Understanding the phenomenon of sexual desire discrepancy in couples.
KEY WORDS: sexual desire; desire discrepancy; sexual satisfaction; low desire
Sexual desire, defined as having an interest in sexual activity that leads the individual to seek out sexual activity and/or be pleasurably receptive to the partner's initiation (Basson, 2008), is a topic of great interest to researchers and lay people alike. In particular, interest has focused on low desire in women. The movement to treat low desire using pharmaceutical interventions has led to potential concerns about pathologizing normal variation in female sexual desire (Bancroft, Loffus, & Long, 2003; Tiefer, 2001). While the current study does not speak directly to this this debate, we focus on the related construct of "sexual desire discrepancy," the difference between two partners' sexual desire levels. Zilbergeld and Ellison (1980) coined this term to reflect a shift in the focus of sex therapy from the individual to the couple in recognition of the fact that partners having different levels of sexual desire does not necessarily indicate that one partner is experiencing atypically high or low (e.g., female sexual interest/arousal disorder or male hypoactive sexual desire disorder) levels of sexual desire.
However, discrepancies in sexual desire are not a phenomenon unique to couples seeking therapy. Given that two individuals are likely to differ in their level of sexual desire and that levels of sexual desire tend to fluctuate over time, influenced by biological factors, internal motivations, and characteristics of people's relationships (e.g., length of the relationship; Levine, 1987), it is inevitable that most individuals will experience some degree of sexual desire discrepancy with their sexual partner and that this will change over the course of the relationship (Herbenick, Mullinax, & Mark, 2014). While there has been considerable theorizing about the impact of low sexual desire or high desire discrepancy in clinical samples, much less is known about the phenomenon of sexual desire discrepancy in non-clinical samples.
Although studies have demonstrated an association between desire discrepancies and relationship and sexual outcomes in non-clinical samples, findings have been inconsistent (Bridges & Horne, 2007; Davies, Katz, & Jackson, 1999; Mark & Murray, 2012; Willoughby & Vitas, 2012). For example, some studies have found that a higher level of actual desire discrepancy is associated with lower sexual satisfaction for men (Mark, 2012), while other studies have failed to find a significant association for men (Davies et al., 1999; Mark & Murray, 2012). The findings for women are similarly inconsistent. Whereas Davies and colleagues (1999) and Mark and Murray (2012) found that greater actual desire discrepancy was associated with lower sexual satisfaction in women, Mark (2012) did not replicate this association.
The extant research on sexual desire discrepancy has had some important limitations that our research seeks to address. With three exceptions (Herbenick, Mullinax, & Mark, 2014; Mark, 2014; Willoughby, Farero, & Busby, 2014), past studies examining sexual desire discrepancies in heterosexual couples have recruited short-term dating samples from college/university populations (Davies, Katz, & Jackson, 1999; Mark & Murray, 2012; Willoughby & Vitas, 2012). Given that sexual desire fluctuates through the stages of a relationship (Byers & Rehman, 2014), it is critical to examine desire discrepancies in both short- and long-term relationships. Indeed, there is evidence to suggest that findings about sexual desire discrepancy from samples of couples in short-term dating relationships may not generalize to couples in long-term dating and marital relationships (Willoughby & Vitas, 2012). Further, in one of the few studies that has focused on more established romantic relationships, Willoughby and colleagues (2014) found that married men and women's greater sexual desire discrepancy was associated with lower relationship satisfaction. A key difference between this study and the current one is in the conceptualization of sexual desire discrepancy. Willoughby et al. (2014) operationalized sexual desire discrepancy as the difference between one's desired and actual frequency of sexual intercourse. Though examining differences in desired and actual frequency of sex represents an informative line of research that may have implications for satisfaction outcomes in relationships, it is a problematic measure of sexual desire discrepancy in the relationship. Research shows that partners are motivated to engage in sex for many reasons and may do so in the absence of sexual desire (Vannier & O'Sullivan, 2010; Wood, Milhausen, & Jeffrey, 2014).
The first goal of the current study was to investigate the association between sexual desire discrepancy and sexual satisfaction in the context of long-term romantic relationships rather than short-term dating relationships. Past studies examining sexual desire discrepancy produce different estimates of the construct depending on whether it is measured by gathering data from one or both partners. Consistent with the work done by Davies et al. (1999), we argue that these two methods lead to the measurement of informative, yet distinct, constructs. Specifically, when an individual is asked to subjec tively compare his/her own level of sexual desire to that of his/her partner, what is being measured is best conceptualized as the individual's perception of sexual desire discrepancy. In contrast, when both partners are asked to report on their own levels of sexual desire and the difference between these two desire scores is computed, the resulting variable is best conceptualized as actual desire discrepancy (1) (e.g., Davies et al., 1999; Mark, 2012; Mark & Murray, 2012).
We used past research on desire discrepancy to inform our hypotheses. Based on the study by Davies and colleagues (1999) showing that, compared to individuals who perceived less desire discrepancy between themselves and their partners, those who perceived greater discrepancy were less sexually satisfied, we predicted that perceiving one's desire as discrepant from one's partner would be associated with lower sexual satisfaction. As noted earlier, the results of the association between actual desire discrepancy and sexual satisfaction are inconsistent. Thus, we did not offer any specific predictions for the association between actual desire discrepancy and sexual satisfaction.
