Young Women's Degree of Control over First Intercourse: An ...

Young Women’s Degree of Control over First
Intercourse: An Exploratory Analysis
By Joyce Abma, Anne Driscoll and Kristin Moore
does not include intercourse to unwant-
ed but voluntary sex and to nonvoluntary
Context: While policymakers and researchers alike often seem to believe that young women’s
decision to initiate sexual intercourse is conscious and free of ambiguity, the actual degree of
intercourse and rape. The issue of early
control that such young women exert over first intercourse has rarely been explicitly examined.
age at first sexual experience is inter-
twined with this topic at all points along
Methods: The 1995 National Survey of Family Growth asked all women who had experienced
the continuum.
intercourse to rate, on a 1–10 scale, the wantedness of their first intercourse; they were then
asked whether the experience was voluntary. Logistic regression analysis of data for women
Unwanted sexual experiences and abuse
aged 15–24 who had experienced first premarital intercourse was performed to test the effect
that do not include intercourse fall on the
of background factors and wantedness scores on contraceptive use at voluntary first intercourse.
less extreme end of the range. A study in
Los Angeles, in which 18% of adolescent
Results: Twenty-four percent of women aged 13 or younger at the time of their first premarital in-
females reported having had an unwant-
tercourse report the experience to have been nonvoluntary, compared with 10% of those aged
19–24 at first premarital intercourse. About one-quarter of respondents who reported their first in-
ed sexual experience, suggests that these
tercourse as voluntary chose a low value (1–4) on the wantedness scale. Women whose first part-
experiences are not uncommon. Moreover,
ner was seven or more years older than themselves were more than twice as likely as those whose
of the experiences reported in the survey,
first partner was the same age or younger to choose a low value (36% vs. 17%). Women whose
39% had occurred before the age of 13, and
partner had been seven or more years older were also less likely than other women to have used
half had occurred between 13 and 16.4
contraceptives at first intercourse. After the introduction of controls for demographic and back-
Unwanted intercourse represents a
ground factors, partner age discrepancy and relationship status, wantedness of voluntary first in-
more extreme form of negative sexual ex-
tercourse was not independently related to the odds of contraceptive use at that intercourse.
perience. A substantial proportion of
Conclusion: Characterizing women’s first intercourse as simply voluntary or nonvoluntary is
young women apparently experience un-
inadequate. Measures that take into account degrees of wantedness may help elucidate rela-
wanted first or early sexual intercourse:
tionships between sexual initiation, contraceptive use and teenage pregnancy. The fact that sub-
In the 1992 National Health and Social Life
stantial numbers of young women voluntarily participated in a first sexual experience about which
Survey (NHSLS), 25% of women report-
they felt ambivalent or negative deserves the attention of program planners and service providers.
ed that while their first intercourse had not
Family Planning Perspectives, 1998, 30(1):12–18
been forced, neither had it been wanted.5
Nonvoluntary or forced sexual inter-
course, including rape, represents the
this event. Consideration of these factors
most extreme category on the spectrum of
is important, especially given our knowl-
Over the past two decades, the per-
centage of teenagers who have ex-
negative sexual experiences. In the
perienced sexual intercourse has
edge that a certain proportion of inter-
NHSLS, 4% of women aged 18–59 de-
grown (although the increase abated in the
course is nonvoluntary, and that this pro-
scribed their first intercourse as “forced.”
early 1990s).1 Earlier age at first intercourse
portion is greater at younger ages.2
A number of risk factors have been
poses increased risks to the health and
According to 1987 national data, 7% of
shown to be associated with these nega-
well-being of adolescents, as the initiation
women and men aged 18–22 had experi-
tive experiences among girls and young
of sexual intercourse marks the beginning
enced forced intercouse, and almost half
women. For unwanted intercourse, one
of exposure to the risk of unintended preg-
of all women’s nonvoluntary experiences
group of factors involves individual be-
nancy and sexually transmitted diseases.
had occurred before age 14.3
haviors and characteristics, sometimes in-
These outcomes have important implica-
Even when women perceive their first
teracting with peer influences. Adolescent
tions for adolescents, and are among the
sexual intercourse as having been volun-
girls who scored high in peer conformity
reasons why programs designed to avert
tary, age differences between young
negative consequences among girls and
women and their partners may imply im-
Joyce Abma is demographer with Reproductive Statis-
young women focus on delaying sexual in-
balances in control over the experience.
tics Branch, National Center for Health Statistics, Hy-
attsville, MD. Anne Driscoll is postdoctoral fellow with
tercourse and promoting the importance
Augmenting what we currently know
and Kristin Moore is executive director of Child Trends,
of contraception.
about sexual initiation with additional in-
Inc., Washington, DC. An earlier version of this article
Much of the pertinent research on
formation on its wantedness is especially
was presented at the annual meeting of the Population
young women’s initiation of sexual ac-
important in light of growing public con-
Association of America, Washington, DC, March 27–29,
tivity assumes that the decision to have
cern about sexual victimization and
1997. The National Survey of Family Growth was joint-
ly planned and funded primarily by the National Cen-
sex is a choice that young women make
teenage pregnancy.
ter for Health Statistics, the National Institute for Child
after weighing the opportunity costs in-
Health and Human Development and the Office of Pop-
volved in such a situation. Analyses of
Background
ulation Affairs, with additional support from the Ad-
young women’s first intercourse rarely
Negative Sexual Experiences
ministration for Children and Families. The authors ded-
take into account the degree of control that
Previous research has focused on several
icate their work to the memory of Kathryn A. London,
whose expertise and dedication were critical to the de-
they have, or perceive themselves as hav-
types of negative sexual experiences, rang-
velopment of the survey instrument and to the innova-
ing, over the timing and circumstances of
ing from unwanted contact or abuse that
tive measures used in this research.
