Send Orders for Reprints to reprints@benthamscience.

net

10 The Open Family Studies Journal, 2013, 5, 10-18

Open Access
Family Influences on Adolescent Sexual Activity and Alcohol Use
Tsui-Sui Annie Kao* and Winifred Ann Carter

University of Michigan, School of Nursing, 400 North Ingalls Building, Ann Arbor, MI 48109-5482, USA

Abstract: Sexual activity and alcohol use continue to have negative effects on the well-being of today’s adolescents. This
study used multiple regression analysis to examine relationships among personal factors, family factors, and adolescent
sexual activity and alcohol use over three time points. Significant protective factors, which were inversely related to risky
behavior, included general adolescent–mother communication, father and family connectedness, and disapproving
parental attitudes towards sex. Permissive parenting was associated with self-efficacy for safe sex, early sexual initiation,
and increased alcohol use. Health care providers should work to promote positive parent–child relationships and familial
protective effects.
Keywords: Parent–child relationship, adolescent risky behaviors, sexual onset, alcohol abuse.

INTRODUCTION had sexual intercourse as those who did not consume

alcohol. The correlation between sexual initiation and
Adolescence is a period of development often marked by alcohol use makes it imperative for the health services
experimentation and engagement in new activities. Many
community to determine potential relationships between
adolescents engage in risky behaviors, including early sexual
these behaviors and an adolescent’s personal demographics
initiation, multiple sex partners, and alcohol use [1-3]. In
and family factors.
order to prevent serious physical and psychological health
problems during adolescence and later in life, it is critical for This study examines the relationship between personal
family members and health care providers to recognize and factors, family influences, self-efficacy for safe sex, and
eliminate factors related to adolescents’ risky health adolescent sexual behaviors and alcohol use over three time
behaviors. points. Because there is limited knowledge about family
influences on adolescents’ development of healthy
For adolescents, early sexual activity and alcohol use are
behaviors, findings from this study will assist community
well documented health problems. In 2007, a report [4] by
health care providers, mental health care providers, school
the U.S. Department of Health and Human Services noted administrators, counselors, and even parents, by identifying
that about 48% of high school students reported having
opportunities for intervention which will encourage
sexual intercourse, 15% of whom reported having four or
adolescents to practice healthy behaviors while avoiding
more sexual partners. A longitudinal study of 884 middle
risky ones.
school students [5] extrapolated that about 6.2% of
adolescents in the United States have had sex before age 13. BACKGROUND AND SIGNIFICANCE
Early sexual initiation has been associated with increased
health risks and negative health outcomes including lack of Research has suggested several significant predictors of
adolescent sexual initiation and alcohol use, including family
condom use, higher rates of sexually transmitted infections
function, family structure, and adolescents’ self-efficacy for
(STIs), and teenage pregnancy [2].
engaging in safe sex [3, 5, 6, 8-10]. In order to promote more
Alcohol use during adolescence also is a concern. effective communication and preventative care, it is
Approximately 16% of 12- to 17-year-olds reported having important to determine the relationships between family
their first drink prior to age 13 [6]. In 2009, about 41.8% of factors and adolescents’ sexual behaviors and alcohol use.
high school students had drunk alcohol [7]. Alcohol use and The specific aims of this study are to determine a) how
sexual activity can have compounded effects. A survey personal demographic factors and family structure are
analysis [3] of 26,023 students in grades 7–12 from one associated with adolescent sexual behaviors and alcohol use,
Midwestern state showed that sexually active male b) how family function is associated with adolescent sexual
adolescents reported a moderate-to-high level of alcohol use. behaviors and alcohol use when controlling for personal
An event history analysis [8] of 457 adolescents in grades 8– factors and family structure, c) how family structure, family
10 from single-mother and two-parent homes found that function, and self-efficacy, influence adolescent sexual
those who consumed alcohol were twice as likely to have behaviors and alcohol use, and d) how sexual behavior and
alcohol outcomes are related in adolescents.
*Address correspondence to this author at the University of Michigan, School Several studies show a relationship between family
of Nursing, 400 North Ingalls Building, Ann Arbor, MI 48109-5482, USA; structure and adolescent sexual activity. Similar to previous
Tel: 734-615-8209; Fax: 734-647-0351; research, for purposes of this study, family structure refers to
E-mails: anniekao@umich.edu, anniekao6493@gmail.com parental income, marital status, and education attained.