The second goal of the current study was to compare categorical versus dimensional approaches to assess perceived sexual desire discrepancy. In the few existing studies that have examined perceptions of desire discrepancy, two types of questions have been used to measure the construct. Some studies have used a dichotomous scale to ask participants to report on whether their desire is approximately the same as or different from that of their partner. For example, Davies et al. (1999) asked the question, "Do you and your partner have roughly similar sexual desire levels?" Participants were given the option to answer either yes or no. In other studies, a categorical response option has been used to assess the direction of the desire discrepancy. For example, Bridges and Horne (2007) asked participants to select which statement best depicted their relationship. The response options included, "My partner desires to have sexual relations more than I do, and this has caused problems in our relationship," "I desire to have sexual relations more often than my partner, and this has caused problems in our relationship," and "My partner and I desire sexual relations to the same degree" (Bridges & Horne, 2007, p. 46). Categorical response scales allow individuals to report whether they believe a desire discrepancy exists in the relationship and to indicate the direction of the discrepancy (i.e., which partner has higher versus lower sexual desire). One important type of response scale that, to our knowledge, has not been used in past research is a continuous, Likert-type response scale that allows participants to describe the magnitude and direction of sexual desire discrepancy in their relationship. We believe that examining both the direction and magnitude of the perceived desire discrepancy will provide a more comprehensive understanding of this construct. Thus, we investigated whether using different response scales would influence reports of sexual desire discrepancy.
In sum, there were two overarching goals for the current study. First, we wanted to compare two different ways of conceptualizing sexual desire discrepancy (actual versus perceived) and to examine how each conceptualization relates to sexual satisfaction. In our study, we focused on sexual satisfaction as the outcome of interest in light of strong, consistent evidence that supports its relevance and importance to overall well-being (e.g., Laumann et al., 2006). As well, data gathered from cross-sectional and longitudinal studies support the strong association between sexual satisfaction and relational quality and stability (see review by Rehman, Fallis, & Byers, 2013). Relatedly, we wanted to investigate whether the direction of the discrepancy (i.e., perceiving one's desire to be greater versus less than one's partner's) matters when examining how desire discrepancies relate to sexual satisfaction. The second goal was to compare the information gleaned by using categorical versus dimensional approaches to measure perceived sexual desire discrepancy.
To achieve our study's goals, we collected and analyzed sexual desire data from two independent samples. Sample 1 consisted of couples in long-term committed heterosexual relationships recruited from the community. Both partners completed measures of their own sexual desire allowing us to calculate a measure of actual sexual desire discrepancy. Sample 2 consisted of individuals in committed heterosexual relationships. Participants completed a series of questions designed to assess perceptions of desire discrepancy. All participants completed a standardized measure of sexual satisfaction.
Sample 1 (Dyadic Data). The first sample consisted of couples who were participating in a larger, longitudinal study of sexuality in relationships. Data for the current study are cross-sectional and were collected during the third phase of the longitudinal study. Participating couples were recruited from Southwestern Ontario using online and newspaper advertisements, posters placed in doctors' and sex therapists' offices, and referrals from doctors and sex therapists. To be eligible for the study, couples had to be in a heterosexual, married or cohabiting relationship. If they were cohabiting they were required to have been living together for at least two years to ensure they were in long-term committed relationships. In addition, both partners had to be between the ages of 21 and 65 at the time they initially participated in the study, be able to speak and read English at a grade 8 level to ensure they could understand and complete study measures, and be willing to participate.
In total, 84 couples participated in the current study. Two couples were excluded from the analyses because at least one partner did not complete the sexual desire questionnaire. Thus, the final sample consisted of 82 couples. Men were an average of 40.6 years old (SD = 11.29) and women were an average of 38.5 years old (SD = 11.36). On average, couples had been in their relationship for 13.2 years (SD = 8.71) and were either married (79%) or cohabiting (21%). Most (92%) participants identified as White.
Sample 2 (Individual Data). This was an online sample recruited using an advertisement on Amazon Mechanical Turk. Recruitment materials indicated that researchers were seeking volunteers for a study of "sexuality and relationships." To be eligible for the study, participants had to be 21 years old and in a committed heterosexual relationship. All participants resided in the United States. A total of 212 individuals completed the study. The data for 19 participants was excluded from analyses because participants did not meet eligibility criteria (i.e., they reported they were not currently in a committed relationship). The data for two more participants was excluded because participants answered two or more validity questions incorrectly (e.g., "Choose Strongly Agree to show that you read this question carefully"). Thus, the final sample consisted of 191 participants (63% female). On average, men were 32.0 years old (SD = 13.33) and had been in their current relationships for 8.5 years (SD = 10.33). Women were, on average, 30.5 years old (SD = 11.65), and had been in their current relationships for 7.5 years (SD = 8.21). They were either married or common law (39%), dating and living with their partner (28%) or dating and living apart from their partner (32%). The majority of participants identified as White (78%), while others identified as Black (9%), Hispanic (6%), Asian (3%), South Asian (2%), and Other (2%).
Measures administered to both samples
Demographic questionnaire. Participants completed a demographic questionnaire designed for the current study to gather background information including age, gender, ethnicity, relationship status, and relationship length.