12
Family Planning Perspectives

---

were more vulnerable to unwanted in-
greater number of partners.15 These can re-
about the circumstances of first inter-
tercourse initiated by their male peers.6 In
sult in increased and more prolonged ex-
course. We focus in this article on non-
the NHSLS, peer pressure was cited by
posure to the risks of sexually transmitted
voluntary first intercourse in a national-
women as the main reason for unwanted
diseases and early pregnancy.
ly representative sample of women,
voluntary first intercourse.7
Nonvoluntary intercourse may also be
examining the prevalence of nonvolun-
As is the case with less extreme forms of
associated with subsequent nonvoluntary
tary first intercourse among young
sexual abuse, younger women are at greater
sexual experiences: A history of sexual
women aged 15–24 in 1995. Among the
risk of experiencing nonvoluntary inter-
abuse in general appears to increase the
young women who described their first
course. Life-table estimates of data from the
odds of sexual coercion in junior and senior
intercourse as voluntary, we explore vari-
National Survey of Children suggest that
high school.16 Although data are scarce and
ations in the degree to which young
women who experienced sex at younger
findings are not entirely consistent, non-
women wanted first intercourse to hap-
ages were more likely to have experienced
voluntary intercourse may be associated
pen at that time. We also investigate the
nonvoluntary sex than were those whose
with a greater likelihood of pregnancy dur-
possibility that the age difference between
first intercourse occurred at older ages.8
ing adolescence.17 The experience of non-
a young woman and her partner may be
Family of origin may also play an im-
voluntary intercourse may increase the
associated with the wantedness of first in-
portant role in the chance of experiencing
likelihood of such risk factors for teenage
tercourse. Finally, we address the poten-
forced or coerced intercourse, at both first
pregnancy as early first sex, contraceptive
tial link between wantedness of voluntary
and subsequent intercourse. White females
nonuse and multiple partners.18
first intercourse and contraceptive use at
who lived apart from their parents before
Sexual abuse, particularly at early ages,
that intercourse. We also perform multi-
age 16 are at higher risk of experiencing
has also been linked to subsequent prob-
variate regression analysis of the effects
nonvoluntary intercourse.9 Parenting style
lems, such as involvement in prostitution,
on contraceptive use of degree of want-
is also relevant: Young women who have
impairments to psychological well-being
edness of voluntary first intercourse, part-
authoritative parents and whose parents
and mental health,19 alcohol abuse and
ner’s age and his relationship with the
monitored their behavior closely are at
suicidal ideation.20 A recent national study
young woman.
lower risk than are other adolescents.10
showed that women who as children ex-
Some parental behaviors also appear to
perienced sexual intercourse with adults
Data and Methods
be related to the risk of nonvoluntary in-
had lower overall well-being than did
The National Survey of Family Growth
tercourse. Daughters of parents who
women who had experienced childhood
(NSFG) is a periodic survey that collects
drank heavily or used illegal drugs dur-
sexual contact with adults that only in-
data on the factors affecting pregnancy,
ing their preschool or elementary years,
cluded fondling or who did not experi-
health and childbearing of women in the
for example, are more likely to experience
ence any such sexual contact.21
United States. Cycle 5 obtained informa-
nonvoluntary intercourse than are daugh-
Several explanations have been offered
tion from a national probability sample of
ters of parents who did not engage in these
for the processes by which nonvoluntary
10,847 noninstitutionalized women aged
behaviors.11 The role played by family in-
intercourse may lead to subsequent nega-
15–44 in the civilian population. Inter-
come and other resources remains unclear,
tive outcomes. One hypothesis is that a high
viewing took place between January and
although family welfare receipt during
level of sexual activity among young
October 1995. Black women and Hispan-
childhood has been found to be unrelat-
women who have been abused, together
ic women were oversampled, which fa-
ed to the odds of nonvoluntary sex once
with feelings of powerlessness about sex,
cilitates the analysis of many phenomena
other factors are controlled.12
puts teenage women at greater risk for preg-
within these racial and ethnic groups. The
Religious participation may also be re-
nancy.22 In addition, young women who
sample also included large numbers of
lated to the likelihood of nonvoluntary
have undergone developmentally inap-
teenagers and young adults.
sex. College women who attended reli-
propriate sexual socialization may have al-
Cycle 5 of the NFSG included a di-
gious services relatively less often were
tered concepts of self and others resulting
chotomous measure of women’s percep-
found to be more likely to report experi-
from adaptations and compromises that
tions of whether their first intercourse was
encing coerced sex than were women who
were needed for psychological survival.
voluntary or nonvoluntary, and a 1–10
attended religious services more often.13
Resulting “hallmarks of sexual victim-
scale representing the degree to which
Other behaviors, such as alcohol abuse
ization” include perceptions of power-
women wanted the first intercourse to
and higher levels of sexual activity, are
lessness, diminished capacity for acade-
occur. Until now, it has been difficult to ad-
also associated with an increased likeli-
mic work and other tasks of adolescence,
dress the issue of varying degrees of want-
hood of experiencing forced sex.14
heightened vulnerability to males, an in-
edness of first intercourse using existing
creased salience of sexual issues, miscon-
data. Inclusion of these two measures in
Outcomes of Abuse and Forced Intercourse
ceptions about sexual norms and the
Cycle 5 reflects an attempt to allow for the
Besides the trauma of the experience of
equation of sex with love and caregiving.23
complexity and multidimensionality of
nonvoluntary sexual intercourse itself,
From a life-course perspective, sexual con-
the concept of nonvoluntary intercourse.
