Introduction

The number of households in the U.S. led by female same-genderFootnote 1 (FSG) couples is estimated to be over 405,000 and continues to grow (American Community Survey, 2014). However, the vast majority of research on romantic relationships has primarily focused on mixed-gender couples (i.e., couples comprised of a woman and a man; Hartwell, Serovich, Grafsky, & Kerr, 2012). Scholars have recently called for empirical investigations of same-gender couples, with an emphasis on developing guidelines for practitioners and creating interventions to support healthy relationship functioning (American Psychological Association, 2011; Markman & Rhoades, 2012).

Across female same-gender and mixed-gender couples, research has established strong bidirectional connections between sexual satisfaction and perceptions of relationship quality (Butzer & Campbell, 2008; Byers, 2005; Christopher & Sprecher, 2000; Henderson, Lehavot, & Simoni, 2009; Holmberg, Blair, & Phillips, 2010). Consequently, relationship interventions often seek to enhance sexual satisfaction within the larger goals of promoting healthy and satisfying relationships. Further, sexual minority women have expressed a desire for relationship services that explicitly address issues related to sex and sensuality in FSG relationships (Scott & Rhoades, 2014). Research has been particularly scarce, however, regarding sexuality within these relationships. Although several similarities have been noted between the sexual behaviors and sexual communication of same- and mixed-gender couples (Holmberg & Blair, 2009), clinicians would be mistaken to assume that sexuality within FSG relationships fits within heteronormative frameworks. For example, sexual encounters in FSG couples are unlikely to center around penile–vaginal penetration, and sexual scripts and desires may also differ for women in FSG relationships (Cohen & Byers, 2014). Limited research has sought to clarify the nature of sexuality within these relationships as well as factors associated with sexual satisfaction in FSG couples. The current study sought to expand the literature by providing descriptive information regarding sexual practices and conceptualizations of sex in FSG relationships and to investigate factors associated with sexual satisfaction in these relationships.

Defining Having Sex and Sexual Norms in Female Same-Gender Couples

A key limitation of much research on the sexual practices of FSG couples is the lack of consistency in defining how sexual minority women conceptualize having sex, which in turn limits the inferences drawn from the literature about factors associated with sexual satisfaction. Indeed, some comparative studies across men and women in same- and mixed-gender relationships have compared any or vaguely defined aspects of sexual activity between female partners to explicitly defined sexual penile–vaginal intercourse, anal penetration, or oral sex between men and women or men and men (e.g., McCabe, Brewster, & Harker Tillman, 2011). Recent studies on conceptualizations of sex have generally found that women in romantic relationships with women are more likely to have broader conceptualizations of sex compared to men in same-gender relationships and compared to both men and women in mixed-gender relationships. For example, female same-gender couples are more likely to consider oral sex, the use of sex toys, or manual stimulation to constitute having sex compared to male same-gender or mixed-gender couples (Horowitz & Spicer, 2013; Sewell, McGarrity, & Strassberg, 2017).

Scholars have also argued that heteronormative biases surrounding sexual practices may make asking about frequency or definitions of having sex phallocentric in nature, possibly influencing how sexual minority women respond to study questionnaires that ask about definitions of sex more generally (Cohen & Byers, 2014). Interestingly, when bisexual women have been asked about sexual definitions, they are more likely to endorse a broader array of sexual acts as constituting “having sex” when considering their sexual encounters with women compared to their sexual encounters with men; thus, the gender composition of partners engaged in sexual acts may be important to consider when asking questions about sexual definitions (Schick et al., 2016). It may therefore be useful to explicitly ask women in FSG relationships how they define having sex within the context of their relationships. Further, no study, to our knowledge, has evaluated how likely women in FSG relationships are to agree with their partners on their conceptualizations of having sex, which is an important consideration when examining dyadic sexual satisfaction. Finally, research is limited regarding evaluations of how often women in FSG relationships engage in specific sexual acts. This information could be clinically useful by helping clinicians better understand normative rates of sexual activity within these relationships. One of the aims of this study was to address these gaps in the literature by explicitly assessing how women in same-gender relationships define having sex within their relationships and to provide descriptive data regarding individual reports of frequency of engagement in specific sexual acts. We used a new measure, the Sexual Activities Questionnaire, to address these questions.

