Abstract Historically, the reasons people have sex have
been assumed to be few in number and simple in nature–to
reproduce, to experience pleasure, or to relieve sexual
tension. Several theoretical perspectives suggest that motives
for engaging in sexual intercourse may be larger in
number and psychologically complex in nature. Study 1
used a nomination procedure that identified 237 expressed
reasons for having sex, ranging from the mundane (e.g., ‘‘I
wanted to experience physical pleasure’’) to the spiritual
(e.g., ‘‘I wanted to get closer to God’’), from altruistic (e.g.,
‘‘I wanted the person to feel good about himself/herself’’)
to vengeful (e.g., ‘‘I wanted to get back at my partner for
having cheated on me’’). Study 2 asked participants
(N = 1,549) to evaluate the degree to which each of the 237
reasons had led them to have sexual intercourse. Factor
analyses yielded four large factors and 13 subfactors,
producing a hierarchical taxonomy. The Physical reasons
subfactors included Stress Reduction, Pleasure, Physical
Desirability, and Experience Seeking. The Goal Attainment
subfactors included Resources, Social Status, Revenge, and
Utilitarian. The Emotional subfactors included Love and
Commitment and Expression. The three Insecurity subfactors
included Self-Esteem Boost, Duty/Pressure, and Mate
Guarding. Significant gender differences supported several
previously advanced theories. Individual differences in
expressed reasons for having sex were coherently linked
with personality traits and with individual differences in
sexual strategies. Discussion focused on the complexity of
sexual motivation and directions for future research.
Abstract Historically, the reasons people have sex have
been assumed to be few in number and simple in nature–to
reproduce, to experience pleasure, or to relieve sexual
tension. Several theoretical perspectives suggest that motives
for engaging in sexual intercourse may be larger in
number and psychologically complex in nature. Study 1
used a nomination procedure that identified 237 expressed
reasons for having sex, ranging from the mundane (e.g., ‘‘I
wanted to experience physical pleasure’’) to the spiritual
(e.g., ‘‘I wanted to get closer to God’’), from altruistic (e.g.,
‘‘I wanted the person to feel good about himself/herself’’)
to vengeful (e.g., ‘‘I wanted to get back at my partner for
having cheated on me’’). Study 2 asked participants
(N = 1,549) to evaluate the degree to which each of the 237
reasons had led them to have sexual intercourse. Factor
analyses yielded four large factors and 13 subfactors,
producing a hierarchical taxonomy. The Physical reasons
subfactors included Stress Reduction, Pleasure, Physical
Desirability, and Experience Seeking. The Goal Attainment
subfactors included Resources, Social Status, Revenge, and
Utilitarian. The Emotional subfactors included Love and
Commitment and Expression. The three Insecurity subfactors
included Self-Esteem Boost, Duty/Pressure, and Mate
Guarding. Significant gender differences supported several
previously advanced theories. Individual differences in
expressed reasons for having sex were coherently linked
with personality traits and with individual differences in
sexual strategies. Discussion focused on the complexity of
sexual motivation and directions for future research.
Keywords Sexual motivation Sexual intercourse
Gender differences
Introduction
Why people have sex is an extremely important, but
surprisingly little studied topic. One reason for its relative
neglect is that scientists might simply assume that the
answers are obvious: to experience sexual pleasure, to
relieve sexual tension, or to reproduce. Previous research
already tells us that the answers cannot be as few or
psychologically simple. Leigh (1989), for example, documented
seven reasons for sex: pure pleasure, to express
emotional closeness, to reproduce, because a partner wants
it, to please a partner, to make a conquest, and to relieve
sexual tension. The most comprehensive existing taxonomy,
framed from a theoretical perspective of dispositional sexual
motives, documented eight reasons: to feel valued by a
partner, expressing value for a partner, obtaining relief from
stress, nurturing one’s partner, enhancing feelings of personal
power, experiencing a partner’s power, experiencing
pleasure, and procreating (Hill & Preston, 1996).
Several theoretical perspectives suggest that reasons for
engaging in sexual intercourse might be even more
numerous and complex than even this previous research
suggests. With the exception of ‘‘to make a conquest,’’ most
of the documented reasons for having sex above implicitly
assume the context of an ongoing romantic relationship or
long-term mateship. Sexual strategies theory (Buss &
Schmitt, 1993) and strategic pluralism theory (Gangestad &
Simpson, 2000), however, propose that humans have a
menu of mating strategies, including long-term, short-term,
and extra-pair mating. Thus, there might be reasons for
having sex with a casual sex partner or extra-pair partner,
such as the desire to experience sexual variety (Symons,
1979) or seeking to improve one’s sexual skills (Greiling &
Buss, 2000) that differ from those that motivate sex in the
context of an ongoing romantic relationship.
