Attitudes, Experience, and Anticipation of Sex Among

Matern Child Health J (2011) 15:S54S64
DOI 10.1007/s10995-011-0879-5
Attitudes, Experience, and Anticipation of Sex Among

5th Graders in an Urban Setting: Does Gender Matter?
Karen M. Anderson Helen P. Koo
Renee R. Jenkins Leslie R. Walker
Maurice Davis Qing Yao M. Nabil El-Khorazaty
Published online: 14 September 2011

Springer Science+Business Media, LLC 2011
Abstract To identify gender differences in correlates of

anticipation and initiation of sexual activity in the baseline
survey of 562 African-American 5th grade students prior to
initiation of a school-based pregnancy prevention intervention curriculum. Students from 16 elementary schools
were administered the baseline questionnaire during classroom periods. Using these data, binary and ordered logistic
regression models were used to analyze the factors affecting
virginity and anticipation of sexual activity separately by
gender, and tests of interaction between each factor and
gender were conducted on the combined sample. More boys
than girls had already had sex (18% vs. 5%) and anticipated
having sexual intercourse in the next 12 months (56% vs.
22%). Boys and girls also differed in the factors that affected
these outcomes. The perception that their neighborhood was
safe reduced the odds that boys anticipated sexual activity
but was not associated with this outcome among girls.
Pubertal knowledge increased the odds of anticipation, but
only among boys. Attitudes favoring abstinence decreased
K. M. Anderson R. R. Jenkins (&)
DC-Baltimore Research Center on Child Health Disparities,
Howard University, Howard University Research Building 1,
Room 214, 1840 7th Street NW, Washington, DC 20001, USA
e-mail: rjenkins@howard.edu
H. P. Koo Q. Yao
RTI International, Research Triangle Park, NC, USA
L. R. Walker
Seattle Childrens Hospital, Seattle, WA, USA
M. Davis
Eunice Kennedy Shriver National Institute for Child Health
and Human Development, Bethesda, MD, USA
M. Nabil El-Khorazaty
RTI International, Rockville, MD, USA
123
anticipation of sex among both genders, but slightly more

among girls than boys. Having more frequent parentchild
communication about sex was associated with increased
anticipation among girls but decreased anticipation among
boys. Curriculum based approaches to adolescent pregnancy
prevention are appropriate for 5th grade elementary students
who may already be anticipating sexual activity in communities with disproportionate rates of teen pregnancy. The
design of the interventions should consider the differences
in motivating factors by gender.
Keywords Pre-adolescents Sexuality Gender
differences Sexual activity Virginity Risk behaviors
Adolescent pregnancy prevention
Introduction
Adolescents in the United States initiate sexual intercourse
at alarmingly young ages. Youth Risk Surveillance data
indicate that 5.9% of high school students initiated sex
before age 13. For Black male students the rate reaches
24.9% [1]. Yet, relatively few comprehensive sex education
programs are offered in elementary school settings. Only
27.5% of all schools sampled in the United States have a
policy that teaches about pregnancy prevention in elementary schools and only 49% require human sexuality education. Of those that teach pregnancy prevention or sexuality,
1.3 classroom hours are required and 10% provide any
classroom activity opportunities for the students to practice
skills [2]. In contrast, Mueller et al. [3] showed evidence that
school sex education can reduce adolescent sexual risk
behaviors when provided before sexual initiation, particularly in youth at high risk for early sexual debut. The literature suggests that having sex education for elementary
school aged children is very limited but may help delay early
sexual debut, especially in high risk populations.
While information on what is taught in schools can be
elicited from national surveillance studies, there is less
information available about the association between preadolescent sexual attitudes, experiences and anticipation of
sexual behavior. The information could provide evidence
for increasing elementary school prevention programs
aimed at decreasing early sexual debut.
Research also shows that gender plays a role in attitudes
toward sexual intercourse in young adolescents. One study
of 4th and 5th graders found that twice as many males as
females anticipated having sex within the next year [4];
however, the study excluded already sexually active children. In addition, research has consistently shown a gender
disparity in age of reported sexual debut, particularly in
black youth [4, 5] and in attitudes toward sex [6, 7].
This paper describes the attitudes toward sexual activity,
anticipation and experience of sexual intercourse among a
sample of 5th graders. In particular, it examines differences
between boys and girls in these factors. Thus, it provides
relatively rare data for a young age group, and could be
used in the promotion and design of interventions to delay
sexual debut among elementary students taking gender
differences into account.
The data were obtained from 5th grade students enrolled
in a pregnancy prevention school-based program, Building
Futures for Youth (BFY), prior to the beginning of the
intervention. The program was delivered over a 3-year
period and involved the parents of the students. In an earlier
study by several of the present authors only 44% of 7th grade
male students in the same school district reported being
virgins and consequently limited the programs ability to
intervene prior to sex debut [8]. Consequently the new BFY
program targeted 5th graders and their parents with the
intent of intervening prior to initiating sex. The significant
gender differences in the proportion of boys versus girls
reporting sexual activity in the prior study also prompted the
investigators to consider tailoring some curriculum modules
to be more gender specific and hold same-sex sessions.
The content of the BFY intervention was based on a
conceptual framework for risk and resilient behavior. The
Social Cognitive Theory [9] informed the framework
because it takes into account individual, environmental and
peer factors in explaining human interpersonal behavior.
To guide the development of the intervention and its
evaluation, we developed a logic model considering these
factors based on the research literature on specific variables
influencing sexual attitudes and behavior of adolescents
and pre-adolescents. Figure 1 represents the variables
chosen for inclusion in the logic model. These are also the
variables included in the present article. A report of the
evaluation of the intervention has been published [10].
S55
Methods
Study Design
The BFY intervention to delay sexual initiation was
implemented and evaluated, starting with 5th graders in 16
Washington, DC, schools during the 20012002 school
year. Eight schools each were randomly assigned to the
intervention and control conditions. The students were
followed longitudinally into the 6th and 7th grades. This
paper uses baseline data from the children at all 16 schools
prior to intervention activities, to examine baseline associations among various sociodemographic factors and the
boys and girls psychosocial attributes, attitudinal and
knowledge characteristics related to sexuality, and their
sexual experience and anticipation of sexual activity.
Sample
The 5th grade students attending these 16 schools were
considered eligible for the study if they were able to
complete the survey in English. Students in special education classes were excluded because the survey was selfadministered.
Fig. 1 Factors associated with sexual experience and anticipated

