Sex Roles, VoL 7, No. 8, 1981
Loving and Leaving: Sex Differences in
Romantic Attachments 1
Ziek Rubin 2
Letitia Anne Peplau
University of California - Los Angeles
Charles T. Hill
University of Washington
We propose a two-part generalization about sex differences in entering into and
giving up romantic attachments: (1) Men tend to fall in love more readily than
women; (2) women tend to fall out of love more readily than men. Evidence in
support of these generalizations is derived from a longitudinal study of 231
college student dating couples. The data suggest that women are more cautious
than men about entering into romantic relationships, more likely to compare
these relationships to alternatives, more likely to end a relationship that seems
ill fated, and better able to cope with rejection. We consider several possible ex-
planations of these sex differences from the standpoints of psychoanalytic
theory, the social and economic context of mate selection, and the socialization
of men and women in the management of their own emotions. To evaluate these
(and any other) explanations, further research might profitably investigate
whether and to what degree these sex differences are found in other segments
of the population.
1This research was supported by National Science Foundation grant GS-27422 to Zick
Rubin. The authors are grateful to Claire Engers, Sherry Ward, and Susan Willard for
their contribution to this research and to Jessie Bernard, Nancy Chodorow, George W.
Goethals, Paul Rosenblatt, Ann Swidler, and Shelley Taylor for their helpful comments.
2 Correspondence should be sent to Zick Rubin, Department of Psychology, Brandeis Uni-
versity, Waltham, Massachusetts 02254.
0360-0025/81/0800-0821503.00/0 © 1981 plenum Publishing Corporation
822 Rubin, Peplau, and Hill
This report compares men's and women's orientations toward beginning and
ending close male-female relationships. Specifically, it considers whether there is
any general difference between the sexes in the propensity or ability to fall in
love in the first place, and in the propensity or ability of one who is in love to
fall out of it.
This is an area of inquiry which has perhaps been explored most thorough-
ly by songwriters, comicbook creators, and other producers of popular culture
both today and in times gone by. It is an area in which stereotypes reign su-
preme. If our reading of the popular wisdom is correct, the most common set of
perceptions holds that of the two sexes, women are the more starry-eyed and
sentimental, while men are the more hardhearted and rational. A woman, accord-
ing to this stereotype, is more likely to fall in love at first sight and to experience
such symptoms as a heightened pulse, a trembling hand, and an itching in her
heart. Meanwhile, the male object of the woman's affection is presumed to re-
main impassive and even unaware of the strange transformation that she is
undergoing. Men have also been known to experience some of these physiolog-
ical symptoms of love, but they are generally thought to be less likely to experi-
ence them than women are. A related pair of stereotypes portray the man as a
ruthless exploiter who falls out of love, if he ever was in it, quickly and casually,
moving on to new conquests while the woman who loves him tearfully watches
him ride off into the distance. Women, according to the stereotype, are the
lovers, men the leavers.
Do these stereotypes contain a kernel of truth? Our research leads us to
propose not only that these stereotypes of female lover and male leaver are un-
justified but also that there is a notable difference between men and women that
goes in precisely the opposite direction. Our hypothesis can best be stated as a
two-part empirical generalization: (a) Men tend to fall in love more readily than
women. (b) Women tend to fall out of love more readily than men. Before we
proceed to the evidence for these hypothesized differences, several specifications
are in order.
First, the terms "fall in love" and "fall out of love" are not being used in
a very special or mysterious way. They simply refer to people's ability or pro-
pensity to enter into and to give up romantic attachments. Several different in-
dicators of falling in and out of love will be introduced as we examine the rel-
Second, these generalizations were suggested primarily by the results of a
longitudinal study of 231 college student dating couples in the Boston area. The
hypothesized differences are seen as being most relevant to such dating or pre-
marital relationships in middle-class America today. They would not necessarily
be found in different times, different cultures, different age groups, different
social class groups, or different sorts of relationships (such as marriage). As we
will suggest later, comparisons with other segments of the population might be
of great interest in formulating different explanations of sex differences in love.
