Content uploaded by Winifred A Gebhardt
Author content
All content in this area was uploaded by Winifred A Gebhardt on Jun 11, 2015
Content may be subject to copyright.
INTERNATIONAL ARTICLE
Need for Intimacy in Relationships and Motives for
Sex as Determinants of Adolescent Condom Use
WINIFRED A. GEBHARDT, Ph.D., LISETTE KUYPER, B.Sc., AND GWEN GREUNSVEN, M.Sc.
Purpose: To investigate the extent to which need for
intimacy in relationships and motives for sex are capable
of distinguishing between adolescents who always have
protected sex and adolescents who do not, with reference
to both steady and casual relationships.
Methods: Seven-hundred-and-one adolescents (424
males and 277 females) aged 15 to 23 years filled out a
questionnaire assessing determinants of sexual behavior
and various outcome behaviors, such as having casual sex
and condom use.
Results: Two-thirds of the participants (470) were sex-
ually experienced. Discriminant functions using con-
structs of the Theory of Planned Behavior (attitude,
subjective norm, self-efficacy), as well as the need for
intimacy in relationships and motives for sex, were
found to distinguish significantly between adolescents
who always had protected sexual intercourse and adoles-
cents who did not. This was valid both for sex within
steady and within casual relationships. Besides a positive
attitude and high perceived subjective norms, protected
sex with a steady partner was characterized by low scores
on the scale for the motive for sex to express love, and on
the scale for the need for intimacy in relationships.
Consistent condom use with casual partners was related
to high self-efficacy, attitude, and perceived subjective
norms, as well as a greater need for intimacy in relation-
ships. Gender differences emerged with respects to de-
terminants of (un)safe sex with casual partner(s), in that
for males the variables of the Theory of Planned Behav-
ior were most important in distinguishing those who had
unsafe casual sex from those who had safe casual sex. For
females, on the other hand, the variables concerning the
meaning attached to having sex or to the relationship
within which sex occurs seemed to be of more signifi-
cance in explaining (un)safe sex with casual partner(s).
However, owing to small subsamples, caution is war-
ranted when interpreting these differences.
Conclusion: In adolescence, the factors which influence
the decision to have (un)protected sex depend on the
kind of relationship that exists between the partners. In
addition, the goals which adolescents pursue with regard
to their relationships and with regard to having sex
fulfill an important role in the decision-making process
whether to use condoms. © Society for Adolescent Medi-
cine, 2003
KEY WORDS:
Adolescence
Casual partner
Condom use
Gender differences
Need for intimacy
Sex motives
Steady partner
The Netherlands
Adolescence is characterized by efforts to consolidate
knowledge of oneself, and to integrate the images of
the self into a personal identity [1]. Information on
the outcome of behavior of others (i.e., vicarious
learning) and information on how the outside world
evaluates one’s own behavior (i.e., feedback), are
essential for the development of a consistent set of
values, beliefs and behaviors, which together shape
this unique identity. Therefore, adolescents typically
tend to experiment with diverse roles in numerous
relationships with adults and peers [2].
One of the important subtasks within the self-
definition is the development of a sexual identity.
Clearly, during the past decades, the Christian moral
Clinical and Health Psychology, Leiden University, Leiden, The
Netherlands.
Address correspondence to: Dr. Winnie Gebhardt, Clinical and
Health Psychology, Leiden University, P.O. Box 9555, 2300 RB Leiden,
The Netherlands. E-mail: gebhardt@fsw.leidenuniv.nl
Manuscript accepted February 14, 2003.
JOURNAL OF ADOLESCENT HEALTH 2003;33:154–164
© Society for Adolescent Medicine, 20031054-139X/03/$–see front matter
Published by Elsevier Inc., 360 Park Avenue South, New York, NY 10010doi:10.1016/S1054-139X(03)00137-X
of not having intercourse before marriage has all but
disappeared in Western society [3]. Magazines, tele-
vision programs and pop songs all underscore the
message that the expression of physical love with a
romantic, nonmarital partner is normal, and desir-
able in adult life. In fact, within most groups of
adolescents one is expected to have a certain amount
of sexual experience before marriage. On the other
hand, it is generally regarded to be morally wrong to
have many sexual partners, especially for females [4].
Adolescents have to balance such views to establish
their sexual identity, and this includes finding an
answer to questions such as “with whom, how, and
when will I be sexually intimate?”
Thus, experimenting with relationships and inti-
mate behavior has increasingly become a normal
pattern of adolescent development. The young ado-
lescent, however, does not yet have a consistent set
of norms or a behavioral repertoire, which makes it
difficult to attain satisfactory sexual experiences. As
a result, young people’s sexual encounters are often
unplanned, sporadic, and sometimes the result of
social pressure or coercion [4,5]. Adolescents tend to
be insufficiently aware of the health risks of physical
intimacy, and they often lack information on how to
prevent such risks [6,7]. Consequently, they are in
danger of contracting sexually transmitted infections
(STIs), which can have serious consequences because
of their young age [8,9].
Although some STIs, such as syphillis, have been
brought to an all-time low, worldwide there has been
an increase in the overall amount of STI [10]. Espe-
cially adolescents and young adults appear to be at
high risk for contracting STI [11]. It is estimated that
there are 15 million new STI cases in the United
States each year. Roughly 25% of these infections are
found in teenagers [9]. In the Netherlands, it was
estimated that annually about 100,000 individuals
contract an STI [12]. Of all the newly diagnosed STI
patients in Dutch municipal health services and STI
clinics in 2000, 40% were younger than age 25 years
[13,14].
Many adolescents who have had sexual inter-
course with more than one partner do not use
condoms [15,16]. Leenaars et al [17] found that 30%
of the sexually active Dutch high school students
never had protected sex, and 45% used a condom
only on an irregular basis. In a study by Dijkstra et al
[18], 53% of teenagers aged 16 to 20 years reported
that they had intercourse without a condom during
their most recent sexual encounter.
