Introduction

Faking orgasm, a sexual strategy used by some women, has been documented for nearly a century. Touted during the Victorian era by physicians as “a justifiable innocent deception” (Robinson, 1917, p. 306) by wives looking to please their husbands, contemporary studies have estimated that 53–67 % of women engage in this behavior (Darling & Davidson, 1986; Kaighobadi, Shackelford, & Weekes-Shackelford, 2012; Muehlenhard & Shippee, 2010; Wiederman, 1997). Although faking orgasm has received much attention in the popular media (e.g., health and lifestyle magazines, movies, television shows, internet blogs), a review of the literature reveals little empirical evidence regarding its prevalence, correlates, or motivational underpinnings. Simply put, faking orgasm is one of the least well-studied human sexual behaviors. The present research was specifically aimed at examining the motivational aspects of faking orgasm in heterosexual women.

An understanding of the female orgasm is necessary to begin an inquiry into its feigned counterpart. Although a review of the physiological mechanisms of female orgasm is beyond the scope of this article (for review, see Komisaruk, Beyer-Flores, & Whipple, 2006; Meston, Hull, Levin, & Sipski, 2004), researchers have long argued that male and female orgasms have separate functions and different evolutionary determinants. Specifically, it is widely understood that female orgasm is not a critical element of reproduction (e.g., Baker & Bellis, 1993; Meston et al., 2004; Smith, 1984; Wallen & Lloyd, 2008). Although 75–90 % of women do not consistently orgasm during sexual activity with a partner (Bancroft, Loftus, & Long, 2003), and approximately 5–10 % of women do not orgasm at all (for review, see Lloyd, 2005), there is no evidence indicating that the absence of female orgasm prevents conception. Some researchers have proposed that the female orgasm may have been an outcome of evolutionary natural selection by offering reproductive advantages. One such possible advantage is that female orgasm may encourage or reinforce sexual behavior necessary for reproduction (for review, see Lloyd, 2005; Zietsch, Miller, Bailey, & Martin, 2011). Consistent with that idea, Brewer and Hendrie (2011) found that female copulatory vocalizations associated with orgasm effectively sped up their partner’s orgasm, which may increase the likelihood of conception.

Evolutionary considerations may be relevant to this research to the extent that natural selection has likely shaped women’s sexual goals (e.g., wanting to retain a mate), cognitions (e.g., thoughts about self and others as a partner), and emotions (e.g., empathy for a partner, feelings of arousal or sexual satisfaction, etc.), all of which are expected to contribute to the decision to fake orgasm. For example, in a study of 453 heterosexual female college students, Kaighobadi et al. (2012) found that women who perceived increased risk of partner infidelity were more likely to fake orgasm during sexual intercourse. They also found that women who faked orgasm were more likely to engage in mate retention strategies, such as direct guarding, intersexual negative inducements, and public possession signals (see Buss, 1988). Other research has found that women who fake orgasm are more likely to engage in intersexual negative inducements, i.e., to behave in less-exclusive and flirtatious ways with other men (Thornhill, Gangestad, & Comer, 1995). These findings suggest that women may fake orgasm either to display commitment, interest, or love to their partner, or to manipulate their partner’s commitment to them (Kaighobadi et al., 2012). In support of this interpretation, participants in Hite’s (1976) survey reported faking orgasm to prevent their partner from “straying” and as a means of fulfilling their nightly “social obligation” (p. 263).

A woman’s decision to fake orgasm may also have a meaningful impact on her male partner. Research has shown that 90 % of men are concerned with whether their female partner experiences orgasm (McKibbin, Bates, Shackelford, Hafen, & LaMunyon, 2010); in turn, Muehlenhard and Shippee’s (2010) study of 101 female college students found nearly 70 % of the women in their sample reported faking orgasm to avoid hurting their partner’s feelings. Other common reasons reported include faking orgasm to please their partner or because their partner’s orgasm seemed imminent. Darling and Davidson (1986) found that women whose partner asked if they experienced orgasm were more likely to fake orgasm; these women were also more likely to feel guilty if they did not orgasm, presumably stemming from concerns about hurting their partner’s feelings (see also Hite, 1976; Muehlenhard & Shippee, 2010). In the absence of female orgasm, faking orgasm may serve to help either or both partners avoid feelings of inadequacy or abnormality (Dove & Wiederman, 2000).

Given the strong possibility that men are often unable to distinguish between real and faked female orgasms, as suggested by popular media (e.g., When Harry Met Sally, Seinfeld, Parenthood), women may view faking orgasm as an effective strategy for coping with pressure to orgasm from themselves or their partners. Indeed, research has found a significant discrepancy between women’s reports of faking orgasm and their partner’s estimates of its frequency. In one study, a woman “reported that she never orgasmed but faked 100 % of the time; her partner reported that she orgasmed 100 % of the time and never faked” (Thornhill et al., 1995, p. 1605). Nearly 12 % of that sample reported significant discrepancies of more than 20 % between women’s reports of faking orgasm and their partner’s estimates of its frequency.