1874-9224/13 2013 Bentham Open

Family Influences on Adolescent Sexual Activity and Alcohol Use The Open Family Studies Journal, 2013, Volume 5 11

Recent research has suggested that parent income and others [19, 20] found no direct link between parent–child
socioeconomic status are significant predictors of early sexual discussion and adolescents’ risky sexual behaviors. In
adolescent sexual initiation [3]. Specifically, higher addition, parent–child communication about alcohol use is
socioeconomic status is associated with delay of sexual not predictive of adolescents’ initiation of alcohol uses [18].
initiation. Additionally, prior research [8] found a significant These mixed results often are attributed to parents’ or
association between single-parent families and early sexual adolescents’ inconsistent reports about their sexual
intercourse among adolescents. Adolescent girls living in discussion, delayed timing of engagement in such
dual-parent families and adolescents boys whose mothers discussions, or lack of in-depth discussion. While these
reported higher levels of education were more likely to factors could interfere with successful sexual discussion
remain abstinent [5]. Although these findings indicate some between parents and adolescents, indirect communication,
associations between family structure and adolescent sexual such as role modeling, monitoring, and adolescents’
debut, the relationship between alcohol use, early sexual perceived connectedness with parents and family are
behaviors, and family structure remains unclear. important to consider when assessing these diminished
effects [16, 17].
We also assessed six variables that address aspects of
family function, specifically: adolescents’ perceptions of An adolescent’s perceived connectedness with his or her
parental attitudes towards sex, degree of permissive parents and family is a documented factor in decision
parenting, maternal sexual discussion, general adolescent– making about risky behaviors. Research [3, 10] found a
mother communication, and adolescent connectedness with strong protective effect of family connectedness in relation
father and family, respectively. Each family function to adolescent sexual debut. Parent and family connectedness
variable is discussed below. has also has been found to be protective against adolescent
alcohol use. Specifically, adolescents who reported less
Research [9, 10] has noted significant associations
between parents’ disapproving attitudes about adolescent frequent use of alcohol also reported higher levels of
connectedness with and support from parents and other
sexual intercourse and alcohol use. Disapproving parental
family members [15, 21, 22]. Additionally, there is an
attitudes toward early sexual activity and alcohol use were
increasing emphasis on understanding and utilizing paternal
significantly correlated with delayed sexual debut and a
influences on adolescent development in risky behaviors [16,
decreased likelihood of using alcohol [9, 10]. While
22, 23]. In this study, we use longitudinal data from a
disapproving parents communicate their disapproval of risky
sexual behavior and alcohol use, a permissive parent acts “in comprehensive dataset of adolescent behavior to explore
evidence of a relationship between adolescents’ perceive
a non-punitive, acceptant, and affirmative manner towards
connectedness with their families and their reported sexual
the child’s impulses, desires, and actions” [11 p. 889].
and alcohol behaviors later in adolescence.
Specifically, a permissive parent allows children to regulate
their own activities as much as possible, avoids the exercise Theoretical Framework and Self-Efficacy
of control, and does not encourage them to obey externally
defined standards. The conceptual model for this study is derived from
social cognitive theory [24], which describes human
Research [12-14] has shown that permissive parenting is behavior as a triadic, dynamic, and reciprocal process in
a significant risk factor for early sexual initiation and alcohol which behavior, personal factors, and environmental
use. Decreased parental involvement and an absence of influences interact [25, 26]. In this study, we operationalize
parental decision making about a child’s health were the triadic model with these measurements:
predictors of early sexual initiation [12]. Less permissive
parenting, on the other hand, was protective against early 1. Personal factors: age, gender, racial/ethnic back-
adolescent sexual initiation. Daughters with mothers who ground (independent variable)
restricted and monitored their participation in external 2. Environmental factors: family structure and family
activities demonstrated a lower risk for sexual activity [13, function (independent variable)
14]. In terms of predicting adolescents’ alcohol use,
however, it is not clear which styles are most protective. 3. Behaviors: sexual initiation and alcohol use (dep-
endent variables)
Although strict parenting practices are not significantly
related to alcohol use, parents’ supportive relationship and According to social cognitive theory [24, 25, 27, 28]
monitoring practices are related to adolescents’ lower perceived self-efficacy-defined as an individual’s belief in
likelihood of using alcohol [9, 15, 16]. his or her ability to produce effects-serves as an important
Although health care providers are encouraged to urge construct in an adolescent’s cognitive processing and
influences his or her decision making. Social cognitive
parents to discuss risks associated with sexual activity with
theory proposes that for adolescents, the interplay of
their adolescents, reported results of the effects of maternal
personal competencies, self-management efficacy, and the
sexual discussion on adolescent sexual initiation and alcohol
prevailing influence of influential peers determine whether
use are inconsistent. Some researchers [5, 17] have
they forsake risky activities or become chronically enmeshed
suggested that increased maternal sexual discussion is
protective against risky health behaviors, whereas others [12, in them.
18] have indicated that maternal sexual discussion is not Cross sectional studies [5, 13] have recognized self-
protective against or predictive of risky behaviors. As efficacy to engage in safe sex as an important predictor of
reported previously [5], when daughters feel comfortable adolescents’ safety in other health behaviors. Self-efficacy is
communicating with their mothers about sex, they are less necessary for behavior change because it directly influences
likely to engage in sex over the following year. However, effort and performance of any given task [25]. Research [5,
12 The Open Family Studies Journal, 2013, Volume 5 Kao and Carter