Factors Associated with Sexual Satisfaction

Research has generally shown that numerous relationship and sexual processes are associated with perceptions of sexual satisfaction, although women and men may value particular aspects differently (De la Garza-Mercer, 2006; Laan & Both, 2008; Montesi, Fauber, Gordon, & Heimberg, 2011; Peplau, 2003). For example, within mixed-gender couples, women have been shown to value the context and emotional quality of their relationships more strongly than men, and report placing more emphasis on the emotional quality of their sexual encounters as opposed to the frequency (Peplau, 2003). Therefore, it is plausible that women in FSG relationships may place particular value on the emotional quality of their sexual encounters, relative to the frequency.

In support of this possibility, Blair and Pukall (2014) found that mean levels of sexual satisfaction across same- and mixed-gender couples were comparable, despite replicating findings that women in same-gender relationships reported having sex less frequently compared to mixed-gender or male same-gender couples (Solomon, Rothblum, & Balsam, 2005). Further, the average length of time that sexual encounters last is longest for FSG couples compared to mixed-gender or male same-gender couples (Blair & Pukall, 2014; Breyer et al., 2010; Cohen, Byers, & Walsh, 2008). Although the authors did not directly measure the quality of the sexual encounters across couple types, their findings may suggest that women in FSG relationships derive sexual satisfaction to a greater degree from the quality or intensity of sexual encounters, while frequency may be more relevant for other couple types.

At the same time, sexual frequency for sexual minority women may still hold importance regarding sexual satisfaction. Scholars have also argued that as societal scripts regarding sexuality continue to change, women may feel more empowered to express sexual desires and initiate sexual encounters compared to several decades ago (Cohen & Byers, 2014). This may be particularly relevant in FSG couples, where women may feel less constrained by social gender roles. Further, some researchers have shown that, on average, women in FSG relationships would like to increase the frequency of sex within their relationships, suggesting that frequency is an important within-group factor to investigate (James & Murphy, 1998; Spitalnick & McNair, 2005). Most recently, scholars have found that the frequency of genital touching, frequency of orgasm, and the strength of desire for sex were each associated with perceptions of sexual satisfaction for women in FSG relationships (Cohen & Byers, 2014; Tracy & Junginger, 2007).

Some scholars have argued that higher levels of emotional intimacy—defined as perceptions of emotional closeness and safety within one’s relationship—in FSG relationships may reduce sexual attraction and desire. The terms merger or fusion have been used to refer to extreme levels of intimacy and co-dependency in which individuality between partners may become blurred (Burch, 1986; Ossana, 2000; Spitalnick & McNair, 2005). This phenomenon has been speculated to occur more often in lesbian couples and to lead to lower desire between partners. In line with this idea, one study demonstrated that environmental mastery—defined as the capacity of individuals to utilize the environment to meet one’s needs—is more important in lesbian relationships compared to gay male relationships (Biss & Horne, 2005). The authors speculated that this finding may be due to environmental mastery leading to increased self-assertiveness, which may be useful in female same-gender couples given the higher rates of closeness and cohesion (Biss & Horne, 2005).

At the same time, scholars have also acknowledged the heterosexist bias present in deeming higher intimacy in FSG relationships as inherently problematic and have demonstrated that high levels of intimacy do not predict lower relationship quality (Ackbar & Senn, 2010). However, research assessing the association between emotional intimacy and sexual satisfaction within FSG relationships has been sparse, with one recent study demonstrating that emotional intimacy and sexual satisfaction are positively associated across sexual orientations for individuals in committed relationships (Mark, Garcia, & Fisher, 2015) Another study suggested that having sex for the purpose of improving intimacy is beneficial to sexual satisfaction across women in both mixed- and same-gender relationships (Sanchez, Moss-Racusin, Phelan, & Crocker, 2011) and qualitative research has similarly demonstrated that women in same-gender relationships find sex rewarding because it enhances both physical and emotional intimacy (Cohen et al., 2008). Given that sexual intimacy—referring to how close, comfortable, and connected individuals feel during or after sexual encounters—has been shown to have stronger associations with sexual satisfaction in FSG couples compared to mixed-gender couples (Birnie‐Porter & Lydon, 2013), it may be useful to assess how emotional intimacy more broadly compared to sexual intimacy specifically is associated with sexual satisfaction in FSG relationships.