other negative outcomes of forced inter-
tact with an adult during childhood may
Cycle 5 also collected data on the char-
course have been documented. Since non-
be the starting point of a sexual trajecto-
acteristics of all first voluntary male sex-
voluntary intercourse occurs more com-
ry in which individuals are more likely to
ual partners and a selected subset of sub-
monly at younger ages than does voluntary
engage in potentially harmful sexual be-
sequent sexual partners. Finally, the
intercourse, it can produce the same neg-
havior that in turn produces adverse long-
survey sought a wide array of information
ative consequences that have been shown
term consequences.24
on the women’s family background char-
to result from early first intercourse—
Although there is some evidence of the
acteristics and other demographic and at-
among them, a higher frequency of subse-
negative consequences of early first in-
titudinal characteristics, allowing con-
quent adolescent sexual activity and a
tercourse, much remains to be learned
sideration of factors that have been shown
Volume 30, Number 1, January/February 1998
13

---

Young Women’s Control over First Intercourse
ings involving the scale to those whose
who chose low scale values (three and
Table 1. Percentage distribution of women
first intercourse was voluntary. In addi-
lower) also reported first intercourse as
aged 15–24 who had premarital first inter-
course, by whether first intercourse was vol-
tion, all analyses are limited to the 2,042
“voluntary.” For example, 45% of the
untary or nonvoluntary, according to wanted-
women who experienced premarital first
young women who chose the lowest scale
ness of first intercourse, National Survey of
intercourse, due to potential differences
value reported their experience had been
Family Growth, Cycle 5, 1995 (N=2,017)
between married and unmarried respon-
“voluntary.” While hardly any respon-
Wanted-
N
Voluntary
Non-
Total
dents in the voluntariness of first sex and
dents who reported that intercourse was
ness
voluntary†
other outcomes, such as contraceptive use.
not voluntary chose high scale values,
1
231
55.4
44.6
100.0
Because this configuration of measures
about 27% who reported voluntary first
2
107
77.4
22.6
100.0
has not been used in any prior survey, in-
intercourse also chose a scale value of four
3
153
89.5
10.5
100.0
4
157
97.0
3.0
100.0
tersurvey comparisons may be difficult.
or lower (not shown).
5
363
97.1
2.9
100.0
However, the measures represent an im-
6
180
99.3
0.7
100.0
portant first attempt to capture on a national
Correlates of Wantedness
7
220
99.8
0.2
100.0
8
205
100.0
0.0
100.0
level a more nuanced sense of the experi-
Overall, 91% of the women reported that
9
120
100.0
0.0
100.0
ence of first sex than items assessing age at
their first intercourse was voluntary (Table
10
281
100.0
0.0
100.0
first intercourse and dichotomous measures
2). Most young women whose first inter-
†Twenty-five women who described their first intercourse as “rape”
of voluntariness can provide alone.
course was voluntary rated the wanted-
were not asked whether first intercourse was voluntary or non-
voluntary, and so are excluded from this table. Note: In this and
ness of this experience in the middle-to-
subsequent tables, wantedness is classified on a 1–10 scale, with
Results
high range: About half classified the
1=least wanted and 10=most wanted.
Correlation of Responses
experience in the range of 5–8, and about
Virtually no respondents who assigned
one-sixth assigned it the highest possible
to be important risk factors for early, non-
scale values at or above the two middle
value (10). Seven percent gave the expe-
voluntary intercourse.
scores (five and six) chose the response
rience the lowest value on the scale.
Before presenting the scale and the di-
“not voluntary” to describe their first pre-
Young black women were more likely to
chotomous measures of wantedness, in-
marital intercourse (Table 1). In addition,
rate the wantedness of their first intercourse
terviewers determined whether each re-
the proportion who reported voluntary
as “one” (13%) than were white women
spondent had had intercourse, as well as
first intercourse increased linearly with ris-
(6%) or Hispanic women (4%). At the same
the date of and her age at first intercourse.
ing scale value, with 100% of the young
time, Hispanic women were more likely to
Interviewers were instructed not to count
women who chose seven, eight, nine, or
rate the wantedness of their first experience
anything other than vaginal heterosexu-
10 also choosing “voluntary.”
as “10” (21%) than were white women
al intercourse.
In contrast, a substantial percentage
(14%) or black women (12%).
Respondents were then asked to rate
the wantedness of their first intercourse.
Table 2. Among women aged 15–24 who had premarital first intercourse, percentage whose
The interviewer directed the respondent
first intercourse was nonvoluntary, and among those who had voluntary first intercourse, per-
to look at a card depicting a numbered
cent distribution by wantedness of first intercourse, all according to selected characteristics
scale and asked, “Which number would
Characteristic
% non-
Wantedness (N=1,859)
you say comes closest to describing how
voluntary†
1
2
3–4
5–6
7–8
9
10
Total
much you wanted that first sexual inter-
(N=2,042)
course to happen? On this scale, a one
All
9.1
6.9
4.1
15.4
28.0
23.6
6.9
15.1
100.0
means that you really didn’t want it to
Age at first intercourse
happen at the time, and a 10 means that
<14
23.6
12.5
6.3
25.4
32.3
12.1
1.0
10.4
100.0
you really wanted it to happen.” If a
14
9.3
10.3
7.4
15.0
28.7
20.9
6.0
11.7
100.0
woman volunteered that it was “rape,” all
15
9.5
7.5
3.2
16.6
31.2
24.7
6.0
10.8
100.0
16
4.7
6.8
4.0
15.1
27.5
25.1
7.0
14.5
100.0
subsequent questions about first inter-
17
5.0
4.0
3.8
14.2
24.7
25.5
9.5
18.4
100.0
course were skipped.
