Content uploaded by Amy Muise
Author content
All content in this area was uploaded by Amy Muise on Apr 21, 2016
Content may be subject to copyright.
Article
Sexual Frequency Predicts Greater
Well-Being, But More is Not Always Better
Amy Muise
1
, Ulrich Schimmack
1
, and Emily A. Impett
1
Abstract
Is it true that engaging in more frequent sex is associated with greater well-being? The media emphasizes—and research
supports—the claim that the more sex you have, the happier you will feel. Across three studies (N¼30,645), we demonstrate
that the association between sexual frequency and well-being is best described by a curvilinear (as opposed to a linear) association
where sex is no longer associated with well-being at a frequency of more than once a week. In Study 1, the association between
sexual frequency and well-being is only significant for people in relationships. In Studies 2 and 3, which included only people in
relationships, sexual frequency had a curvilinear association with relationship satisfaction, and relationship satisfaction mediated
the association between sexual frequency and well-being. For people in relationships, sexual frequency is no longer significantly
associated with well-being at a frequency greater than once a week.
Keywords
sexuality, well-being, relationships
Sex is like money; only too much is enough.
John Updike
When it comes to sex, is it true that one can never have
enough? Or is there an optimal sexual frequency after which
sex is no longer associated with greater well-being? Popular
messages in the media often imply that engaging in more fre-
quent sex is better for relationship quality. In one example, a
New York Times article reported on two couples who ‘‘kick-
started their marriage’’ by having sex every day for a year
(Gardner, 2008). Several studies have documented a positive
linear association between sexual frequency and romantic
relationship satisfaction (Brezsnyak & Whisman, 2004;
Byers, 2005; Call, Sprecher, & Schwartz, 1995). Having
more frequent sex is also associated with greater overall
well-being (Cheng & Smyth, 2015; Laumann, Gagnon,
Michael, & Michaels, 1994). For example, data from the
U.S. General Social Survey (GSS) have documented a posi-
tive linear association between self-reported sexual fre-
quency and happiness—the more sex people reported, the
happier they felt (Blanchflower & Oswald, 2004).
One explanation for a linear association between sexual
frequency and well-being is that engaging in sex is associated
with positive emotions and therefore, the more sex you have,
the happier you feel. In a study where people provided
reports of their daily activities and associated affect, sex
was the activity rated as producing the most positive affect
(Kahneman & Krueger, 2006; Kahneman, Krueger, Schkade,
Schwartz, & Stone, 2004), and one strategy to maximize
well-being may be for people to reallocate their time to
engage in the particular activities that hold the greatest poten-
tial to enhance well-being (see review by Lyubomirsky &
Layous, 2013). However, for couples with busy lives, work
responsibilities, and children to care for, feeling the pressure
to engage in sex as frequently as possible may be daunting
and even stressful. Perhaps the popular perception and previ-
ous research evidence suggesting that sex will continue to
enhance well-being at higher frequencies is misguided; peo-
ple may be able to engage in sex frequently enough to max-
imize their well-being without aiming to engage in sex as
frequently as possible. As such, the link between sexual fre-
quency and well-being might be better characterized by a
curvilinear association—where sex is no longer significantly
associated with well-being after a certain frequency—than a
linear association—where the association between sex and
well-being is consistent across frequencies—although this
possibility has never been empirically tested.
Our prediction that the association between sexual fre-
quency and well-being is curvilinear draws upon an important
model that emerges from clinical perspectives in sex therapy.
1
Department of Psychology, University of Toronto Mississauga, Mississauga,
Ontario, Canada
Corresponding Author:
Amy Muise, Department of Psychology, University of Toronto Mississauga,
3359 Mississauga Road, Mississauga, Ontario, Canada, L5L 1C6.
Email: amy.muise@utoronto.ca
Social Psychological and
Personality Science
2016, Vol. 7(4) 295-302
ªThe Author(s) 2015
Reprints and permission:
sagepub.com/journalsPermissions.nav
DOI: 10.1177/1948550615616462
spps.sagepub.com
by guest on April 4, 2016spp.sagepub.comDownloaded from
The ‘‘good enough sex’’ model acknowledges that it is impor-
tant for couples to engage in sexual intimacy to maintain
satisfying romantic relationships but also to hold realistic
expectations about their sex life (McCarthy & Metz, 2008;
Metz & McCarthy, 2007). Applied to the current research,
if couples are having sex frequently enough to maintain their
intimate connection, then they might be maximizing their
well-being. In a recent study in which a group of couples
was asked to double their sexual frequency, no increases in
well-being were observed for these couples compared to
those in a control group who maintained their current fre-
quency (Loewenstein, Krishnamurti, Kopsic, & McDonald,
2015). Since the couples in this study were already engaging
in sex regularly (i.e., about 5 times per month on average), it
is possible that they had already maximized the benefits for
their well-being.