The state of the current literature suggests that the field would benefit from a more thorough and precise understanding of how aspects of the sexual relationship are associated with sexual satisfaction. Thus, a second aim of the current study was to address how general emotional intimacy, emotional intimacy regarding sex, physical pleasure (e.g., orgasm), desired frequency of sex, and actual sexual frequency are associated with sexual satisfaction in FSG couples. Studies that have evaluated correlates of sexual satisfaction in sexual minority women have generally evaluated these associations at the individual level (Cohen & Byers, 2014; Tracy & Junginger, 2007). Given that sexual satisfaction happens within a relational context in FSG couples and that women may be particularly attuned to and especially value the interpersonal quality of their relationships (Peplau, 2003), it is important to account for the dyadic nature of sexual satisfaction within FSG relationships. Actor–partner interdependence models (APIMs; Kenny, Kashy, & Cook, 2006) are equipped to provide effect estimates regarding how both one’s own functioning in a particular relationship factor (actor effects) as well as one’s partner’s functioning (partner effects) are associated with one’s own experience on given outcomes (e.g., one’s own sexual satisfaction). Thus, the current study utilized APIMs to assess the dyadic nature of sexual satisfaction related to the second aim.

Hypotheses

No hypotheses were made a priori regarding the exploratory, descriptive goals of the study (Aim 1), including definitions of sex, agreement across partners, and frequency of sex. Regarding models associated with sexual satisfaction (Aim 2), it was predicted that significant, positive associations would be found for actor and partner effects of actual sexual frequency, desired sexual frequency, frequency of orgasm, emotional intimacy, and sexual intimacy.

Method

Participants

Participants included 103 FSG couples (N = 206 individuals). Eligibility criteria included that both partners were over 18 years of age, identified their genders as female, were English-speaking and capable of completing the study questionnaires, and that the couples had been in their romantic relationships for a minimum of 2 months. Participants who failed to meet any of these criteria were excluded from final analyses. Participants were recruited through advertisements in a lesbian-focused newsletter, Web site, and Facebook advertisements, and recruitment tables and flyers at lesbian-friendly events.

Measures

Sexual Activities Questionnaire

The Sexual Activities Questionnaire was a new measure developed for this study to measure conceptualizations and frequency of specific sexual activities, as well as overall actual frequency of sex, desired frequency of sex, and frequency of orgasm in FSG couples.

Sex Definitions

Frequency of sexual activities was measured by 14 items that assess a variety of sexual activities ranging from cuddling and kissing to oral sex, genital-to-genital touching, and the use of sex toys. Most questions also specified whether the participant pleased her partner or was pleased by her partner. Participants were then asked to indicate which of these sexual activities they considered to constitute “having sex” within FSG relationships (yes or no; Table 2).

Sexual Frequencies

Following the list of specific behaviors, a single item assessed actual frequency of sex with their partner: “Considering all of your sexual activities with your partner, about how often do you and your partner have sex?” Participants were also asked about the frequency of having an orgasm in the presence of their partner. Desired frequency of sex was measured by the question “How often would you ideally like to have sex with your partner?” All frequency items used the following scale: 1 = never, 2 = more than 6 months ago, 3 = every other month, 4 = once a month, 5 = every other week, 6 = once a week, 7 = more than once a week, 8 = once a day, 9 = more than once a day.

Sexual Satisfaction

Overall sexual satisfaction was assessed with a single item stating “We have a satisfying sensual or sexual relationship” with a 7-point Likert scale ranging from 1 (strongly disagree) to 7 (strongly agree). The average score on this measure was 5.5 (SD = 1.7).