Several studies have suggested that women’s willingness to engage in sexual acts often stems from considerations that are not strictly sexual, but are more closely related to the rewards associated with an emotional connection with their partner or the need to avoid negative sexual outcomes (e.g., Basson, 2000, 2006; Graham, Bancroft, Doll, Greco, & Tanner, 2007; Mah & Binik, 2001). For example, research has shown that important sexual outcomes for women, such as perceiving sexual acts as pleasurable (Haavio-Mannila & Kontula, 1997), orgasmic responsiveness, and sexual satisfaction (King, Belsky, Mah, & Binik, 2011; Laumann, Paik, & Rosen, 1999; Mah & Binik, 2005), are associated with higher levels of emotional intimacy and relationship satisfaction. It is not surprising then that approximately 68 % of heterosexual women would stay with a partner, even if she had never experienced an orgasm with that partner (Brewer & Hendrie, 2011). Thus, insofar as faking orgasm enhances the perception of sex as pleasurable to a woman and/or her sexual partner, doing so may contribute to feelings of sexual satisfaction, as well as relationship satisfaction and maintenance, intimacy enhancement, and avoidance of conflict, which are often goals of female sexual behavior.

Many feminist scholars have argued that both popular culture and the tendency toward “medicalization” of the female sexual response has had a detrimental effect on perceptions of “normal” female sexual functioning (e.g., Basson, 2001; Heiman, 2007; Potts, 2000; Tiefer, 1991, 2001). Pervasive erotic images of women in Western culture and controversial diagnostic criteria in the DSM (e.g., Female Orgasmic Disorder) (for review, see Graham, 2010) have created an unrealistic expectation that women must achieve orgasm during sexual acts. Media exposure has further emphasized the necessity of being a “good lover,” i.e., able to respond sexually to, and satisfy the needs of, their partner (Darling & Davidson, 1986; Potts, 2000; Tiefer, 1991; Waterman & Chiauzzi, 1982). In their study, Muehlenhard and Shippee (2010) concluded that faking orgasm suggests a discrepancy between what is actually happening and what a woman thinks should be happening based on the expectations of society and her relationship. Ultimately, women’s choice to fake orgasm may be an attempt to avoid being viewed as physiologically and/or psychologically abnormal.

Given these influences, the female orgasm may be erroneously viewed by both women and their partners as evidence of their sexual competence or adequacy. In the absence of orgasm, faking orgasm may serve to help either or both partners avoid feelings of inadequacy or abnormality and feel better about themselves sexually and generally (Costa & Brody, 2007; Dove & Wiederman, 2000). These potential motives are further supported by research suggesting that women fake orgasm to increase the real or perceived excitement of sex for themselves or their partner (Hite, 1976; Wiederman, 1997). Participants in Hite’s (1976) survey indicated that faking orgasm was motivated by reasons related to perceived sexual competence, including wanting to appear sexy and reinforcing their partner for behavior they liked. Providing further support for this notion, in a study of 161 female college students, Wiederman (1997) found that women who faked orgasm had significantly higher levels of sexual self-esteem, i.e., as a result of their “responsiveness,” they tended to view themselves more positively as a sexual partner.

Although the studies noted here have suggested potential motives for women’s decision to fake orgasm—such as to protect their partner’s feelings, avoid feeling or being viewed as abnormal, and to maintain or strengthen their relationship with a partner—those ideas were not subjected to systematic inquiry. Much of the previously cited research suffered from important methodological limitations.

Darling and Davidson’s (1986) survey of 805 professional nurses included both open-ended responses that were coded as well as closed-form items regarding sexual attitudes, sexual behavior, and female sexual response. Although the researchers detailed the process through which items were created, no established scales were used for validation purposes and psychometric properties were not reported. Furthermore, Darling and Davidson did not disclose how it was determined whether or not participants had faked orgasm. In the Wiederman (1997) study, only one item asked participants whether they “have, at one time or another, pretended to have an orgasm during sexual intercourse” (p. 134); faking orgasm during other sexual encounters was not explored. Muehlenhard and Shippee (2010), much like Darling and Davidson (1986), asked participants to complete a 5-part open-ended and closed-form item survey. Psychometric properties of items were not reported, so it is unclear if they were adequate. In all three studies, frequency of faking orgasm was not assessed and only Muehlenhard and Shippee (2010) attempted to identify potential reasons why women faked orgasm.