Females are more likely to have unprotected in-
tercourse than males [19]. Abraham and Sheeran [20]
argue that adolescent females may feel that it is their
partner’s responsibility to decide whether a condom
will be used. They may also believe that they lack the
skills to negotiate safer sex practices, and may fear
termination of the relationship or coercion in case
they do not agree to have unprotected sex.
Psychological research on determinants of
(un)safe sexual behavior has mostly applied a deci-
sion-making framework [21]. It includes the applica-
tion of theoretical models, such as the Health Belief
Model [22,23], the Protection Motivation Theory [24],
the Theory of Reasoned Action [25], and the Theory
of Planned Behavior [26]. Within this framework,
one assumes that individuals will adopt a healthy
behavior (in this case, the use of condoms), once they
are convinced that the benefits will outweigh the
drawbacks. Thus, when the estimated benefits of
using condoms (e.g., reduced chance of acquiring
STI) are higher than the expected disadvantages (e.g.,
reduced satisfaction), the chance that precautions are
taken increases. Attitude, beliefs and motivations
with regard to health protection and disease preven-
tion are therefore expected to explain whether or not
protected sex is practiced.
To date, research based on this framework has
had some success. In particular, the Theory of
Planned Behavior has been fairly successful in ex-
plaining condom use [27]. However, the decision-
making framework does not take into account that
behavior may fulfill a variety of individual goals
unrelated to health protection or disease avoidance.
For example, a more relevant goal may be personal
development or a wish to belong and to be accepted
by others [28,29].
Several studies support the notion that the mean-
ing of sexual behavior can be an important determi-
nant of condom use. For example, Cooper et al [30]
reported that certain motives for sex could explain,
in part, the occurrence of unprotected sex among
adolescents. For instance, having sex to cope with
negative emotions, or to please one’s partner, or to
feel better about oneself were all negatively related
to condom use. On the other hand, having sex to
share intimacy was positively related to protected
sex. In a study among university students, who
intended to use condoms with a new partner, it was
found that the importance attached to having sexual
intercourse was a significant determinant for con-
dom use [31]. An intervention study by Sanderson
and Cantor [32] showed that adolescents who pursue
identity goals in their relationships (i.e., goals related
to self-reliance and self-exploration) are more likely
to have protected sex after technical training (e.g.,
September 2003 NEED FOR INTIMACY IN RELATIONSHIPS AND MOTIVES FOR SEX 155
how to use a condom) than after training in commu-
nication skills (e.g., how to introduce the subject of
condom use). On the other hand, adolescents who
pursue intimacy goals (i.e., goals related to mutual
dependence and open communication) appeared to
profit more from communication skills training than
from technical instructions.
The present study investigates whether a model
which incorporates constructs related to the signifi-
cance of sex and the relationship within which sex is
practiced, as well as variables from the Theory of
Planned Behavior, can successfully explain variance
in adolescents’sexual behavior. It examines whether
adolescents who have protected sex differ from those
who have unprotected sex, with respect to their need
for intimacy in relationships and the goals which are
pursued by having sexual intercourse, as well as
their attitude toward consistent condom use, their
perceived social pressure to use condoms, and their
feelings of self-efficacy regarding condom use.
Methods
Participants
The 701 students who participated in our study were
recruited from five Dutch vocational secondary
schools. They were in training to become health care
assistants, cooks, carpenters, plumbers, technicians,
etc. The group of respondents consisted of 424 (60%)
males, and 277 (40%) females. They were between
the age of 15 and 23 years, with a mean age of 18
years (S.D. ⫽1.5). The majority (91%) of the partici-
pants still lived at home. Eighty-nine percent of the
participants were born in the Netherlands, whereas
the other participants had a Turkish, Moroccan,
Surinam, or Dutch Antilles ethnic background.
Eighty-six percent reported having had a relation-
ship at one time, whereas 43% indicated that they
were in a relationship at the time of the investigation.
Sixty-seven percent was sexually experienced in that
they had had sexual intercourse in the past. Of those
with experience, 34% had had one sexual partner,
30% two sexual partners, 15% three sexual partners,
and 21% more than three sexual partners. Ninety-
three percent reported having sex with one or more
steady partner(s), whereas 46% had had sex with one
or more casual partner(s).
One-hundred-twenty-eight of the males and 103
of the females reported that they had never had
sexual intercourse. Important reasons for abstinence
were: (a) not yet having found the right partner
(85%), (b) worries about STI or AIDS (19%), (c)
worries about pregnancy (10%), and (d) not approv-
ing of sex (10%). These respondents were excluded
from further analyses, because they had no experi-
ence with condom use. Thus, the reported analyses
are based on the data provided by the 470 sexually
experienced adolescents.
Measures
Sexual partners. The sexual experience with steady
or casual partners was investigated by means of
three questions: “With how many different partners
have you had sex in your life?”(open-ended ques-
tion); “Have you ever had sexual intercourse with a
steady partner?”(“yes”or “no”); and “With how
many casual partners (i.e., someone with whom you
do not have a relationship, ‘one-night-stands’) have
you had sex?”(open-ended question).
Usual condom use. Usual condom use was as-
sessed with two questions: “Do you use condoms
when having sex with steady partner(s)?”and “Do
you use condoms when having sex with casual
partner(s)?”The answer to both questions could be
given on a five-point Likert scale, ranging from
“never”to “always.”Those who indicated that they
did not use condoms consistently were asked to
indicate by means of a checklist their reason(s) for
the omission. These were: “we used the pill as a
contraceptive method,”“it was our first time,”“we
have been tested for STI and no diseases were
diagnosed,”“we have always had safe sex before,”
“we have known each other for a long time,”“no
condoms were available,”“it did not cross our
mind,”“we were under the influence of alcohol or
drugs,”and “we want to have children.”
Attitude. Respondents were asked to indicate on a
five-point semantic differential scale whether they
thought that the use of a condom each time they had
sex would be: purposeful, pleasant, good, healthy,
comfortable, and safe [25]. The total score for attitude
was computed by the addition of the score of each of
the six items. A high score indicates a more positive
attitude toward protected sex. The scale had a Cron-
bach alpha coefficient of .78.