Global Measure of Sexual Satisfaction (GMSEX; Lawrance & Byers, 1995). The GMSEX is a 5-item measure of sexual satisfaction that requires participants to rate their satisfaction with their sexual lives overall using 7-point scales with adjective pairs at each anchor (e.g., Very Bad-Very Good). Scores on the GMSEX range from 5 to 35 with higher numbers indicating greater sexual satisfaction. Lawrance and Byers (1995) have found evidence of high reliability ([alpha] = .96) and good concurrent validity. The measure showed excellent internal consistency in the current study (Sample 1: Cronbach's alpha = .97 for men and .95 for women; Sample 2: Cronbach's alpha = .96 for men and .97 for women).
Sample 1 only (dyadic data)
Hurlbert Index of Sexual Desire (HISD; Apt & Hurlbert, 1992). The HISD is a 25-item measure that was used to assess participants' sexual desire for their partners (e.g., "I look forward to having sex with my partner") and their general sexual desire (e.g., "I daydream about sex"). Participants responded to items using a 5-point Likert scale ranging from 0 (All of the Time) to 4 (None of the Time). Scores range from 0 to 100, with higher scores indicating higher overall sexual desire. The measure has sound psychometric properties, with evidence of strong validity and reliability ([alpha] = .89; Hurlbert, Apt & Rombough, 1996). Their measure demonstrated excellent internal consistency in the current sample (Cronbach's alphas = .92 for men and .96 for women).
Sample 2 only (individual data)
Perceived sexual desire discrepancy. To facilitate comparisons between the current study and those that have used a single-item measure of desire discrepancy (e.g., Davies et al., 1999; van Anders, Hipp, & Kane Low, 2013), we created an item that stated, "In general, how does your sexual desire level compare to that of your partner?" Participants answered by selecting one of the following responses: "I have a higher level of sexual desire than my partner," "My partner has a higher level of sexual desire than me," or "My partner and I have equal levels of sexual desire."
To address the limitations associated with measuring desire discrepancy using only a categorical response scale, we developed a second item to assess the degree of partners' sexual desire discrepancy using a continuous response scale. This item, adapted from Ard's (1977) study on marital sexual experiences, asked "How different would you say your sexual desire level is from that of your partner at the present time?" Participants responded using a 5-point scale ranging from 1 (No Different) to 5 (Very Different).
Finally, we created a third item that assessed both the direction and magnitude of an individual's perceived sexual desire discrepancy. Item 3 stated, "Using the scale below, rate which statement best captures yours and your partner's desire levels." Responses were indicated on a 5-point scale: 2 (My desire level is much higher than my partner's), 1 (My desire level is slightly higher than my partner's), 0 (My desire level is equal to my partner's), -1 (My partner's desire level is slightly higher than mine), and -2 (My partner's desire level is much higher than mine).
Sample 1 (Dyadic Data). The study took approximately 3 hours to complete. At the outset of the study, couples came into the lab together. Two research assistants worked with each couple. The research assistants reviewed an information letter with each couple, who were provided the opportunity to ask questions and then gave written consent to participate. After giving consent, couple members were separated into different rooms to complete several questionnaires as well as a semi-structured interview and a discussion task that were not relevant to the current study. The questionnaires unrelated to this study included measures of couples' communication, relationship commitment, and sexual functioning. During the semi-structured interview, participants discussed sexual and relationship problems they have with their partner with a research assistant. For the discussion task, each partner chose a sexual problem to discuss with their partner. Participants completed a background questionnaire first and engaged in the discussion with each other midway through the study. The remaining questionnaires and interviews were completed in random order before and after the discussion to ensure that order effects did not systematically influence the results. At the end of the study, couples were reunited, debriefed, and provided with a list of sexual health resources. Each partner received $50.00 for participating in the study.
Sample 2. After reading an information letter about the study, participants gave consent to participate by clicking a radio button. They were then linked to an online survey, which took approximately 15 minutes to complete. Participants were paid in their Amazon account for participating in the study.
Descriptive Statistics for Actual Desire Discrepancy (Sample 1--Dyadic Data)
To measure actual sexual desire discrepancy, we subtracted women's Hurlbert Index of Sexual Desire (HISD; Apt & Hurlbert, 1992) scores from men's HISD scores. The resulting measure of actual sexual desire discrepancy represented the difference between partners' scores, with positive numbers indicating the male partner had higher desire and negative numbers indicating the female partner had higher desire. A score of 0 represented no sexual desire discrepancy.
Of the 82 couples, 30% fell within 1 SD of 0 and 70% had a sexual desire discrepancy that was greater than 1 SD. Among the 70% of couples whose desire discrepancy score fell more than 1 SD away from 0, 11% were couples in which the female partner reported higher sexual desire and 59% were couples in which the male partner had higher desire. The magnitude of couples' desire discrepancies was significantly greater in couples in which the male partner had higher desire relative to couples in which the female partner had higher desire, ([M.sub.Male Greater] = 26.44, SD 16.13, [M.sub.Female Greater] -11.65, SD = 11.69) t(79) = -9.74, p < .001. In sum, our sample consisted mainly of couples that had sexual desire discrepancies in their relationships and men tended to be the higher desire partners.
The average sexual satisfaction for the men and women in Sample 1 was 26.54 (SD = 7.21) and 26.25 (SD = 6.57) respectively. Thus, on average, our sample was characterized by high levels of sexual satisfaction.