Although the information gained from these studies can begin to shape our understanding of why women choose to fake orgasm, the methodological limitations of these studies highlight the need for a more in-depth quantitative examination. To systematically build this base of knowledge, a psychometrically sound measure of women’s specific motivation(s) for faking orgasm is needed. Building on the relevant literature and expanding the research focus to include oral sex and sexual intercourse will allow us to build a strong foundation for future exploration of faking orgasm. The current study aimed to develop a scale to assess the dimensions underlying women’s decision to fake orgasm, as “fake orgasm has many practitioners but few champions” (Jagose, 2010, p. 518).

Study 1

Method

Participants

Undergraduate females enrolled in introductory psychology and marketing courses at a large urban university in southeastern Pennsylvania (N = 481) were recruited through a web-based management system and through IRB-approved flyers posted around campus. Participants were required to meet two inclusion criteria: (1) having engaged in sexual intercourse and/or received oral sex and (2) having faked an orgasm during one or both of those activities. Inclusion criteria were included in the informed consent, questionnaire instructions, and were specifically queried following collection of demographic data. Participants ranged in age from 18 to 32 years (M = 20.33 years, SD = 2.48), were racially and ethnically diverse (White: 53 %, Black/African-American: 34 %, Other: 13 %; Hispanic: 8 %), and primarily heterosexual in orientation (heterosexual: 92 %; homosexual: 3 %; bisexual: 4 %). Of the total sample, 94.2 % endorsed engaging in sexual intercourse, 97.9 % reported receiving oral sex, and 89.5 % reported ever having experienced an orgasm. More than 60 % of participants reported receiving oral sex from 2 to 5 partners; one-third of participants reported having penile–vaginal intercourse with 4–7 partners, and over 15 % reported having intercourse with more than 10 partners. Nearly 65 % of participants were married or in a committed relationship and approximately 20 % of women in the study reported participating in an uncommitted sexual relationship (e.g., “hooking up” or “friends with benefits”).

To be included in the analyses for the Sexual Intercourse portion of the scale, participants had to endorse being predominantly heterosexual in orientation and having faked an orgasm at least once during sexual intercourse. This resulted in a sample of 352 women who ranged in age from 18 to 31 years (M = 20.35 years, SD = 2.46) and were racially and ethnically diverse (White: 57 %, Black/African-American: 30 %, Other: 13 %; Hispanic: 7 %). A comparable sample of 340, who had endorsed both a predominantly heterosexual orientation and having faked orgasm while receiving oral sex, were included in the analyses for the Oral Sex portion of the scale. Participants ranged in age from 18 to 31 years (M = 20.34 years, SD = 2.35) and were racially and ethnically diverse (White: 55 %, Black/African-American: 32 %, Other: 13 %).

Procedure

Data were collected in small groups of six using self-report questionnaires administered on the computer at laboratory computer stations. Data collection was monitored by a research assistant in a separate room.

Measures

Demographics

Prior to completion of the questionnaires, participants reported their age, ethnicity, predominant sexual orientation, and sexual experience. In addition, participants were asked to choose their most applicable relationship for purposes of completing the questionnaires (i.e., to select a current or past relationship in which they most frequently and/or consistently faked orgasm), indicate the status of that relationship, and respond to the remainder of the measures in the context of that particular relationship.

Faking Orgasm Scale for Women (FOS) Items

Based on previous research in the field (e.g., Darling & Davidson, 1986; Muehlenhard & Shippee, 2010; Wiederman, 1997) and focus group interviews with women in the sample population, a pool of 73 self-report itemsFootnote 1 was generated to examine women’s motivation for faking orgasm during sexual intercourse and oral sex. Each item assessed a singular motivational element and was written in a straightforward manner, using simple language, and avoiding trendy expressions (Clark & Watson, 1995). The items sought to address three major areas of inquiry: (1) beliefs about orgasm and faking orgasm, masturbation, relationship factors, and feelings of responsibility; (2) behaviors and beliefs regarding oral sex; and (3) behaviors and beliefs regarding sexual intercourse.

A pilot study of 117 female undergraduates at a small liberal arts college in Ohio, ranging in age from 18 to 22 (age M = 19.5 years; SD = 1.14 years), completed a paper-and-pencil questionnaire in groups of 20–30 participants as part of an independent study project of the first author (Cooper & Fenigstein, 2004). Participants were predominantly heterosexual (93 %); those indicating homosexual (6 %) or bisexual (1 %) orientation were omitted from analysis, resulting in a total sample of 108. Nearly 76 % of participants had engaged in sexual intercourse; over 83 % reported giving oral sex and approximately 82 % endorsed receiving oral sex; more than 84 % of participants reported having experienced orgasm at some point; and approximately 69 % of participants in this study endorsed masturbating.