Sex Differences in Romantic Attachments 823
Third, even among the sample of dating couples that we will be consider-
ing, the sex differences to be reported are not massive ones. On all the measures
to be discussed, there is a great deal of overlap between the distributions of the
two sexes, and the overall differences are modest ones. Nevertheless, the various
strands of evidence combine to suggest that the postulated differences are real
ones, if they are viewed as actuarial propositions about a preponderance of cases.
These actuarial propositions may be of considerable interest not as social facts in
their own right, but for what they imply about the socialization of the two sexes
for close relationships in contemporary America.
With these specifications in mind, let us turn to the evidence for our two-
part hypothesis. First, we will describe the research program that provides our
main source of data. Then we will discuss the evidence bearing on each of the
two parts of our empirical generalization. Finally, we will consider several pos-
sible lines of explanation for the observed differences.
THE BOSTON COUPLES STUDY
Through a series of letters and advertisements in the spring of 1972, we
recruited a sample of 231 couples who were "dating" or "going together" at
four colleges in the Boston area. The four colleges were chosen with a view to-
ward diversity. They included a small private college, a large private university,
a Catholic university, and a state college enrolling commuter students. The large
majority of participants came from middle-class backgrounds. About half of the
participants' fathers had graduated from college and about one-fourth of the
fathers held graduate degrees. The modal couple consisted of a male junior and
a female sophomore who had been dating for about eight months. Almost all the
couples were dating one another exclusively, but few had any concrete plans
about marriage. Further details of the sampling procedure and characteristics of
the sample have been reported elsewhere (Hill, Rubin, & Peplau, 1976).
We proceeded to follow up these couples through extensive questionnaires
in fall 1972, spring 1973, and (by mall) in spring 1974. The response rates on
the follow-up questionnaires remained high, even though by 1974 many of the
participants had left the Boston area. In 1974 the mailed questionnaires were
returned by 83% of the women and 75% of the men in the initial sample. In all
cases, each partner was asked to complete the questionnaire individually. Sub-
jects were assured that their responses would be kept in strict confidence, and
that their replies would never be revealed to their partners. 3
3 In many cases, the partners decided to discuss their responses with one another after they
turned in their questionnaires. This was one way in which participation in the research
had an effect on couples' relationships, in some respects similar to the effects of couples
counseling. For a discussion of "couples research as couples counseling," see Z. Rubin
and Mitchell (1976).
824 Rubin, Peplau, and Hill
The questionnaires covered a wide range of events, experiences, attitudes,
and feelings. On the follow-up questionnaires we also asked participants to reflect
on changes in the relationship over time. These questions are of particular interest
in cases in which the couple had broken up in the interim. We also interviewed
a subset of the participants more intensively. Of special relevance to this report
is a series of interviews conducted in fall 1972 with 18 people whose relation-
ships had ended since the previous spring.
Proposition 1: Men Tend to Fall in Love More Readily than Women
In spite of prevailing stereotypes about "romantic" women, or women
"out to catch a man," there is converging evidence that men tend to fall in love
more readily than women do. Men have consistently been found to have higher
scores than women on measures of "romanticism" (Hobart, 1958; Knox & Spora-
kowski, 1968; Z. Rubin, 1969). In the Boston Couples Study men were again
found to score significantly higher than their girlfriends on such a romanticism
scale (paired t = 4.10, df = 230, p < .001). This scale assesses the degree to which
a person adheres to such tenets of romantic ideology as the belief that love
strikes at first sight and overcomes bars of race, religion, and economics. Men's
greater belief in this romantic ideology suggests that they may be more ready than
women to fall in love quickly and with a wider range of partners, while women
may tend to be more deliberate and discriminating about entering into a roman-
This difference in ideology may be paralleled by a difference in dating
goals. We asked participants in the Boston Couples Study to indicate how impor-
tant each of a variety of goals was as a reason for entering their relationship. Sur-
prisingly, in light of the prevailing stereotype of romantic women, men rated the
"desire to fall in love" as a significantly more important reason for entering the
relationship than did women (paired t = 2.21, df = 227, p < .05).