While no existing empirical work has tested a possible
curvilinear association between sexual frequency and well-
being, recent research in positive psychology and the
affective sciences suggests that there are some cases in which
seemingly positive experiences and emotions are associated
with neutral or even negative consequences (Gruber, 2011;
Mauss, Tamir, Andersen, & Savino, 2011; McNulty & Finc-
ham, 2012). For example, although happiness is associated
with several benefits, including better health and greater
social connection, extremely elevated positive emotions can
have negative consequences (Gruber, 2011), and high levels
of happiness valuation can actually lead to less happiness
(Mauss et al., 2011). Even highly pleasurable activities such
as socializing with friends have been judged as having more
value at moderate levels than at higher amounts (Diener, Ng,
& Tov, 2008), and people at higher levels of income do not
report significantly greater happiness than those making
moderate incomes (Kahneman, Krueger, Schkade, Schwartz,
& Stone, 2006). In terms of sex, evidence suggests that the
benefits of engaging in sex are not limited to momentary
increases in positive affect, but that positive feelings are car-
ried over, at least to the next day (Burleson, Trevathan, &
Todd, 2007), and sexual experiences are associated with
broader feelings of satisfaction with the relationship (Brezs-
nyak & Whisman, 2004; Byers, 2005; Call et al., 1995; for a
review see Impett, Muise, & Peragine, 2014). It is unlikely
then that couples would have to engage in sex daily in order
to maximize the benefits. Instead, if couples are engaging in
sex frequently enough to feel satisfied with their relation-
ship, they are optimizing well-being. What we do not yet
know from previous work is at what sexual frequency, on
average, there is no longer a significant association with
greater well-being.
In the current set of studies, we expect sexual frequency to
have a curvilinear association with well-being where greater
sexual frequency is associated with greater satisfaction, but that
this association is no longer significant at higher frequencies.
We also expected a curvilinear association with romantic rela-
tionship satisfaction and for this to moderate the association
between sexual frequency and well-being as relationship
satisfaction tends to be closely associated with overall
well-being (for a review see Heller, Watson, & Iles, 2004).
In Study 1, we test whether a curvilinear association
better explains the relationship between sexual frequency and
well-being than a linear association, and whether the effect is
consistent for people who are in romantic relationships and
those who are single. In Studies 2 and 3, we test the prediction
that there is a curvilinear association between sexual frequency
and romantic relationship satisfaction, and that relationship
satisfaction mediates the association between sexual frequency
and well-being for people in relationships. Across all studies,
we also test whether the association between sexual frequency
and well-being changes as a function of other important predic-
tors of sexual frequency in romantic relationships, including
gender, age, and relationship duration.
Study 1
Method
Study 1 includes data from the GSS, a high-powered, nationally
representative survey conducted almost annually in the United
States for the last 40 years (there have been 29 GSS surveys
since 1972). Our analyses included all 14 GSS time points from
1989 to 2012 in which our key variables of interest (sexual fre-
quency and happiness) were measured. The data are a replicat-
ing cross-sectional design in which a new sample of
participants was selected for each survey year (Smith, Mars-
den, Hout, & Kim, 2013). Our analyses included 25,510 parti-
cipants (11,285 men and 14,225 women) who completed all of
our key variables of interest across the 14 time points. Partici-
pants range in age from 18 to 89 (M¼45.13, SD ¼16.94). The
variables included in the analyses were sexual frequency in the
last year (‘‘About how often did you have sex during the last 12
months?’’) rated on a 7-point scale (0 ¼not at all,1¼once or
twice,2¼once a month,3¼2–3 times a month,4¼weekly,5
¼2–3 times per week, and 6 ¼4 or more times per week; M ¼
2.85, SD ¼1.97) and general happiness (‘‘Taken all together,
how would you say things are these days—would you say that
you are very happy, pretty happy, or not too happy?’’). We
reverse coded this item so that higher values represent more
happiness (1 ¼not too happy,2¼pretty happy, and 3 ¼very
happy;M¼2.19, SD ¼.63). Participants were also asked about
their marital status each year, and in 1996 and 1998, unmarried
participants were asked whether they had a current romantic
involvement. For the purpose of testing a moderation of our
effects by relationship status, we selected people who reported
either being currently married or having a current romantic
involvement (coded as 1; N¼16,935) and people who reported
that they were unmarried without a current romantic involve-
ment (coded as 1; N¼7,856). The GSS includes data on the
length of marital relationships (computed by subtracting parti-
cipants’ current age from the age they reported marrying their
spouse), participants’ marital length ranged from 1 year to 73
years (M¼23.01, SD ¼15.08).