Descriptive Statistics for Perceived Desire Discrepancy (Sample 2--Individual Data)
To assess perceptions of sexual desire discrepancy, participants responded to three items. Response frequencies for these items are presented in Tables 1 to 3. The first item stated, "In general, how does your sexual desire level compare to that of your partner?" To examine whether men and women differed in the degree to which they endorsed having desire higher than, lower than, or equal to their partners, we conducted a 2 (gender) x 3 (desire discrepancy level) chi-square test. The test revealed that the degree to which participants endorsed each Item 1 category differed significantly by gender, [chi square] (1, N = 191) = 19.24, p < .001.
For Item 2, which assessed the perceived magnitude of sexual desire discrepancy, participants were asked "How different is your sexual desire level from that of your partner at the present time?" Responses were reported on a 5-point scale ranging from 1 (No different) to 5 (Very different). On average, women rated the magnitude of sexual desire discrepancy in their relationship as 2.50 (SD = 1.24) and men rated the magnitude of discrepancy as 2.54 (SD = 1.22). To examine whether men and women differed in their responses to this item, we conducted an independent-samples f-test. Results showed that men's and women's reports did not differ significantly, f(188) = .23, ns. Overall, men and women reported a similar degree of sexual desire discrepancy in their relationships. Importantly, in comparing the frequencies of the different types of sexual desire discrepancy as reported by women in response to Item 1 to Item 2, we note that 10% more women reported a sexual desire discrepancy in response to Item 2. Results from a one-sample z-test of proportions indicated that this difference in rates of reported discrepancy across items was statistically significant (z = 2.28, p < .05).
Item 3 assessed the direction and magnitude of desire discrepancy in participants' relationships. A breakdown of response frequencies for each of the 5 categories is provided in Table 3. Men and women's reports of direction and magnitude of desire discrepancy differed significantly, t(188) = 3.56, p < .001. Specifically, men (M = 0.50, SD = 1.10) reported that their own desire was higher than their partners' desire more often than women (M = -.09, SD = 1.11).
Actual Desire Discrepancy and Sexual Satisfaction (Sample 1--Dyadic Data)
To determine the association between the direction of actual desire discrepancy on men and women's sexual satisfaction, dyadic sexual desire discrepancy was recalculated. Men's self-reported desire scores were subtracted from women's self-reported desire scores when predicting outcomes for women, and women's self-reported desire scores were subtracted from men's when predicting outcomes for men. This allowed values greater than one to equal self-greater-than-partner discrepancies and values less than one to equal partner-less-than-self discrepancies in all analyses. To test the association between the actual sexual desire discrepancy score and men's and women's sexual satisfaction, a multiple regression analysis was conducted that included linear, quadratic, and cubic terms for the desire discrepancy score. The discrepancy variable was centred at 0 (equaling no difference between partners) before computing the quadratic and cubic terms. The combination of linear, quadratic, and cubic terms in the model allowed for testing whether the magnitude and/or direction of the association between actual sexual desire discrepancy and sexual satisfaction changed as the discrepancy in sexual desire changed from partner-higher, to equal, to self-higher. The linear, squared, and quadratic terms were all nonsignificant for both men ([[beta].sub.linear] = - .19, [[beta].sub.quadratic] = .01, [[beta].sub.cubic] = .24, all ns) and women ([[beta].sub.linear] = .14, [[beta].sub.quadratic] = .04, [[beta].sub.cubic] = .14, all ns), suggesting a lack of association between actual sexual desire discrepancy and sexual satisfaction.
Perceived Desire Discrepancy and Sexual Satisfaction (Sample 2--Individual Data)
Item 1. We examined the association between the categorical item (Item 1) designed to assess perceptions of sexual desire discrepancy ("In general, how does your sexual desire level compare to that of your partner?") and sexual satisfaction. Item 1 assessed which, if any, partner had higher or lower desire in the relationship. Given that Item 1 had a categorical response scale, our first step was to create a set of dummy codes for the variable before conducting a multiple regression analysis. One set of dummy codes was created in which "equal desire" was used as the reference group. A second set of dummy codes was created in which "higher desire than partner" was used as the reference group. We conducted two multiple regression analyses, with each analysis testing a different set of Item 1 dummy codes as the predictor variables and GMSEX as the criterion variable. Results showed that there were no significant differences in sexual satisfaction for women whose desire was higher than their partner (mean difference = .65, ns), compared to those who had equal desire with their partner. There was, however, a significant difference in sexual satisfaction for women who had lower desire than their partner compared to those who had equal desire with their partner (mean difference = - 3.30, p < .01). Women with lower desire than their partner were less sexually satisfied than those with equal desire. Women who had higher desire than their partner did not differ significantly in sexual satisfaction from those who had lower desire than their partner (mean difference = 1.36, ns). For men, there was no significant difference in sexual satisfaction between those whose sexual desire was lower than their partner and those who had equal desire with their partner (mean difference = -1.46, ns). In contrast, there was a significant difference in sexual satisfaction between men whose desire was higher than their partner and those who had equal desire with their partner (mean difference = - 3.21, p < .05) or less desire than their partner (mean difference = 5.64, p < .01).