sexual activity
123
S56
Parents of all eligible students signed consent forms for

their children to participate in the study. Of 793 eligible
students, 620 (89.5%) received parental consent. Of these,
562 (90.6%) completed the students baseline survey.
Analysis of the sociodemographic characteristics of these
students showed that they are representative of the students
at the study schools in being mostly African American
(99% vs. 97% in the 16 schools), more female (54% female
vs. 52% female in the 16 schools), and low income, as
represented by receipt of free or reduced priced school
lunch (75% vs. 76%).
Students were administered the baseline questionnaire
during classroom periods. Students marked a hard-copy
version of the 70-item questionnaire while listening to an
audiotape of the questions and answer choices using
audiocassette players with earphones. Pretesting found that
the use of audiocassette players decreased student interactions and distraction and provided more privacy while
completing the questionnaire.
Survey Instrument
The baseline questionnaire measured background and risk
factors related to early sexual debut and adolescent pregnancy, as well as outcomes targeted by the intervention
(Fig. 1). Intermediate outcomes included knowledge and
attitudinal factors that are relevant to delaying sexual initiation and reducing pregnancy. We developed two genderspecific questionnaires to tailor the wording to boys versus
girls. We included new questions and questions from previously validated survey instruments, including the
National Longitudinal Study of Adolescent Health [11], the
Teen Activities and Attitudes Study [12], and the questionnaires used in a predecessor study [8, 13]. The instrument was tested and revised twice using cognitive
interviews and pilot tests. The institutional review boards
of the contributing institutions and the National Institute of
Child Health and Human Development (NICHD) approved
the final questionnaires and all study procedures.
Measures
Dependent Variables
Experience of Sexual Intercourse This was a dichotomous variable (yes and no) based on responses to the
question, Have you ever had sex? (The questionnaire
stated that having sex means the males penis is inside the
females vagina).
Anticipated Level of Sexual Activity in the Next
12 Months This variable was based on three questions
asking whether the students thought they would, in the next
123
12 months, kiss or touch under the clothes if they were

going out with someone they like a lot, and have sex. We
classified students who did not expect to engage in any of
these behaviors as having low risk anticipation, those
who reported that they might kiss or touch under the
clothes (answered yes or dont know) as having
moderate risk anticipation, and those who responded
that they might have sexual intercourse (yes or dont
know) as having high risk anticipation.
Independent Variables
Background Characteristics We identified six background factors as independent variables: age; mother
works; free or reduced-price lunch (as a measure of poverty); frequency of student attendance at religious services;
student feels their neighborhood is safe; and status of the
childs pubertal development. The girls were asked: Have
you started growing breasts or body hair? and the boys
were asked: Have you started growing body hair or has
your voice started to change? Boys or girls answering
yes to their question were classified as having experienced pubertal signs. (Details on the puberty variables can
be found in another publication) [14].
School and Self Variables The self confidence variable
was based on responses to two statements: Other kids my
age like me a lot and My parents are very proud of me.
The two numbers were summed to give a self confidence
score. The liking school variable was similarly constructed, summing responses to three statements: I
get along with most of the students at my school; I like
most of the teachers at my school; and I like going to my
school. The extracurricular activities variable represents the number of activities that students indicated they
had participated in. We constructed three categories of
educational aspiration from responses to two questions
about plans to finish high school and seek education
beyond high school.
Attitudinal and Knowledge Variables Three variables,
attitude toward sexual abstinence, attitude toward resisting
pressure to have sex, and parentchild communications
about sexual topics, were scales created through factor
analysis of groups of questions intended to measure these
constructs. Higher scores on these scales represented
greater favorability toward abstinence and refusal of sex,
and more frequent parentchild communication about
sexual topics. See Rose et al. [15] for details regarding the
development and psychometric properties of these scales.
We based the variable, whether parent would be upset if
the student were having sex and the parent found out, on a
direct question. The knowledge of puberty variable was
based on the number of pubertal signs that students