296 Social Psychological and Personality Science 7(4)
by guest on April 4, 2016spp.sagepub.comDownloaded from
Data Analytic Strategy
We computed a curvilinear sexual frequency variable by squar-
ing the sexual frequency variable (centered around the mean)
and entered both the linear and curvilinear sexual frequency
variables as predictors of general happiness in a regression
model. A significant curvilinear term in these analyses indi-
cates a nonlinear association between sexual frequency and
happiness. Then, to determine whether the effect is consistent
across various demographic characteristics of our sample, we
tested whether either the linear or curvilinear association
between sexual frequency and happiness is moderated by rela-
tionship status (1¼single and 1 ¼in a relationship), gender
(1¼woman and 1 ¼man), age, or relationship duration.
Results
The results did, in fact, demonstrate a significant linear associ-
ation between sexual frequency and happiness, b¼.12,
t(23,249) ¼17.81, p< .001. When we included the curvilinear
(squared) term in the equation, the linear effect remained sig-
nificant, b¼.14, t(23,249) ¼19.23, p< .001, but the curvi-
linear term was also significant, b¼.05, t(23,249) ¼
7.12, p< .001. This pattern of results indicates that a curvi-
linear association best explains the association between sexual
frequency on happiness. Both the linear and quadratic effects of
sexual frequency on happiness, however, were moderated by
relationship status, b¼.05, t(17,352) ¼5.42, p< .001 and
b¼.03, t(17,352) ¼2.28, p¼.02, respectively. As
depicted in Figure 1, for single people, neither the linear,
b¼.02, t(5,658) ¼1.27, p¼.21, nor quadratic effects, b¼
.02, t(5,658) ¼1.14, p¼.25, of sexual frequency on happiness
were significant, whereas both the linear, b¼.10, t(11,693) ¼
9.18, p< .001, and quadratic effects, b¼.03, t(11,693) ¼
2.21, p¼.03, were significant for people in relationships.
We also conducted these analyses using only data from
1996 and 1998, two time points of the GSS where relation-
ship status was more accurately assessed (i.e., participants
reported their relationship status, not just their marital status,
in these years) and the results were consistent. Using a tech-
nique described by Nelson and Simonsohn (2014), we con-
firmed that there is no longer an association between
sexual frequency and satisfaction with life at higher frequen-
cies, which, based on graphing the results, was at a fre-
quency of more than once a week (see Figure 1). That is,
there was a significant linear relationship between sexual
frequency and well-being for people having sex once a week
or less, b¼.09, t(8,278) ¼8.34, p< .001, and no association
for people having sex more than once a week, b¼.01,
t(3,414) ¼.34, p¼.74.
Finally, we examined the consistency of this effect across
demographic characteristics of the sample including gender,
age, and relationship duration. Relationship duration was
only reported for married participants, so the analyses includ-
ing relationship duration only include married participants.
Although women, t(25,508) ¼17.26, p< .001, people who
are older (r¼.46, p< .001), and those in longer relation-
ships (r¼.60, p< .001) reported engaging in less frequent
sex, the results held when we controlled for gender, age, and
relationship duration, and there were no significant modera-
tions by any of these variables (all ps > .13). These results
suggest that our findings are consistent across younger and
older people, for both men and women and for people in mar-
riages of longer and shorter duration.
Study 2
Method
In Study 2, we sought to replicate the effects from Study 1 in an
independent sample of people in romantic relationships as well
as test our prediction that one reason why people in relation-
ships report greater well-being when having more frequent sex
is because sexual frequency is associated with greater relation-
ship satisfaction. In Study 2, we also improved on the measure-
ment of well-being by using the Satisfaction with Life Scale
(Diener, Emmons, Larsen, & Griffin, 1985), a validated
multi-item measure of well-being. Also, in order to convey the
magnitude of the effects in a more practical way, we compared
the association between sexual frequency and well-being to the
association between income and well-being.
Participants currently involved in a romantic relationship
were recruited through Amazon’s Mechanical Turk. We
recruited 395 participants, but 16 participants (4%)reported
that they were not currently in a romantic relationship, and
an additional 44 participants (12%) did not pass an attention
check embedded in the survey, therefore their data were not
included in the current analyses. The final sample included
335 participants (138 men and 197 women). In Study 2, we
1
1.5
2
2.5
3
Once a
month
2-3 times
per
month
Not at all Once or
twice
Weekly 2-3 times
per week
4+ times
per week
Romantic Relationship Single
Happiness
Sexual Frequency in the Past 12 Months
Figure 1. Moderating effect of relationship status on the quadratic
association between sexual frequency in the past 12 months and self-
reported happiness (Study 1). Note. Bars represent 95% confidence
intervals.
Muise et al. 297
by guest on April 4, 2016spp.sagepub.comDownloaded from
have 80%power to detect a small effect (R
2
¼.03) in a
model with two predictors (i.e., the linear and curvilinear
effects) at an aof .05. Participants ranged in age from 18
to 64 years (M¼31.0, SD ¼9.1) and comprised a diverse
range of ethnic backgrounds; 65%were European, 10%
were African American, 9%were Asian, 4.5%were Latino
or Mexican, 2%were Native American, 1.5%were Indian,
and 8%self-identified as ‘‘other.’’ Most participants were
married or cohabitating (84%), and the majority of partici-
pants (90%) identified as heterosexual. Participants had
been in their current relationship for between 4 months and
30 years (M¼7.5 years, SD ¼8.4 years). Each participant
was paid US$.60 for completing the 20-minute online
survey.