Item 2. To examine the relationship between perceived magnitude of sexual desire discrepancy and sexual satisfaction (irrespective of direction of discrepancy), we examined zero-order correlations between responses on Item 2 ("How different is your sexual desire level from that of your partner at the present time?") and GMSEX scores. Item 2 responses significantly correlated with GMSEX for women, r= -.52, p < .001, and men, r = -.53, p < .001. To further examine the relationship between individual sexual desire discrepancy and sexual satisfaction for men and women, we conducted a multiple regression analysis with Item 2 as the predictor variable and GMSEX as the outcome variable. Results showed that the degree of sexual desire discrepancy perceived by participants predicted 27% of the variance in sexual satisfaction for women, F(l,119) = 42.49, p < .001, and 28% of the variance in sexual satisfaction for men F(1,67) = 25.84, p < .001. For men and women, greater individual sexual desire discrepancies predicted lower sexual satisfaction.
Item 3. To examine relationship between Item 3 ("Using the scale below, rate which statement best captures yours and your partners' desire levels," with responses ranging from -2, partner desire much higher than mine, to +2, my desire much higher than partner, with 0 equaling no difference in desire) and sexual satisfaction, we conducted a multiple regression analysis, as described in the results for Sample 1. The analysis included the linear, quadratic, and cubic terms for the perceived desire discrepancy score, as measured by Item 3. The combination of linear, quadratic, and cubic terms in the model allowed for testing whether the magnitude and/ or direction of the association between perceived sexual desire discrepancy and sexual satisfaction changed as the discrepancy in sexual desire changed from partner-higher, to equal, to self-higher. Sex differences were tested by creating interaction terms between sex and each of the three terms. The model was run twice, once including all predictors, and again dropping any nonsignificant terms aside from lower order terms (e.g., linear) that had to be included to interpret higher order terms (e.g., quadratic). In the final model predicting sexual satisfaction, only the quadratic term ([beta] = -.47, p < .001) was significant. The main effects of gender and Item 3 sexual desire discrepancy and all gender-interactions were nonsignificant, and, with the exception of Item 3 sexual desire discrepancy, were removed from the model. The final regression model explained 21% of the variance in sexual satisfaction.
To facilitate the interpretation of the quadratic function linking perceived sexual desire discrepancy and sexual satisfaction, the quadratic function was plotted over a scatter plot of Item 3 sexual desire discrepancy scores and GMSEX sexual satisfaction scores (see Figure 1). The figure suggests that perceiving one's own sexual desire as being much lower or much higher than one's partner's is associated with lower levels of sexual satisfaction. In addition, in an effort to better understand the magnitude and significance of differences in sexual satisfaction at different levels of perceived desire discrepancy, a spline model was fit using dummy codes for each level of desire discrepancy, such that differences in sexual satisfaction were tested at each level of perceived desire discrepancy, with zero discrepancy as the reference group. Results from this post-hoc analysis indicated that only individuals with ratings of -2 (My partner's desire is much higher than mine) or +2 (My desire is much higher than my partner's) were significantly different in sexual satisfaction relative to those perceiving zero discrepancy in sexual desire (mean difference = -9.47, p < .001 and -6.80, p < .001, respectively), with both groups scoring lower.
The overall purpose of the current study was to better understand the phenomenon of sexual desire discrepancy in normative, non-clinical samples of individuals in long-term committed relationships. The first goal was to compare two different ways of conceptualizing desire discrepancy, actual versus perceived desire discrepancy, and to test whether sexual satisfaction relates differently to perceptions of desire discrepancy as compared to actual levels of desire discrepancy. We measured actual desire discrepancy using couples' data by creating a difference score between partners' self-reported sexual desire levels. In contrast, we measured perceptions of sexual desire discrepancy by asking individual participants to report on their subjective experience of differences in sexual desire between themselves and their partners in three distinct ways.
We did not detect a significant association between actual sexual desire discrepancy and sexual satisfaction for either men or for women. In addition to testing the linear association between these variables, we also tested for quadratic and cubic effects. In contrast to the findings for couples' actual desire discrepancy, individuals' perceptions of desire discrepancy related significantly to both men and women's sexual satisfaction. Our findings for actual desire discrepancy suggest that variation in desire levels between partners is not necessarily a problem in the relationship. Just as partners learn to deal with differences in values, goals, and priorities in other domains of life, most couples have likely developed adaptive ways to handle desire discrepancies without such discrepancies adversely impacting their sexual satisfaction. Our findings are consistent with the notion advanced by Herbenick and colleagues (2014) that for most couples, desire discrepancies should be viewed as a normal and expected part of the sexual relationship, rather than as a "bug" that needs to be fixed.
We would like to emphasize that it is important to be cautious when interpreting our results for actual desire discrepancy. Our sample was characterized by individuals who were, overall, sexually satisfied (see descriptives for Sample 1). Thus, in interpreting the findings from the current study, the most appropriate conclusion regarding actual sexual desire discrepancy is that in a sample characterized by fairly high levels of sexual satisfaction, the level of discrepancy between partners in their sexual desire does not relate to their sexual satisfaction. However, we cannot generalize these findings to more distressed samples.