13] has shown that adolescent males and females who had Measures
the self-efficacy to refuse sex also were less likely to engage
Specific variables were extracted from Waves I, II, and
in sex. Building on these findings, our study examined if
III of the Add Health dataset and factor analysis was
adolescent self-efficacy to engage in safe sex is related to
performed to estimate how much variability was due to
current or past family structure or family function, and ways
in which their perceived self-efficacy impacts future sexual common factors prior to establishing reliability. Z-scores
then were calculated for each item and summarized to form
initiation or alcohol use.
the independent variables for the regression modeling.
Adolescents’ involvement in risky health behaviors is a
Dependent variables were drawn from Wave III data,
complex issue that can lead to devastating health
which assessed age of sexual initiation (higher value reflects
consequences and costs to both the individual and the overall
health care system. A more explicit understanding of familial earlier sexual debut), total number of sex partners (1–50),
and total alcohol use. The Add Health study measured five
influences on adolescents’ health risk behaviors will assist
aspects of adolescent alcohol use:
health providers who are involved in developing efficacious
family-centered intervention programs for use at and 1. sexual regret after alcohol use;
individual and community level.
2. total days of alcohol use during the past year;
The review of literature suggests several hypotheses
3. the frequency of intoxication;
which can be tested through statistical analysis of the Add
Health dataset: a) family structure is related to adolescents’ 4. binge drinking within the previous two weeks; and
risky health behaviors; b) family function is predictive of 5. mean alcohol consumption per drinking episode
sexual behaviors as well as alcohol use when personal during the past year.
factors are held constant; c) stronger family structure and
function are associated with increased self-efficacy for safe Independent variables were drawn from Waves I and II
sex; and d) alcohol use and sexual behaviors are significantly data. Wave I variables assessed adolescent demographic
associated in adolescents. For purposes of this study, family factors (age, gender, and race), family structure (parent
structure includes the variables parental income, marital income, marital status, and education level), and family
status, and education; family function includes disapproving function (adolescents’ perceived parental attitudes toward
parental attitudes about sex, the degree of permissive sex, permissive parenting style, mother–adolescent
parenting, parent-adolescent communication, family relationships, and adolescent connectedness with their
connectedness, and mother-adolescent communication. fathers and their families). Personal demographic factors
were included in the analysis and subsequently controlled in
METHODS the multiple regression models. To assess family structure,
In this study, we conducted a secondary analysis of the parent income and education levels were grouped into
nationally representative, longitudinal Adolescent Health several categories, with higher values representing higher
(Add Health) dataset [29] across the first three waves of data income and education levels. Parent marital status was
collection. Prior to conducting the data analysis, we received recorded into two categories: married and
approval for exemption from IRB review through the single/divorced/widowed.
University of Michigan Health Science Institutional Review Perceived parental attitudes about sex consisted of four
Board (HUM00031244). responses from adolescents. Higher values represented
Sample disapproving parental attitudes about their adolescents
engaging in sexual behaviors at Wave IThis measure had a
The study sample used data from the University of North good reliability (Cronbach’s
=.904).
Carolina’s Add Health database, which sampled over
Permissive parenting style consisted of seven items that
100,000 adolescents. The longitudinal data selected for this
asked adolescents if their parents allowed them to make their
secondary analysis consisted of data spanning three time
own decisions about curfews, choosing friends, what to
points, defined as Waves I, II, and III. This analysis included
wear, watching television, bedtimes, and diet. In other
a combination of questionnaires and in-home interviews with
words, permissive parents allowed their adolescents to make
adolescents and parents conducted at Wave I (1994–1995),
follow-up interviews with the adolescents at Wave II (1996), decisions freely. While the reliability of the measure was low
(Cronbach’s
=.658), we decided to keep the measure in the
and in-home questionnaires and interviews with adolescents
model because it is frequently suggested that parenting style
and parents at Wave III (2001–2002). For this analysis,
is an important factor to consider when exploring parental
inclusion criteria at Wave I consisted of adolescents in
influences [9, 11, 30, 31].
grades 7-12 who also participated in Waves II and III. The
Add Health researchers considered the mother to be the Mother–adolescent relationships were assessed via two
preferred parent for this study. Our sample included a total domains: sexual discussion and general communication
of 9,067 adolescents and mothers. In terms of the adolescent between mothers and adolescents. Mother–adolescent sexual
population, slightly more females than males were included, discussion was assessed with five items that asked mothers
53.8% to 46.2%. Their mean age at Wave I was 14.50 years (or a mother figure) questions about how much they talked
(SD = 1.72 years). See Table 1. with their children about birth control and sex, particularly
Family Influences on Adolescent Sexual Activity and Alcohol Use The Open Family Studies Journal, 2013, Volume 5 13