18
4.8
2.3
1.5
13.1
28.7
26.4
8.4
19.7
100.0
All remaining respondents were next
19–24
9.7
4.5
1.5
7.6
22.5
28.2
10.8
24.9
100.0
asked: “Would you say then that this first
Race/ethnicity
sexual intercourse was voluntary or not
White non-Hispanic
7.9
6.0
3.5
14.9
27.2
25.8
8.4
14.2
100.0
Black non-Hispanic
10.8
12.9
5.8
20.9
28.8
14.8
4.5
12.3
100.0
voluntary?” Respondents who chose
Hispanic
10.8
4.3
5.6
11.9
29.5
23.8
4.0
20.8
100.0
“voluntary” were asked subsequent ques-
tions concerning first sexual partners and
Childhood living arrangement
Both biological/adoptive
contraception at first intercourse. As in the
parents from birth
7.9
5.5
3.7
14.7
26.8
25.6
8.6
15.1
100.0
case of those who identified the first in-
Single parent from birth
11.8
11.9
3.0
17.2
30.1
16.3
0.7
20.9
100.0
tercourse as “rape,” those who chose “not
Both biological/adoptive
parents, then one parent
9.4
5.6
3.1
16.2
28.9
23.6
6.9
15.6
100.0
voluntary” (or “don’t know” or “refused”)
Stepparent (ever)
8.9
8.5
3.7
15.2
32.1
23.0
5.7
11.9
100.0
were not asked subsequent questions
about their first intercourse.
Age first lived away from parents
Never
8.8
6.1
3.4
14.5
29.8
22.5
8.1
15.6
100.0
Because we sought to study these issues
≤16
12.3
12.2
7.3
20.7
31.6
15.3
1.5
11.4
100.0
among a contemporary cohort, we re-
17
10.5
9.8
6.9
11.0
31.2
21.2
3.6
16.3
100.0
stricted our analysis to the 2,933 women
18
8.4
6.3
3.5
16.7
21.2
27.9
8.3
16.1
100.0
≥19
8.2
5.3
3.2
15.1
31.6
24.3
7.1
13.5
100.0
aged 15–24 at the time of the survey. To
maintain consistency of the analysis sub-
†Includes 25 women who reported first intercourse to have been “rape.” Notes: Percentages may not total 100% due to rounding. See
note to Table 1.
sample, we limit the presentation of find-
14
Family Planning Perspectives

---

Age at first intercourse also showed a
Table 3. Among women aged 15–24 who had voluntary, premarital first intercourse, percent-
strong and regular pattern of differences in
age who chose wantedness values for first intercourse of less than 5 on a 1–10 scale, by part-
wantedness scale ratings. Thirteen percent
ner’s age relative to the woman’s, according to selected background characteristics (N=1,859)
of those who had voluntary first intercourse
at age 13 or younger chose the lowest want-
Characteristics
All
Same age
1–2 years
3–4 years
5–6 years
≥7 years
partners
or younger
older
older
older
older
edness value, compared with only 5% of
those aged 19–24 at first intercourse. At the
Total (standard error)
26.2 (1.11)
17.2 (2.09)*
26.5 (1.71)
29.3 (2.65)
32.7 (4.26)
36.2 (5.54)
opposite end of the scale, 10% of women
Age at first intercourse
aged 13 or younger at first intercourse chose
≤14
37.5
33.1
33.9
40.4
40.2
45.2
15
27.5
20.7
27.3
27.3
29.5
42.3
the highest wantedness value, compared
16
25.6
12.8
25.3
29.9
45.5
45.1
with 25% of those who were 19–24.
17
21.4
17.1
24.8
13.2
21.3
44.3
Nine percent of women aged 15–24 who
18
17.0
19.9
19.9
7.0
15.1
16.1
≥19
13.2
7.5
17.7
18.0
16.0
4.1
had premarital first intercourse classified
their experience as having been nonvol-
Race/ethnicity
untary or rape.* Nonvoluntary first in-
White non-Hispanic
24.2
15.7
23.2
29.7
30.7
37.6
Black non-Hispanic
39.5
18.4
45.3
39.1
47.3
53.0
tercourse occurred most frequently among
Hispanic
22.0
30.5
19.7
15.5
27.0
23.5
those aged 13 or younger at first inter-
*Differences between this subgroup and other age-difference categories were statistically significant at p<.05.
course (24%). Hispanic and black women
experienced higher rates of nonvoluntary
first intercourse (11% each) than did white
This pattern was most pronounced
Contraceptive Use at First Intercourse
women (8%). Living with both biological
among women who were 16 or younger
Altogether, about one-quarter of the
or adoptive parents from birth until leav-
at voluntary first intercourse, and was par-
women aged 15–24 were not using a con-
ing home was associated with a lower pro-
ticularly marked among those aged 16 at
traceptive method when they first had sex
portion of nonvoluntary first intercourse
first intercourse: These women were more
(Table 4). Nonuse at first voluntary inter-
(8%) than was living with a single parent
than three times as likely to have given a
course decreased gradually, however, with
from birth until leaving home (12%). Re-
low scale rating to describe voluntary first
rising wantedness of first intercourse. Thir-
spondents who had left home at younger
intercourse with a man seven or more
ty-nine percent of young women who
ages had a higher likelihood of nonvol-
years older as they were when describing
scored the wantedness of their first inter-
untary first intercourse.