Why would perceived desire discrepancy be associated with sexual satisfaction when there is no association between actual desire discrepancy and sexual satisfaction? Research has shown that couples' perceptions of different aspects of their relationship tend to be influenced by their satisfaction with the relationship (Weiss, 1980). This phenomenon, termed "global sentiment override" (Weiss, 1980), creates a sort of halo-effect in that different aspects of the relationship are perceived positively or negatively based on the perceiver's feelings about his/her overall relationship. Supporting this notion, Lemay and Neal (2014) have found that partners who are satisfied with their relationship perceive that their partners are more supportive than those with less relationship satisfaction, regardless of the partner's actual level of support. Applying this notion to perceptions of desire discrepancy in the relationship, it is possible that couples who are more sexually satisfied are less likely to perceive a sexual desire discrepancy in their relationship, even if such a discrepancy exists. The current study is aimed at describing the phenomenon of sexual desire discrepancy in established romantic relationships and we are not investigating any causal link between sexual desire discrepancy (actual or perceived) and sexual satisfaction. When causal models are tested, it will be important to investigate the pathway from desire discrepancy to sexual satisfaction as well as the opposite pathway. An understanding of the causal link between these variables will inform further theory development on how perceptions of desire discrepancies form and change over time.
The second goal of our study focused on methodological challenges in assessing the construct of perceived desire discrepancy. In past studies, researchers have tended to use categorical measures to assess perceived desire discrepancy (Bridges & Horne, 2007; Davies et al., 1999). In the current study, in addition to a categorical item that asked participants to report on whether their sexual desire was greater to, less than, or equal to that of their partner (Item 1), participants were asked to rate the magnitude of the discrepancy using a dimensional Likert scale (Item 2). Our results showed that women reported experiencing a sexual desire discrepancy more frequently (i.e., 10% more) when describing the discrepancy on a more nuanced continuous, rather than categorical, response scale. This difference may be attributed to the well-documented advantage of moving from categorical scales to dimensional scales and the resulting increase in statistical power. Peters and Van Voorhis (1940) demonstrated that if a normally distributed score is trichotomized into three groups, the variance explained is 26% less than would be explained by the continuous score.
The first two items that we used to assess perceptions of desire discrepancy are limited because they either measure the direction of the discrepancy (Item 1) or the magnitude of the discrepancy (Item 2) while ignoring the other dimension. This makes it difficult to interpret the results. For example, the results for Item 1 showed that women who perceived their sexual desire to be lower than their partners' desire were less sexually satisfied as compared to women who perceived no differences in desire between themselves and their partners. However, there were no significant differences in the sexual satisfaction of women who perceived their sexual desire to be higher than their partners as compared to women who perceived no difference. As our results suggest, these categorical differences were driven by the magnitude of discrepancy in the two groups (i.e., women who perceive lower desire were characterized by a greater discrepancy in desire between themselves and their partners, as compared to women who were higher on desire). To address the issue of magnitude and directionality, we asked participants a question that simultaneously assessed the direction of the sexual desire discrepancy and the magnitude of the discrepancy (Item 3). By taking into account both dimensions, the results for this item provide a more complete picture of sexual desire discrepancy in the relationship.
When predicting sexual satisfaction using the item that accounted for both magnitude and direction of the perceived desire discrepancy, the results showed that perceiving one's sexual desire as much higher or much lower than one's partner's was associated with lower sexual satisfaction. It is possible that the individual with much lower desire may feel pressured to engage in sexual activity by his/her partner or through an internalized sense of guilt. Conversely, the individual who perceives his/her own desire to be much greater, may feel that his/her needs are not being met and/or may experience negative emotions related to frequently being the initiator of sexual activity or feeling rejected by his/her partner. In this way, both the individual who perceives the self as having much higher or much lower desire than the partner may experience lower sexual satisfaction. These possibilities would need to be systematically investigated in future research.
When we examined the link between perceived desire discrepancy and sexual satisfaction using the categorical item (Item 1), the results showed that men who reported having higher desire than their partners were less satisfied with their sexual relationship than those with desire equal to or less than that of their female partner. In contrast, women who reported experiencing lower levels of desire, as compared to their partners, were less sexually satisfied than women who reported equal or greater desire than their male partner. The results for item 3, a dimensional item used to assess perceptions of desire discrepancy, suggested that there were no gender differences in the association between perceptions of desire discrepancy and sexual satisfaction. Both men and women who experienced their own desire as much higher or lower than their partner's desire experienced lower levels of sexual satisfaction. How do we explain the inconsistency in the results for gender differences? One possibility, as suggested by our data, is that couples where the female partner has lower desire than the male partner are characterized by a larger magnitude of desire discrepancy, compared to couples where the female partner has greater desire than the male partner. Thus, when the magnitude of the difference is taken into account, the group differences are no longer significant. However, it would be premature to conclude that there are no gender differences in how perceptions of desire discrepancy relate to sexual satisfaction. First, the current findings have to be replicated. Second, it is important to consider that if most couples who have a discrepancy experience the direction of the discrepancy to be higher male and lower female desire and that this pairing is associated with negative sexual outcomes, as compared to couples with little to no discrepancy or couples where the female partner has greater sexual desire than the male partner, then the gender difference may be meaningful and worthy of further research.
One limitation of the current study is that we used two different participant samples to examine actual versus perceived desire discrepancy. Although the two samples were similar in age, ethnicity, and relationship length, caution should be taken when making direct comparisons between the two samples. Ideally, we would have been able to compare actual versus perceived discrepancy in the same couples, allowing us to more definitively comment on how perceived discrepancy and actual discrepancy relate differently to sexual satisfaction. Furthermore, we could examine the association between actual and perceived discrepancy and test whether perceived discrepancy relates to ratings of sexual satisfaction, even after accounting for actual desire discrepancy. Another limitation of the current study was that we used different methods to assess actual and perceived discrepancy. Whereas the Hurlbert Index of Sexual Desire, used to compute actual desire discrepancy scores in our dyadic sample, is a multi-item measure of sexual desire, our measures of perceived desire discrepancy were single items. In future work, it would be ideal not only to have a dyadic sample and assess both actual and perceived desire discrepancy in the same sample, but to also assess actual versus perceived desire discrepancy using a similar methodology. One way to do this would be to have both members of a dyad complete two versions of the Hurlbert Index of Sexual Desire--one to rate their own desire and the second to rate their partner's desire.