answered in two questions about the changes occurring
during puberty in boys and to girls.
Statistical Analyses
Because the level of sexual activity experienced and
anticipated, sexual attitudes, and relationships among some
factors were expected to differ between boys and girls, we
conducted analyses separately by gender. We first examined for each gender the bivariate relationships of background, school/self, and attitudinal characteristics of
students with the two dependent variables. For categorical
independent variables, we used cross-tabulations and chisquare tests of significance. For cross-tabulations containing small expected numbers of cases, we performed the
Fishers exact test. For continuous independent variables
(various scales), we computed their means for each category of the dependent variables and tested the significance
of differences between means. We then included any
independent variable that was significantly related (at
P \ 0.10) to either dependent variable, for either gender, in
the subsequent multivariable analysis, so that the set of
independent variables would be comparable across the set
of multivariable models. (We included age in the multivariable models, even though it was not significantly
related to the dependent variables in the bivariate analysis,
to distinguish the effects of age from puberty.) Out of the
562 students who completed the survey, 63 were dropped
from the analyses because they did not respond to all
questions that were used in multivariable analyses.
In the multivariable analysis of experience of sexual
intercourse, we used binary logistic regression to model the
likelihood of having had sex (vs. not having had sex). For
the three-level variable, anticipated sexual activity, we
estimated ordered logistic regression models. We modeled
the likelihood of being in the higher risk groups over being
in the lower risk groups (highest vs. moderate and low; and
highest/moderate vs. low). For each dependent variable, we
used all three sets of independent variables to assess which
background, school/self, and attitudinal factors had independent associations with each dependent variable when all
three sets were taken into account.
These regression models were estimated separately by
gender. To test whether results differed by gender, we
estimated the models for the two genders combined, and
tested, one at a time, the interaction between gender and
each independent variable. A significant interaction (at
P \ 0.10) indicated that the different results observed in
the separate regressions for the two genders were statistically significant.1
In the multivariable models, we re-scaled the three
attitudinal variables from the actual scores to standard
S57
deviation units by dividing each individuals score by the

standard deviation of the variable for that gender. The
effect of a change of one point in the score could not be
expected to be sizeable, given that the scales ranged
approximately from 0 to 100. In contrast, a change of one
standard deviation can be expected to have a meaningful
effect in the outcome.
Results
Sample Characteristics by Gender
Boys were significantly older than girls; 42% of the boys
were aged 11 or older, compared with 29% of the girls
(Table 1). (Ninety percent of students were either age 10 or
11.) Approximately three quarters of the mothers worked,
three quarters of the students received free or reduced-price
lunch, over half attended religious services sometimes,
and nearly 70% felt their neighborhoods were safe. Significantly more girls (79%) than boys (57%) reported
having experienced signs of puberty.
Girls had significantly higher self-confidence scores,
more extracurricular activities, and higher educational
aspirations than boys. The majority of both boys (68%) and
girls (78%), thought they would finish high school and
continue to some form of higher education. Boys and girls
liked their schools equally.
Boys and girls differed in all five attitudinal/knowledge
variables. Girls had more favorable attitudes toward
abstinence and toward resisting having sex, as well as
reporting more frequent communication with a parent
about sexual topics. Significantly more girls than boys
reported that their parents would be upset if they found out
that the youths were having sex (83% vs. 66%). Girls had
better knowledge about pubertal changes. Finally, more
boys (18%) than girls (5%) reported having had sex, and
more boys (56%) than girls (22%) anticipated having sex in
the next 12 months.
Bivariate Relationships
Among boys, pubertal development was significantly
related to having had sex (P = 0.011) and with anticipating
sexual activity (P = 0.051) (Table 2). Perceiving ones
neighborhood as unsafe was associated with being at higher
risk of anticipating sexual activity (P = 0.037). Liking
school less was associated with having sex (P = 0.007)
1
We use the significance level of P \ 0.10 throughout the paper, to

provide information on results that nearly reach the more conventional level of P \ 0.05. By noting the variables with P values
between 0.05 and 0.10, we succinctly point out results that may be
fruitfully considered for further exploration in other studies.
123
S58
Table 1 Student characteristics, by gender

Variable
Males N (%)
Females N (%)
P-value
97 (41.8%)
78 (29.2%)
0.003
135 (58.2%)
189 (70.8%)
Total N (%)
Background characteristics
Age**
11 years old or older
10 years old or younger
175 (35.1%)
324 (64.9%)
Mother works
No
54 (23.6%)
63 (24.0%)
Yes
175 (76.4%)
200 (76.0%)
No
57 (24.6%)
64 (24.0%)
Yes
175 (75.4%)
203 (76.0%)
0.923
117 (23.8%)
375 (76.2%)
Free/reduced-price lunch
0.876
121 (24.2%)
378 (75.8%)
Frequency of religious attendance

Never
52 (22.8%)
42 (16.0%)
Sometimes, but not every week
117 (51.3%)
153 (58.2%)
270 (55.0%)
59 (25.9%)
68 (25.8%)
127 (25.9%)
Once a week or more
0.134
94 (19.1%)
Safe neighborhood
No
71 (30.6%)
87 (32.6%)
Yes
161 (69.4%)
180 (67.4%)
0.635
158 (31.7%)
99 (42.7%)
56 (21.0%)
133 (57.3%)
211 (79.0%)
Self-confidence (mean, SD)*
1.3 (1.0)
1.5 (0.8)
0.042
Liking school (mean, SD)
1.7 (1.7)
1.9 (1.5)
0.272
1.8 (1.6)
Extracurricular activities (mean, SD)***
2.5 (1.5)
3.2 (2.0)
\0.001
2.9 (1.8)
341 (68.3%)
Pubertal development***
No signs
Breasts/body hair or voice change/body hair
0\.001
155 (31.0%)
344 (68.9%)
School and self variables