Even within the context of an ongoing mateship, there
could be numerous reasons for having sex beyond those
already documented. For example, sex might be used to
reward a partner or as a favor in exchange for something the
partner has done. Or sex might be used to punish a partner,
such as when someone engages in a retaliatory affair in
order to exact revenge on a partner for having committed
some violation within the relationship (Greiling & Buss,
2000). Also, within an ongoing relationship, sex might be
used to intensify the relationship, escalate the level of
commitment within the relationship, or turn a short-term
relationship into a long-term relationship (Buss, 2003). In
the clinical literature, Basson (2000) described how women
may engage in sexual intercourse for the ‘‘spin-offs’’ they
receive, such as emotional closeness, bonding, commitment,
love, affection, acceptance, tolerance, and closeness.
From yet another perspective, people might use sex as a
form of ‘‘mate guarding’’ (Buss & Shackelford, 1997). This
could function in one of several possible ways. First, satisfying
a partner sexually might function to deter the partner
from seeking sexual gratification elsewhere. Second, this
strategic use of sexuality might send signals to potential
mate poachers, perhaps by rendering the partner less ‘‘open’’
to extra-pair liaisons, causing potential mate poachers to
choose other potential targets (Schmitt, 2004; Schmitt &
Buss, 2001).
Another perspective comes from the literature on sperm
competition (Baker & Bellis, 1995; Shackelford, Pound,
Goetz, & LaMunyon, 2005). From this perspective, a man
whose partner might have been sexually unfaithful might
seek sex, which functions to displace the sperm of the rival
male. Or a woman might deplete the sperm of her partner,
leaving few available for insemination of rival women.
None of these hypothesized functions, of course, need
operate through conscious psychological mechanisms.
More generally, sex can be viewed as a fungible resource–
something that one person has the potential to give
and something that another person may want. As a soughtafter
resource, sex can be exchanged for other resources.
Exchanging sex for money, as in the case of prostitution, is
one obvious example (Burley & Symanski, 1981). Sex
could also be exchanged for meat, as occurs among many
traditional hunter-gatherer groups such as the Ache of
Paraguay (Hill & Hurtado, 1996). Sex could be exchanged
for favors, special privileges, a preferred job, or indeed for
any resource.
Finally, the psychology of sex does not occur merely
between the individual partners directly involved. Sex
occurs within a broader social and cultural context, with
implications for prestige, status, and reputation (Buss,
2003). Having sex with a high status individual, for
example, might raise a person’s status within the group.
Within some groups, having sex with numerous partners
might enhance a person’s reputation, providing the motivational
impetus for initiating sex. Sex, of course, can
sometimes damage a person’s status and reputation, providing
reasons for avoiding it or concealing it from others
in the group. In sum, because sex has consequences for
status and reputation that can act as incentives (or deterrents),
a person might be motivated to have sex for social
reasons that have nothing to do with the personal relationship
within which it occurs. All of these diverse theoretical
perspectives, when taken together, point to a
singular conclusion: The reasons people have sex are likely
to be far more numerous and psychologically complex than
previous taxonomists have envisioned.
The current research had several primary goals: (1) to
identify a broader array of potential reasons that motivate
people to engage in sexual intercourse using a nomination
procedure designed to survey the wider domain of
reasons; (2) to develop an organized taxonomy of reasons
for sex using a large sample of women and men; (3) to
provide a more comprehensive research tool that can be
used by sex researchers; (4) to identify whether women
and men differ in their expressed reasons for engaging in
sexual intercourse; and (5) to examine whether individual
differences in sexual strategies, as measured by the
Sociosexuality Inventory (Simpson & Gangestad, 1991),
are linked to individual differences in reasons for having
sex.
Regarding gender differences, previous researchers have
explored this issue in a delimited way. Some have found
that men are more motivated by purely physical reasons,
such as physical release or simply because they are ‘‘horny,’’
whereas women are more motivated by emotional
reasons, such as to become psychologically closer to a
partner (e.g., Carroll, Volk, & Hyde, 1985; Denney, Field,
& Quadagno, 1984; Leigh, 1989). Others have found that
men, more than women, have sex in order to provide relief
from stress and to enhance their feelings of personal power
(Hill & Preston, 1996). Several evolution-based theories
suggest that men will be more motivated by the desire for
sexual variety (Symons, 1979), the chance for an opportunistic
copulation (Buss, 2003), the physical appearance
of a potential partner (Buss & Schmitt, 1993), and that
emotional factors, such as expressing love or intensifying
psychological commitment, would figure more prominently
in women’s reasons for having sex (Buss, 2003).