The suggestion that men tend to fall in love more readily than women is
also supported by more direct reports of attraction in the early stages of relation-
ships. In an extensive study of engagement and marriage conducted by Burgess
and Wallin (1953) in the 1930s and 1940s, manymore men than women reported
that they had been strongly attracted to their eventual fianc~es at their first
meeting or shortly thereafter. In a "computer-dance" study conducted at Iowa
State University in the 1960s (Coombs & Kenkel, 1966), men were more satisfied
with their randomly assigned partners on all criteria; indicated that they felt
more "romantic attraction" toward them; and, when asked to speculate about the
possibility that they could have a happy marriage, were more optimistic. In a
study conducted at the University of Michigan in 1968-1969, Z. Rubin adminis-
tered a self-report "love scale" to a large sample of dating couples. This scale is a
9-item self-report attitude scale calling on the respondent to assess the degree to
Sex Differences in Romantic Attachments 825
which he or she feels attached to, cares about, and feels intimate with a particular
other person (Z. Rubin, 1970). In the total sample the average love scores of men
and of women proved to be approximately equal. But among the 40 couples
who had been dating for only a short time (up to three months), boyfriends'
love scores were significantly higher than those of their girlfriends (Z. Rubin,
1969). All these findings seem consistent with the proposition that men tend to
fall in love more readily than women. In the earliest stages of a relationship,
men tend to report greater attraction and love for their girlfriends than they
receive in return. This is true both in relationships that later become more in-
timate (as in the Burgess and Wallin study) and in relationships that typicaUy
never get beyond an initial date (as in the Iowa State study).
Proposition 2: Women Tend to Fall Out of Love More Readily than Men
Proposition 2 refers specifically to the ending of close relationships. We
were able to keep in touch with at least one member of all but 10 of the 231
couples in the Boston Couples Study over a two-year period, from spring 1972 to
spring 1974. By spring 1974, 20% of the couples about whom we had informa-
tion had married, 33% were still dating or going together, and the remaining 47%
(103 couples) had broken up. In considering the evidence for the second proposi-
tion, we will focus on the couples who had ended their relationship. We will con-
sider evidence from several different domains- participants' self-report love
scores before and after the breakup, their perceptions of problems in the rela-
tionship, reports concerning which of the partners most wanted to break up,
reports of emotional reactions to the breakup, and reports of whether or not the
former partners "stayed friends" after the breakup.
Love Scale Scores. Z. Rubin's love scale was administered to men and
women in the sample at two points in time, separated by one year (spring 1972
and spring 1973; the love scale was not readministered in 1974). Within this
interval, approximately one-third of the total sample of 231 couples had broken
up. Figure 1 presents average love scores of men and women in couples who
stayed together over the course of the year (Togethers) and in couples who
broke up (Breakups). The members of Together couples were reporting their
feelings of attachment, caring, and intimacy toward a current dating partner at
each of two points in time. The members of Breakup couples were responding to
a current dating partner in 1972 and a former dating partner in t973. Not sur-
prisingly, the love scores of men and women in Breakup couples plummeted over
the one-year period, while those of men and women in Together couples stayed
at high levels (Time X Together-Breakup interaction F = 107, df = 1, 129, p <
.001). Of greater relevance are the differences between the average scores of men
and women. Among couples who stayed together over the one-year period,
women's scores were initially slightly higher than their boyfriends' scores. Among
826 Rubin, Peplau, and Hill
1-- ........ Men - Together
Fig. 1. Women's and men's mean love scoresin 1972 and 1973 for those
who stayed together and those who broke up during this interval. The
means are based on individuals who filled out the love scale in both
years (121 women and 118 men in couples who stayed together, 31
women and 39 men in couples who broke up). The maximum possible
score is 117.