Participants responded to open-ended questions and indicated their agreement with the preliminary Faking Orgasm Scale items on a 5-point Likert-type scale, ranging from “Never” to “Always.” Analyses of pilot data revealed adequate variability in the responses to FOS items, and there was no redundancy revealed in the item correlation matrix (based on a criterion of r > .80). Thus, all items were retained for future scale development. Open-ended responses were also content-analyzed with the aim of using the information collected to create and modify self-report items for future scale development. Generally, results of the pilot study suggested women may employ a variety of motives in their decision to fake orgasm, and provided a strong foundation for further scale development [i.e., exploratory factor analysis (EFA)].

Faking Orgasm Scale for Women (FOS)

To expand the breadth of content from the pilot study, 90 new items were constructed based on the results of the pilot study content analysis and additional focus group interviews with women in the sample population. A total pool of 163 self-report items (80 sexual intercourse items and 83 oral sex itemsFootnote 2) was administered to participants. Instructions defined “faking orgasm” as “acting or pretending as if you have had an orgasm when you have not, through vocal confirmation and/or muscular contraction, regardless of the reason,” as these were the major elements identified by participants in the pilot study.

Statistical Analyses

EFA was used to develop a more systematic understanding of the underlying reasons women faked orgasm during oral sex and sexual intercourse. If used properly, EFA is a valid method for reducing the number of items in scale construction, for analysis of factor structures, and for analysis of the validity of the items to be included in the scale (Comrey & Lee, 1992; Reise, Waller, & Comrey, 2000). Factor structure was assessed using parallel analysis (PA) (see Allen, Hayton, & Scarpello, 2004) and examination of the scree plot. All analyses were conducted using the SPSS statistical software package. Assumptions for EFA were tested. Although some items were positively skewed, EFA has been found to be relatively robust against violations of normality (Gorsuch, 1983). In both samples, multivariate outliers were examined and singularity sensitivity analyses revealed no differences in overall factor structures; thus, outlying cases were left in each of the total samples. Furthermore, a visual examination of the correlation matrices suggested that the variances appeared sufficiently similar to proceed with EFA.

EFA was conducted on each of the samples of data collected to reduce the number of items for each scale (80 items of the FOS-Sexual Intercourse and 83 items of the FOS-Oral Sex2). Researchers have identified a minimum sample size of 300 as desirable for EFA, with even fewer cases needed for solutions that have several high loading marker variables (factor loadings >0.80) (Comrey & Lee, 1992; Guadagnoli & Velicer, 1988). Thus, both the Sexual Intercourse and Oral Sex samples were large enough for stable factor structures to be revealed.

Results of the PA and examination of the scree plot guided factor extraction. Principal axis factor (PAF) extraction method was used to determine the number of underlying dimensions or factors. Following the extraction of factors, a Promax oblique rotation was used as the factors were expected to be correlated. Despite allowing for correlation among the factors, using Promax rotation results in maximization of the simple structure by clarifying those variables that do and do not correlate with each factor (Tabachnick & Fidell, 2007).

Results

Exploratory Factor Analysis–Sexual Intercourse Subscale

As there was no a priori hypothesis on the number of factors to be extracted, a number of criteria were used to choose the number of factors to retain. PA suggested an extraction of no more than four factors. After the initial PAF extraction, an examination of the scree plot showed a clear break between the eigenvalues of the fourth and fifth factors. Although eight factors had eigenvalues greater than 1.0, those beyond four were not readily interpretable. The total variance explained by the first four factors was 51.50 %. Nine iterations of removing non-loading, cross-loading, and items loading below <.60 from analyses revealed 35 items and 62.05 % total variance explained. Removal of items from analyses resulted in a sample size of 352 participants without missing data.

The four-factor structure contained 35 of the original 80 items on the Sexual Intercourse Subscale. The first factor was labeled Altruistic Deceit (SIAD) and consisted of 15 items, accounting for 39.3 % of the variance. It contains items that described faking orgasm out of concern for a partner’s feelings and reliability analysis of this factor revealed a coefficient α of .96. The second factor was labeled Fear and Insecurity (SIFI) and consisted of 10 items, accounting for 10.3 % of the variance (α = .93). SIFI contains items that described avoidance of negative emotions. The third factor was labeled Elevated Arousal (SIEA) and consisted of seven items, accounting for 8.42 % of the variance (α = .93). It contains items that described a woman’s attempt to increase her own arousal through faking orgasm. The fourth factor was labeled Sexual Adjournment (SISA) and consisted of three items, accounting for 3.98 % of the variance (α = .76). This factor contains items that described faking orgasm to quickly end sexual intercourse. See Table 1 for descriptive statistics and factor loadings for each factor. Additionally, as hypothesized, the oblique Promax rotation revealed significant correlations among most factors (see Table 2).