a pairing with a partner who was the same
course as 1–2 had not used a method, com-
age or younger (45% vs. 13%). Caution is
pared with 22% of those who chose a want-
Age of Partner and Wantedness
warranted when interpreting the patterns
edness value of 9–10. Differences were
The extent of age difference between the
by single year of age at first intercourse,
relatively small in magnitude, but statisti-
partner and the respondent was inverse-
however, as cell sizes become quite small.
cally significant: Calculation of confidence
ly and significantly associated with the
The association between partner-re-
intervals showed that the prevalence of
level of wantedness of voluntary first in-
spondent age difference and wantedness
contraceptive use associated with scale
tercourse (Table 3).† The wantedness score
of intercourse differed by race and ethnic
values 1–2 and 3–4 differed significantly
decreased as the difference between the
background. Among white women, the
from that for scale values 7–8 and 9–10.
woman’s age and that of her partner in-
proportion with low scale ratings in-
The partner’s condom use accounted
creased. The proportion who assigned a
creased steadily with the partner’s age dif-
for most of the reported contraceptive use
wantedness rating of four or lower to their
ference. Among black women, a slightly
at first intercourse. The proportion of
first intercourse was twice as great among
different pattern emerges: The proportion
women who reported condom use by
women who first had sex with a man
choosing low scale ratings increased
their first partner increased from 43%
seven or more years older as it was among
sharply with an age difference of one year,
among the group that selected the lowest
those who had first intercourse with a man
and remained relatively high, peaking at
wantedness scale values (1–2) to 56%
of the same age or younger. The differ-
53% among women whose first partner
among the women who assigned the next
ences in the proportions between those
was seven or more years older. Hispanic
two values (3–4), and then remained at
with a partner who was the same age or
women had a somewhat more erratic dis-
about this level for all of the remaining
younger and all other age-difference cat-
tribution: The highest
egories were statistically significant.
proportion giving a low
Table 4. Percentage distribution of women aged 15–24 who had
scale rating was among
voluntary, premarital first intercourse, by contraceptive use at first
*Subsequently, “nonvoluntary” first intercourse refers
women whose first
intercourse, according to wantedness of first intercourse (N=1,859)
to intercourse that the respondent reported as “rape” or
partner had been the
as “not voluntary.”
Wantedness
Any method
Pill
Condom
Other†
None
Total
same age or younger.
(and standard
†The scale was dichotomized to efficiently and mean-
There was no consistent
error)
ingfully present the examination of partner age dis-
increase in the percent-
All
73.0 (1.12)
11.8
55.0
6.3
27.0
100.0
crepancy by characteristics of the young woman. The
age reporting low scale
1–2
61.5 (4.03)*
11.1
42.9
7.5
38.5
100.0
group of ratings representing “low” scale values (1–4)
3–4
70.3 (2.80)*
8.8
56.3
5.2
29.7
100.0
was chosen on the assumption that respondents would
values with increasing
5–6
72.2 (2.15)
11.5
55.0
5.7
27.8
100.0
tend to rate less wanted experiences lower than the nu-
age difference. Again,
7–8
76.6 (2.28)
10.9
60.5
5.3
23.4
100.0
9–10
77.9 (2.37)
15.5
54.3
8.1
22.1
100.0
merical middle of the scale—in this case, five and six.
small cell sizes may
Analyses using different groupings of values (for ex-
have compromised the
*Significantly different from categories “7–8” and “9–10” at p <.05. †Includes female condom,
ample, 1–2 vs. 3–10) did not reveal substantially differ-
foam, jelly or cream, suppository, diaphragm, implant, injectable, IUD, female sterilization, va-
stability of some of the
ent or more pronounced patterns. Moreover, isolating
sectomy, natural family planning, rhythm, withdrawal, spermicidal sponge, morning-after pill,
cervical cap and other unspecified methods; women who reported their partner was sterile are
only the lowest values resulted in small cell sizes, com-
percentages for the His-
also included. Notes: Percentages may not total 100% due to rounding. See note to Table 1.
promising the value of detailed breakdowns of the data.
panic women.
Volume 30, Number 1, January/February 1998
15

---

Young Women’s Control over First Intercourse
scale values. The percentage of women
ceptives at first intercourse, and women
The women who described their first in-
who used the pill remained at a fairly
whose mother had not finished high
tercourse as voluntary varied in their as-
steady 11% across the wantedness scale.
school were about half as likely as those
sessments of how much they had wanted
We next examined the possible role of
whose mother had completed four or
the experience to happen. About one-
women’s age, race or ethnicity and moth-
more years of college to have done so.
quarter of those who described their first
er’s education in the relationship between
When we controlled for these demo-
intercourse as voluntary nevertheless gave
wantedness and contraceptive use at first
graphic and background characteristics, we
low wantedness scores (1–4), indicating
voluntary intercourse.* Consistent with
found that the women’s relationship with
that while they had consented, they had
findings from Table 4, the data in Table 5
their partner at voluntary first intercourse
not wanted the intercourse to happen. The
indicate that the proportion using con-
was significantly related to the odds of
reason for these combinations of re-
traceptives was significantly greater for
practicing contraception at that time. Those
sponses is not currently understood, and
the women reporting scale values of five
who were in a more casual relationship
future qualitative research might prof-
or more than for those reporting lower val-
with their first partner were significantly
itably explore the meanings women as-
ues (p≤.001 in a chi-square test). Lower
less likely than women who had been
cribe to the circumstances in which they
levels of contraceptive use among women
going steady with or had been engaged to
first experience intercourse.
with lower wantedness scores were evi-
their partner to have used contraceptives.