1.4 (0.9)
Educational aspirations**
Not finish high school
Finish high school
Beyond high school
33 (14.2%)
31 (11.6%)
41 (17.7%)
158 (68.1%)
25 (9.4%)
211 (79.0%)
68.1 (25.1)
83.7 (16.8)
0.010
64 (12.8%)
66 (13.2%)
369 (74.0%)
Attitudinal and knowledge variables

Attitudes toward abstinencemean (SD)***
\0.001
76.5 (22.4)
Attitude toward refusing sexmean (SD)***
66.4 (27.4)
86.8 (18.3)
\0.001
77.3 (25.1)
Parentchild communicationmean (SD)***
45.3 (29.8)
60.5 (31.1)
\0.001
53.4 (31.4)
Yes
154 (66.4%)
221 (82.8%)
\0.001
375 (75.2%)
No
23 (9.9%)
12 (4.5%)
Parents upset if found child had sex?***

35 (7.0%)
Not sure
55 (23.7%)
34 (12.7%)
Knowledge of both genders puberty mean (SD)***
7.2 (5.1)
8.9 (5.2)
89 (17.8%)
Yes
40 (17.5%)
12 (4.5%)
52 (10.6%)
No
188 (82.5%)
253 (95.5%)
441 (89.4%)
23 (10.0%)
90 (33.7%)
79 (34.2%)
118 (44.2%)
197 (39.6%)
129 (55.8%)
59 (22.1%)
188 (37.8%)
\0.001
8.1 (5.2)
Dependent variables
\0.001
Ever had sex***
Anticipated sexual activity***

Low risk
Medium risk
High risk
* P \ 0.05, ** P \ 0.01, *** P \ 0.001
123
\0.001
113 (22.7%)
and anticipating sexual activity (P = 0.030). Having lower

educational aspirations was related to higher levels of
anticipated sexual activity (P = 0.048). Having less
favorable attitudes toward abstinence and refusing sexual
activity and not thinking that parents would be upset about
their having sex were associated with higher probabilities
of having had sex and anticipating sexual activity
(P \ 0.01). Having fewer communications about sexuallyrelated topics also was associated with anticipating sex
(P = 0.006). Having greater knowledge about puberty was
also significantly related to having had sex (P = 0.012)
and anticipating sexual activity (P = 0.089)
Girls who received a free or reduced-price school lunch
and those who had experienced pubertal changes were
more likely to anticipate having sex (P \ 0.05, Table 3).
Liking school less was related to sexual experience
(P = 0.068) and to anticipating sexual activity
(P = 0.018). Girls who did not expect to finish high school
were more likely to have had sex (P = 0.011). Girls with
more favorable attitudes toward abstinence and refusing
sex were less likely to have had sex (P \ 0.05) and less
likely to anticipate sexual activity (P \ 0.01). Girls with
more knowledge of puberty were more likely to anticipate
sexual activity (P = 0.042).
Multivariable Models
In the multivariable models (Table 4), boys with more
favorable attitudes toward abstinence had lower odds of
having had sex (OR = 0.38 for a standard deviation
change in the score), and those who thought their parents
would be upset about their having sex were less likely to
have had sex than boys who thought otherwise (OR =
0.38). The factors that were significantly related to virginity differed for girls. Girls who were more favorable
toward refusing sex had reduced odds of having had sex
(OR = .62 for one standard deviation change in scores).
Tests of interactions of the various factors with gender in
the combined sample did not show any of the interactions
to be significant, even when the main effects of gender and
each variable, and the interaction term were included in the
regression. There were only 12 girls who have had sex. It is
likely that this is too small a number to support the
detection of significant interaction effects in the models.
Boys who felt their neighborhood was safe were less
likely to anticipate sexual activity (OR = 0.46) (Table 4).
Boys who had experienced signs of puberty had 2.7 times
the odds of anticipating sexual activity than boys without
pubertal signs. Boys favorable toward abstinence were less
likely to anticipate sexual activity (OR = 0.38 for one
standard deviation change in score), and those with more
communication with their parents about sexual topics were
also less likely to anticipate sexual activity (OR = 0.70 for
S59
one standard deviation change in score). Boys who thought