Table 1 Descriptive statistics and factor loadings for Faking Orgasm Scale: Sexual Intercourse Subscale
Full size table
Table 2 Correlations among Faking Orgasm Scale: Sexual Intercourse Subscale Factors
Full size table

Exploratory Factor Analysis–Oral Sex Subscale

As there was again no a priori hypothesis regarding the number of factors to be extracted, similar methods as those used to factor analyze the sexual intercourse scale were used. PA suggested the extraction of no more than six factors. After the initial PAF extraction, an examination of the scree plot showed a clear break between the eigenvalues of the third and fourth factors and nine factors had eigenvalues greater than 1.0. However, those factors beyond four were uninterpretable as the items did not make theoretical sense together. The total variance explained by the first four factors was 47.37 %. Seven iterations removing non-loading, cross-loading, and low-loading (<.60) items from analyses revealed 26 items, 62.28 % total variance explained, and a sample size of 331 participants without missing data.

The four-factor structure retained 26 of the original 83 items on the Oral Sex Subscale. The first factor was labeled Altruistic Deceit (OSAD) and consisted of 10 items, accounting for 35.14 % of the variance (α = .94). It contains items that described faking orgasm out of concern for a partner’s feelings. The second factor was labeled Insecure Avoidance (OSIA) and consisted of seven items, accounting for 11.95 % of the variance (α = .89). OSIA contains items that described faking orgasm to avoid negative, anxiety-like emotions during oral sex. The third factor was labeled Elevated Arousal (OSEA) and consisted of five items, accounting for 8.79 % of the variance (α = .91). It contains items that described a woman’s attempt to increase her own arousal through faking orgasm. The fourth factor was labeled Fear of Dysfunction (OSFD) and consisted of 4 items, accounting for 6.38 % of the variance (α = .86). This factor contains items that described faking orgasm to prevent negative emotions associated with sexual health or inadequate sexual response. See Table 3 for descriptive statistics and factor loadings for each factor. Additionally, as hypothesized, the oblique Promax rotation revealed significant correlations between factors on the Oral Sex Subscale (see Table 4).

Table 3 Descriptive statistics and factor loadings for Faking Orgasm Scale: Oral Sex Subscale
Full size table
Table 4 Correlations among Faking Orgasm Scale: Oral Sex Subscale Factors
Full size table

Discussion

The Faking Orgasm Scale was designed to assess women’s self-reported motives for faking orgasm during sexual intercourse and oral sex. Results of the EFA revealed a four-factor scale during both sexual intercourse and oral sex. The Sexual Intercourse Subscale included Altruistic Deceit, Fear and Insecurity, Elevated Arousal, and Sexual Adjournment factors. The Oral Sex Subscale consisted of Altruistic Deceit, Insecure Avoidance, Elevated Arousal, and Fear of Dysfunction. Participant responses in previous studies utilizing interview and/or coded open-ended response items support the interpretation of these factors (Darling & Davidson, 1986; Hite, 1976; Muehlenhard & Shippee, 2010; Wiederman, 1997).

Each factor was labeled and a description of the constructs they were hypothesized to measure was created based on a face-valid evaluation of the items. Altruistic Deceit is a partner or relationship-centric construct, concerned with faking orgasm to avoid hurting a partner’s feelings and/or making a partner feel good about himself. Elevated Arousal, however, was self-focused, defined as faking orgasm to increase one’s own arousal, excitement or interest in the sexual act. Fear and Insecurity was also a primarily self-focused construct representing faking orgasm to avoid negative emotions and/or self-evaluation. Sexual Adjournment was a factor reflecting a lack of relationship-oriented communication and concerns faking orgasm to bring an end to the sexual act. Insecure Avoidance was also a primarily self-focused factor. Each of the factors for both the Sexual Intercourse and Oral Sex Subscales demonstrated excellent internal consistency.

It is interesting that, in the context of oral sex, Fear and Insecurity separated into seemingly emotion-focused and physiologic-focused factors. Insecure Avoidance taps into women’s desire to avoid anxiety-like responses to a partner’s focus on her genitals. In turn, Fear of Dysfunction appears to represent women who fake orgasm to appear “normal” in their sexual functioning. Of further interest was the finding that the Sexual Adjournment factor was not present in the Oral Sex Subscale. Previous research has characterized faking orgasm behavior as a woman’s method of adhering to sexual scripts (Muehlenhard & Shippee, 2010). Sexual scripts may also explain the lack of a Sexual Adjournment factor or faking orgasm to end the sexual experience during oral sex. In many heterosexual relationships, oral sex may be viewed as a precursor to sexual intercourse; thus, women may not employ faking orgasm to end the sexual encounter because they may be looking to intercourse for additional pleasure and/or achieving orgasm. Alternatively, oral sex may last for a shorter duration than sexual intercourse and thus women may not reach frustration with their oral sex experience and desire for it to end in the same manner.