Intimacy

The Intimate Safety Questionnaire (ISQ; Cordova, Gee, & Warren, 2005) is a 13-item measure designed to assess each partner’s emotional intimacy within her relationship as well as intimacy associated with sexual encounters specifically. Items were assessed on a 5-point Likert scale ranging from 1 (never) to 5 (always). The full scale has demonstrated high reliability and validity in other studies (e.g., Cordova et al., 2005). The 8-item emotional intimacy subscale demonstrated adequate reliability (α = .75). Example items from this subscale include, “I feel comfortable telling my partner things I would not tell anyone else,” and “When I need to cry I go to my partner.” The 5-item sexual intimacy subscale demonstrated low reliability (α = .54) but was improved to adequate reliability with the removal of the item “I feel uncomfortable talking to my partner about our sexual relationship,” (reverse coded; final α = .70). Example items from this subscale include “I feel close to my partner during and after lovemaking,” and “I feel comfortable telling my partner my likes and dislikes while we are making love.”

Procedure

After contacting the lead PI and confirming eligibility criteria, 107 couples completed research sessions at a university in which both partners completed surveys and videotaped communication tasks (not used in the current study). After providing written consent, participants were separated from their partners in order to complete their questionnaires in private. Each individual was provided $25 for their participation. Data from four couples were eventually removed from analysis due to the couples disclosing that they did not meet an eligibility criterion during their research sessions (two couples were together less than 2 months; one couple had a partner who identified as a transgender man; one couple self-reported as having cognitive disabilities that prevented them from completing the protocol questionnaires). All research procedures were approved by a university Institutional Review Board.

Data Analytic Plan

We used guidelines developed by Kenny et al. (2006) for dyadic data analysis. These guidelines recommend using multilevel modeling (MLM) to take into account the nested nature of the data in which individuals (level 1) are nested within couples (level 2). Because the couples in this study were composed of same-gender partners, partners were randomly assigned as either Partner 1 or Partner 2 and the data were treated as indistinguishable dyads. Additionally, compound symmetry was utilized to force “the degree of unexplained variance for the dyad members to be equal” (Kenny et al., 2006, p. 91). APIMs for indistinguishable dyads therefore consist of 5 general parameters including (1) the intercept, (2) fixed actor effects, (3) fixed partner effects, (4) sum of the intercept and residual variance (e ij ), and (5) correlation between the error terms between partners, representing the ICC of the outcome after the predictors are taken into account (Cov(e 1 , e 2); Kenny et al., 2006). For all analyses, results are presented with standardized estimates.

Control Variables

Research has demonstrated that age, relationship length, and cohabitation status are associated with sexual satisfaction (Cohen & Byers, 2014; Rhoades, Stanley, & Markman, 2012; Scott et al., 2017; Tracy & Junginger, 2007); therefore, all models included these as control variables.

Hierarchical Approach

Our analyses used a hierarchical approach designed to test the additional predictive power of actor and partner effects separately. Specifically, Model 1 serves as the baseline model and includes only the control variables. Model 2 adds actor effects for all predictor variables, and finally, Model 3 adds partner effects for predictors.

Results

Sample Characteristics

The majority of participants identified their sexual orientation as lesbian (81.1%), followed by 11.2% selecting “other” (most common write-in response was “queer”), and 7.8% identifying as bisexual. Participant characteristics included an average age of 33.7 years (SD = 9.0), median income range of $30,000–39,999, and median educational level of 16 years (approximately 4 years of college). The average relationship length was 47 months (3.9 years) with a median relationship length of 26 months (2.2 years), indicating that relationship length was positively skewed in this sample due to some couples who had been together for a long time. Seventy-eight percent of the sample was currently cohabiting and 23% had a least one child living in the home. Twenty-three percent of couples endorsed having participated in some form of commitment ceremony (legal marriage, civil union, or commitment ceremony).Footnote 2 Race and ethnicity demographics were as follows: 73.3% Caucasian/European American, 13.6% Hispanic/Latina, 5.8% African-American/Black, 2.4% Native American, 2.4% Asian/Pacific Islander, and 2.4% Multiracial.

Preliminary Results

Correlations Between Outcome Variables

Given the non-independence of the data, basic bivariate correlations were run separately for Partner 1 and Partner 2 (see Table 1). Correlations for Partner 1 are presented above the diagonal and correlations for Partner 2 below the diagonal. Within the text, we present the mean correlation across Partner 1 and Partner 2. As expected, sexual frequency, desired sexual frequency, frequency of orgasm, emotional intimacy, and sexual intimacy were all positively associated with sexual satisfaction. The majority of these factors were also significantly and positively associated with one another, including that higher desired sexual frequency was positively associated with higher actual sexual frequency.