couples who were to break up over the one-year period, in contrast, the women's
love scores were somewhat lower than their boyfriends' scores. Thus, it would
have been possible to do a better job of predicting whether a couple would stay
together on the basis of women's than of men's scores. In addition, among cou-
ples who broke up over the one-year period, the women's scores remained lower
than the men's - and dropped even more sharply. (The overall Sex × Together-
Breakup interaction is significant; F = 7.83,
= 1, 129, p < .01.) The total
pattern of scores points to two related conclusions: First, a woman's love was a
better predictor or barometer of the continuation of a relationship than a man's
love; and, second, as a prelude to and/or as a consequence of the ending of a
close relationship, women's love tended to diminish more than men's. Both these
conclusions seem consistent with the proposition that women tend to fall out of
love more readily than men do.
Perception of Problems in the Relationship.
One likely reason for - or con-
comitant of-falling out of love is the perception that one's relationship is
beset with problems. We gave the members of couples who had broken up a list of
Sex Differences in Romantic Attachments 827
13 common problem areas and asked them to indicate which of these problems
had contributed to the breakup. We found that women indicated more problems
had contributed to the breakup than men did (paired t = 3.15,
77, p < .003).
In particular, more women than men cited "differences in interests," "differences
in intelligence," "conflicting ideas about marriage," "my desire for indepen-
dence," and "my interest in someone else" as contributing factors. The only
problem that men cited more frequently than women was "living too far apart."
These reports are retrospective and clearly susceptible to distortion. Nevertheless,
they suggest that women tended to be more sensitive than men to problem areas
in their relationship, and that women were more likely than men to compare
the relationship to alternatives, whether hypothetical or actual. These tendencies
seem consistent with the postulated tendency of women to fall out of love more
readily than men.
Who Precipitated the Breakup?
If women tend to fall out of love more
readily than men, we would expect them to precipitate the breakups (i.e., to
play the role of "breaker-upper") more frequently than men. Combining men's
and women's independent reports of "who wanted to break up more," we were
able to estimate that the woman was more interested in breaking up in 51% of
the couples, the man in 42%, and the breakup was reported as being mutual in
7%. This preponderance is a small one and could be written off as a statistical
accident. It is given greater credence, however, by the fact that our participants'
reports of the breakups of their closest opposite-sex relationship prior to our
study (with over 200 cases for each sex) suggested a similar preponderance of
female-initiated breakups (Hill, 1974). And in his study at the University of
Michigan, Z. Rubin (1969) found that 17 of 25 nonmutual breakups had been
precipitated by women.
This preponderance of female-precipitated breakups might be easily under-
stood if women tended to have been less involved in their relationships than men.
This was not the case. Once the relationships of the couples in our study had
proceeded beyond their early stages, the women were by all indications at least
as involved as the men. Combining the two partners' reports before the time of
the breakup, women were categorized as the more involved partner in 45% of all
couples and men asmore involved in 36%; in the remaining 19% the two partners
were classified as being equally involved. As Peter Blau has suggested, a relation-
ship in which there is unequal involvement will not always be ended by the less
involved partner: "Whereas rewards experienced in the relationship may lead to
its continuation for a while, the weak interest of the less committed,
or the frus-
trations of the more committed
probably will sooner or later prompt one or the
other to terminate it" (Blau, 1964, p. 84, italics added). Our data, presented in
Table I, suggest that both of Blau's postulated patterns describe a substantial
number of breakups precipitated by women. Many women ended their relation-
ship when they were the less involved partner and wanted to move on to better
alternatives. But in a substantial minority of cases, relationships were ended by
Table I. Who Wanted to End the Relationship, as a Function of the
Couple's Relative Involvement Before the Breakupa
Man was Woman was
more involved more involved Total
Man wanted to end 7 26 33
Mutual ending 4 1 5
Woman wanted to end 23 16 39
Total 34 43
a Measures were derived by averaging men's and women's independent
reports; 18 cases in which the two partners were "equally involved"
women who had been the more involved partner, when they finally realized that
their commitment was not reciprocated. When breakups were precipitated by
men, in contrast, only the first pattern was common. Relationships were fre-
quently ended by men when they were the less involved partner, but only rarely
when they were the more involved. In asymmetrical situations in which one's
own love was not reciprocated, women seemed to be more able than men to
relinquish their love and to take the initiative in ending the relationship.