Subjective norm. The pressure of the social envi-
ronment to use condoms was assessed for two so-
called referents, i.e. the peers and parents [25]. Par-
ticipants were asked to indicate on a five-point scale
(“agree”to “disagree”) to what extent respondents
believed that the referent group would want them to
156 GEBHARDT ET AL JOURNAL OF ADOLESCENT HEALTH Vol. 33, No. 3
use a condom every time they had sex. They were
also asked to what extent they valued the opinion of
peers and parents. The belief that their peers would
want them to use a condom was multiplied by the
value attached to this opinion, and a similar multi-
plication was performed for the parents as a group of
reference. These two variables (peers perceived
norm, and parent perceived norm) had a Cronbach
alpha of .71. A total score for the subjective norm was
calculated by adding the two variables. A higher
score indicates a greater perceived social pressure to
use condoms.
Self-efficacy. Respondents were asked to indicate
on a five-point scale (“agree”to “disagree”) how
certain they felt that they could use a condom under
four difficult situations, i.e., being drunk, being
highly aroused, the partner does not want to use a
condom, or sex is unexpected [25]. The Cronbach
alpha coefficient of the scale, which was a computa-
tion of the four items, was .76. A high score on the
scale for self-efficacy reflects a high degree of cer-
tainty regarding the use of a condom.
Need for intimacy in relationships. The 13-item ques-
tionnaire on dating goals within relationships [32]
was translated into Dutch by a translation/back-
translation procedure. Principal Component Analy-
ses yielded two components. The nine items with
high loadings on the first component all involved the
need for intimacy (e.g., “in my dating relationships I
try to take care of my girl/boyfriend”;“in my dating
relationships I try to spend a substantial amount of
time with my girl/boyfriend”;or“in my dating
relationships I share my most intimate thoughts and
feelings”). The four items of the second component
were all related to the need to remain independent
(e.g., “in my dating relationships I try to maintain a
strong sense of independence”;or“in my dating
relationships I try to keep my individual identity”).
A scale based on the nine intimacy items had a
satisfactory Cronbach alpha coefficient of .78. How-
ever, a scale based on the four identity items had a
Cronbach alpha coefficient of only .46. The two
subscales were positively correlated, implying that
those high on the scale for intimacy also scored high
on the scale for independence. Therefore, it was
decided to use only the scale for intimacy. A high
score on this scale indicates a strong inclination to
pursue intimacy within a romantic relationship.
Motives for sex. Eighteen of the original 29 items in
the questionnaire measuring motives for sex [30]
were translated into Dutch by means of a transla-
tion/back-translation procedure. We selected three
of the highest loading items for each of the proposed
subscales. The items tapping the motives for sex
could again be answered on a five-point scale (“of-
ten”to “never”). A Principal Component Analyses
suggested four underlying factors, which were inter-
preted as follows: (a) the motive to please others, (b)
the motive to enhance one’s mood, (c) the motive to
experience pleasure, and (d) the motive to express
love. The scale measuring the motive to please others
consisted of six items (e.g., “I have sex because
otherwise my partner will not love me anymore”;or
“I have sex because others may otherwise gossip
about me”). It had a Cronbach alpha coefficient of
.89. The motive to enhance mood encompassed three
items (e.g., “I have sex to cheer me up”;or“I have
sex to feel better when I am down”). This scale had a
Cronbach alpha coefficient of .71. The motive to
experience pleasure was based on three items (e.g., “I
have sex because sex feels good”;or“I have sex
because sex is exciting”). It had a Cronbach alpha
coefficient of .69. The motive to express love also
consisted of three items (e.g., “I have sex to feel closer
to my partner”;or“I have sex to feel more emotion-
ally connected to my partner”), and this had a
Cronbach alpha coefficient of .66. For all “motive”
scales, a higher score indicates a higher frequency of
having sex for that particular reason.
Intercorrelations between determinants. The determi-
nant scales appeared to be relatively independent
with all intercorrelations lower than .50.
Procedures
Various vocational high schools were approached
either by mail or by phone. Six schools agreed to
cooperate in the research. The questionnaires were
handed out by the researchers during normal class
hours, or after an examination. Students were in-
formed that the questionnaire would take approxi-
mately 20 minutes to complete. They were all told
that participation in the study was on a voluntary
basis, and that their answers would be treated con-
fidentially. Furthermore, they were instructed that
they could ask questions before, during, or after
completion of the questionnaire. Only one student
refused to fill out the questionnaire.
On completion, students dropped their question-
naire in a closed box placed in front of the classroom.
At one of the schools, participants received a
“healthy snack”during the time they filled out the
September 2003 NEED FOR INTIMACY IN RELATIONSHIPS AND MOTIVES FOR SEX 157
questionnaire. At the other five schools, no rewards
were given. Institutional Review Board approval was
not required.
Data Analyses
The description of sexual behavior and of condom
use was analyzed by means of frequencies and
descriptive statistics. Likewise, Pearson correlations
were calculated to determine the relationships be-
tween the predictor variable and the lifetime number
of sexual partners, and for those who had had casual
sex the lifetime number of casual sex partners. Dif-
ferences between those who had practiced casual sex
and those who had not were studied by means of
Student’st-tests. Two separate discriminant analyses
for sex with steady partners, and for sex with casual
partners, were conducted to identify variables capa-
ble of distinguishing between those who always
used a condom, and those who did not. Student’s
t-tests were applied to assess whether differences
among groups of (non-)users were significant on a
univariate level. Finally, discriminant analyses were
performed to investigate gender differences in deter-
minants of (un)safe sex with steady or casual part-
ner(s).
Results
Frequency of Condom Use
Twenty-three percent of the group with a steady
partner always used condoms, whereas 16% used
them “most of the time,”14% “sometimes,”and 47%
“rarely”or “never.”The main reasons mentioned for
not consistently using condoms were “we used the
pill as contraceptive method”(77%), or “we have
known each other for a long time”(35%).