Study 2

Following initial factor extraction and data reduction, confirmatory factor analysis (CFA) was used to evaluate decisions made during EFA and confirm hypothesized factor structures (for review, see Finch & West, 1997; Floyd & Widaman, 1995; Reise et al., 2000). CFA procedures tend to work best with simple factor structures (each item loads highly on one and only one factor) (Reise et al., 2000), such as that found in Study 1. It was hypothesized that the factor structure of each subscale found during EFA would be supported in a new sample of participants. Further, it was expected that the CFA would reveal good to excellent model fit indices of the factor structure for each subscale of the FOS revealed in Study 1 (Sexual Intercourse: 35 items; Oral Sex: 26 items).

Method

Participants

Data for CFA of the Faking Orgasm Scale were collected from undergraduate females enrolled in introductory psychology and marketing courses at a large urban university in southeastern Pennsylvania. A total of 398 heterosexual women completed the study. Participants ranged in age from 18 to 32 years (M = 20.52, SD = 2.55) and were racially and ethnically diverse (White: 68 %, Black/African-American: 23 %, Other: 14 %; Hispanic: 7 %). Of the total sample, 94.2 % endorsed engaging in sexual intercourse, 95.5 % reported receiving oral sex, and 87.4 % reported ever having experienced an orgasm. Approximately 60 % of participants reported receiving oral sex from 2 to 5 partners, over 27 % of participants reported having penile–vaginal intercourse with 4–7 partners, and almost 15 % reported having intercourse with more than 10 partners. More than 65 % of participants were married or in a committed relationship and approximately 15 % of women reported participating in an uncommitted sexual relationship (e.g., “hooking up” or “friends with benefits”).

To be included in the sample for the Sexual Intercourse CFA, participants had to endorse a predominantly heterosexual orientation and having faked an orgasm during sexual intercourse. The sample for analyses of the Sexual Intercourse Subscale was 312 students. Participants ranged in age from 18 to 32 years (M = 20.47, SD = 2.52), and were racially and ethnically diverse (White: 70 %, Black/African-American: 22 %, Other: 13 %; Hispanic: 6 %).

Participants endorsing a predominantly heterosexual orientation and faking orgasm while receiving oral sex were included in the sample for the Oral Sex CFA. Data for analyses of the Oral Sex Subscale came from 335 students; participants ranged in age from 18 to 32 years (M = 20.52, SD = 2.57) and were racially and ethnically diverse (White: 69 %, Black/African-American: 23 %, Other: 14 %; Hispanic: 6 %).

Procedure

Participants were recruited through a department-approved web-based research management system and through Institutional Review Board approved flyers posted outside classrooms and inside women’s restrooms. Sampling procedures and inclusion criteria were identical to those detailed in Study 1.

Measures

Demographics

Participants completed the same demographics information as in Study 1.

Faking Orgasm Scale for Women (FOS)

Following data reduction of the EFA, the FOS consisted of 61 self-report items (Sexual Intercourse Subscale: 35 items; Oral Sex Subscale: 26 items) designed to examine women’s motivation for faking orgasm during sexual intercourse and oral sex. All other aspects of the FOS, including instructions and basic organization, remained the same.

Analyses

CFA was conducted using the EQS statistical software package (Bentler, 2008). Performing CFA using EQS offers goodness of fit statistics between the hypothesized factor structure and the factor structure within the data, allowing for exploration of error-free constructs. Data were analyzed using Maximum Likelihood (ML) estimation. The primary method for evaluation of goodness-of-fit is the χ 2 test statistic, which is dependent on sample size. As sample sizes increase (i.e., N > 200), χ 2 may falsely detect small divergences as significant (Bentler & Bonnet, 1980). Therefore, there was little reason to rely solely on this measure of goodness of fit so other fit indices were used to determine an acceptable model. Comparative Fit Index (CFI) and the Root-Mean Square Error of Approximation (RMSEA) were used as additional measures of goodness-of-fit. The Akaike Information Criterion (AIC) was used to determine relative goodness of fit of the model following modifications to model parameters (Kline, 1998).

Results

Confirmatory Factor Analysis–Sexual Intercourse Subscale

Initial screening indicated that the data were suitably normally distributed and thus robust corrections were not necessary. All measured variables loaded significantly (p < .01) onto their hypothesized latent variables. The CFA model had excellent fit statistics, Satorra–Bentler χ 2(553, N = 312) = 1008.29, p < .001, CFI = .94, RMSEA = .051 (90 % CI = .046–.056).

To reduce number of items and improve model fit, one item (“so your partner doesn’t feel inadequate if you don’t have a real orgasm?”) was removed from analyses. Excellent fit statistics were achieved and improved slightly, Satorra–Bentler χ 2(521, N = 312) = 887.51, p < .001, CFI = .95, RMSEA = .048 (90 % CI = .042–.053). A lower AIC value indicated the modification resulted in the best fitting model. Table 5 shows descriptive statistics and factor loadings for the final version of the Sexual Intercourse Subscale. Correlations between factors are shown in Table 6.