Also, our samples consisted of mainly White participants who were sexually satisfied. Ceiling effects may limit response variance in samples of couples who are highly satisfied in their relationships. Before generalizing these findings, it will be important to replicate this research with couples from a diverse range of backgrounds in both satisfied and distressed relationships.
Understanding how sexual desire discrepancy relates to outcomes for couples is a vital step in discovering the factors that contribute to the success of committed relationships. The current study examined several methods of measuring sexual desire discrepancy and showed that the use of various measurement techniques influences participant responses and, in turn, the association between sexual desire discrepancy and relationship outcomes. Further, our research shows that perceived sexual desire discrepancy is a key factor in predicting sexual satisfaction for men and women in long-term relationships. Future research exploring the mechanisms that impact the association between sexual desire discrepancy and sexual satisfaction will help to explain how desire discrepancies impact relational outcomes.
(1) We use the term "actual sexual desire discrepancy" to denote the actual difference between two partners' sexual desire levels. This term is not used to imply that actual sexual desire discrepancy is more legitimate than perceived sexual desire discrepancy as both categories may have important implications for how partners experience their relationship.
Apt, C.V., & Hurlbert, D.F. (1992). Motherhood and female sexuality beyond one year postpartum: A study of military wives. Journal of Sex Education and Therapy, 18, 104-114.
Ard, B.N. (1977). Sex in lasting marriages: A longitudinal study. Journal of Sex Research, 13(4), 274-285. http://dx.doi.org/ 10.1080/00224497709550985
Bancroft, J., Loftus, J., & Long, J.S. (2003). Distress about sex: A national survey of women in heterosexual relationships. Archives of Sexual Behavior, 32(3), 193-208. http://dx.doi.org/10.1023/ A:1023420431760 Medline: 12807292
Basson, R. (2008). Women's sexual function and dysfunction: Current uncertainties, future directions. International Journal of Impotence Research, 20(5), 466-478. http://dx.doi.org/10.1038/ ijir.2008.23 Medline: 18548081
Bridges, S.K., & Horne, S.G. (2007). Sexual satisfaction and desire discrepancy in same sex women's relationships. Journal of Sex & Marital Therapy, 33(1), 41-53. http://dx.doi.org/10.1080/ 00926230600998466 Medline: 17162487
Byers, S., & Rehman, U.S. (2014). Sexual well-being. In D.T. Tolman & L.M. Diamond (Eds.), APA Handbook on Sexuality and Psychology (pp. 317-337). Washington, D.C.: APA Press, http:// dx.doi.org/10.1037/14193-011.
Davies, S., Katz, J., & Jackson, J.L. (1999). Sexual desire discrepancies: Effects on sexual and relationship satisfaction in heterosexual dating couples. Archives of Sexual Behavior, 28(6), 553-567. http://dx.doi.org/10.1023/A: 1018721417683 Medline: 10650441
Herbenick, D., Mullinax, M., & Mark, K. (2014). Sexual desire discrepancy as a feature, not a bug, of long-term relationships: Women's self-reported strategies for modulating sexual desire. Journal of Sexual Medicine, 11(9), 2196-2206. http://dx.doi.org/ 10.1111/jsm. 12625 Medline:25052706
Hurlbert, D.F., Apt, C., & Rombough, S. (1996). The female experience of sexual desire as a function of sexual compatibility in an intimate relationship. Canadian Journal of Human Sexuality, 5, 7-14.
Laumann, E.O., Paik, A., Glasser, D.B., Kang, J.H., Wang, T., Levinson, B., Moreira, E.D., Nicolosi, A., & Gingell, C. (2006). A cross-national study of subjective sexual well-being among older women and men: Findings from the Global Study of Sexual Attitudes and Behaviors. Archives of Sexual Behavior, 35(2), 1451b 1. http://dx.doi.org/10.1007/s 10508-005-9005-3 Medline: 16752118
Lawrance, K., & Byers, E.S. (1995). Sexual satisfaction in long-term heterosexual relationships: The interpersonal exchange model of sexual satisfaction. Personal Relationships, 2(4), 267-285. http:// dx.doi.org/10.1111/j. 1475-6811.1995.tb00092.x
Lemay, E.P., Jr., & Neal, A.M. (2014). Accurate and biased perceptions of responsive support predict well-being. Motivation and Emotion, 38(2), 270-286. http://dx.doi.org/10.1007/sll031-0139381-2
Levine, S.B. (1987). More on the nature of sexual desire. Journal of Sex & Marital Therapy, 13(1), 35-44. http://dx.doi.org/10.1080/ 00926238708403877 Medline:3573051
Mark, K.P. (2012). The relative impact of individual sexual desire and couple desire discrepancy on satisfaction in heterosexual couples. Sexual and Relationship Therapy, 27(2), 133-146. http:// dx.doi.org/10.1080/14681994.2012.678825
Mark, K.P. (2014). The impact of daily sexual desire and daily sexual desire discrepancy on the quality of the sexual experience in couples. Canadian Journal of Human Sexuality, 23(1), 27-33. http://dx.doi.org/10.3138/cjhs.23.1.A2
Mark, K.P., & Murray, S.H. (2012). Gender differences in desire discrepancy as a predictor of sexual and relationship satisfaction in a college sample of heterosexual romantic relationships. Journal of Sex & Marital Therapy, 38(2), 198-215. http:// dx.doi.org/10.1080/0092623X.2011.606877 Medline:22390532
Peters, C.C., & Van Voorhis, W.R. (1940). Statistical procedures and their mathematical bases. New York: McGraw-Hill, http:// dx.doi.org/10.1037/13596-000.