Measures
Sexual frequency. Participants were asked to indicate how fre-
quently, on average, they engaged in sex with their romantic
partner (1 ¼less than a once a month, 2¼about once a month,
3¼2–3 times per month, 4¼once a week, 5¼multiple times
per week, and 6 ¼daily;M¼4.03, SD ¼1.37).
Income. Participants indicated their annual household income
using the following categories: ‘‘under US$15,000,’’
‘‘US$15,001–US$25,000,’’ ‘‘US$25,001–US$35,000,’’
‘‘US$35,001–US$50,000,’’ ‘‘US$50,001–US$75,000,’’
‘‘US$75,001–US$100,000’’ and ‘‘over US$100,000.’’ The
median income level reported was US$35,001–US$50,000.
Satisfaction with life. Participants responded to the 5-item Satis-
faction with Life Scale (Diener et al., 1985), which included
items such as: ‘‘I am satisfied with my life.’’ Items were rated
on a 7-point scale from 1 ¼strongly disagree to 7 ¼strongly
agree (a¼.90, M¼4.49, SD ¼1.46).
Relationship satisfaction. We assessed participants’ relationship
satisfaction with the 5-item satisfaction subscale (a¼.97,
M¼6.94, SD ¼2.04) of the Investment Model Scale (Rusbult,
Martz, & Agnew, 1998). Items from this measure, such as ‘‘I
feel satisfied with our relationship,’’ were rated on a 9-point
scale (1 ¼do not agree to 9 ¼agree completely).
Data Analytic Strategy
We created a curvilinear sex frequency variable by squar-
ing the linear variable (centered around the mean). To test
our mediation models, we constructed a 95%confidence
interval (CI) for the indirect effect using bootstrapping
techniques with 5,000 resamples using the INDIRECT
SPSS macro (Preacher & Hayes, 2008; Preacher & Selig,
2010). The indirect effect is present when the CI does
not include zero. Finally, we tested whether our effects
were moderated by gender, age, or relationship duration.
Table 1 displays the bivariate correlations between all study
variables.
Results
As we predicted and consistent with the results from our first
study, sexual frequency had a positive linear association with
satisfaction with life, b¼.16, t(322) ¼2.30, p¼.02,aswell
as a significant curvilinear association, b¼.15, t(322) ¼
2.17, p¼.03. Sexual frequency also had a positive linear
association with relationship satisfaction, b¼.35, t(322) ¼
5.42, p< .001, and a significant curvilinear association,
b¼.20, t(322) ¼3.18, p¼.002. As predicted, there was
a significant indirect effect of sexual frequency (curvilinear)
on well-being through relationship satisfaction (95%CI ¼
[.09, .02]); when relationship satisfaction was included
in the model, it significantly predicted satisfaction with life,
b¼.51, t(319) ¼9.18, p< .001, and both the linear and cur-
vilinear associations between sexual frequency and well-
being dropped to nonsignificance, b¼.05, t(319) ¼
.79, p¼.43 and b¼.01, t(319) ¼.20, p¼84, respec-
tively. As depicted in Figure 2 and consistent with Study 1,
additional analyses (Nelson & Simonsohn, 2014) provided
evidence that there was no association between sexual fre-
quency and well-being at a frequency greater than once a
week. That is, for people having sex once a week or less,
there was a significant linear association between sexual
frequency and satisfaction with life, b¼.35, t(157) ¼
4.59, p< .001, and between sexual frequency and relation-
ship satisfaction, b¼.50, t(157) ¼7.13, p< .001, but no sig-
nificant association for people having sex more than once a
week, b¼.05, t(164) ¼.65, p¼.52 and b¼.11,
t(164) ¼1.36, p¼.17, respectively. The effects remained
significant when controlling for gender, age, and relationship
duration, and were not moderated by any of these variables
(ps > .21), suggesting that the pattern of results held for men
and women, people of different ages, and those in both longer
and shorter relationships.
Next, we conducted reverse mediation analyses to test alter-
native directions of the effects. We did not find support for a
model in which sexual frequency mediates the link between
relationship satisfaction and satisfaction with life (95%CI ¼
[.01, .02]. We do, however, find support for a model in which
satisfaction with life mediates the link between sexual fre-
quency and relationship satisfaction (95%CI ¼[.10,
.003] and a top-down model in which relationship satisfac-
tion mediated the link between satisfaction with life and sexual
frequency (95%CI ¼[.17, .34]). In these models, the mediator
accounts for 100%and 95%, respectively, of the association
between the predictor and the outcome variable, and in our pre-
dicted model, the mediator accounts for 100%of the effect. It is
important to note that the top-down model only includes the
linear sexual frequency variable since it is the dependent vari-
able, and this model does not include the curvilinear effect.