Table 1. Demographic Descriptions of Sample at Wave I (N=9067)

Age (Mean =14.50; SD= 1.172) Number Percentage

11 3 .0
12 373 4.1
13 1731 19.1
14 2126 23.4
15 2613 28.8
16 2221 24.5

Gender
Male 4188 46.2
Female 4879 53.8
Racial/Ethnic Background
Hispanic 1300 14.3
African American 2104 23.2
Asian 591 6.5
Caucasian 5817 64.2
Parental Education ((Mean= 4.65; SD= 1.250
(1) Never went to school 7 .1
th
(2) 8 grade or below 315 3.5
th
(3) Completed 8 grade without completing high school or without finishing business/vocational school 767 8.5
(4) High school diploma or GED 2090 23.1
(5) Business or trade school after high school, or college without graduating 2072 22.9
(6) Graduated from college 1009 11.1
(7) Professional training after obtaining 4-year college degree 608 6.7
Parent Income (Mean=3.54; SD= 1.694)
(1) 0–10,000 630 6.9
(2) 10,001–35,000 1358 15.0
(3) 35,001–40,000 1423 15.7
(4) 40,001–50,000 789 8.7
(5) 50,001–65,000 831 9.2
(6) 65,001–1000,000 857 9.5
(7) 100,001–999,000 228 2.5

Parent Marital Status

Single, divorced, or widowed 2319 25.6
Married 8148 64.3
Total Sample N = 9067 100%
SD: Standard Deviation.