their parents would be upset about their having sex also had
reduced odds of anticipating sexual activity (OR = 0.55).
The factors significantly related to sexual anticipation
differed in some respects for girls (Table 4). The more girls
liked their school, the less likely they were to anticipate
sexual activity (OR = 0.85). Girls who were more favorable toward abstinence were also less likely to anticipate
sexual activity (OR = 0.37 for one standard deviation
change in score). Girls with more frequent communication
with their parents about sexual subjects were more likely to
anticipate sexual activity than those with less communication (OR = 1.39 for one standard deviation change in
score). Girls with better knowledge about puberty had
higher odds of anticipating sexual activity than those with
less knowledge (OR = 1.05).
Some of the observed differences for anticipating sexual
activity between boys and girls were found not to be significant in the tests of interactions of the various variables
with gender in the combined sample (Table 4). Four factors
differed significantly by tests of interaction in their relationships to anticipation of sexual activity between boys
and girls. For boys safe neighborhoods related to less
anticipation of sexual activity, and the interaction test
showed this to be significant (P = 0.028). Attitude toward
abstinence was significantly related to sexual anticipation
for both boys and girls. The effect for girls was larger than
for boys (OR of 0.37 for girls, 0.38 for boys). The interaction test showed this difference was significant (P =
0.075). For girls, more frequent parentchild communication about sexual topics was positively associated with
anticipation of sexual activity, whereas for boys, the
association was negative. The interaction of gender with
communication was significant (P = 0.0006). Knowledge
about puberty was positively related to anticipation of
sexual activity among girls, but was not significantly
associated for boys. This difference was significant in the
interaction test (P = 0.044).
Discussion
This study demonstrates that some preadolescents in upper
elementary school have had sexual intercourse and anticipate having sexual intercourse in the near future. The levels
of sexual activity are consistent with those found by the DC
Middle School Youth Risk Behavior Survey, in which
17.8% of males reported experiencing intercourse by age
11 or younger [16].
The factors associated with sexual experience and
anticipation of sexual activity were markedly different in
boys and girls in the multivariable analysis. For boys, the
only variables significantly related to virginity were a more
123
S60
Table 2 Bivariate relationships of background, school, and attitudinal/knowledge variables to experience of sexual intercourse and anticipated
sexual activity among male students
Variable
Ever had sex

No
Anticipated sexual activity

Yes
P-value
Low risk
Medium risk
High risk
P-value
Age
11 years old and older
10 years old and younger
77 (80.2%)
111 (84.1%)
19 (19.8%)
21 (15.9%)
0.447
0.736
11 (11.5%)
12 (8.9%)
27 (28.1%)
52 (38.5%)
58 (60.4%)
71 (52.6%)
0.252
0.891
Mother works
No
43 (81.1%)
10 (18.9%)
Yes
143 (83.1%)
29 (16.9%)
5 (9.4%)
17 (32.1%)
31 (58.5%)
17 (9.7%)
62 (35.4%)
96 (54.9%)
5 (8.9%)
15 (26.8%)
36 (64.3%)
18 (10.3%)
64 (36.6%)
93 (53.1%)
No
42 (76.4%)
13 (23.6%)
Yes
146 (84.4%)
27 (15.6%)
0.173
0.331

Never
42 (80.8%)
10 (19.2%)
19 (36.5%)
25 (48.1%)
97 (85.1%)
17 (14.9%)
0.458
11 (9.5%)
8 (15.4%)
38 (32.8%)
67 (57.8%)
0.407
Once a week or more
45 (77.6%)
13 (22.4%)
3 (5.1%)
22 (37.3%)
34 (57.6%)
No
55 (78.6%)
15 (21.4%)
5 (7.1%)
17 (24.3%)
48 (68.6%)
Yes
133 (84.2%)
25 (15.8%)
18 (11.2%)
62 (38.5%)
81 (50.3%)
88 (89.8%)
10 (10.2%)
15 (15.3%)
34 (34.7%)
49 (50.0%)
100 (76.9%)
30 (23.1%)
8 (6.0%)
45 (33.8%)
80 (60.2%)
Self confidencemean (SD)
1.3 (1.0)
1.4 (0.9)
0.516
1.2 (1.0)
1.5 (0.9)
1.2 (1.0)
0.208
Liking schoolmean (SD)
1.8 (1.6)
1.1 (1.9)
0.007**
2.2 (1.2)
2.0 (1.5)
1.5 (1.8)
0.030*
Extracurricular activitiesmean (SD)
2.4 (1.5)
2.8 (1.3)
0.116
2.0 (0.8)
2.7 (1.6)
2.4 (1.4)
0.138
7 (21.9%)
22 (68.8%)
0.048*a
8 (19.5%)
28 (68.3%)
64 (40.5%)
79 (50.0%)
Safe neighborhood
Pubertal development
No changes
Breasts/body hair or voice change/
body hair
0.305
0.011*
0.037*
0.051#
Educational aspirations
Finish high school
Beyond high school
Attitude toward abstinencemean (SD)
25 (75.8%)
8 (24.2%)
33 (80.5%)
8 (19.5%)
130 (84.4%)
24 (15.6%)
0.463
3 (9.4%)
5 (12.2%)
15 (9.5%)
73.0 (22.9)
46.7 (23.7)
0.000***
88.9 (12.6)
76.4 (20.5)
59.3 (25.6)
0.000***
Attitude toward refusing sexmean

(SD)
69.7 (26.4)
49.4 (26.5)
0.000***
75.7 (24.2)
75.1 (24.8)
59.5 (27.6)
0.000***
Parentchild communicationmean
(SD)
45.5 (30.4)
45.2 (27.1)
0.960
58.9 (30.8)
49.2 (28.8)
40.1 (29.2)
0.006**
Parents upset if found child had sex?