Table 1 Correlations between sexual satisfaction, sexual frequency, desired sexual frequency, frequency of orgasm, emotional intimacy, and sexual intimacy
Full size table

Descriptive Statistics Regarding Sexual Practices

Conceptualization of Sexual Acts

Percentages of participants reporting that they consider the following sexual acts as “having sex” were: 97% hand-to-genital stimulation, 96% oral sex, 96% using sex toys, 94% genital-to-genital contact, 87% anal stimulation, 62% joint masturbation, and 5% French kissing/making out (see Table 2). Thus, results suggest that the vast majority (>85%) of participants considered any form of genital touching from one partner to the other to constitute having sex.

Table 2 Sexual activities considered to be having sex, agreement across partners, ever engaged in by participants, and descriptive scores of sexuality frequency measures
Full size table

Discrepancies in Conceptualizations of Sex Between Partners

Forty-three percent of couples reported perfect (100%) agreement across partners regarding conceptualizations of all seven sexual acts. The largest agreement was on hand-to-genital touching, in which 93% of partners agreed, and the largest disagreement was reported on joint masturbation, in which 61% of partners agreed. The mean number of acts disagreed upon by partners was 1.0 (SD = 1.1) out of 7 physical acts. Table 2 shows the average percent of agreement and disagreement per sexual act within couples.

Frequency of Physical and Sexual Behaviors

Regarding engagement in various sexual acts, the least common sexual act that couples reported ever engaging in was anal stimulation or penetration (29%). More participants reported ever engaging in joint masturbation (63%), using sex toys (77%), or genital-to-genital touching (81%), and the vast majority of couples (>93%) reported engaging in oral sex or hand-to-genital touching with their current partner at some point in their relationship (Table 2). Participants reported engaging most frequently in physical acts involving non-genital touching, including kissing on the lips (Mdn = 9; more than once a day), cuddling (M = 9; more than once a day), and making out/French kissing (Mdn = 7; more than once a week).

The most frequent sexual acts involving genital touching were hand-to-genital stimulation/penetration (Mdn = 6; once a week), followed by oral sex (Mdn = 5; every other week), genital-to-genital touching (Mdn = 4; once a month), and using sex toys (M = 3; every other month). Anal stimulation/penetration (M = 1; never) was the sexual act least frequently engaged in by participants, followed by joint masturbation (M = 2; more than six months ago).

When asked how often participants had sex with their partner, the median frequency was 6 representing once per week, while the average desired sex frequency was higher with a median score of 7, representing more than once per week. Participants also reported a median frequency of 6 in terms of orgasming in the presence of their partner, indicating that participant reports of frequency of orgasm were similar to their frequency reports of having sex, at approximately once a week. Overall, 69% of participants reported a higher desired frequency of sex compared to their reports of actual sex frequency.

Factors Associated with Sexual Satisfaction

Sexual Satisfaction

MLM was utilized to investigate how various relationship and sexual factors were related to overall sexual satisfaction and examined the additive effects of each partner’s predictors in the model. Model 1 explored baseline associations between control variables and sexual satisfaction; Model 1 included actor age modeled at level 1 and the couple’s relationship length and cohabitation status modeled at level 2. Results showed that relationship length was negatively associated with sexual satisfaction (β = −.25; SE = .10; p = .017), suggesting that couples who were together longer reported lower sexual satisfaction. Actor age (β = .07; SE = .09; p = .42) and the couple’s cohabitation status (β = −.11; SE = .09; p = .24) were not significantly associated with sexual satisfaction.

Model 2 added the actor predictor variables of desired sexual frequency, emotional intimacy, sexual intimacy, and sexual frequency to the baseline model described above. All actor predictor variables were modeled simultaneously at level 1. Sexual frequency and frequency of orgasm were highly correlated (r = .91), introducing problems with multicollinearity; thus, only sexual frequency was included in the model and frequency of orgasm was excluded. Additionally, the intra-class correlation (ICC) between partner reports of sexual frequency was also very high (r = .87). To address this, we took the average report of sexual frequency between partners within each couple and included this score as a level-2 predictor.