the negative consequences of pregnancy, the dangers of an wrong, and if they were satisfied with the way they
STI, and the moral/ethical issues of sexual intercourse. These communicated with their mothers. These items were reverse
questions consisted of responses on a 1–4 scale, with a coded to associate a higher number with a higher level of
higher number representing more frequent discussion communication. An adolescent without a mother or maternal
between mother and adolescent. Acceptable reliability was figure was assigned a zero value for this measurement.
noted (Cronbach’s
=.899). General adolescent–mother Reliability was adequate (Cronbach’s
=.834).
communication consisted of three items assessing whether
We assessed two domains of connectedness to evaluate
adolescents would initiate discussion with their mothers adolescents’ perceived relationship with their fathers and
about things happening in school, when something went
their families. Adolescent–father connectedness consisted of
14 The Open Family Studies Journal, 2013, Volume 5 Kao and Carter

four specific questions that asked adolescents about their Personal Factors
perceived warmth, satisfaction, closeness, and caring
Age was the most significant personal factor. At Wave I,
relationships with their fathers. All items regarding father
the adolescent was between grades 7 and 12, approximately
connectedness were reverse-coded to associate a higher score
11 – 18 years old. Wave II occurred one year later [29]. The
with a higher level of connectedness. The reliability for this
measure was good (Cronbach’s
=.850). Family older age of adolescents at Wave I was more likely to report
a later sexual initiation ( = -.17, p
.001) at Wave III and
connectedness was assessed via four questions that asked
more likely to report a lower average number of drinks
adolescents about their perceptions of parental care, their
consumed in the year prior ( = -.07, p
.001). Older ages
families’ understanding of them, time spent together as a
were associated with more days of alcohol use ( =.03,
family, and degree of attention paid to each other. A higher
p
.05). Male gender was significantly related to a higher
score reflected a higher level of connectedness. Reliability
for family connectedness measure was acceptable degree of self-efficacy to engage in safe sex as reported at
Wave 2 ( =.16, p
.001). Male gender was found to be a risk
(Cronbach’s
=.758).
factor for a higher number of sex partners, sexual regret after
The final independent variable, self-efficacy for safe sex, alcohol use, a higher number of days of alcohol use, a higher
was drawn from responses at Wave II. Because the dataset number of days of intoxication, consuming five or more
did not specifically assess self-efficacy to delay sexual drinks on one occasion, and higher average number of drinks
initiation or abstain from alcohol use, self-efficacy to engage in the past year.
in safe sex was used as a proxy for general self-efficacy.
Adolescents of Asian descent were more likely to report
This measure of self-efficacy was then regressed to Wave III
later sexual initiation ( = -.05, p
.05). African American
variables assessing sexual initiation and alcohol-related
adolescents experienced the same age of sexual onset and
behaviors. Self-efficacy was assessed via questions about an
similar number of sexual partners compared with adolescents
adolescent’s ability to interrupt behaviors leading to sex in
order to use birth control, his or her ability to plan for the use of other racial groups. In addition, African American
adolescents were less likely to report sexual regret after
of birth control, and his or her ability to resist intercourse in
alcohol use and less likely to report using alcohol at Wave
the absence of birth control. These items were re-coded on a
III. Adolescents of Hispanic descent ( = -.05, p
.05)
1–5 scale, with a higher number reflecting a higher level of
reported lower than average levels of self-efficacy for safe
assurance. The reliability for self-efficacy was acceptable
sex, whereas Caucasian adolescents ( =.06, p
.05) were
(Cronbach’s
=.728).
more likely to report a higher degree of self-efficacy for safe
Data Analysis sex.
SPSS Statistics 18 was used for data analysis. Multiple Family Structure
regressions were first performed to explore relationships
All family structure variables were significantly
among personal factors, family factors, and reported
protective against early sexual initiation. Parent income was
behaviors. Dependent variables were regressed on
associated with higher self-efficacy for safe sex ( =.07,
independent variables to determine the respective variances.
After determining a full model, block regression was p
.001), being protective against early sexual initiation ( =
-.04, p
.0), regret of sexual activity due to alcohol use (
performed to determine if the independent variables of
=.10, p
.001), total days of alcohol use ( = .16, p
.001),
family function and self-efficacy remained significant to
total days of intoxication (=.16, p
.001), consuming five or
each dependent variable when the independent variables of
more drinks at one time ( =.10, p
.001), and high average
family structure and personal factors were held constant.
number of drinks in one year ( =.07, p
.001). Parent
Finally, a correlation analysis was conducted to determine if
alcohol use correlated with age of sexual initiation and total education was associated with higher self-efficacy for safe
sex ( =.04, p
.05), being protective against early sexual
number of sex partners.
initiation ( = -.05, p
.05), regret of sexual activity due to
RESULTS alcohol use ( =.09, p
.001), total days of alcohol use (
=.11, p
.001), total days of intoxication ( =.11, p
.001),
The results of our multiple regressions (see Table 2)
and consuming five or more drinks at one time ( =.05,
showed that highly rated personal factors, family structure,
p
.05). Marital status was found to be associated with being
and family function contribute to delay of early sexual protective against early sexual initiation ( = -.08, p
.001)
initiation and alcohol use. Contrary to our hypothesis, self-
but unlike the other factors, adolescents with married parents
efficacy for safe sex was significantly associated with early
were associated with fewer days of alcohol use ( = -.04,
sexual initiation, higher levels of sexual regret after alcohol
p
.05).
use, and increased days of alcohol use. None of the family
structure variables were significantly related to total number Family Function
of sex partners. Consistent with our hypothesis, a permissive
parenting style was found to be predictive of high self- Every family function variable was significantly
associated with adolescent sexual initiation. Disapproving
efficacy, early sexual initiation, and higher level of alcohol
parental attitudes, family– and father–adolescent connected-
use. Mother–adolescent sexual discussion, father–adolescent
ness, and general adolescent–mother communication were
connected-ness, and general adolescent-mother communi-
found to be protective against early sexual initiation.
cation were not significant predictors of any of the alcohol
Permissive parenting style ( =.04, p
.05) and mother–
outcome variables. The following are the results organized
for each group of independent variables. adolescent sexual discussion at Wave I ( =.12, p
.001) were
related to adolescents’ early sexual initiation at Wave 3.
Family Influences on Adolescent Sexual Activity and Alcohol Use The Open Family Studies Journal, 2013, Volume 5 15