No
54 (71.1%)
22 (28.9%)
Yes
134 (88.2%)
18 (11.8%)
6.8 (5.1)
9.0 (5.0)
Knowledge of both genders puberty

mean (SD)
0.001**
0.012*
5 (6.5%)
17 (22.1%)
55 (71.4%)
18 (11.7%)
62 (40.3%)
74 (48.1%)
5.1 (4.7)
7.8 (5.3)
7.2 (5.1)
Due to small expected cell counts, results from Fishers exact test are shown instead of chi-square test
P \ 0.10, * P \ 0.05, ** P \ 0.01, *** P \ 0.001
favorable attitude toward abstinence and thinking their

parents would be upset if they found their child was having
sex. For girls, the only (marginally) significant association
with virginity was a favorable attitude toward refusing sex.
123
0.003**
0.089#
The genders also differed in the ways factors were

related to anticipation of sexual activity. Among both boys
and girls, a more favorable attitude toward abstinence was
associated with less anticipation of sexual activity, but the
S61
Table 3 Bivariate relationships of background, school, and attitudinal/knowledge variables to experience of sexual intercourse and anticipated
sexual activity among female students
Variable
Ever had sex

No

Yes
P-value
Low risk
Medium risk
High risk
P-value
Age
76 (97.4%)
177 (94.7%)
2 (2.6%)
10 (5.3%)
0.518a
28 (35.9%)
62 (32.8%)
37 (47.4%)
81 (42.9%)
13 (16.7%)
46 (24.3%)
0.389
No
58 (92.1%)
5 (7.9%)
0.169a
26 (41.3%)
27 (42.9%)
10 (15.9%)
0.209
Yes
191 (96.5%)
7 (3.5%)
62 (31.0%)
89 (44.5%)
49 (24.5%)
No
62 (98.4%)
1 (1.6%)
18 (28.1%)
37 (57.8%)
9 (14.1%)
Yes
191 (94.6%)
11 (5.4%)
72 (35.5%)
81 (39.9%)
50 (24.6%)
40 (95.2%)
2 (4.8%)
16 (38.1%)
14 (33.3%)
12 (28.6%)
148 (97.4%)
4 (2.6%)
55 (35.9%)
63 (41.2%)
35 (22.9%)
62 (92.5%)
5 (7.5%)
17 (25.0%)
39 (57.4%)
12 (17.6%)
No
82 (94.3%)
5 (5.7%)
28 (32.2%)
40 (46.0%)
19 (21.8%)
Yes
171 (96.1%)
7 (3.9%)
62 (34.4%)
78 (43.3%)
40 (22.2%)
55 (98.2%)
1 (1.8%)
28 (50.0%)
20 (35.7%)
8 (14.3%)
198 (94.7%)
11 (5.3%)
62 (29.4%)
98 (46.4%)
51 (24.2%)
Self confidencemean (SD)
1.5 (0.8)
1.2 (1.2)
0.197
1.4 (0.8)
1.4 (0.9)
1.6 (0.7)
0.483
Liking schoolmean (SD)
1.9 (1.5)
1.1 (1.6)
0.068#
2.2 (1.3)
1.7 (1.6)
1.6 (1.5)
0.018*
Extracurricular activitiesmean (SD)
3.2 (2.0)
3.0 (1.8)
0.677
3.3 (2.2)
3.3 (1.9)
3.0 (1.8)
0.505
11 years old and older

10 years old and younger
Mother works
0.304a
0.034*

Never
Once a week or more
0.208a
0.110
Safe neighborhood
No changes
Breasts/body hair or voice change/
body hair
0.537a
0.470a
0.911
0.013*
25 (83.3%)
5 (16.7%)
Finish high school
25 (100.0%)
0 (0%)
Beyond high school

Attitude toward abstinencemean (SD)
203 (96.7%)
0.011*a
7 (3.3%)
9 (29.0%)
13 (41.9%)
9 (29.0%)
9 (36.0%)
10 (40.0%)
6 (24.0%)
72 (34.1%)
95 (45.0%)
44 (20.9%)
0.864
84.3 (16.5)
71.6 (20.2)
0.010*
92.1 (9.6)
84.1 (14.8)
70.2 (20.5)
0.000***
Attitude toward refusing sexmean

(SD)
87.5 (17.6)
70.8 (24.6)
0.002**
88.2 (16.5)
89.0 (15.1)
79.9 (24.4)
0.005**
Parentchild communicationmean
(SD)
60.6 (31.4)
57.8 (27.8)
0.761
58.6 (32.3)
62.2 (29.4)
59.8 (32.7)
0.700

No
44 (95.7%)
2 (4.3%)
Yes
209 (95.4%)
10 (4.6%)

mean (SD)
8.7 (5.1)
10.4 (6.0)
1.00a
0.267
12 (26.1%)
22 (47.8%)
12 (26.1%)
78 (35.3%)
96 (43.4%)
47 (21.3%)
7.7 (5.0)
9.4 (5.1)
9.5 (5.4)
Due to small expected cell counts, results from Fishers exact test are shown instead of chi-square test
P \ 0.10, * P \ 0.05, ** P \ 0.01, *** P \ 0.001
effect was larger for girls than boys. Thus, it appears

important to get these pre-teens to accept the abstinence
message. Consistent and clear messaging, whether with
respect to abstinence or contraception, is one of the
0.467
0.042*
characteristics of effective teen pregnancy prevention

