; 1998. 15 Lawhead RA, Majmudar B. Early diagnosis of vulvar neoplasia as a result of vulvar self-examination. J Reprod Med 1990;35: 1134–7. 16 Herbenick D. The development and validation of a scale to measure attitudes toward women’s genitals. Int J Sex Health 2010;21:153–66. 17 Morrison TG, Bearden A, Ellis SR, Harriman R. Correlates of genital perceptions among Canadian post-secondary students. Electronic J Human Sexuality 2005. Available at: http://www. ejhs.org/volume8/GenitalPerceptions.htm (accessed November 5, 2009). 18 Braun V, Wilkinson S. Socio-cultural representations of the vagina. J Reproduct Infant Psychol 2001;19:17–32. 19 Waskul DD, Vannini P, Wiesen D. Women and their clitoris: Personal discovery, signification, and use. Symbol Interactism 2007;30:151–74. 20 Braun V, Kitzinger C. “Snatch”, “hole” or “honey pot”? Semantic categories and the problem of nonspecificity in female genital slang. J Sex Res 2001;38:146–58. 21 Braun V, Kitzinger C. Telling it straight? Dictionary definitions of women’s genitals. J Sociolinguistics 2002;5:214–32. 22 Snell RS. Clinical anatomy for medical students. 3rd edition. London: Little, Brown and Co.; 1986. 23 Williams PL. Grays anatomy: The anatomical basis of medicine and surgery. 38th edition. Edinburgh: Churchill Livingstone; 1996.
Herbenick and Reece 24 Bancroft JH, Loftus J, Long JS. Distress about sex: A national survey of women in heterosexual relationships. Arch Sex Behav 2003;32:193–208. 25 Downe-Wamboldt B. Content analysis: Method, applications, and issues. Health Care Women Int 1992;13:313–21. 26 Kondracki NL, Wellma NS, Amundson DR. Content analysis: Review of methods and their applications in nutrition education. J Nutr Educ Behav 2002;34:224–30. 27 Graneheim UH, Lundman B. Qualitative content analysis in nursing research: Concepts, procedures and measures to achieve trustworthiness. Nurs Educ Today 2004;24:105–12. 28 Rosen R, Brown C, Heiman J, Leiblum S, Meston C, Shabsigh R, Ferguson D, D’agostino R. The Female Sexual Function Index (FSFI): A multidimensional self-report instrument for the assessment of female sexual function. J Sex Marital Ther 2000;26:191–208. 29 Ter Kuile MM, Brauer M, Laan E. The Female Sexual Function Index (FSFI) and the Female Sexual Distress Scale (FSDS): Psychometric properties within a Dutch population. J Sex Marital Ther 2006;32:289–304. 30 Verit FF, Verit A. Validation of the female sexual function index in women with chronic pelvic pain. J Sex Med 2007;4: 1635–41. 31 Wiegel M, Meston C, Rosen R. The Female Sexual Function Index (FSFI): Cross-validation and development of clinical cutoff scores. J Sex Marital Ther 2005;31:1–20. 32 Witting K, Santtila P, Jern P, Varionen M, Wager I, Höglund M, Johansoon A, Vikström N, Sandnabba NK. Evaluation of the Female Sexual Function Index in a population based sample from Finland. Arch Sex Behav 2008;37:912–24. 33 Lowenstein L, Gamble T, Sanses TV, van Raalte H, Carberry C, Jakus S, Pham T, Nguyen A, Hoskey K, Kenton K. Changes in sexual function after treatment for prolapse are related to the improvement in body image perception. J Sex Med 2009;6:2286–91. 34 Miklos JR, Moore RD. Labiaplasty of the labia minora: Patients’ indications for pursuing surgery. J Sex Med 2008; 5:1492–5. 35 Burrows LJ, Shaw HA, Goldstein AT. The vulvar dermatoses. Review. J Sex Med 2008;5:276–83. 36 Goodman MP, Placik OJ, Benson RH, Miklos JR, Moore RD, Jason RA, Matlock DL, Simopoulos AF, Stern BH, Stanton RA, Kolb SE, Gonzalez F. A Large Multicenter Outcome Study of Female Genital Plastic Surgery. J Sex Med 2009; Nov 12 [Epub ahead of print] doi: 10.1111/j.1743-6109.2009. 01573.x.
Appendix A Female Genital Self Image Scale
The following items are about how you feel about your own genitals (the vulva and the vagina). The word vulva refers to a woman’s external genitals (the parts that you can see from the outside such as the clitoris, pubic mound, and vaginal lips). The word vagina refers to the inside part, also sometimes called the “birth canal” (this is also the part where a penis may enter or where a tampon is inserted). Please indicate how strongly you agree or disagree with each statement. Please mark an “X” in the box to indicate how strongly you agree or disagree with each statement. Strongly Disagree I I I I I I I
feel positively about my genitals. am satisfied with the appearance of my genitals. would feel comfortable letting a sexual partner look at my genitals. think my genitals smell fine. think my genitals work the way they are supposed to work. feel comfortable letting a healthcare provider examine my genitals. am not embarrassed about my genitals.
J Sex Med 2010;7:1822–1830
Disagree
Agree
Strongly Agree