Rehman, U.S., Fallis, E., & Byers, E.S. (2013). Sexual satisfaction in heterosexual women. In D. Castaneda (Ed.), An essential handbook of women's sexuality (Vol. 1, pp. 25-46). Santa Barbara, CA: Praeger Publishers.
Tiefer, L. (2001). A new view of women's sexual problems: Why new? Why now? Journal of Sex Research, 38(2), 89-96. http:// dx.doi.org/10.1080/00224490109552075
van Anders, S.M., Hipp, L.E., & Kane Low, L. (2013). Exploring co-parent experiences of sexuality in the first 3 months after birth. Journal of Sexual Medicine, 10(8), 1988-1999. http:// dx.doi.org/10.1111/jsm. 12194 Medline:23911123
Vannier, S.A., & O'Sullivan, L.F. (2010). Sex without desire: Characteristics of occasions of sexual compliance in young adults' committed relationships. Journal of Sex Research, 47(5), 429-439. http://dx.doi.org/10.1080/00224490903132051 Medline: 19662565
Weiss, R.L. (1980). Strategic behavioral marital therapy: Toward a model for assessment and intervention. In J.P. Vincent (Ed.), Advances in Family Intervention, Assessment and Theory (Vol. 1, pp. 229-271). Greenwich, CT: JAI Press.
Willoughby, B.J., Farero, A.M., & Busby, D.M. (2014). Exploring the effects of sexual desire discrepancy among married couples. Archives of Sexual Behavior, 43(3), 551-562. http://dx.doi.org/ 10.1007/s 10508-013-0181 -2 Medline:24045904
Willoughby, B.J., & Vitas, J. (2012). Sexual desire discrepancy: The effect of individual differences in desired and actual sexual frequency on dating couples. Archives of Sexual Behavior, 41(2), 477-486. http://dx.doi.org/10.1007/s 10508-011 -9766-9 Medline:21573707
Wood, J.R., Milhausen, R.R., & Jeffrey, N.K. (2014). Why have sex? Reasons for having sex among lesbian, bisexual, queer, and questioning women in romantic relationships. Canadian Journal of Human Sexuality, 23(2), 75-88. http://dx.doi.org/10.3138/ cjhs.2592
Zilbergeld, B., & Ellison, C.R. (1980). Desire discrepancies and arousal problems in sex therapy. In S.R. Leiblum 8; L.A. Pervin (Eds.), Principles and Practice of Sex Therapy (pp. 65-101). New York: Guilford Press.
Siobhan E. Sutherland,  Uzma S. Rehman,  Erin E. Fallis,  and Jackson A. Goodnight 
 Department of Psychology, University of Waterloo, Waterloo, ON
 Department of Psychology, University of Dayton, Dayton, OH
Correspondence concerning this article should be sent to Siobhan E. Sutherland, Department of Psychology, University of Waterloo, 200 University Ave West, Waterloo, ON N2L 3G1 Email: Siobhan.firstname.lastname@example.org
Table 1. Results for Item 1 (In general, how does your sexual desire level compare to that of your partner?) Female Male Higher than partner 25% 56% (N = 30) (N = 40) Lower than partner 39% 20% (N = 47) (N = 14) Equal to partner 36% 24% (N = 43) (N = 17) Table 2. Results for Item 2 (How different is your sexual desire level from that of your partner?) Female Male No difference 26% 23% (N = 31) (N = 16) Slightly different 28% 32% (N = 34) (N = 22) Somewhat different 24% 21% (N = 29) (N = 15) Quite different 13% 17% (N = 16) (N = 12) Very different 8% 7% (N = 10) (N = 5) Table 3. Results for Item 3 (Rate which statement best captures yours and your partner's desire levels) Female Male Partner's desire is much higher 12% 6% (N = 14) (N = 4) Partner's desire is slightly higher 23% 13% (N = 28) (N = 9) Desire levels are equal 35% 24% (N = 42) (N = 17) Own desire is slightly higher 22% 40% (N = 27) (N = 28) Own desire is much higher 8% 17% (N = 9) (N = 12)
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Research Presented at the 41st Annual Meeting of the Canadian Sex Research Forum, Kingston, Ontario, October 23-25, 2014|
|Author:||Sutherland, Siobhan E.; Rehman, Uzma S.; Fallis, Erin E.; Goodnight, Jackson A.|
|Publication:||The Canadian Journal of Human Sexuality|
|Date:||Aug 1, 2015|
|Previous Article:||Condom use behavioural skills mediate the relationship between condom use motivation and condom use behaviour among young adult heterosexual males:...|
|Next Article:||The sexual side of castration narratives: fiction written by and for eunuchs and eunuch "wannabes".|