Results from Model 2 indicated that all four actor effects, including desired sexual frequency, emotional intimacy, sexual intimacy, and sexual frequency were significantly and uniquely associated with sexual satisfaction (Table 3). Specifically, emotional intimacy (β = .13; p = .014), sexual intimacy (β = .18; p = .002), and sexual frequency (β = .80; p < .001) were all positively associated with sexual satisfaction. Unexpectedly, desired sexual frequency was negatively associated with sexual satisfaction (β = −.18; p < .001), indicating that after accounting for the other variables, higher desired sexual frequency was associated with lower sexual satisfaction. However, bivariate correlations demonstrated that desired sexual frequency and sexual satisfaction were positively correlated (r = .67), as well as desired frequency and actual sexual frequency (r = .70). Thus, individuals with higher sexual desire were generally engaging in more sexual activity and were also more sexually satisfied, suggesting that the negative effect of desired sexual frequency on sexual satisfaction was only significant after accounting for actual sexual frequency. Additionally, after including the actor predictor variables, the effect of relationship length became nonsignificant (β = −.07; p = .34).

Table 3 Associations between actor desired sexual frequency, sexual intimacy, and emotional intimacy, and couple actual sexual frequency and sexual satisfaction
Full size table

Model 3 included the addition of partner effects at level 1, including partner desired sexual frequency, partner emotional intimacy, and partner sexual intimacy. After including these variables, all actor predictors remained significant, including actor emotional intimacy, actor sexual intimacy, actor desired sexual frequency, as well as couple-level sexual frequency (Table 4). Additionally, partner sexual intimacy was also significantly associated with sexual satisfaction (β = .12; p = .043), suggesting that after accounting for the effects of actor predictors and couple-level sexual frequency, one’s partner’s perception of intimacy with sex was uniquely associated with one’s own levels of sexual satisfaction. Further, r-to-z transformations demonstrated that sexual frequency was a significantly stronger predictor of sexual satisfaction compared to all other predictors (p < .001).

Table 4 Associations of actor and partner desired sexual frequency, sexual intimacy, and emotional intimacy, and couple actual sexual frequency on sexual satisfaction
Full size table

Discussion

The current study provided novel descriptive information regarding how FSG couples conceptualize sex, the frequency of engagement in specific sexual activities, and which relationship and sexual factors are associated with sexual satisfaction. Overall, findings suggest that women in FSG relationships may conceptualize having sex in broader terms compared to what is typically assumed in research that uses sexual intercourse as the standard definition. Further, findings demonstrated that sexual satisfaction was associated with numerous relationship and sexual factors. Our methodology also illustrated the dyadic nature of sexual satisfaction, particularly when considering sexual intimacy of both partners.

Conceptualizations of and Engagement in Sexual Activities

This study provided descriptive information regarding conceptualizations of sex and sexual practices within FSG couples. Findings indicated that sexual acts involving partnered genital touching—including oral sex, genital-to-genital touching, hand-to-genital touching, using sex toys, and anal stimulation/penetration—were all considered acts that constituted having sex by the vast majority of participants (>85%). By contrast, one partner masturbating in the presence of her partner (joint masturbation) was only considered having sex by approximately 60% of participants, suggesting that a substantial proportion of couples defined sex as exclusive to partners touching each other. Further, non-genital touching such as making out/French kissing was considered having sex by very few participants (<5%). It is interesting to consider how across all sexual acts, at least some participants did not consider each sexual activity to constitute having sex, and on average, partners disagreed on how they conceptualized one out of seven sexual acts. Hence, conceptualizations of sex by women in FSG relationships cannot be completely reduced to an absolute definition and may not be shared by partners within the same relationship. Sanders et al. (2010) found that less than half of participants from a national survey of adults considered manual stimulation of genitals from one partner to another to be having sex, and less than three-fourths considered oral sex to be having sex. Therefore, in line with previous research, FSG couples may conceptualize having sex in broader terms compared to the general population (Horowitz & Spicer, 2013; Sewell et al., 2017). Indeed, within Sanders and colleagues’ study, the only sexual acts that were considered having sex by the vast majority of participants involved penile–vaginal penetration. Thus, definitions of having sex within the general public may be informed by heterosexual bias that prioritizes penile–vaginal intercourse above other sexual acts. In terms of engagement in these various sexual activities, results suggest that FSG couples may vary in the type of sexual activities they typically engage in while also indicating that some sexual acts, such as hand-to-genital stimulation and oral sex, are more common within FSG relationships.