The women’s responses demonstrate the
dent for most age-groups, including the
The age difference between the young
complexity of characterizing sexual expe-
two youngest. Among those aged 15 and
woman and her partner also had a signifi-
riences. We found that nonvoluntary ex-
17 at first intercourse, however, the dif-
cant effect. Women who had voluntary first
periences were rated lowest on the scale of
ferences were negligible or nonexistent.
intercourse with a man seven or more years
wantedness, while voluntary experiences
The pattern of lower contraceptive use
older were less than one-half as likely as
were assigned a range of wantedness val-
corresponding to lower scale values was
those whose first partner was the same age
ues. A dichotomy of voluntary versus not
particularly pronounced for young white
or younger to have practiced contraception.
voluntary is insufficient to capture the
women. For young Hispanic women,
The association between wantedness of
meaning of the first intercourse experience.
however, the pattern was reversed: Sixty-
voluntary first intercourse and contracep-
Our analysis provides more evidence that
one percent of those who chose a low
tive use at first intercourse was very small
research on first intercourse and contra-
wantedness rating had practiced contra-
and fell short of statistical significance when
ception would benefit from being reframed
ception at first intercourse, compared with
other factors were controlled.† Thus, while
to allow for young women’s ambivalence
47% of those who gave first intercourse a
the partner’s relative age and the young
and different degrees of decision-making,
high wantedness rating.
women’s evaluation of the wantedness of
rather than universally assuming clear-cut
Mother’s education was associated
first intercourse appear to be related, only
situations and calculated responses. Future
with an increased likelihood that the re-
the partner’s relative age had an indepen-
research should attempt to elucidate the
spondent practiced contraception at first
dent effect on the odds of contraceptive use.
mechanisms that lead to nonvoluntary, un-
intercourse, but only when the wanted-
wanted and early first intercourse, as well
ness of first intercourse was rated as five
Discussion
as the effects of these experiences on con-
or higher. Among the group who rated the
Our findings provide new information on
traceptive use, pregnancy and later ado-
wantedness of first intercourse as four or
critical issues surrounding young
lescent development.
less, there was no relationship between
women’s first intercourse. While most of
It is generally assumed that a young
mother’s education and contraceptive use
the women reported that their first pre-
at first intercourse.
marital intercourse had been voluntary, a
Table 5. Among women aged 15–24 who had
small but important proportion of U.S.
voluntary, premarital first intercourse, per-
Logistic Regression Analysis
women aged 15–24 (9%) said that their
centage who practiced contraception at first
We performed logistic regression analy-
first such experience had been nonvol-
intercourse, by wantedness of first intercourse,
according to selected characteristics (N=1,859)
sis to determine whether age, race and eth-
untary. Consistent with the findings of
nicity, mother’s education, relationship
previous studies, we found this to be the
Characteristic
Wantedness
with partner, age discrepancy with part-
case more frequently among women who
1–4
5–10
ner and wantedness score affect contra-
were younger at first intercourse.
Total
66.6
75.3
ceptive use at first voluntary intercourse
We also found that women who as chil-
(Table 6). Women aged 13 or younger at
dren had lived with a single parent from
Age at first intercourse
≤
first intercourse were less likely than
birth or had left home at ages younger than
13
48.8
56.2
14
61.3
71.5
women who were 20 or older at first in-
16 were more likely than others to have ex-
15
75.2
75.2
tercourse to have practiced contraception
perienced nonvoluntary first intercourse.
16
63.8
75.6
17
81.5
83.6
at that time. Black and Hispanic women
These results parallel those of other stud-
18–24
74.9
77.6
had lower odds of having used contra-
ies that have shown the family of origin
Race/ethnicity
and parental presence to be an important
White non-Hispanic
69.2
82.7
*Since the consideration of demographic factors added a
influence on the odds of a young woman’s
Black non-Hispanic
60.8
64.7
level of complexity to the analysis, we returned to our pre-
experiencing nonvoluntary intercourse.25
Hispanic
61.1
47.0
vious classification of scale values as “low” (1–4) and “high”
(5–10) to facilitate interpretation of the relationships.
Further research is needed to clarify the dy-
Mother’s education
namics by which these associations with
<high school graduate
60.0
56.0
†We tried alternate models to test different ways of ana-
High school graduate
70.5
76.7
lyzing the scale of wantedness of first sex. We tested mod-
the family of origin arise. Where possible,
Some college
67.8
83.8
els that contained a continuous form of the scale, as well
future investigations should focus on
≥4 years of college
62.2
85.5
as a dichotomous version of the scale with values 1–2 form-
whether family or residential changes pre-
ing one category and values 3–10 forming another. The
Note: See note to Table 1.
findings were very similar across the alternative models.
ceded nonvoluntary intercourse.
16
Family Planning Perspectives

---

woman’s lack of control over a sexual ex-
larger age discrepancies.
Table 6. Coefficients (and standard errors) and odds ratios from
perience reduces the likelihood that con-
Because our analysis
logistic regression showing effects of selected characteristics on
traceptives will be used during that ex-
excluded women who
contraceptive use at first intercourse (N=1,766)
perience, particularly coitus-specific
reported first intercourse
methods. The spontaneous nature of some
as nonvoluntary, our
Characteristic
Coefficient
Odds ratio
nonvoluntary intercourse may reduce
findings on the relation-
Age at first intercourse
young women’s opportunities for fore-
ship between wanted-
<13
–1.08 (0.33)
0.34**
14–15
–0.35 (0.31)
0.70
thought or planning, and so lessen even
ness of first intercourse
16–17
–0.27 (0.31)
0.76
further the likelihood that contraceptives
and partner age and con-
18–19
–0.20 (0.32)
0.82
≥20
ref
1.00
will be used under such circumstances.
traceptive use are con-
Thus, the relatively high proportions of
servative. We anticipate
Race/ethnicity
sexual experiences that are nonvoluntary
that these patterns would
White non-Hispanic
ref
1.00
Black non-Hispanic
–0.69 (0.14)
0.50***
among young adolescent women com-
be amplified in the pop-
Hispanic
–1.19 (0.18)
0.30***
pared with adult women may explain in
ulation of young women
Mother’s education
part why teenagers are especially unlike-
that includes those
<high school
–0.73 (0.22)
0.48**
ly to practice contraception26 and why
whose first intercourse
High school graduate
–0.31 (0.20)
0.73
women who have first intercourse before
was nonvoluntary.