Table 2. Results of Multiple Regression

Wave I Wave II Wave III Outcome Variables

Days of Five or Average
Self- Sexual Number of Sexual Days of
Demographic data Alcohol More Number of
Efficacy Initiation Sex Partners Regret Intoxication
Use Drinks Drinks

Personal Factors
Age -.17** .03* -.07**
Gender, 1=female, 2-male .16** -.09** -.08** -.13** -.15** -.18** -.17**
African American -.12** -.17** -.18** -.16** -.18**
Hispanic -.05* -.05*
Asian -.05* -.05* -.04* -.06*
Caucasian .06* .06*

Family Structure
Parent Income .07** -.04* .10** .16** .16** .10** .07**
Parent Education Level .04* -.05* .09** .11** .11** .05*
Parent Marital Status -.08** -.04*

Family Function
Disapproving Parental Attitudes -.18** -.12** -.04* -.05* -.03* -.038* -.04*
Degree of Permissive Parenting Style .05** .04* .03* .06** .06** .04* .03*
Mother–Adolescent Sexual Discussion with Adolescent .06** .12** .06**
Adolescent Connectedness to Father -.06**
Adolescent Connectedness to Family -.07** -.06* -.060* -.039* -.044*
General Adolescent–Mother Communication .04* -.05* -.06**
Model 1 R
.05 .05 .027 .064 .132 .142 .080 .077
Model 2 R
.06 .12 .059 .074 .144 .150 .085 .083
R
change .01** .07** .032** .010** .011** .008** .005* .006**
*p
.05; **p
.001.