Additionally, in line with previous research (Solomon et al., 2005), 69% of partners in this study reported wanting a higher desired frequency of sex (approximately more than once per week) compared to how often they were actually having sex with their partner (approximately once a week). At the same time, sexual frequency and frequency of orgasms were highly correlated, and a low percentage (3%) of women in our sample reported never experiencing an orgasm in the presence of their partners. These findings support the notion that broad conceptualizations of sex within FSG couples may serve as a particular strength by giving partners numerous ways to experience sexual satisfaction. Indeed, estimates place the average frequency of women orgasming during sex between 20 and 50% in the general population (Bancroft, Loftus, & Long, 2003; Puts, Welling, Burriss, & Dawood, 2012), and orgasm frequency is positively related to sexual satisfaction in the general population (Costa & Brody, 2007). Taken together, these results suggest that when women in FSG relationships engage in sex, their sexual encounters tend to be sexually fulfilling, consistent with recent literature suggesting the lesbian women have higher frequencies of orgasm compared to heterosexual women (Frederick, John, Garcia, & Lloyd, 2017; Garcia, Lloyd, Wallen, & Fisher, 2014), while also suggesting that FSG couples are generally engaging in sex less frequently than desired.

Factors Associated with Sexual Satisfaction

This study utilized dyadic actor–partner interdependence models in order to make a unique contribution to the literature on sexual satisfaction in FSG couples. Results provided evidence that numerous factors, including sexual frequency, the emotional quality of the relationship, and intimacy specific to sexual experiences, were positively associated with sexual satisfaction within FSG relationships, as expected. It was unexpected to find a negative association between desired sexual frequency and sexual satisfaction given previous research (Cohen & Byers, 2014; Tracy & Junginger, 2007); however, it is important to emphasize that this negative association was only found after accounting for actual sexual frequency, which was generally lower than desired frequency, suggesting that this finding reflects this discrepancy. Further, sexual satisfaction, actual sex frequency, and desired sexual frequency all had positive bivariate correlations with one another. Thus, in general, couples with more sexual desire coupled with high sexual frequency are likely the most satisfied. It is also important to note that despite high correlations, each of the four actor predictors, as well as partner sexual intimacy, predicted unique variance in sexual satisfaction. Taken together, these findings support that the notion that sexual satisfaction is a complex, multi-faceted phenomenon that may be best understood by the interplay of different aspects of the sexual relationship.

Furthermore, although relationship length was negatively associated with sexual satisfaction when only including control variables in the model, this effect was diminished to nonsignificance when taking into account other factors more specifically related to the quality of the sexual relationship. The term “lesbian bed death” has been a colloquial term used to imply that FSG couples stop having satisfying sexual experiences after being together for a substantial period of time (Cohen & Byers, 2014). Findings from the current study strongly suggest that it is not relationship length itself that negatively impacts sexual satisfaction; instead, it is likely the concurrent decline in sexual frequency, emotional intimacy, and sexual intimacy that contribute to this association. Of note, mixed-gender couples have also shown this pattern (Scott et al., 2017), suggesting that this may be a normative pattern across couple types.

Findings regarding actor emotional intimacy and actor and partner intimacy with sex are particularly important given that for decades, many scholars conceptualized high intimacy in lesbian relationships as pathological and damaging to sexual satisfaction (e.g., Burch, 1986; Ossana, 2000). This project adds to the growing body of evidence that posit that these assumptions are likely based within a heteronormative framework, as both emotional and sexual intimacy were positively associated with sexual satisfaction (cf. Ackbar & Senn, 2010; Hardtke, Armstrong, & Johnson, 2010). Thus, higher levels of intimacy likely reflect deeper feelings of emotional safety and connection that make sexual experiences more satisfying.