Seeing It Coming. Apparently related to women's ability to give up love
more readily than men do is the fact that women tended to see the breakup
coming sooner than men did. There was an overall tendency for women to report
that the breakup was more "gradual" (as opposed to "abrupt") than their boy-
friends reported (paired t = 2.10, df = 76, p = .039). As the means in Table II
indicate, this difference was most clear when the woman was the breaker-upper
and when the breakup was mutual. When the man was the breaker-upper, the
breakup was perceived by both sexes as being about equally abrupt (interaction
F = 3.02, df = 2, 70, p = .056). On the whole, then, women seem to have had
Table II. Mean Ratings of Perceived Gradualness of the Breakup, as a Function
of Who Wanted to Break Up a
Man wanted Both wanted Woman wanted
to break up to break up to break up
(N = 31) (N = 5) (N = 37)
Women's perceived 4.23 6.00 5.05
Men's perceived 4.29 3.80 3.89
aOnly couples in which we have reports from both partners are included.
Scale: 1 = ending was "extremely abrupt"; 9 = ending was "extremely
Sex Differences in Romantic Attachments 829
more a warning that a breakup was coming than men did, and perhaps more time
to prepare themselves to fall out of love.
Emotional Reactions to Breaking Up. Our proposition about sex differ-
ences in falling out of love would suggest that breaking up tends to be a more
traumatic experience for men than for women. Unfortunately, the quantitative
data available to test this proposition are limited to the 15 couples from whom
we obtained reports of emotional reactions to the breakup from both partners
on the one-year follow-up. These data suggest that men were hit harder by the
breakup than were women. In the wake of the breakup, men tended to report
that they felt more depressed, more lonely, less happy, and less free than did
their former girlfriends (paired ts median p = .15). Even women who had been
more involved in the relationship than their boyfriends tended to feel greater
equanimity after the breakup than did men in comparable situations. For ex-
ample, one woman ended her relationship with her boyfriend when she could
no longer tolerate his continuing neglect of her. Ruth later told us that she had
no regrets about her relationship with David or its ending. "It's probably the
most worthwhile thing that ever happened to me in my 21 years," she said,
"so I don't regret having the experience at all. But after being in the supportive
role, I want a little support now." Ruth added that "I don't think I ever felt
romantic [about David] - I felt practical. I had the feeling that I'd better make
the most of it because it won't last that long." In contrast, men in such situa-
tions were more likely to regret that they had not been able to give the relation-
ship another chance, and more likely to react with incredulity to indications
that their love was not reciprocated. Women who were rejected were also likely
to react with considerable grief and despair, but they seemed less likely to retain
the hope that their rejectors really loved them after all.
In this connection, it is of interest to consider the clinical impressions of
George W. Goethals (1973), based on his experience counseling young people:
"The notion that the young adult male is by def'mition a heartless sexual predator
does not bear examination. In point of fact some of the most acute cases of de-
pression I have ever had to deal with occurred in attempting to help young men
with their betrayal by a young woman in whom they had invested a great deal
and who had, as the relationship developed, exploited them rather ruthlessly"
Our proposition about a sexdifference in the readiness to give up love may
provide part of the explanation for Goethals' observation. It is unlikely that
women are by nature any more "exploitative" or "ruthless" than men are. But
if men are in fact less ready or able to give up love, men may be particularly like-
ly to be mystified, hurt, and ultimately crushed by rejection.
Staying Friends. If men fred it more difficult than women to renounce
their love, we might also expect relations between former partners to be more
strained after the woman has rejected the man than vice versa. Whereas a rejected
woman may be able to redefine her relationship with her boyfriend from "love"
830 Rubin, Peplau, and Hill
Table III. "Staying Friends," as a Function of Who Wanted
to Break Up a
Who wanted to end the relationship?