Finally, to convey the magnitude of the effects in more prac-
tical terms, in Figure 2 we graphed the association between
sexual frequency and satisfaction with life with the association
between income and satisfaction with life. We calculated effect
sizes for the mean difference between people who reported
298 Social Psychological and Personality Science 7(4)
by guest on April 4, 2016spp.sagepub.comDownloaded from
having sex less than once a month compared to once a week
(d¼.91, 95%CI ¼[.46, 1.38]) and between people who
reported making US$15–US$25,000 per year compared to
US$50–US$75,000 (d¼.42, 95%CI ¼[.06, .79]). As
depicted in Figure 2, the increase in well-being gained from
engaging in sex less than once a month compared to once a
week is larger than the increase in well-being gained from
making between US$15–US$25,000 per year and making
US$50–US$75,000 per year.
Study 3
Method
In Study 3, we use data from the National Survey of Families
and Households (NSFH) to conduct a third test of our predic-
tion that a curvilinear association better describes the rela-
tionship between sexual frequency and well-being than a
linear association for people in romantic relationships. The
NSFH is a three-wave, 14-year national study of married cou-
ples in the United States. Our analyses included all married
(mixed-sex) couples who completed at least one wave of the
NSFH (N ¼2,400 couples), and 1,321 couples (55%)com-
pleted all three waves. We have 99%power to detect a small
effect with two predictors at an aof .05. A detailed explana-
tion of the content and design of the NSFH is reported in
Sweet, Bumpass, and Call (1988). The variables included
in our analyses were both partners’ reports of sexual fre-
quency (‘‘About how often did you and your husband/wife
have sex in the past month?’’ and participants entered the
number of times; M
T1
¼8.06, M
T2
¼6.48, and M
T3
¼
5.12). They also completed measures of relationship satisfac-
tion (‘‘Taking things all together, how would you describe
your marriage?’’ from 1 ¼very unhappy to 7 ¼very happy;
M
T1
¼6.09, M
T2
¼5.90, and M
T3
¼6.14) and happiness
(‘‘Taking all things together, how would you say things are
these days?’’ from 1 ¼very unhappy to 7 ¼very happy;
M
T1
¼5.67, M
T2
¼5.51, and M
T3
¼5.75).
Data Analytic Strategy
We used multi-level modeling to account for the nonindepen-
dence in the data. We conducted a three-level model with ran-
dom intercepts where time point was nested within person and
person was nested within couple. In this study, we report the
standard errors for each effect (as opposed to the CIs) and the
unstandardized bs. Romantic partners’ reports of sexual fre-
quency were highly correlated with each other across all three
time point (rs¼.59 to .71, ps < .001), so we created a couple-
level sexual frequency variable for each time point using the
mean of partners’ ratings. This variable centered represented
the linear effect of sexual frequency. We then squared the cen-
tered variable to represent the curvilinear sexual frequency
variable. Both the linear and curvilinear sexual frequency vari-
ables were entered as predictors of relationship satisfaction and
general happiness. To test our prediction that relationship satis-
faction mediates the association between sexual frequency and
happiness, we used the Monte Carlo Method for Assessing
Mediation (Selig & Preacher, 2008). Table 2 displays the cor-
relations among all variables.
Results
Consistent with our predictions and the results of our previous
two studies, we found that sexual frequency had a significant
linear association with relationship satisfaction, b¼.04,
SE ¼.004, t(5,664.20) ¼10.77, p< .001, and a significant
Table 1. Bivariate Correlation Matrix (Study 2).
1. 2. 3. 4. 5. 6. 7.
1. Sexual frequency — .48*** .26*** .03 .34*** .34*** .09
2. Relationship satisfaction — .53*** .10 .22 .19** .03
3. Satisfaction with life — .14* .16** .11 .15**
4. Gender — .05 .08 .01
5. Age — .70*** .30***
6. Relationship length — .27***
7. Income —
Note. Gender is coded as 0 ¼men and 1 ¼women.
***p< .001. **p< .01. *p< .05.
Figure 2. Curvilinear association between sexual frequency and
satisfaction with life compared to the assocaition between income and
satisfaction with life (Study 2). Note. Bars represent 95% confidence
intervals.