Parental disapproving attitudes and permissive parenting also reported an earlier age of sexual initiation at Wave III.
style emerged as the most significant predictors for all sexual We also found self-efficacy for safe sex to be associated with
behavior and alcohol use outcomes. Disapproving parental higher levels of sexual regret after alcohol use (
=.03,
attitudes were protective against early sexual initiation (
= - p
.05) and increased days of alcohol use (
=.05, p
.05).
.18, p
.001), total number of sex partners ( = -.12, p>.001),
sexual regret after alcohol use (
= -.04, p
.05), days of Correlation Among Dependent Variables
alcohol use (
= -05, p
.05), days of intoxication (
= -.03, All alcohol outcome variables were positively correlated
p
.05), five or more drinks on one occasion (
= -.04, with total number of sex partners. That is, adolescents who
p
.05), and average number of drinks in the past year (
= - reported any form of excessive drinking (defined by
.04, p
.05). Permissive parenting style was associated with duration, amount, intoxication, or on average) were more
higher self-efficacy specific to engage in safe sex (
=.05, likely to have multiple sex partners (r =.124–.160, p
.05).
p
.001), but was a risk factor for early sexual initiation (
Sexual regret after alcohol use was closely related to total
=.04, p
.05), sexual regret after alcohol use (
=.03, p
.05), number of sex partners at Wave III (r =.221, p
.05).
days of alcohol use (
=.06, p
.001), days of intoxication (
Conversely, early sexual initiation was only related to sexual
=.06, p
.001), five or more drinks on one occasion (
=.04, regret after alcohol use (r =.037, p
.05) and excessive
p
.05), and average number of drinks in the year prior (
drinking, five or more drinks either within 2 weeks (r =.041,
=.03, p
.05). Disapproving parental attitudes, mother– p
.05) or 12 months (r =.041, p
.05). Number of days
adolescent sexual discussion, and general adolescent–mother drinking and number of days of intoxication were not found
communication at Wave I were positively associated with to be statistically significant with regard to early sexual
adolescent self-efficacy to engage in safe sex at Wave II. initiation.
As an independent variable, self-efficacy to engage in Variances Explained
safe sex was found to be significantly associated with early
sexual initiation (
=.05, p
.05). Adolescents who reported a After controlling for personal and family structure
high level of self-efficacy for engaging in safe sex at Wave II factors, our models explained 3–14% of the variance over all
16 The Open Family Studies Journal, 2013, Volume 5 Kao and Carter