Did the couple
"stay friends"? Man Mutual Woman Total
Yes 28 5 23 56
(70%) (71%) (46%)
No 12 2 27 41
(30%) (29%) (54%)
a "Staying friends" was assessed by pooling the two partners'
reports. If either partner said "No," the couple was cate-
gorized as nonfriends.
to "friendship" - which, as Davis (1973) notes, is often a euphemism for ac-
quaintanceship - a rejected man may find such a redefinition more difficult to
accomplish. In such cases, "staying friends" is likely to be impossible. The data
are clearly consistent with this expectation. As shown in Table III, a couple was
much more likely to report that they stayed friends if the man had been the one
who precipitated the breakup or if the breakup had been mutual than if the
woman had precipitated the breakup (×2 = 5.83, p < .06).
Having summarized the evidence for both parts of our generalization-
men tend to fall in love more readily than women, and women tend to fall out
of love more readily than men - let us turn to possible explanations for it. We
will consider three lines of explanation for the observed differences, from the
standpoints of psychoanalytic theory, the social and economic context of mate
selection, and the socialization of men and women in the management of their
own emotions. 4
When we reported these findings to our psychoanalytically oriented col-
leagues, they welcomed them, because they provide support for a psychoanalytic
4 Another possible line of explanation derives from sociobiological speculations about sexual
selection. Our finding that women tended to be more "selective" than men, being more
cautious about entering a romantic relationship and quicker to extricate themselves from a
relationship that seemed ill fated, might be seen as an instance of the general tendency for
females in the large majority of animal species to be more selective than males in their
choice of mate. See, for example, Barash (1977). It would involve a rather large jump from
the data at hand, however, to argue that the human sex difference under discussion is a
product of natural selection.
Sex Differences in Romantic Attachments 831
notion about sex differences in the capacity for love. This is the notion that
because of the strength of their initial love for their mothers, men have a greater
capacity for complete heterosexual commitment than women do. Women, in
contrast, are believed to handle the Oedipal conflict in a more gradual way, first
shifting their sexual interest from mother to father, and later from father to
substitute objects of love. Working from this set of assumptions, Nancy Cho-
dorow (1976) argues that "Women have a richer, ongoing inner world to fall
back on, and the man in their life does not represent the intensity and exclusivity
which the woman represents to the man" (p. 463). Because of their richer
reserve of internalized emotional objects and their lesser emotional dependence
on men, women may tend to enter romantic relationships more cautiously, to
leave them more readily, and to rebound more easily after a rejection. We remain
skeptical of this psychoanalytic position, because there is very little firm evidence
for the set of early childhood events that it presumes to occur. Nevertheless, the
psychoanalytic position may have a grain of truth, and it suggests that the sex
difference that we have been discussing may be general, extending far beyond
the specific domain of campus courtship. There just may be some deeply rooted
aspects of men's and women's personality that make it likely for men to make
more complete heterosexual commitments than women - or, at least, to do so
more quickly - and that make it harder for men to get over the loss of such a
A different explanation of our propositions derives from an examination
of the social and economic context of mate selection. Rather than presupposing
any deep personal or emotional differences between men and women, this ex-
planation focuses on the peculiarities of the institution of courtship in Western
society. According to this approach, women must be more cautious, practical,
and realistic than men in the process of mate selection for simple social and eco-
nomic reasons. In most marriages, the wife's status, income, and life chances are
far more dependent on her husband's than vice versa. As a result, in a "free
choice" system of mate selection (like that in contemporary America) the woman
must be especially discriminating. She cannot allow herself to fall in love too
quickly; nor can she afford to stay in love loo long with the "wrong person."
The woman must carefully evaluate her partner's strengths and weaknesses and
must compare him to potential alternative partners, in order to be sure that she
is getting the best possible "bargain" in the marriage market. Men, on the other
hand, being in a position of greater power both in the larger society and in the
marriage market, do not need to worry so much about such rational calculations.