Muise et al. 299
by guest on April 4, 2016spp.sagepub.comDownloaded from
curvilinear association, b¼–.01, SE ¼.0003, t(5,493.39) ¼
–3.70, p< .001. Sexual frequency also had a significant linear
association with happiness, b¼.02, SE ¼.004, t(5,271.49) ¼
4.76, p< .001, but the curvilinear effect of sexual frequency on
happiness did not reach significance, b¼–.0005, SE ¼.0004,
t(5,410.47) ¼–1.27, p¼.21. There was, however, a significant
indirect effect for both the linear effect (CI ¼[.01, .02]) and
curvilinear effect (CI ¼[.005, .003]) of sexual frequency
on happiness through relationship satisfaction. Relationship
satisfaction predicted overall happiness, b¼.36, SE ¼.01,
t(5,403.84) ¼28.74, p< .001, and when entered into the model
with both sexual frequency variables, the associations between
linear sexual frequency, b¼.004, SE ¼.003, t(5,112.36) ¼
1.13, p¼.26, and curvilinear sexual frequency, b¼–.00001,
SE ¼.0003, t(5,352.27) ¼–.15, p¼.88, with happiness were
significantly reduced. As shown in Figure 3, we confirmed
that sex and relationship satisfaction were not significantly
associated at a frequency of more than once a week. For cou-
ples having sex approximately weekly (6 times or less per
month), there was a significant linear association between sex-
ual frequency and relationship satisfaction, b¼.04, SE ¼.01
t(3,537.98) ¼4.24, p< .001, whereas for those having sex more
often, the association was not significant, b¼.01, SE ¼.01
t(1,473.48) ¼1.15, p¼.25. None of the effects of sexual fre-
quency on well-being were moderated by gender (all ps > .50)
or by time point (these analyses would also account for age and
relationship duration for couples who participate in all three
time points; all ps > .10).
Finally, as in Study 2, we conducted reverse mediation
analyses. We found some support for a model in which sexual
frequency mediates the link between relationship satisfaction
and happiness (95%CI ¼[.007, .003]), however the indirect
effect is small, sexual frequency accounts for 7%of the asso-
ciation between relationship satisfaction and happiness. In our
predicted model, relationship satisfaction accounts for 95%of
the association between sexual frequency and satisfaction
with life. There is no significant curvilinear association
between sexual frequency and happiness, so we do not find
support for a model where happiness mediates that association
between sexual frequency and relationship satisfaction. As in
Study 2, we find support for a top-down model in which
relationship satisfaction mediated the link between satisfac-
tion with life and sexual frequency (95%CI ¼[.06, .19]);
in this model, the mediator accounts for 14%of the associa-
tion and only includes the linear (and not the curvilinear) sex-
ual frequency variable.
Discussion
The current set of studies help dispel the notion that sex has
limitless benefits for well-being and, instead, indicate that
at least for people in romantic relationships, sexual frequency
is no longer significantly associated with well-being at a fre-
quency greater than once a week. Consistent with our theore-
tical rationale, the current findings suggest that one reason
why greater sexual frequency is associated with greater
well-being for people in relationships is that having more fre-
quent sex (up to about once a week) is associated with greater
relationship satisfaction. In terms of single people, in Study 1
we found no linear or curvilinear association for people not
currently in romantic relationships. Likely, there are
Table 2. Bivariate Correlation Matrix (Study 3).
1. 2. 3. 4. 5. 6. 7. 8. 9.
1. Sexual frequency (T1) — .56*** .43*** .14*** .07*** .04* .11*** .03 .05**
2. Sexual frequency (T2) — .47*** .11*** .16*** .08*** .07*** .12*** .07***
3. Sexual frequency (T3) — .04* .06*** .11*** .05*** .02 .08***
4. Relationship satisfaction (T1) — .34*** .20*** .47*** .25*** .16***
5. Relationship satisfaction (T2) — .30*** .26*** .45*** .23***
6. Relationship satisfaction (T3) — .15*** .20*** .35***
7. Happiness (T1) — .32*** .26***
8. Happiness (T2) — .28***
9. Happiness (T3) —
Note. T1 ¼Time 1; T2 ¼Time 2; T3 ¼Time 3.
***p< .001. **p< .01. *p< .05.
4
5
6
7
Not at all in last
month
1-2 in the last
month
Weekly 2-3 per week 4 or more
times per wee
k
Relationship Satisfaction
Sexual Frequency in the Last Month
Figure 3. Curvilinear association between sexual frequency in the last
month and romantic relationship satisfaction (Study 3). Note. Bars
represent 95% confidence intervals. We created sexual frequency
categories for graphing purposes only, and the continuous sexual
frequency variable was used in the analyses. The y-axis does not
include the whole scale.
300 Social Psychological and Personality Science 7(4)
by guest on April 4, 2016spp.sagepub.comDownloaded from
important moderators that influence the association between
sexual frequency and well-being for single people (for a
review of casual sex and well-being see Vrangalova, 2015),
an area that is ripe for future research.