The current research was capable of testing these theories,
as well as providing a more comprehensive atheoretical
description of a potential panoply of gender differences in
the reasons expressed for having sex.
Study 1: Initial Item Generation
Method
Participants were 203 men and 241 women ranging in age
from 17 years to 52 years who were recruited from upper
and lower level psychology classes, graduate classes in
psychology, and from community volunteers who were
participating in several other ongoing studies in the Sexual
Psychophysiology Laboratory at the University of Texas.
Potential volunteers were told the purpose of the study was
to make a comprehensive list of all the reasons why people
engage in sexual intercourse.
Persons who agreed to participate in the study were
asked the following open-ended prompt, ‘‘Please list all the
reasons you can think of why you, or someone you have
known, has engaged in sexual intercourse in the past.’’
Participants were encouraged to list as many responses as
possible. Efforts were made to ensure confidentiality by
having participants sit at a distance from other participants
such that other’s responses could not be easily read. Anonymity
was assured by having participants include only
their age and gender on their answer sheets.
Results
A total of 715 reasons were collected. The authors each
independently reviewed the reasons and noted those that
were either identical or similar with minor wording changes.
Reasons given that both authors marked as similar
were deleted or compiled into one combined response,
resulting in 237 distinct reasons. The 237 reasons (see
Table 3) obtained were listed using a conventional questionnaire
format with each item presented as a brief
descriptive statement to which subsequent participants
rated the likelihood that a given reason has or would lead
them to engage in sexual intercourse. The questionnaire
was termed the Why Have Sex? (YSEX?) questionnaire.
The following instructional set was given at the beginning
of the instrument, ‘‘People have sex (i.e., sexual intercourse)
for many different reasons. Below is a list of some
of these reasons. Please indicate how frequently each of
the following reasons led you to have sex in the past. For
example, if about half of the time you engaged in sexual
intercourse you did so because you were bored, then you
would circle ‘‘3’’ beside question 3. If you have not had sex
in the past, use the following scale to indicate what the
likelihood that each of the following reasons would lead
you to have sex. I have had sex in the past because...’’ The
response choices were listed on a 5-point Likert scale, with
scale interval anchors being None of my sexual experiences
(1), A few of my sexual experiences (2), Some of my sexual
experiences (3), Many of my sexual experiences (4), and All
of my sexual experiences (5).
Study 2: Psychometric Analyses
Method
Participants
A total of 1,549 undergraduate students (503 men, 1046
women) participated in this study in exchange for course
credit. Participants were distinct from those who took part
in Study 1. Participants were enrolled in either the 2000–
2003 Fall sessions of Introductory Psychology courses
(September–December) or the 2001–2004 Spring Introductory
Psychology sessions (January–May). Sample sizes
varied between cohorts (2000–2001, n = 570; 2001–2002,
n = 341; 2002–2003, n = 341; 2003–2004, n = 297). Participants
ranged in age from 16 years to 42 years (96%
between the ages of 18 and 22), and the mean age was
19 years.
The sample consisted of 62% Caucasian, 4% African
American, 15%Hispanic,15% Asian American,<1% Native
American, and 3% ‘‘other ethnicity’’ participants. Ethnicity
was determined with the question, ‘‘What ethnicity do you
most identify with?’’ Region of heritage was also assessed
based on countries of birth or countries of parents’ birth.
Across ethnicities, most participants reported North American
heritage (96% of Caucasians, 91% of African Americans,
94% of Hispanics, 71% of Asian Americans, 67% of
Native Americans, and 59% of ‘‘other’’ ethnicities). The
second most substantial reported heritage was Asian heritage
(23% of Asian Americans and 33% of ‘‘other’’ ethnicities).
Less than 5% of each ethnicity reported South American,
African, Middle Eastern, Western European, Eastern European,
or South Pacific heritage. Religious affiliations varied,
with 36% Fundamentalists, 25% Catholics, 14% Protestants,
14% Atheists/Agnostics, 5% Jewish, 4% Hindus, 2% Buddhists,
1% Muslims, and <1% Pagans/Spiritualists.
Sexual experience data were available for 910 women
and 343 men. Of these, 73% of women and 68% of men
reported having experienced sexual intercourse in the past.