In the group that practiced casual sex, 48% indi-
cated that they always used condoms, whereas 28%
reported that they used condoms most of the time.
Ten percent made occasional use of condoms, and
14% reported that they practically always had un-
protected sexual intercourse. The main reasons for
unsafe sex with casual partners were “we used the
pill as a contraceptive method”(54%), “no condom
was available”(43%), “we were under the influence
of alcohol or drugs”(37%), or “it did not cross our
mind”(21%). Unsafe behavior is far more likely to
occur with steady partners than with casual partners
(77% vs. 52%). However, also in the latter group, half
have experienced unprotected casual sex.
Women were far less likely to have engaged in
casual sex than men (39% vs. 56%). They were more
likely than men to practice unsafe sex, particularly
with their steady partner(s) (85% vs. 68%).
Table 1 provides the cross-tabulations of three
types of sex experiences with steady partner(s)
(“never,”“safe,”and “unsafe”) with three types of
sex experiences with casual partner(s) (“never,”
“safe,”and “unsafe”).
Relationship Between Determinants and Number
of (Casual) Sex Partners
As shown in Table 2, the number of sexual partners
(i.e. both steady and casual partners) in the past was
unrelated to the attitude toward condom use, the
subjective norm, and the motive for sex to express
love. There was a small but significant, negative
relationship between the number of sexual partners
and condom-use self-efficacy (r ⫽⫺.13; p⬍.01), and
the need for intimacy in relationships (r ⫽⫺.13; p⬍
.01). Furthermore, weak positive relationships ex-
isted between the number of sexual partners and the
motives for sex to enhance mood (r ⫽.15; p⬍.01), to
experience pleasure (r ⫽.14; p⬍.01), and to please
others (r ⫽.10; p⬍.05).
Student’st-tests revealed that those who had
experience with casual sex differed from those who
had never had such encounters with respect to
self-efficacy (t⫽2.24; p⬍.05) (Table 3). This implies
that those who had never had casual sex reported
stronger feelings of self-efficacy to use condoms. The
Table 1. Distribution of the Population With Respect to
Sexual Experience and Condom Use in Numbers of
Participants
Casual Sex
Never Safe
b
Unsafe Total
Never sex steady
partner
12 8 20
Steady sex safe 58
1
22 5 85
Steady sex unsafe 192
2
66 95 353
Total 250 100
3
108
4
458
a
a
Owing to missing data on condom use the total number
included in the Table is 458 (and not 470).
Group 1: Steady safe group: those who always used condoms
with their steady partner(s) and who had never had casual sex.
Group 2: Steady unsafe group: those who did not always use
condoms with their steady partner(s) and who had never had
casual sex.
Group 3: Casual safe group: those who always used condoms
with their casual partner(s).
Group 4: Casual unsafe group: those who did not always use
condoms with their casual partner(s).
158 GEBHARDT ET AL JOURNAL OF ADOLESCENT HEALTH Vol. 33, No. 3
differences between the two groups were greater
with respect to the need for intimacy in relationships
and three of the four motives for sex. To be specific,
the “casual sex”group scored lower on the scale
measuring need for intimacy in relationships (t⫽
3.37; p⬍.001) than those who had never had this
experience. At the same time this group scored
higher on the motive for sex to please others (t⫽
⫺4.16; p⬍.001), the motive for sex to enhance mood
(t⫽⫺3.65; p⬍.001), and the motive for sex to
experience pleasure (t⫽⫺2.84; p⬍.01). No signifi-
cant differences between the two groups were found
for the attitude toward condom use, the subjective
norm, or the motive for sex to express love.
Within the group of those who had experience
with casual sex, the number of casual partners was
significantly related to the motive for sex to enhance
mood (r ⫽.23; p⬍.001), and the motive for sex to
please others (r ⫽.20; p⬍.01). No significant
relations were found among the number of casual
partners and the other determinant variables.
Discriminant Analyses: Steady Partners
In the first discriminant analysis, we included those
respondents who had had sexual intercourse with a
steady partner, and never had intercourse with a
casual partner, resulting in a total number of 250
participants (Group 1 and Group 2) (Table 1). Forty-
three respondents were found to have missing val-
ues on one or more of the independent variables. Of
the remaining 207 subjects, 52 had indicated that
they always used condoms, and 155 had indicated
that they did not always do so. The discriminant
function derived from the analysis was significant. In
Table 4 the function coefficient for each determinant
is shown. The results of the comparison of the two
subgroups by Student’st-tests are also given in the
table. Attitude (r ⫽.77), subjective norm (r ⫽. 65), the
motive for sex to express love (r ⫽⫺.46) and the
need for intimacy (r ⫽⫺.29) had the strongest
correlations with the function. Furthermore, on a
univariate level, the two groups appeared to differ
with respect to the motive to have sex to enhance
mood.
Based on the discriminant function, 78% of the
participants were correctly classified (i.e., 85% of
“safe sex group,”and 76% of the “unsafe sex
group”). Cross-validation analyses resulted in a total
of 73% of the participants being correctly classified.
Table 2. Determinants’Relationship With Number of
Lifetime Sexual Partners (Steady and Casual Partners),
and Number of Lifetime Casual Partners
Lifetime No.
Sex Partners
Lifetime No.
Casual Sex
Partners
1. Attitude ⫺.08 ⫺.09
2. Subjective norm ⫺.02 ⫺.04
3. Self-efficacy ⫺.13** ⫺.06
4. R intimacy ⫺.13** ⫺.11
5. M please others .10* .20**
6. M enhance mood .15** .23**
7. M experience pleasure .14** .00
8. M express love .01 ⫺.02
*p⬍.05; ** p⬍.01.
R⫽relationship goal of; M ⫽motive for sex to.