Table 5 Descriptive statistics and factor loadings for Faking Orgasm Scale: Sexual Intercourse Subscale Confirmatory Factor Analysis
Full size table
Table 6 Correlations among Faking Orgasm Scale: Sexual Intercourse Subscale Factors
Full size table

Confirmatory Factor Analysis–Oral Sex Subscale

Initial data screening indicated that the data were suitably normally distributed and thus robust corrections were not necessary. Table 7 shows the factor loadings for the hypothesized latent factors and the means and standard deviations of all measured variables in the CFA Oral Sex model. All measure variables loaded significantly (p < .01) onto their hypothesized latent variables. Correlations between factors are shown in Table 8. However, the CFA model had unacceptable fit statistics, Satorra–Bentler χ 2(293, N = 398) = 780.80, p < .001, CFI = .91, RMSEA = .072 (90 % CI = .065–.078).

Table 7 Descriptive statistics and factor loadings for final Faking Orgasm Scale: Oral Sex Subscale Confirmatory Factor Analysis
Full size table
Table 8 Correlations among Faking Orgasm Scale: Oral Sex Subscale Factors
Full size table

To improve model fit, modifications to the model were made based on factor interpretability and avoidance of redundancy. One item (“because you think there may be something physically wrong with you if you don’t orgasm?”) was removed from analyses; however, this adjustment did not result in adequate fit, Satorra–Bentler χ 2(269, N = 328) = 703.59, p < .001, CFI = .92, RMSEA = .070 (90 % CI = .064–.077). Two items (“because you are afraid your partner is not enjoying it?” and “to reinforce certain behaviors/techniques/actions of your partner?”) were removed from analyses, but did not result in adequate fit, Satorra–Bentler χ 2(224, N = 331) = 484.15, p < .001, CFI = .94, RMSEA = .059 (90 % CI = .054–.069). One additional item (“because you feel guilty your partner is giving you oral sex?”) was subsequently removed from analyses and adequate fit statistics were achieved, Satorra–Bentler χ 2(203, N = 335) = 432.00, p < .001, CFI = .95, RMSEA = .058 (90 % CI = .050–.066). A reduction in AIC value indicated this was the best fitting model. Table 7 shows descriptive statistics and factor loadings for the final version of the Oral Sex Subscale. Correlations between factors are shown in Table 8.

General Discussion

The present research attempted to develop a reliable and valid measure, the FOS, to assess the motivations behind a woman’s decision to fake an orgasm. The FOS is the first scale of its kind. CFA replicated its factor structure with excellent model fit and minimal modifications to each subscale.

The current studies indicated that the decision to fake orgasm during sexual intercourse and oral sex was often motivated by important psychological goals or functions. In contrast to previous arguments that faking orgasm is disingenuous or potentially problematic to individual and relationship health (e.g., Darling & Davidson, 1986; Jagose, 2010) or—as some feminist theorists have argued—that faking orgasm is an act that reflects women’s submission to the androcentric societal and cultural expectations placed upon them and their sexuality (e.g., Fahs, 2010; Jackson & Scott, 2002; Potts, 2000), the present research suggested that some motives for faking orgasm can be conceptualized as relationship-promoting and sexual pleasure-enhancing and that a broader and less negativistic view of this practice is in order.

Factor structure differences between the Sexual Intercourse and Oral Sex Subscales highlighted the need for further investigation into the potential psychological mechanisms operating during varying sexual contexts and the implications of these mechanisms. The research suggested that women may fake orgasm to alleviate increased feelings of anxiety, self-consciousness, and physiological abnormality when their partner’s focus was explicitly centered on their genitals. Societal expectations may also help to explain this increased awareness of one’s sexual responding during oral sex, as both the media and research community have asserted that women are more likely to experience an orgasm following clitoral stimulation (e.g., Masters & Johnson, 1966). This increased attention to the clitoris may have resulted in increased pressure on women to perform when stimulated.

There were several limitations to these studies. First, these studies were based on a convenience sample of mostly single heterosexual female college students, which may limit the generalizability of the findings. It is unclear to what extent women in long-term committed relationships fake orgasm and, if they do, what motives determine their decision to do so. Previous studies have found older single women to be more likely to fake orgasm than their younger married counterparts (Darling & Davidson, 1986); further research in women across all age groups is necessary to explore the stability of the FOS’s factor structure and potential developmental factors. It is also unknown what, if any, differences exist between heterosexual and lesbian or bisexual women’s prevalence of and motives for faking orgasm. Preliminary studies suggest that lesbian and bisexual women may fake orgasm more frequently than heterosexual women; however, differences in motives have yet to be explored (Cooper, Conner, & Fauber, 2010). Additional research is needed to explore the stability of the factor structure in lesbian, bisexual, and/or older populations.