Some college
0.09 (0.27)
1.09
≥4 years of college
ref
1.00
age 19 are less likely to have used a
We found that want-
method at first intercourse than are those
edness of first inter-
Relationship with partner
Just met/just friends/
who were older at first intercourse.27
course, after controlling
went out once in a while
–0.31 (0.14)
0.73***
Since data on contraceptive use during
for other relevant situa-
Going steady/engaged
ref
1.00
nonvoluntary first intercourse were un-
tional and background
Partner’s age
available, we examined associations be-
factors, means no
Same age or younger
ref
1.00
tween contraceptive use and varying de-
greater nor lesser chance
1–2 years older
–0.16 (0.16)
0.85
grees of wantedness of voluntary first
of having that first in-
3–4 years older
–0.09 (0.19)
0.91
5–6 years older
–0.28 (0.28)
0.76
intercourse experiences. While maternal
tercourse protected.
≥7 years older
–0.82 (0.25)
0.44**
education beyond the high school level
Rather, the critical pre-
Wantedness of first intercourse
was positively associated in bivariate
dictors were those cor-
1–4
–0.20 (0.14)
0.82
analyses with contraceptive use at first in-
related with wanted-
5–10
ref
1.00
tercourse when wantedness was high, this
ness—an older partner,
Constant
2.30 (0.37)
10.02***
background characteristic did not affect
a casual relationship and
contraceptive use when first intercourse
young age at first sex. To
*Difference is significant at p<.05. **Difference is significant at p<.01. ***Difference is signifi-
cant at p<.001. Notes: See note to Table 1. ref=reference category.
was classified as relatively unwanted. This
the extent that unwant-
finding is consistent with research on in-
ed intercourse is spon-
terpersonal power and the indicators of
taneous and wanted intercourse is more
partner decrease the likelihood of contra-
pregnancy risk that found female power
likely to be planned, wantedness may have
ceptive use at first intercourse.
in a relationship to be positively correlat-
stronger implications for those methods
Some questions remain about the mean-
ed with contraceptive use.28 This pattern
that require premeditated effort, such as the
ing of the scale values. It is unrealistic to ex-
suggests that factors known to be related
pill, than for coitus-specific methods such
pect that all young women would interpret
to contraceptive use at first sex may only
as the condom. The multivatiate analysis
the scale the same way and evaluate their
operate if the young woman consciously
considered all methods together, but fur-
past experiences by a common standard.
chooses and actively wants the experience.
ther insight may be gained from analyzing
Undoubtedly, there are differences in in-
In multivariate regression analysis, the
methods separately in the future.
terpretation and evaluation, possibly dis-
effects of young women’s race and ethnic-
Despite the lack of an independent re-
tinctive by subculture. Such lack of unifor-
ity, age at first intercourse and mother’s ed-
lationship between wantedness of inter-
mity could be a factor in the interpretation
ucation were generally consistent with pre-
course and contraceptive use at voluntary
of the scale, in the process of remembering
vious studies, in which these factors have
first intercourse, our findings on wanted-
first intercourse, in defining life experiences
all been shown to be related to contracep-
ness of first intercourse are nonetheless
with regard to control and in having expe-
tive use at first intercourse.29 The finding
important. Scale values were correlated
riences subsequent to the one in question
that women who had had first intercourse
with other significant circumstantial fac-
that influence how it is evaluated.
with a man seven years or older were less
tors, and could serve as an indicator of
The purpose of this article was to pro-
than half as likely to use contraceptives as
how often and under what circumstances
vide descriptive information on nonvol-
women whose first partner had been the
young women go through a critical expe-
untary first intercourse and its correlates,
same age or younger suggests an exception
rience they would rather have avoided.
using measures that have not been used
to the usual pattern of higher levels of con-
The fact that substantial numbers of
before in a national survey. However, the
traceptive use with increased age: In rela-
young women voluntarily participated in
NFSG focuses on the onset of sexual ac-
tionships with large age discrepancies, the
a first sexual experience about which they
tivity as an indicator of the beginning of
partner’s advanced age in relation to the
felt ambivalent or negative deserves at-
women’s exposure to the risk of pregnan-
young woman works against contraceptive
tention. It is especially salient for research
cy and sexually transmitted diseases. The
use. Partners who are closer in age, even
and service programs, given that these
primacy of this mission precluded special
when both are young, are more likely to
“voluntary but unwanted” experiences
focus on nonvoluntary intercourse. In ad-
take precautions and actively prevent ad-
are most prevalent among young women
dition, since the NFSG collected informa-
verse consequences than are couples with
whose youth or age difference with their
tion only on vaginal heterosexual inter-
Volume 30, Number 1, January/February 1998
17

---

Young Women’s Control over First Intercourse
course, our research did not consider other
ily, 1993, 55(4):941–952.
lems in adolescence: exploring the effects of moderating
types of sexual activity. Thus, many issues
variables, Journal of Marriage and the Family, 1997,
7. Laumann EO, 1996, op. cit. (see reference 2).
remain for future surveys and develop-
59(1):131–142; and Putnam F, Urinary catecholamine ex-
8. Moore KA, Nord CW and Peterson JL, 1989, op. cit.
cretion in sexually abused girls, Journal of the American
mental work to address.