Although the results of three studies with over 30,000 parti-
cipants point resoundingly to the conclusion that the associa-
tion between sexual frequency and well-being is curvilinear,
what is not clear from the current research is why sexual fre-
quency is no longer associated with well-being at frequencies
greater than once a week. The average amount of sex reported
in established relationships is approximately once a week
(Blanchflower & Oswald, 2004; Call et al., 1995; Laumann
et al., 1994), so perhaps this tends to be the average because
engaging in sex more frequently is no longer associated with
well-being. It is also possible that couples feel satisfied as long
they think they are engaging in the amount of sex that is con-
sidered to be average for couples of their relationship status and
duration. Consistent with this possibility, one study found that
happiness was positively associated with one’s own frequency
but was negatively associated with the actual sexual frequency
of one’s peers (Wadsworth, 2014). It is not clear from this
work, however, if people are aware of the average sexual fre-
quency and feel better if they believe they are at or above this
frequency. It is also possible that the point at which sex is no
longer associated with greater well-being differs based on
demographic factors or individual differences, such as a per-
son’s ideal sexual frequency. Although the curvilinear effect
was not moderated by age, gender, or relationship length in the
current studies, it is possible that the point at which there is no
longer an association between sexual frequency and well-being
could differ based on these factors.
It is important to note that the current set of studies all exam-
ined links between naturally occurring sexual frequency and
well-being. We cannot make causal claims and, in fact, we find
evidence for both top-down and bottom-up effects for the asso-
ciation between sexual frequency and well-being, consistent
with research on other indicators of well-being (Brief, Butcher,
George, & Link, 1993; Nakasato, Schimmack, & Oishi, 2011).
Experimental research could provide evidence for the direction
of this effect, however, in a recent study couples did not report
greater well-being when they were instructed to double their
sexual frequency (Loewenstein et al., 2015). The authors sug-
gested that the directive of being asked to increase sexual fre-
quency removed partners’ intrinsic motivation to engage in sex
and therefore made sex less enjoyable. The couples in this
study, however, were already having sex about once a week
(i.e., 5 times a month) at baseline. An interesting avenue for
future research would be to test whether increasing sexual fre-
quency benefits couples who are having sex less frequently
than once a week, but the research by Loewenstein et al. high-
lights the challenges of conducting experimental work in the
domain of sexuality.
In John Updike’s statement at the opening of the article, he
suggests that there are limitless benefits to engaging in sex (and
making more money) in that a person can never get enough.
Our research demonstrates, however, that although greater
sexual frequency is associated with greater well-being, more
is not always better. Instead, sex may be like money—only too
little is bad.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to
the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for
the research, authorship, and/or publication of this article: This work
has been supported by a Social Science and Humanities Research
Council (SSHRC) Banting Postdoctoral Fellowship awarded to the
first author, a SSHRC Insight Grant awarded to the second author, and
a SSHRC Insight Grant and Insight Development Grant awarded to the
last author.
References
Blanchflower, D. G., & Oswald, A. J. (2004). Money, sex and happi-
ness: An empirical study. Scandinavian Journal of Economics,
106, 393–415. doi:10.1111/j.1467-9442.2004.00369.x
Brezsnyak, M., & Whisman, M. A. (2004). Sexual desire and relation-
ship functioning: The effects of marital satisfaction and power.
Journal of Sex & Marital Therapy,30, 199–217.
Brief, A. P., Butcher, A. H., George, J. M., & Link, K. E. (1993). Inte-
grating bottom-up and top-down theories of subjective well-being:
The case of health. Journal of Personality and Social Psychology,
64, 646–653.
Burleson, M. H., Trevathan, W. R., & Todd, M. (2007). In the mood
for love or vice versa? Exploring the relations among sexual activ-
ity, physical affection, affect, and stress in the daily lives of mid-
aged women. Archives of Sexual Behavior,36, 357–368.
Byers, E. S. (2005). Relationship satisfaction and sexual satisfaction:
A longitudinal study of individuals in long-term relationships.
Journal of Sex Research,42, 113–118.
Call, V., Sprecher, S., & Schwartz, P. (1995). The incidence and fre-
quency of marital sex in a national sample. Journal of Marriage
and Family,57, 639–652.
Cheng, Z., & Smyth, R. (2015). Sex and happiness. Journal of Eco-
nomic Behavior & Organization,112, 26–32.
Diener, E. D., Emmons, R. A., Larsen, R. J., & Griffin, S. (1985). The
satisfaction with life scale. Journal of Personality Assessment,
49, 71–75.
Diener, E., Ng, W., & Tov, W. (2008). Balance in life and declining
marginal utility of diverse resources. Applied Research in Quality
of Life,3, 277–291.
Gardner, R., Jr. (2008, June 8). Yes, dear. Tonight again. The New
York Times. Retrieved from www.nytimes.com/2008/06/08/fash-
ion/08nights.html
Gruber, J. (2011). Can feeling too good be bad? Positive emotion
persistence (PEP) in bipolar disorder. Current Directions in Psy-
chological Science,20,217–221.
Heller, D., Watson, D., & Iles, R. (2004). The role of person versus
situation in life satisfaction: A critical examination. Psychological
Bulletin,130, 574–600.
Muise et al. 301
by guest on April 4, 2016spp.sagepub.comDownloaded from
Impett, E. A., Muise, A., & Peragine, D. (2014). Sexuality in the con-
text of relationships. In L. Diamond & D. Tolman (Eds.), APA
handbook of sexuality and psychology (Vol. 1, pp. 269–316).