Among both men and women, 88% reported having engaged
in oral sex and 95% reported having engaging in
some form of sexual petting in the past. Four percent of
the women and 2% of men were married. Six percent of
women and 5% of men were living with a sexual partner.
Measures
Reasons for sexual intercourse Motivators for engaging
in sexual intercourse were measured using the 237 item
YSEX? questionnaire that was developed in Study 1.
Personality The Big Five Inventory-short form (BFI) is
a 44 item inventory developed by John, Donahue, and
Kentle (1991) to assess each of the ‘‘Big Five’’ personality
dimensions: Neuroticism, Extraversion, Openness to
Experience, Agreeableness, and Conscientiousness. The 44
self-descriptive statements were measured on a 5-point
scale, with responses ranging from disagree strongly (1) to
agree strongly (5).
Sexual strategies The following five items from the
Sociosexual Orientation Inventory (SOI; Simpson &
Gangestad, 1991) were used to analyze unrestricted sexual
behavior and fantasy: ‘‘With how many partners have you
had sex (intercourse or oral sex) in the past year?’’; ‘‘With
how many partners will you probably have sex with
(intercourse or oral sex) over the next five years?’’; ‘‘With
how many partners have you had sexual intercourse, or oral
sex, on one and only one occasion?’’; ‘‘When in a stable
and committed relationship, how often would you fantasize
about sex with someone other than your current partner?’’;
and ‘‘Sex without love is okay.’’ An SOI composite was
computed by averaging the sum of the five z-transformed
scores.
Procedure
Participants were administered a demographics questionnaire,
the YSEX? questionnaire derived from Study 1, the
short form version of the Big Five Inventory (John et al.,
1991), selected items from the Sociosexual Orientation
Inventory (Simpson & Gangestad, 1991), and a number of
additional sexually relevant questionnaires not reported
here. In each cohort, participants were administered questionnaires
in small groups of 5–10 same-sex individuals.
To preserve privacy, participants were given either partitions
or several feet of space from other participants. Participants
were informed of the sexual nature of the study
before they applied to participate. Same-sex researchers
informed participants of the sexual material, obtained
consent, administered questionnaire packets, and answered
participant questions that arose during testing. Confidentiality
was protected by assigning each participant with a
randomized code number connected to all of his or her
data, and consent forms were kept in a separate file. Upon
completion, questionnaire packets were inserted into a large ‘‘drop box’’ as participants exited the testing room.
Questionnaire data were stored in locked file cabinets, and
all files connected to the study were password-protected.
The Institutional Review Board at the University of Texas
at Austin approved this research from 2000 to 2004.
Results
Item analyses
Most and least frequently endorsed reasons for having
sex There were nine themes that appeared to characterize
the most frequently endorsed reasons for having intercourse:
(1) pure attraction to the other person in general;
(2) experiencing physical pleasure; (3) expression of love;
(4) having sex because of feeling desired by the other; (5)
having sex to escalate the depth of the relationship; (6)
curiosity or seeking new experiences; (7) marking a special
occasion for celebration; (8) mere opportunity; and (9) sex
just happening due to seemingly uncontrollable circumstances.
Table 1 shows the 50 reasons for having sex receiving
the highest endorsements for men and women separately.
The most frequently endorsed reason why both men and
women reported engaging in sexual intercourse was because
they were attracted to the person. As can be seen in
Table 1, out of a pool of 237 items, 8 of the top 10, and 20
of the top 25 reasons given for engaging in sexual intercourse
were similar for men and women.
Table 2 presents the 50 most infrequently endorsed
reasons for having sexual intercourse out of the possible
237 reasons. The least frequently endorsed reasons for
having intercourse could be categorized into five broad
themes. One theme involved wanting to harm another
person, either a partner (e.g., ‘‘I was mad at my partner, so I
had sex with someone else’’), a rival (e.g., ‘‘I wanted to
hurt an enemy’’) or a stranger (e.g., ‘‘I wanted to give
someone else a sexually transmitted disease (e.g., herpes,
AIDS)’’. A second theme involved attaining resources such
as a job, a promotion, money, drugs, or gifts. A third
common theme revolved around enhancing social status
(e.g., ‘‘I wanted to be popular’’; ‘‘I wanted to have more sex
than my friends’’). A fourth theme centered on having sex
as a means to a seemingly unrelated end, such as getting rid
of a headache or menstrual cramps, and a fifth theme
involved having sex out of duty or pressure.
Gender differences in reasons for having sex To examine
gender differences at the item level, given that 237 t-tests
would be conducted, we set the significance threshold at a
conservative level of p < .005. Thus, roughly one difference
that showed significance at this level would be