Table 3. Differences Among Adolescents With, and
Without, Experience With Casual Sex
No
Casual Sex
Mean
Casual Sex
Mean t-Value
1. Attitude 20.51 20.00 1.14
2. Subjective norm 19.30 19.83 ⫺.42
3. Self-efficacy 14.33 13.43 2.24*
4. R intimacy 32.07 30.31 3.37***
5. M please others 9.78 11.95 ⫺4.16***
6. M enhance mood 8.91 10.03 ⫺3.65***
7. M experience pleasure 12.98 13.53 ⫺2.84**
8. M express love 12.66 12.56 .48
*p⬍.05; ** p⬍.01; *** p⬍.001.
R⫽relationship goal of; M ⫽motive for sex to.
Table 4. Discriminant Analysis and Student’st-tests:
Predictors of Condom Use With Steady Partner(s)
Correlation
With
Function
a
Mean
Scores
Group
b
t-Value
Attitude .77 safe M ⫽24.15 7.83***
unsafe M ⫽19.35
Subjective norm .65 safe M ⫽28.29 5.94***
unsafe M ⫽16.02
M express love ⫺.46 safe M ⫽11.69 ⫺4.18***
unsafe M ⫽13.12
R intimacy ⫺.29 safe M ⫽31.11 ⫺2.64**
unsafe M ⫽35.06
M enhance mood ⫺.24 safe M ⫽8.06 ⫺2.20**
unsafe M ⫽9.19
Self-efficacy .16 safe M ⫽15.54 1.88
†
unsafe M ⫽14.30
M experience
pleasure
⫺.19 safe M ⫽12.58 ⫺1.76
†
unsafe M ⫽13.14
M please others .07 safe M ⫽9.77 .59
unsafe M ⫽9.39
†
p⬍.10; ** p⬍.01; *** p⬍.001.
a
Wilks’Lambda ⫽.713;
2(8)
⫽67.98 (p⬍.001).
b
Safe group; those who always used condoms with their
steady partner(s) (N ⫽52; 37 males and 15 females).
Unsafe group: those who did not always use condoms with
their steady partner(s) (N ⫽155; 87 males and 68 females).
R⫽relationship goal of; M ⫽motive for sex to.
September 2003 NEED FOR INTIMACY IN RELATIONSHIPS AND MOTIVES FOR SEX 159
Gender Differences in Determinants of (Un)safe
Sex With Steady Partner(s)
Separate analyses were conducted for males and
females to assess gender differences in discriminat-
ing variables between those who have safe and those
who have unsafe sex with steady partner(s). The
variables that correlated the highest with the func-
tion which distinguished males who had safe sex (N
⫽37), from males who had unsafe sex (N ⫽68) with
their steady partner(s), were: attitude (r ⫽.88),
subjective norm (r ⫽.46), and the motive to express
love (r ⫽⫺.31). Eighty-one percent of the male
participants were correctly classified (i.e., 87% of the
safe sex group, and 78% of the unsafe sex group).
Similarly, the variables correlating the highest with
the function distinguishing females who practiced
safe sex (N ⫽15) from females who practiced unsafe
sex (N ⫽87) with their steady partner(s) were:
subjective norm (r ⫽.80), attitude (r ⫽.51), and the
motive to have sex to express love (r ⫽⫺.38).
Furthermore, the variables measuring self-efficacy
(r ⫽.38) and the motive to have sex to experience
pleasure (r ⫽⫺.32) were related to the function
which differentiated the two groups of women. Sev-
enty-eight percent of the female participants were
correctly classified (i.e., 80% of the safe sex group
and 78% of the unsafe sex group) (Table 5).
Thus, both males and females who had safe sex
had a more positive attitude, a higher subjective
norm, and were more likely to have sex to express
love than those who had unsafe sex. In addition,
females who had safe sex were more likely to have a
high self-efficacy to use condoms, and less likely to
have sex to experience pleasure.
Discriminant Analyses: Casual Partners
A group of 208 respondents reported to have had
experienced casual sex (Group 3 and Group 4). Of
this group, 28 participants had missing values on one
or more of the independent variables. The remaining
180 subjects could be divided into a group of 84
participants who consistently used condoms with
their casual partners, and a group of 96 participants
who did not consistently use condoms. The function
that was derived from the analysis was significant.
The function coefficient for each determinant is pre-
sented in Table 6, together with the results of Stu-
dent’st-tests. The highest correlations with the func-
tion were those for self-efficacy (r ⫽.80), attitude
(r ⫽.56), subjective norm (r ⫽.39), and the need for
intimacy (r ⫽.34).
Classification based on the discriminant function
correctly identified 68% of the respondents (i.e., 64%
of the “safe sex group,”and 72% of the “unsafe sex
group”). Sixty-four percent of the participants were
correctly classified after cross-validation analyses.
Gender Differences in Determinants of (Un)safe
Sex With Casual Partner(s)
Another set of analyses was performed to determine
gender differences in discriminating variables be-
tween those who have safe and those who have
unsafe sex with casual partner(s). The variables that
correlated the highest with the function which dis-
tinguished males who had safe sex (N ⫽67) from
males who had unsafe sex (N ⫽68) with their casual
partner(s), were: attitude (r ⫽.62), self-efficacy (r ⫽
.60), and subjective norm (r ⫽.35). Sixty-six percent
of the male participants were correctly classified (i.e.,
66% of the safe sex group, and 67% of the unsafe sex
group). The variables correlating the highest with the
function which distinguished females who practiced
safe sex (N ⫽17) from females who practiced unsafe
sex (N ⫽28) with their casual partner(s) were: the
motive to have sex to express love (r ⫽.49), the
motive to have sex to please others (r ⫽⫺.40), self
efficacy (r ⫽.33), and the need for intimacy in
relationships (r ⫽.33). Based on the discriminant
function, 76% of the female participants were cor-
rectly classified (i.e., 82% of the safe sex group and
71% of the unsafe sex group) (Table 7).
Thus, males who had safe sex had a more positive
attitude, stronger feelings of self-efficacy, and a
higher subjective norm than males who had unsafe
sex. Females who had safe sex, when compared with
females who had unsafe sex, had higher scores on
the scale for the motive to have sex to express love,
the scale for self-efficacy, and the scale for the need
for intimacy in relationships, and lower scores on the
scale for the motive to have sex to please others.