Clinical Implications

Practitioners can integrate findings of this study into practice by acknowledging the breadth of possible sexual behaviors within FSG couples and by encouraging couples to develop shared expectations in terms of how sex is conceptualized. Although the touching of genitals from one partner to the other was the most commonly endorsed behavior that participants considered to be having sex, there was no consensus for any sexual act. Additionally, clinicians may benefit from emphasizing how flexible conceptualizations of sex may be considered a particular strength of FSG relationships. Indeed, sex therapy interventions commonly used with mixed-gender couples often encourage couples to de-centralize penile–vaginal intercourse and to increase the breadth of sexual behaviors (e.g., McCarthy & Thestrup, 2008).

Findings from this study suggest that numerous areas may be targeted in interventions aimed to increase sexual functioning in FSG couples. Practitioners should be sensitive to the possibility that FSG couples may believe, internally or explicitly, the negative stereotype of lesbian bed death. Practitioners may refer to findings of this study suggesting that it is not necessarily relationship length itself that leads to declines in sexual satisfaction, and instead emphasize that relationship and sexual factors are more important to address. Additionally, at face value, descriptive and model findings may suggest that increasing sexual frequency should be the primary focus of interventions aimed to increase sexual satisfaction. However, given that the emotional quality of the relationship as well as intimacy specific to sex were also uniquely associated with sexual satisfaction, it remains unclear whether increases in sexual frequency alone would improve sexual satisfaction. Findings further reveal the dyadic nature of sexuality within these relationships, particularly regarding significant actor and partner effects of sexual intimacy on sexual satisfaction after accounting for other variables, including frequency of sex.

Practitioners would be best advised help couples with low sexual satisfaction to increase their frequency of sexual activity while also addressing broader relationship issues and the emotional quality of their relationship and sexual encounters. For example, practitioners could help couples more openly discuss their desires for sexual experiences and problem solve around any barriers or disagreements. These discussions could be facilitated by either structured skills-based communication (Markman, Stanley, & Blumberg, 2010) or through other therapy approaches that help partners discuss vulnerable experiences and emotions (Hardtke et al., 2010; Jacobson & Christensen, 1998; Johnson, 2008). Approaches such as sensate focus may also help couples engage in sensual and sexual acts in ways that enhance intimacy and communication (Masters & Johnson, 1970; Weiner & Avery-Clark, 2014).

Limitations and Future Directions

All the data from this study were cross-sectional, limiting the ability to make directional conclusions. Longitudinal research will therefore be necessary to evaluate how these factors work over time and in order to make causal inferences. The sample was also selected from a large, Western metropolitan area, so caution must be taken in generalizing these findings to other areas of the country. This might be particularly important when considering how the cultural acceptance of same-gender couples may vary by region and political climate. Participants in this study were predominantly white, middle class, and college educated, limiting interpretations of how these findings will translate to other demographic groups. Research has demonstrated that Black and Latino sexual minorities face a compounded risk for discrimination and often receive less acceptance from others for their sexual orientations (O’Donnell, Meyer, & Schwartz, 2011), and it is unclear how these experiences may impact sexual functioning. Consequently, the intersections of identities within FSG relationships will be important to investigate. Additionally, consistent with limitations in many LGBT-focused studies, non-probability sampling makes it unclear why some couples may have participated compared to others who did not, possibly biasing results and limiting external validity (Meyer & Wilson, 2009).

The current study also used a new measure to better evaluate conceptualizations and frequency of sex. Future analyses with this measure could evaluate how specific sexual activities (oral sex, using sex toys, etc.) relate to overall sexual satisfaction and relationship quality. It may also be useful to parse apart how the giving or receiving of pleasure might relate to sexual satisfaction. Studies with this measure may want to consider revising the conceptualization of sex questions to a Likert scale (as opposed to yes/no) as has been used in other studies to more thoroughly assess how context may affect how women in these relationships define sex (Horowitz & Spicer, 2013; Sewell et al., 2017). Finally, within-couple moderation analyses may also be beneficial, such as how discrepancies in levels of desire or conceptualizations of sex predict sexual satisfaction.