Instead, the man can better afford the luxury of being "romantic." This socio-
logical explanation is by no means a new one. It was stated most bluntly by
Willard Waller (1938) in the 1930s: "There is this difference between men and
women in the pattern of bourgeois family life. A man, when he marries, chooses
a companion and perhaps a helpmate, but a woman chooses a companion and at
832 Rubin, Peplau, and Hill
the same time a standard of living. It is necessary for a woman to be mercenary"
We suspect that there is a great deal of truth in this analysis. In spite of the
recent movement toward more egalitarian sex-role attitudes, the marriage market-
place in the 1980s is still characterized to a large extent by the same basic
inequities that characterized it in the 1930s. This socioeconomic explanation re-
mains incomplete, however, because it does not deal with the links between the
social and economic requirements of mate selection and the differences in emo-
tional capacities that we have been discussing. To flU out the explanation, there-
fore, we also need to focus on what men and women learn about the experience
and management of emotions, s Evidence from a variety of studies suggests that
women come to be more socially sensitive than men. Women tend to be more
empathic than men (Hoffman, 1977) and more sensitive to nonverbal com-
munication (Hall, 1978). In addition, women have been found to make sharper
distinctions between interpersonal sentiments, such as those of "liking" and
"loving" than men do (Z. Rubin, 1970). These sex differences are in accord with
the traditional assignment of women to the role of social-emotional specialists,
while men are the traditional task specialists (Parsons & Bales, 1955). The em-
phasis on social-emotional matters in women's socialization may lead them to
be more sensitive than men to the quality of their interpersonal relationships,
both in the present and projecting into the future. Thus, women may evaluate
their relationships more carefully than men do, and their criteria for falling in
love - and for staying in love - may be higher than men's.
In addition, Hochschild (Note 1) has argued convincingly that women
come to be more adept than men at cognititively managing their own feelings. In
self-report accounts of emotional experiences that Hochschild collected, women
were more likely than men to write about actively managing their own feelings,
using such terms as "I made myself feel .... " "I snapped myself out of it," and
"I tucked my feelings in." Men seemed to be less closely in touch with their own
feelings and to take a less active stance with regard to them. Hochschild's analysis
suggests that women are also more likely to exert cognitive control over such
events as falling in and out of love. Such greater cognitive control would help to
explain our findings that women tended to be (1) less likely to be swept off their
s It is, of course, possible that some aspects of sex differences in the experience and manage-
ment of emotions build on differences in genetic predispositions. For example, Hoffman
(1977) has summarized evidence that newborn female infants are more likely than new-
born males to cry in response to another infant's cry, suggesting the possibility of a con-
stitutional precursor of sex differences in empathy. That sex differences in the experience
and management of emotions build on such constitutional differences remains highly
speculative, however. In contrast, it is hard to .doubt that social learning in early and later
life plays a major role in the development of emotional differences.
Sex Differences in Romantic Attachments 833
feet into a deep love relationship; (2) more likely to perceive the problems of a
relationship and, if necessary, to end it; (3) better able to get over their feelings
of loss when a relationship ends; and (4) when rejected, better able to accomplish
the transition from love to friendship. Women presumably develop this greater
cognitive control as a result of socialization experiences which emphasize that
they have a considerable degree of power in the emotional domain, whereas such
emotional socialization is neglected for men. It can also be argued that a greater
degree of control for women in the domain of their own emotions is a necessary
adaptation to their lesser degree of power and control in other domains.
SUMMARY AND CONCLUSION
We proposed a two-part generalization about sex differences in love: (1)
Men tend to fall in love more readily than women; and (2) women tend to
fall out of love more readily than men. We then presented data from a longitud-
inal study of 231 student dating couples in support of these propositions. The
data suggest that women were less "romantic" than men, more cautious about
entering into romantic relationships, more sensitive to the problems of their
relationships, more likely to compare their relationships to alternatives, more
likely to end a relationship that seemed ill fated, and better able to cope with
rejection. Psychoanalytic theorists might account for these differences in terms
of underlying differences between men and women in the capacity for com-
plete heterosexual commitment. A socioeconomic explanation, in contrast,
focuses on the need for women to be more practical and discriminating than men
in the marriage marketplace. A final explanation refers to socialization experiences
through which women learn to control and manage their own emotions more
effectively than men do.