Table 5. Classification of Condom Use With Steady
Partner(s)
Actual Group
a
Predicted Group
Safe Unsafe
Safe 44 (85%) 8 (15%)
Unsafe 38 (25%) 117 (76%)
Total correct classification 78%
Cross-validated classification 73%
a
Safe group: those who always used condoms with their
steady partner(s) (N ⫽52; 37 males and 15 females)
Unsafe group: those who did not always use condoms with
their steady partner(s) (N ⫽155; 87 males and 68 females)
160 GEBHARDT ET AL JOURNAL OF ADOLESCENT HEALTH Vol. 33, No. 3
Discussion
Two-thirds of the participants had experience with
sexual intercourse. Of those, 93% had had sex with
(a) steady partner(s), and 46% had had sex with (a)
casual partner(s). Females were far less likely to have
experienced casual sex than males, although they
had more frequently engaged in unprotected sex,
particularly with their steady partner(s). This is in
line with research of, for example, Vogels et al [19]
who suggest that this difference between the genders
may be related to the fact that females are more likely
than males to have intercourse with older partners
within the context of a steady relationship.
The participants who had experienced casual sex
were less likely to pursue intimacy in relationships
than those who never had casual sex. Furthermore,
those who had engaged in casual sex were more
inclined to have sex either to please others, or to
enhance their mood, or to experience pleasure. Hav-
ing sex to please others and having sex to enhance
mood were also positively related to the number of
casual partners. This is consistent with previous
research, which indicates that individuals consider
casual relationships less intimate and non-exclusive
than steady relationships [33]. Sprecher and McKin-
ney [34] also found that those who have casual
relationships tend to focus on lust and physical
release and not on commitment and intimacy.
Almost one-fifth of those who had sex with a
steady partner and almost one-half of those who had
sex with casual partners used a condom at each
contact. The finding that condoms are more consis-
tently used with casual partners than with steady
partners is in agreement with earlier reports in the
literature [35–38]. Those who had unprotected sex
with their steady partner frequently reported that
they did not make use of a condom because they
used the pill to prevent conception. The next most
frequently mentioned reason for not using a condom
was that the relationship had already existed for a
longer period of time. This implies that when a
relationship has become more stable, adolescents
have less inclination to protect themselves against
STIs, because they feel increasingly secure about the
relationship itself. In agreement with this, Kelly et al
mentioned that in an unpublished study by Williams
et al [5], it was found that the students’decision to
engage in safe sex with a steady partner is related to
perceived attributes of the partner such as warmth
and kindness. Variables related to actual risk, such as
information concerning the partner’s sexual history,
are much less likely to influence sex practices within
the context of a stable relationship [39]. In our study,
discriminant analyses also revealed that those who
wish to express love through having sex and those
who are in search of intimacy in their relationships
are less likely to use condoms with their steady
partner(s). The desire for intimacy and for emotional
closeness within a relationship, in which mutual
trust prevails, appears therefore to diminish the need
for protected sex.
Consistent with earlier results [27], the discrimi-
nant analysis also revealed that attitude and subjec-
tive norm are important variables for the character-
izations of the group who had safe and the group
who had unsafe sex with their steady partner(s).
Table 6. Discriminant Analysis and Student’st-tests:
Predictors of Condom Use With Casual Partner(s)
Correlation
With
Function
a
Mean
Scores
Groups
b
t-
Value
Self-efficacy .80 safe M ⫽14.77 4.50***
unsafe M ⫽12.07
Attitude .56 safe M ⫽21.04 3.12**
unsafe M ⫽18.92
Subjective norm .39 safe M ⫽22.36 2.19*
unsafe M ⫽21.11
R intimacy .34 safe M ⫽31.30 1.93
†
unsafe M ⫽26.71
M please others ⫺.10 safe M ⫽11.28 ⫺.56
unsafe M ⫽11.76
M enhance mood ⫺.09 safe M ⫽9.95 ⫺.48
unsafe M ⫽10.17
M experience
pleasure
⫺.08 safe M ⫽13.54 ⫺.46
unsafe M ⫽13.66
M express love .07 safe M ⫽12.64 .37
unsafe M ⫽12.52
†
p⬍.10; * p⬍.05; ** p⬍.01; *** p⬍.001.
a
Wilks’Lambda ⫽.86;
2(8)
⫽25.19 (p⬍.001).
b
Safe group: those who always used condoms with their
casual partner(s) (N ⫽84; 67 males and 17 females).
Unsafe group: those who did not always use condoms with
their casual partner(s) (N ⫽96; 68 males and 28 females).
R⫽relationship; M ⫽motive to.
Table 7. Classification of Condom Use With Casual
Partner(s)
Actual Group
a
Predicted Group
Safe 54 (64%) 30 (36%)
Unsafe 27 (28%) 69 (72%)
Total correct classification 68%
Cross-validated classification 64%
a
Safe group: those who always used condoms with their
casual partner(s) (N ⫽84; 67 males and 17 females).
Unsafe group: those who did not always use condoms with
their casual partner(s) (N ⫽96; 68 males and 28 females).
September 2003 NEED FOR INTIMACY IN RELATIONSHIPS AND MOTIVES FOR SEX 161
Those who always had safe sex had a more positive
attitude and a higher perceived subjective norm than
those who did not consistently use condoms. Thus,
within steady relationships, a more positive attitude
toward the use of condoms, and the impression that
others are of the opinion that one should use con-
doms, increase the likelihood of their being used.
The most common reason for not always having
protected sex with a casual partner was again the use
of the contraceptive pill. Other frequently mentioned
causes of unsafe sex were: the unavailability of a
condom, the influence of alcohol or drugs, and
negligence. Discriminant analysis showed that all
three constructs of the Theory of Planned Behavior
and need for intimacy in relationships were able to
differentiate between those who always had safe sex
and those who did not. The consistent condom users
had a more positive attitude, a higher perceived
subjective norm, and stronger feelings of self-efficacy
to use condoms when compared with the inconsis-
tent users. Interestingly, participants who did not
engage in casual sex had greater self-efficacy for
condom use, whereas those who had casual sex were
actually more likely to use them.