It is likely that motives for faking orgasm are influenced by sociocultural circumstances and individual differences not considered in the current study. Future studies might include individuals from diverse backgrounds, which may provide important information about the generalizability of the current findings. Recent research has indicated that mental health symptoms differentially impact the prevalence of faking orgasm across racial and ethnic groups: depressive symptoms in men and women of European or African-American descent predicted the likelihood they had faked orgasm, but no such relationship was found for Asian-American men or women (Cooper, Morrison, & Heimberg, 2012). Studies exploring potential differences in the frequency of, or motives for, faking orgasm among racial and ethnic groups could provide a better understanding of cultural influences on this behavior.

Another possible limitation of the current study was the self-report method used, as some concern has been raised regarding the reliability of self-reported sexual behavior because of intentional distortion, inaccurate recall, and estimation of behavior. Studies utilizing multiple methods have strengthened confidence in self-report as an acceptable method for collecting data regarding sexual behaviors that occur naturally outside of a laboratory setting (McFarlane & St. Lawrence, 1999). Further, individuals may not be able to accurately identify the reasons they fake orgasm in a given context. Although steps were taken to reduce this error (e.g., asking participants to specify and answer items regarding a particular sexual partner), the potential for inaccurate reporting remains. Some participants may have difficulty reporting on their motivational drives; however, it is likely that given the conscious nature of the decision to fake orgasm, most women will be able to accurately report on their motives when choosing to fake orgasm during oral sex or sexual intercourse. It may be that women who fake orgasm frequently will have greater insight into why they fake orgasm, as it is a choice they are thinking about and making often. Alternatively, women who engage in this behavior may habitually fake orgasm and thus lack insight into their motives for doing so. In general, objective verification of retrospective self-reported behavior is problematic for sex research due to logistical and ethical limitations of observing sexual behavior.

Furthermore, validity of the FOS must be established by examining the relationship between the FOS factors and other constructs related to sexual behaviors and attitudes. Test–retest studies are also necessary to explore the temporal stability of faking orgasm. As previous research has focused exclusively on the prevalence rather than frequency of this behavior, details regarding the occurrence of faking orgasm is lacking. Such research may clarify the clinical implications of faking orgasm and inform the need for potential interventions. There remains much controversy in the field around how to best conceptualize and understand normal female sexual functioning (for review, see Graham, 2010); given the high prevalence rates of faking orgasm and orgasmic difficulties reported in previous studies (Bancroft et al., 2003; Darling & Davidson, 1986; Kaighobadi et al., 2012; Lloyd, 2005; Muehlenhard & Shippee, 2010; Wiederman, 1997), it stands to reason that faking orgasm may play a role in or be an adaptive response to sexual dysfunction. Little is known about the level of distress associated with women’s decision to fake orgasm, regardless of motive for doing so, as this has yet to be studied. In clinical practice, research has shown that only 14–17 % of physicians explicitly inquire about sexual functioning and behavior with their female patients (Nusbaum, Helton, & Ray, 2004), despite calls that treatment providers ascertain orgasmic capacity while taking routine sexual histories (Sinha & Palep-Singh, 2008). Without data to indicate that faking orgasm is not problematic, further exploration into these potential relationships is warranted and may provide new insights to the sexual difficulties of women.

The data in these studies indicated that women employed both self-focused and relationship-oriented motives when faking orgasm, suggesting that, in addition to both cognitive and affective factors, contextual influences play an important role in a woman’s determination to fake orgasm. The multidimensional construct of motivation for faking orgasm provides a basis for future research on relationship satisfaction, sexual desire, arousal, sexual satisfaction, and sexual functioning. Development of the FOS has permitted identification of theoretically important motivational constructs that are likely to influence a variety of sexual behaviors.

In addition to continued validation of the FOS, future research should explore motives for faking orgasm in men. Previous research has indicated that approximately 25 % of heterosexual men fake orgasm during sexual intercourse (Muehlenhard & Shippee, 2010) and no studies have explicitly explored this phenomenon among gay men. Faking orgasm is an understudied aspect of women’s sexual behavior and seems to have been overlooked almost entirely as a male sexual practice.

These findings indicate that the decision to fake orgasm must be conceptualized outside the narrow assumptions of sexual dissatisfaction and deceit; future research must focus on broadening our understanding of these dispositional differences and the manner in which each type of motivation uniquely influences the course and outcome of naturally occurring sexual behavior and partner interactions. The FOS is the first assessment tool designed to explore why women fake orgasm and should facilitate further research into this phenomenon and may help examine the clinical implications of this behavior as it pertains to the individual and relationships. Such information may ultimately lead to the development of clinical interventions to improve the quality of sexual experiences, as well as sexual and romantic relationships. Results of the current study underscore the need for continued research into this newly identified construct and continued work on establishing the psychometric properties of the FOS.