programs [17]. However, it is also important to consider
that the context of remaining abstinent may differ by
gender and address different underlying motivations. A
123
S62
Table 4 Results of binary logistic regressions of sexual experience and ordered logistic regressions of anticipated sexual activity by gender
Variables
Ever had sex
Males OR (95% CI)
Females OR (95% CI)
Males OR (95% CI)
Females OR (95% CI)
11 years old or older
1.10 (0.492.48)
0.49 (0.102.48)
0.96 (0.541.70)
0.88 (0.521.49)
10 years old or younger
1.00
1.00
1.00
1.00
Age
No
Yes
Safe neighborhood
1.00
1.00
1.00
1.00
0.73 (0.301.78)
3.17 (0.3627.90)
0.81 (0.411.59)
0.97 (0.551.72)
No
1.00
1.00
1.00
1.00
Yes
0.84 (0.352.06)
1.05 (0.264.23)
0.46 (0.240.89)*b
1.22 (0.712.08)b
No pubertal signs
1.00
1.00
1.00
1.00
Breasts/body hair or voice change/body hair
1.76 (0.664.68)
2.41 (0.2721.78)
2.68 (1.405.12)**
1.49 (0.802.78)
0.88 (0.701.12)
0.81 (0.541.21)
0.88 (0.731.06)
0.85 (0.711.00)#
1.00
1.00a
1.00
1.00
Finish high school
0.82 (0.223.04)
Beyond high school
0.86 (0.292.49)
0.38 (0.230.62)***
School variables
Liking school
0.98 (0.332.89)
0.69 (0.242.00)
0.47 (0.121.88)a
0.67 (0.281.60)
0.92 (0.422.02)
0.75 (0.411.35)
0.38 (0.260.57)***b
0.37 (0.270.49)***b
1.07 (0.761.52)
0.70 (0.530.93)*b
0.89 (0.681.16)
1.39 (1.071.81)*b

Attitude toward abstinence (SD units)
Attitude toward refusing sex (SD units)
Parentchild communication (SD units)
1.00 (0.611.65)
1.27 (0.831.92)
0.62 (0.381.03)
1.26 (0.642.48)

No/not sure
1.00
1.00
1.00
1.00
Yes
0.38 (0.160.89)*
0.91 (0.174.81)
0.55 (0.281.06)#
0.64 (0.321.27)
1.06 (0.971.17)
1.04 (0.911.19)
0.96 (0.901.02)b
1.05 (1.001.11)*b
Results in the Ever Had Sex columns are the odds of having had sex relative to the odds of not having had sex. Results in the Anticipated
Sexual Activity columns are the odds of being in the higher risk groups: high-risk group versus moderate/no-risk group and odds of being in the
high/moderate-risk group versus no-risk group
#
P \ 0.10, * P \ 0.05, ** P \ 0.01, *** P \ 0.001
In the Ever had sex model, for girls, the categories for Not finish high school and Finish high school were collapsed
The test of interaction of gender with this variable was significant: with safe neighborhood, P = 0.028; with attitude toward abstinence,
P = 0.075; with parentchild communication, P = 0.0006; with pubertal knowledge, P = 0.044
study of predominantly white 8th graders supports this

possibility, finding that boys initiate sexual activity in the
context of seeking status with their peers, whereas girls
initiate sex primarily in the context of a boyfriend/girlfriend relationship [18].
In national studies, non-virgin teens are somewhat more
likely to talk to their parents about sex than virgin teens,
with girls being more likely to have conversations about
how to say no to sex [19]. The differences we observed
with girls who anticipate sex reporting more parental
communication in contrast to boys suggest that younger
boys delay communication. Girls had significantly more
123
knowledge about puberty than boys, however, for boys

more knowledge about puberty was not associated with
anticipating sex, but for girls, it was positively related to
anticipating sexual activity. These results suggest that girls
may use knowledge about puberty in relation to their personal plans differently than boys, or, alternatively, that girls
accumulate pubertal knowledge differently as they gain
sexual interest or experience.
We anticipated an association between school connectedness and delaying sexual activity because of studies in
older adolescents [20]. In our study, the liking school
variable was associated with a lower likelihood of
anticipating sexual activity in girls only. For both genders,