Washington, DC: American Psychological Association.
Kahneman,D.,&Krueger,A.B.(2006).Developmentsinthemea-
surement of subjective well-being. The Journal of Economic
Perspectives,20, 3–24.
Kahneman, D., Krueger, A. B., Schkade, D. A., Schwarz, N., & Stone,
A. A. (2004). A survey method for characterizing daily life expe-
rience: The day reconstruction method. Science,306, 1776–1780.
Kahneman, D., Krueger, A. B., Schkade, D. A., Schwarz, N., & Stone,
A. A. (2006). Would you be happier if you were richer? A focusing
illusion. Science,312, 1908–1910.
Laumann, E. O., Gagnon, H. J., Michael, R. T., & Michaels, S. (1994).
The social organization of sexuality: Sexual practices in the United
States. Chicago, IL: The University of Chicago Press.
Loewenstein,G.,Krishnamurti,T.,Kopsic,J.,&McDonald,D.
(2015). Does increased sexual frequency enhance happiness?
Journal of Economic Behavior & Organization,116, 206–218.
Lyubomirsky, S., & Layous, K. (2013). How do simple positive activ-
ities increase well-being? Current Directions in Psychological
Science,22, 57–62.
Mauss, I. B., Tamir, M., Andersen, C. L., & Savino, N. S. (2011). Can
seeking happiness make people unhappy? Paradoxical effects of
valuing happiness. Emotion,11, 807–815.
McCarthy, B., & Metz, M. (2008). The ‘‘Good Enough Sex’’ model: A
case illustration. Sexual and Relationship Therapy,23, 227–234.
McNulty, J. K., & Fincham, F. D. (2012). Beyond positive psychol-
ogy? Toward a contextual view of psychological processes and
well-being. American Psychologist,67, 101–110.
Metz, M. E., & McCarthy, B. W. (2007). The ‘‘Good-Enough Sex’’
model for couple sexual satisfaction. Sexual and Relationship
Therapy,22, 351–362.
Nakasato, N., Schimmack, U., & Oishi, S. (2011). Effects of changes
in living condition on well-being: A prospective top-down bottom-
up model. Social Indicators Research,100, 115–135.
Nelson, L., & Simonsohn, U. (2014, September 14). Thirty-
somethings are shrinking and other u-shaped challenges.
Retrieved from http://datacolada.org/2014/09/17/27-thirty-some-
things-are-shrinking-and-other-u-shaped-challenges/
Preacher,K.J.,&Hayes,A.F.(2008).Asymptoticandresampling
strategies for assessing and comparing indirect effects in multiple
mediator models. Behavior Research Methods,40, 879–891.
Preacher, K. J., & Selig, J. P. (2010). Monte Carlo method for asses-
sing multilevel mediation: An interactive tool for creating confi-
dence intervals for indirect effects in 1-1-1 multilevel models
[Computer software]. Retrieved from http://www.quantpsy.org
Rusbult, C. E., Martz, J. M., & Agnew, C. R. (1998). The investment
model scale: Measuring commitment level, satisfaction level,
quality of alternatives, and investment size. Personal Relation-
ships,5, 357–387.
Selig, J. P., & Preacher, K. J. (2008, June). Monte Carlo method for
assessing mediation: An interactive tool for creating confidence
intervals for indirect effects [Computer software]. Retrieved from
http://quantpsy.org/
Smith,T.W.,Marsden,P.,Hout,M.,&Kim,J.(2011).General Social Sur-
veys, 1972–2010 [Machine-readable data file]. Sponsored by National
Science Foundation. Chicago, IL: National Opinion Research Center.
Sweet, J. A., Bumpass, L. L., & Call, V. (1988). The design and con-
tent of the National Survey of Families and Households. Madison:
University of Wisconsin, Center for Demography and Ecology.
Vrangalova, Z. (2015). Does casual sex harm college students’ well-
being? A longitudinal investigation of the role of motivation.
Archives of Sexual Behavior,44, 945–959.
Wadsworth, T. (2014). Sex and the pursuit of happiness: How other
people’s sex lives are related to our sense of well-being. Social Indi-
cators Research,116, 115–135. doi:10.1007/s11205-013-0267-1
Author Biographies
Amy Muise is a postdoctoral fellow at the University of Toronto Mis-
sissauga. Her research focuses on sexuality in romantic relationships.
Ulrich Schimmack is a professor at the University of Toronto Missis-
sauga. His research aims to contribute to the scientific understanding
of happiness.
Emily A. Impett is an associate professor at the University of Toronto
Mississauga. Her research focuses on how close relationships contrib-
ute to happiness and well-being.
Handling Editor: Simine Vazire
302 Social Psychological and Personality Science 7(4)
by guest on April 4, 2016spp.sagepub.comDownloaded from