To evaluate these (and any other) explanations of sex differences in love,
it would be valuable to investigate the generalizability of the present findings to
other segments of the population. Our sample was restricted to college students
of predominantly middle-class background. Almost all these students deferred
marriage until or beyond the end of their college years. Would similar sex dif-
ferences be found among working-class couples who do not attend college and
who typically marry earlier? To the extent that the sex differences we observed
reflect deeply rooted aspects of men's and women's personality, we might expect
them to generalize beyond the boundaries of social class. It might be, however,
that women in working-class couples tend to feel a greater pressure to marry
early and are less likely, as a result, to exhibit the degree of caution and cognitive
control in their love relationships that the women in our sample displayed (cf.
L. B. Rubin, 1976). It would also be interesting to determine whether the sex
differences we observed would also be found among older unmarried couples or
834 Rubin, Peplau, and Hill
among previously married men and women who are considering remarriage. 6
It is hoped that future research will extend the present investigation to couples
of different social class backgrounds, ages, and marital histories, as well as of dif-
ferent cultures and historical periods. Such comparative research would help us
to choose more knowledgeably among the various explanations of sex differ-
ences in love that have been or might be offered. Such research would also help
to create a fuller appreciation of the ways in which psychological, social struc-
tural, and cultural forces join to shape intimate relationships between men and
1. Hochschild, A. R.
Attending to, codifying and managing feelings: Sex differences in love.
Paper presented at the meeting of the American Sociological Association, San Francisco,
2. Zeiss, A. M., & Zeiss, R. A.
Sex differences in initiation of and ad/ustment to divorce.
Paper presented at the meeting of the Western Psychological Association, San Francisco,
Barash, D. P.
Sociobiology and behavior.
New York: Elsevier, 1977.
Blau, P. M.
Exchange and power in social life.
New York: Wiley, 1964.
Burgess, E. W., & Wallin, P.
Engagement and marriage.
Philadelphia: Lippincott, 1953.
Chodorow, N. Oedipal asymmetries and heterosexual knots.
Coombs, R. H., & Kenkel, W. F. Sex differences in dating aspirations and satisfaction with
Journal of Marriage and the Family,
1966, 28, 62-66.
Davis, M. S.
New York: Free Press, 1973.
Goethals, G. W. Symbiosis and the life cycle.
British Journal of MedicalPsychology,
Hall, J. A. Gender effects in decoding nonverbal clues.
Hill, C. T. The ending of successive opposite.sex relationships.
Doctoral dissertation, Harvard
Hill, C. T., Rubin, Z., & Peplau, L. A. Breakups before marriage: The end of 103 affairs.
Journal of Social Issues,
1976, 32(1), 147-168.
Hobart, C. W. The incidence of romanticism during courtship.
Hoffman, M. L. Sex differences in empathy and related behaviors.
Knox, D. H., Jr., & Sporakowski, M. J. Attitudes of college students toward love.
of Marriage and the Family,
6 Zeiss and Zeiss (Note 2) have recently reported data on sex differences in initiating and
adjusting to divorce that closely parallel our findings of sex differences in falling out of
Sex Differences in Romantic Attachments 835
Parsons, T., & Bales, R. F. Family, socialization, and interaction processes. Glencoe, II1.:
Free Press, 1955.
Rubin, L. B. Worlds of pain: Life in the working-class family. NewYork: Basic Books, 1976.
Rubin, Z. The social psychology of romantic love, Doctoral dissertation, University of
Rubin, Z. Measurement of romantic love. Journal of Personality and Social Psychology,
1970, 16, 265-273.
Rubin, Z., & Mitchell, C. Couples research as couples counseling: Some unintended effects
of studying close relationships. American Psychologist, 1976, 31, 17-25.
Wailer, W. The family: A dynamic interpretation. New York: Dryden, 1938.