neither liking school nor educational aspiration was significantly associated with virginity. These differences from
the other studies could reflect differences in the samples
and measurement, but also possibly the differences in the
ages of the subjects. The association of neighborhood
safety with anticipation of sex in boys is similar to Add
Health data on older teens noting a modest association of
neighborhood context to sexual initiation in boys [21]. Our
study suggests the neighborhood variable should be further
studied and may be especially relevant for males.
Our findings are limited to fifth graders in selected areas
of Washington, DC, so they might not be generalizable to
all fifth graders. Our data are based on self-reports and
actively consented students, raising the possibility of bias.
We do not know the context of the reported sexual
experiences.
Conclusions
Although other researchers have studied the factors we
investigated in older adolescents, the present data are
unique in their focus on preadolescents. The gender differences we found in fifth graders prior and anticipated
sexual experiences are similar to those that others have
found in older teenagers. The findings suggest that some
aspects of interventions to delay sexual activity should be
tailored to differing underlying attitudes and needs of preand early adolescent males and females and that young
males are especially important targets of interventions to
delay sexual debut. Our findings also support the development of interventions to prevent teenage pregnancy or delay
sexual debut prior to the middle school years, particularly in
communities with high levels of teenage pregnancy.
Acknowledgments The authors would like to thank the members of
the Building Futures for Youth research team of the NIH-DC Initiative to Reduce Infant Mortality for their roles in supporting this
research. The authors would also like to thank Allison Rose and
Deborah Schwartz for providing valuable comments on earlier drafts
of this manuscript. Funding was provided by grants from the Eunice
Kennedy Shriver National Institute of Child Health and Human
Development (NICHD) Grants 3U18HD030445, 3U18HD030447,
5U18 HD31206, 3U18HD031919, and 5U18HD036104 and the
National Center on Minority Health and Health Disparities
(NCMHHD). Participating institutions include Howard University,
Georgetown University, RTI International, and NICHD.
References
1. Centers for Disease Control and Prevention. (2010) Youth risk
behavior surveillance, 2009. MMWR Surveillance Summer, 59
(No. SS-5).
S63
2. Kann, L., Telljohann, S. K., & Wooley, S. F. (2007). Health
education results from school health policies and programs study
2006. The Journal of School Health, 77(8), 408434.
3. Mueller, T. E., Gavin, L. E., & Kulkrani, A. (2008). The association between sex education and youths engagement in sexual
intercourse, age at first Intercourse, birth control use at first sex.
Journal of Adolescent Health, 42, 8996.
4. Forehand, R., Gound, M., Kotchick, B. A., Armistead, L., Long,
N., & Miller, K. S. (2005). Sexual intentions of black preadolescents: Associations with risk and adaptive behaviors. Perspect
Sex Reproduction Health, 37, 1318.
5. Upchurch, D. M., Levy-Storms, L., Sucoff, C. A., & Aneshensel,
C. S. (1998). Gender and ethnic differences in the timing of first
sexual intercourse. Family Planning Perspect, 30(3), 121127.
6. Cuffee, J. J., Hallfors, D. D., & Waller, M. W. (2007). Racial and
gender differences in adolescent sexual attitudes and longitudinal
associations with coital debut. Journal of Adolescent Health,
41(1), 1926.
7. Petersen, J. L., & Hyde, J. S. (2010). A meta-analytic review of
research on gender differences in sexuality, 19932007. Psychological Bulletin, 136(1), 2138.
8. Raine, T. R., Jenkins, R., Aarons, S. J., Woodward, K., Fairfax, J.
L., El-Khorazaty, M. N., et al. (1999). Sociodemographic correlates of virginity in seventh-grade black and Latino students.
Journal of Adolescent Health, 24(5), 304312.
9. Bandura, A. (1986). Social foundations of thought and action: A
social cognitive theory. Englewood Cliffs, N.J.: Prentice-Hall.
10. Koo, H. P., Rose, A., El-Khorazaty, M. N., Yao, Q., Jenkins, R.
R., Anderson, K. M., et al. (2011). Evaluation of a randomized
intervention to delay sexual initiation among 5th graders followed through 6th grade. Sex Education: Sexuality, Society and
Learning, 11(1), 2746.
11. Udry, J. R. (2003). The National Longitudinal Study of adolescent
health (Add Health), Waves I & II, 19941996; Wave III,
20012002 [machine-readable data file and documentation].
Chapel Hill, NC: Carolina Population Center, University of North
Carolina at Chapel Hill.
12. Mathematica Policy Research Inc. (1999). Teen activities and
attitudes study, Wave 1 questionnaire. Princeton, NJ: Mathematica Policy Research.
13. Aarons, S. J., Jenkins, R. R., Raine, T. R., El-Khorazaty, M. N.,
Woodward, K. M., Williams, R. L., et al. (2000). Postponing
sexual intercourse among urban junior high school studentsa
randomized controlled evaluation. Journal of Adolescent Health,
27(4), 236247.
14. Koo, H. P., Rose, A., Bhaskar, B., & Walker, L. R. (2011).
Relationships of pubertal development among early adolescents
to sexual and nonsexual risk behaviors and caregivers parenting
behaviors. The Journal of Early Adolescence. doi:10.1177/
0272431611409746.
15. Rose, A., Koo, H. P., Bhaskar, B., Anderson, K., White, G., &
Jenkins, R. R. (2005). The influence of primary caregivers on the
sexual behavior of early adolescents. Journal of Adolescent
Health, 37(2), 135144.
16. 2007 Youth Risk Behavior Survey Results, District of Columbia
middle school survey. Summary table-weighed data. www.
dccampaign.org/content/view/142/89/. Accessed September 11,
2011.
17. The National Campaign to Prevent Teen and Unplanned Pregnancy. (2010). What works 2010: Curriculum-based programs
that help prevent teen pregnancy. www.TheNationalCampaign.
org. Accessed September 11, 2011.
18. Little, C. B., & Rankin, A. (2001). Why do they start it?
Explaining reported early-teen sexual activity. Sociology Forum,
16(4), 703729.
123
S64
19. The National Campaign to Prevent Teen and Unplanned Pregnancy. (2006). Sceince says: Parent-child communication about
sex and related topics. 25:16. www.teenpregnancy.org. Accessed September 10, 2011.
20. Resnick, M. D., Bearman, P. S., Blum, R. W., Bauman, K. E.,
Harris, K. M., Jones, J., et al. (1997). Protecting adolescents from
harm: Findings from the National Longitudinal Study on adolescent health. JAMA, 278(10), 823832.
123

21. Cubbin, C., Santelli, J., Brindis, C. D., & Braveman, P. (2005).
Neighborhood context and sexual behaviors among adolescents:
Findings from the National Longitudinal Study of adolescent
health. Perspectives on Sexual and Reproductive Health, 37(3),
125134.

Attitudes, Experience, and Anticipation of Sex Among

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Attitudes, Experience, and Anticipation of Sex Among

Uploaded by

Copyright:

Available Formats

Matern Child Health J (2011) 15:S54S64