Furthermore, it should be noted that in steady
relationships the pursuit of intimacy was negatively
related to safe sex. For casual partners, however, it
was positively related to safe sex. This is contrary to
expectations of Sanderson and Cantor [32] that safe
sex is most likely to occur in situations that corre-
spond with the primary relationship goals of the
individual. Therefore, according to Sanderson et al,
individuals who pursue intimacy goals would be
more inclined to use condoms in steady relation-
ships, and individuals who pursue identity goals
would prefer to use condoms in casual relationships.
The results of our study, however, indicate that
while in steady relationships the need for intimacy
interferes with condom use, it facilitates condom use
with casual partners. It appears that those who attach
importance to intimacy and mutual trust are more
likely to consider the consequences of unsafe casual
sex. In other words, they seem to take into consider-
ation a possible future steady relationship, in which
they would like to experience exactly those qualities.
According to Abraham et al [31], the rated impor-
tance of having sexual intercourse is a significant
determinant of condom use among those who intend
to use this contraceptive with a new partner. In
agreement with Cooper et al [30], we found that the
motives one has for having sex influences sexual
behavior (such as the number of sexual partners,
whether one engages in casual sex, and condom use).
It is likely that the motives for sex are closely linked
to the importance attached to sexual intercourse.
Simultaneously, the motives for sex provide insight
into why sex is of such importance to the individual.
We performed some analyses on gender differ-
ences in determinants of (un)safe behavior. It ap-
peared that for steady relationships no major differ-
ences existed between males and females. However,
with respect to casual partners, it seemed that for
males the more rational decision-making model ap-
plies, including the variables attitude, subjective
norm, and self-efficacy. For females, however, the
contextual factors (i.e., the meaning attached to the
encounter and to the relationship) appeared to be
more crucial in understanding their condom use
behavior. However, the sample sizes of the sub-
groups in this study were in some instances very
small, particularly for women who have safe sex. It is
advisable that future studies further explore gender
differences.
Limitations
Some methodological issues should also be consid-
ered. One important limitation of our study is that it
is based on a cross-sectional design, which prevented
us from determining the causal direction of the
observed associations. The sample consisted of 701
adolescents attending vocational high school, of
which 470 appeared to be sexually active. Vocational
education is the lowest educational level in The
Netherlands. Future studies, involving students
from other types of education, and high school
dropouts should establish whether the results can be
generalized to the total population of Dutch adoles-
cents. Furthermore, participants were asked to report
on their past sexual behavior and condom use. The
recollection of the past behavior may not have been
completely accurate, and may have been colored by
the subsequent events, beliefs, and emotions. A gen-
eral tendency to answer in a socially desirable man-
ner may also have influenced the response pattern of
the participants. However, a substantial proportion
of the respondents admitted to having unsafe sex,
which does not agree with the norm they expected
from the researchers.
These considerations do not invalidate our results,
which clearly indicate that expanding the Theory of
Planned Behavior with constructs related to the
meaning of sex and relationships, can increase our
understanding of condom use with steady and ca-
sual partner(s). The new model, incorporating vari-
ables from these two perspectives, led to a model
162 GEBHARDT ET AL JOURNAL OF ADOLESCENT HEALTH Vol. 33, No. 3
which could correctly classify approximately 68%
(for sex with casual partners) or 78% (for sex with
steady partners) of all participants.
Conclusion
The findings have implications for health promotion
activities concerning safe sex. It is evident that inter-
ventions should incorporate insights derived from
the Theory of Planned Behavior. The data reported
here support the notion that improving attitudes
toward the use of condoms, developing stronger
subjective norms, and enhancing self-efficacy have a
positive influence on the practice of safe sex in
adolescence. Our results also indicate that interven-
tions that target specific subgroups and as such take
into account the type of relationships (e.g., steady or
casual) and the meaning of the relationship and sex
itself would be even more effective in promoting safe
sex. For example, one may argue that those who have
a steady affectionate relationship should learn to
regard safe sex as a way of demonstrating love and
concern for their partner. Possibly the slogan of a
Dutch health campaign well-known to a large part of
the population: “I will either have safe sex, or I will
not have sex at all,”should be replaced for this
subgroup by “I care about you, therefore I will use a
condom.”For those who engage in casual sex, a
more adequate motto may be ‘feel good, buy con-
doms, carry them with you, and use them: You are
worth it!’
The results of our study indicate that the goals
which adolescents pursue by having a relationship
and by having sex are related to their sexual behavior
as expressed by the number of sexual partners,
whether they are engaging in casual sex, and con-
dom use. If these considerations are incorporated in
the decision-making framework for the development
of interventions, it may well increase the practice of
safe sex among adolescents.
References
1. Erikson EH. Childhood and Society. New York: Norton, 1950.
2. Kroger J. Identity in Adolescence: The Balance Between Self
and Other, 2nd Edition. London: Routledge, 1996.
3. DuBois-Reymond M, Te Poel Y, Ravesloot J. Jongeren en hun
Keuzes. Bussum: Coutinho, 1998, [Youngsters and Their
Choices].
4. Lear D. Sexual communication in the age of AIDS: The
construction of risk and trust among young adults. Soc Sci
Med 1995;41:1311–23.
5. Kelly JA, Kalichman SC. Increased attention to human sexu-
ality can improve HIV-AIDS prevention efforts: Key research
issues and directions. J Consult Clin Psychol 1995;63:907–18.
6. Buysse A. Adolescents, young adults and AIDS: A study of
actual knowledge vs. perceived need for additional informa-
tion. J Youth Adolesc 1996;25:259–71.
7. Minichello V, Paxton S, Cowling V, et al. University students’
knowledge of STDs: Labels, symptoms and transmissions. Int
J STD AIDS 1996;7:353–8.
8. World Health Organization. Young People and Sexually
Transmitted Diseases: Fact Sheet Number 186. Geneva: World
Health Organization, 1997.