(see reference 2).
Academy of Child and Adolescent Psychiatry, 1994,
Given the newness of these measure and
33(30):329–327.
9. Ibid.
the relative scarcity of national data with
21. Browning CR and Laumann EO, Sexual contact be-
10. Small SA and Kerns D, 1993, op. cit. (see reference 6).
which to compare results, our analyses
tween children and adults: a life course perspective,
should be regarded as exploratory and the
11. Moore KA, Nord CW and Peterson JL, 1989, op. cit.
American Sociological Review, 1997, 62(4):540–560.
(see reference 2).
findings as suggestive. Nevertheless, these
22. Donaldson PE, Whalen MH and Anastas JW, Teen
measures fill a gap in evidence about first
12. Ibid.
pregnancy and sexual abuse: exploring the connection,
intercourse, and they augment existing ev-
Smith College Studies in Social Work, 1989, 59(3):289–300.
13. Mynatt CR and Allgeier ER, Risk factors, self-attri-
idence about first intercourse among young
butions, and adjustment problems among victims of sex-
23. Musick JS, Young, Poor and Pregnant, New Haven, CT,
ual coercion, Journal of Applied Social Psychology, 1990,
women in the United States. They also re-
and London: Yale University Press, 1993.
20(2):130–153.
inforce the idea that research on the initia-
24. Browning CR and Laumann EO, 1997, op. cit. (see
tion of sexual activity would be enhanced
14. Ibid.; and Small SA and Kerns D, 1993, op. cit. (see
reference 21).
reference 6).
by considering possible nonconsent or am-
25. Moore KA, Nord CW and Peterson JL, 1989, op. cit.
bivalence surrounding first intercourse.
15. Koyle P et al., Comparison of sexual behaviors among
(see reference 2).
adolescents having an early, middle, and late first inter-
course experience, Youth and Society, 1989, 20(4):461–475;
26. Brown SS and Eisenberg L, The Best Intentions: Un-
References
and Seidman SN, Mosher WD and Aral SD, Predictors
intended Pregnancy and the Well-Being of Children and Fam-
1. Abma JC et al., Fertility, family planning, and women’s
of high-risk behavior in unmarried American women:
ilies, Washington, DC: National Academy Press, 1995.
health: new data from the 1995 National Survey of Family
adolescent environment as a risk factor, Journal of Ado-
Growth, Vital and Health Statistics, 1997, Series 23, No. 19.
27. Peterson LS, Contraceptive use in the United States:
lescent Health, 1994, 15(2):126–132.
1982–1990, Advance Data from Vital and Health Statistics,
2. Laumann EO, Early sexual experiences: how volun-
16. Small SA and Kerns D, 1993, op. cit. (see reference 6).
1995, No. 260; and Mosher WD and McNally JW, Con-
tary? How violent? in Smith MD et al., eds., Sexuality and
traceptive use at first premarital intercourse: United
American Social Policy, Menlo Park, CA: Henry J. Kaiser
17. Roosa MW et al., The relationship of childhood sex-
States, 1965–1988, Family Planning Perspectives, 1991,
Family Foundation, 1996; and Moore KA, Nord CW and
ual abuse to teenage pregnancy, Journal of Marriage and
23(3):108–116.
Peterson JL, Nonvoluntary sexual activity among ado-
the Family, 1997, 59(1):119–130; Laumann EO, 1996, op.
lescents, Family Planning Perspectives, 1989, 21(3):110–114.
cit. (see reference 2); and Boyer D and Fine D, Sexual abuse
28. Jorgensen SR, King SL and Torrey BA, Dyadic and
as a factor in adolescent pregnancy and child maltreat-
social network influences on adolescent exposure to preg-
3. Moore KA, Nord CW and Peterson JL, 1989, op. cit.
ment, Family Planning Perspectives, 1992, 24(1):4–11 & 19.
(see reference 2).
nancy risk, Journal of Marriage and Family, 1980,
42(1):141–155.
18. Stock JL et al., Adolescent pregnancy and sexual risk-
4. Erickson PI and Rapkin AJ, Unwanted sexual expe-
taking among sexually abused girls, Family Planning Per-
riences among middle and high school youth, Journal of
29. Mauldon J and Luker K, The effects of contraceptive
spectives, 1997, 29(5):200–203 & 227.
Adolescent Health, 1991, 12(4):319–325.
education on method use at first intercourse, Family Plan-
ning Perspectives, 1996, 28(1):19–24; Peterson LS, 1995, op.
19. Miller BC, Monson BH and Norton MC, The effects
5. Laumann EO, 1996, op. cit. (see reference 2).
cit. (see reference 27); Mosher WD and McNally JW, 1991,
of forced sexual intercourse on white female adolescents,
6. Small SA and Kerns D, Unwanted sexual activity
op. cit. (see reference 27); and Forrest JD and Singh S, The
Child Abuse and Neglect, 1995, 19(10):1289–1301.
among peers during early and middle adolescence:
sexual and reproductive behavior of American women,
incidence and risk factors, Journal of Marriage and the Fam-
20. Luster T and Small S, Sexual abuse history and prob-
1982–1988, Family Planning Perspectives, 1990, 22(5):206–214.
18
Family Planning Perspectives

---