9. Centers for Disease Control and Prevention. Tracking the
Hidden Epidemics: Trends in STDs in the United States.
Atlanta, GA: Centers for Disease Control and Prevention,
2000.
10. UNAIDS Report on the global HIV/AIDS epidemic . Avail-
able at: http://www.unaids.org., 2000. Accessed June 10,
2001.
11. U.S. Department of Health and Human Services. Sexually
Transmitted Disease Surveillance, 1999. Atlanta, GA: Centers
for Disease Control and Prevention, 2000.
12. SOA-foundation. Epidemiologie van SOA in Nederland: Al-
gemeen [Epidemiology of STI in the Netherlands]. Available
at: http://www.soa.nl2002. Accessed June 10, 2002.
13. Haks K, Van de Laar MJW. Stijging soa niet beperkt tot
risicogroepen [Increase in STI not limited to risk groups].
Infectieziekten Bulletin 2001;12:412–5.
14. Van de Laar M, Haks K, Coenen T. Sterke stijging in het aantal
soa-diagnoses in 2000 [Strong increase in the number of
STI-diagnoses in 2000]. Infectieziekten Bulletin 2001;12:285–8.
15. Fife-Schaw CR, Breakwell GM. Estimating sexual behaviour
parameters in the light of AIDS: A review of recent UK studies
of young people. AIDS Care 1992;4:187–201.
16. Vogels T, Van der Vliet R. Jeugd en Seks. The Hague: SDU,
1990, [Youth and Sex].
17. Leenaars PEM, Rombouts R, Lammerdin AG, Zwetsloot LM.
Determinanten van condoomgebruik onder jongeren [Deter-
minants of condom use among youngsters]. Gedrag Gezond
1991;19:77–87.
18. Dijkstra A, Garnefski N, Gebhardt W, et al. Gedrag en
gezondheid 2000: Resultaten van een onderzoek naar gedrag
en gezondheid bij leerlingen van een MBO [Behaviour and
Health 2000]. Leiden University, Department of Clinical and
Health Psychology, 2000.
19. Vogels T, Brugman E, Van Zessen G. AIDS-related knowl-
edge, attitudes, and behavior: A comparison of Dutch students
and dropouts. Adolescence 1999;34:369–79.
20. Abraham C, Sheeran P. Modelling and modifying young
heterosexuals’HIV-preventive behaviour: A review of theo-
ries, findings and educational implications. Patient Educ
Couns 1994;23:173–86.
21. Harden A, Ogden J. Condom use and contraception non-use
amongst 16–19 year olds: A within subjects comparison.
Psychol Health 1999;14:697–709.
22. Brown LK, DiClemente RJ, Reynolds LA. HIV prevention for
adolescents: Utility of the health belief model. AIDS Educ Prev
1991;3:50–9.
23. Mahoney CA, Thombs DL, Ford OJ. Health belief and self-
efficacy models: Their utility in explaining college student
condom use. AIDS Educ Prev 1995;7:32–49.
24. Van der Velde FW, Van der Pligt J. AIDS-related health
behavior: Coping, protection motivation, and previous behav-
ior. J Behav Med 1991;14:429–51.
September 2003 NEED FOR INTIMACY IN RELATIONSHIPS AND MOTIVES FOR SEX 163
25. Morrison DM, Rogers Gillmore M, Baker SA. Determinants of
condom use among high-risk heterosexual adults: A test of the
theory of reasoned action. J Appl Soc Psychol 1995;25:651–76.
26. Godin G, Maticka-Tyndale E, Adrien A, et al. Cross-cultural
testing of three social cognitive theories: An application to
condom use. J Appl Soc Psychol 1996;26:1556–86.
27. Sheeran P, Abraham C, Orbell S. Psychosocial correlates of
heterosexual condom use: A meta-analysis. Psychol Bull 1999;
125:90–132.
28. Maes S, Gebhardt WA. Self-regulation and health behaviour:
The health behaviour goal model. In: Boekaerts M, Pintrich
PR, Zeidner M (eds). Handbook of Self-Regulation. San Diego,
CA: Academic Press, 2000:343–68.
29. Gebhardt WA, Maes S. Integrating social-psychological frame-
works for health behavior research. Am J Health Behav
2001;25:528–36.
30. Cooper ML, Shapiro CM, Morrison AM. Motivations for sex
and risky sexual behavior among adolescents and young
adults: A functional perspective. J Pers Soc Psychol 1998;75:
1528–58.
31. Abraham C, Sheeran P, Norman P, et al. When good inten-
tions are not enough: Modeling postdecisional cognitive cor-
relates of condom use. J Appl Soc Psychol 1999;29:2591–2612.
32. Sanderson CA, Cantor N. Social dating goals in late adoles-
cence: Implications for safer sexual activity. J Pers Soc Psychol
1995;68:1121–34.
33. Snyder M, Simpson JA. Orientations toward romantic rela-
tionships. In: Perlman D, Duck S (eds). Intimate Relationships:
Development, Dynamics and Deterioration. London: Sage,
1987:45–62.
34. Sprecher S, McKinney K. Sexuality. London: Sage, 1993.
35. Plichta SB, Weisman CS, Nathanson CA, et al. Partner-specific
condom use among adolescent women clients of a family
planning clinic. J Adolesc Health 1992;13:506–13.
36. Pilkington CJ. Is safer sex necessary with a “safe”partner?
Condom use and romantic feelings. J Sex Res 1994;31:203–10.
37. Reisen CA, Poppen PJ. College women and condom use:
Importance of partner relationship. J Appl Soc Psychol 1995;
25:1485–98.
38. Ellen JM, Cahn S, Eyre SL, Boyer CB. Types of adolescent
sexual relationships and associated perceptions about condom
use. J Adolesc Health 1996;18:417–22.
39. Buysse A. Safer sexual decision making in stable and casual
relationships: A prototype approach. Psychol Health 1998;13:
55–66.
164 GEBHARDT ET AL JOURNAL OF ADOLESCENT HEALTH Vol. 33, No. 3