outcome variables. After entry of family function, the model his/her behavior in order to obtain a desired outcome,.
explained an additional .5–7% (R
change =.005–.72). Ultimately, managing these pressures and resisting risky
Although the R
change values were small, they were all activity requires the adolescent to use a range of self-
significant (p
.05, see Table 2). regulation and self-efficacy techniques [28].
DISCUSSION Adolescent family connectedness was also found to be
highly protective against risky behaviors. Feeling part of a
While personal factors and family structure relate to close family may help fulfill an adolescent’s need to beloved
adolescent risky health behaviors, this study shows several or connected. Early sexual relationships may substitute for
modifiable family function variables also contribute to family closeness when the latter does not exist [3]. Families
adolescent behavior. These modifiable factors include operate as multiple social systems, with multiple interlocking
disapproving parental attitudes, structured parenting style, relationships [35]. As such, it is important to assess how the
father and family connectedness, and general adolescent– family functions as a whole. It has been suggested [25, 27,
mother communication. More importantly, when personal 28] that parental efficacy and family efficacy may play an
factors and family structure are held constant, the effect of important role in adolescents’ development of various
family function is significant over time. An understanding of aspects of self-efficacy. For example, our results showed that
the modifiable family function variables provides health care while adolescents’ perceptions of general communication
workers meaningful points of discussion with adolescents with their mothers was protective against early sexual
and their families. Providers may be able to offer initiation and number of sex partners, mother–adolescent
interventions that maximize a family’s protective effects sexual discussion was not similarly protective. It is possible
thereby decreasing adolescents’ risky behaviors. For that a mother’s openness to allowing her adolescent to
example, working with families to increase parental communicate feelings through general communication may
disapproval about adolescent sexual activity may be an be more effective than lecturing a teenager on the negative
effective way to decrease an adolescent’s likelihood of consequences of sex. Consequently, a closer mother–
engaging in risky behaviors. Although parents’ marital status adolescent relationship may facilitate adolescents’
and income might contribute to adolescents’ early sexual mindfulness to mothers’ expectations and enhance
initiation, this research echoes previous reports [3] which acceptance of their parents’ values and beliefs.
suggest that more frequent parental supervision may be
beneficial in delaying adolescent sexual initiation. Questions remain concerning the relationship between
alcohol use and sexual behaviors. Although a causal
Our results show that while higher parent income and relationship could not be established by this study, the
education levels are protective against early sexual initiation, associations among family function, drinking patterns, and
they are also risk factors for adolescent alcohol use. This sexual behaviors cannot be ignored. Congruent with the
might be attributed to differences in perceived social norms literature [36], we not only found that disapproving parental
about drinking: more affluent parents may inadvertently attitudes about sex were protective against early sexual
model drinking behavior as an acceptable social activity. initiation and multiple sex partners, but we also found them
Such adolescents may also have more access and more to have a protective effect on drinking behavior over time.
opportunities to consume alcohol. Since adolescents’ alcohol This verifies the importance of parental influence,
use is highly correlated with adolescents’ multiple sexual particularly perceived parental attitudes, on the development
partners and is a co-factor for HIV, it is important to develop of an adolescent’s health behaviors, including decisions
strategies to minimize such risk [32-34]. about when to have sex and consume alcohol. It is critical for
A higher degree of permissive parenting style was found health care providers to encourage parents to connect with
to be associated with early sexual initiation and alcohol use their children and help parents learn effective ways to
at Wave III. If we understand permissive parenting to be express their attitudes, values, expectancies, and beliefs
indicative of a decrease in structured parenting practices (i.e., about health behaviors without fear.
parental monitoring), this finding correlates with the impact Across all outcome measures, the R
change was found to
of protective parental monitoring with regards to delaying be low. To explore this further, other factors may need to be
adolescent sexual debut [13, 16]. Since we did not find added to the analysis. Furthermore, the variables in the Add
adolescent self-efficacy for engaging in safe sex to be Health dataset were not specifically designed for our study.
protective against risky sexual behaviors, we suspect that Each variable should be explored in-depth to determine its
adolescents’ confidence in their ability to stop sexual activity specific factors and their contribution to risky behaviors.
when they do not have condoms or are highly aroused might Additionally, parenting style should be further explored
be overestimated, particularly for those who were not because of the low reliability. Different dimensions of
sexually active at an earlier time point (Wave II). parenting style should be examined, such as degree of parent
It is possible that this aspect of self-efficacy is not demandingness (i.e., a parent’s willingness to set behavior
sufficient to explain adolescents’ risky behaviors over time. expectations for his or her child) [11].
Self-efficacy is a domain-specific competency [28]. Though Although the Add Health study surveyed adolescents and
self-efficacy beliefs can increase and sustain an individual’s families between 1994 and 2002 and findings may not be
motivation, efficacy beliefs alone will not produce desired directly applicable to contemporary adolescents, the
behavioral outcomes if other sub-skills needed for such longitudinal data is especially useful for understanding
action are lacking. Thus, it is important for an adolescent to parental influences on adolescents’ behavior concerning their
develop other elements of efficacy, such as self-regulation health choices. The time span of the original dataset, coupled
[25, 28], which is an individual’s ability to take control of with its size and scope, provide insight into the long term
Family Influences on Adolescent Sexual Activity and Alcohol Use The Open Family Studies Journal, 2013, Volume 5 17

effect of parental and family influences upon a range of study design. Information on how to obtain the Add Health
adolescent behaviors. More research is needed to help data files is available on the Add Health website
explain how self-efficacy specific for safe sex predicted (http://www.cpc.unc.edu/addhealth). No direct support was
early sexual initiation, sexual regret after alcohol use, and received from grant P01-HD31921 for this analysis.
more frequent alcohol use. To help explain this finding,
future studies should address adolescent health by designing REFERENCES
more specific tools which can specifically measure self- [1] Centers for disease control and prevention. National youth risk
efficacy. behavior survey overview. Atlanta, GA: U.S. Department of Health
and Human Services, Division of Adolescent and School Health
CONCLUSIONS 2009.
[2] Eaton DK, Kann L, Kinchen S, et al. Youth risk behavior
Risky adolescent behaviors are a serious health issue. surveillance — United States, 2009. USA: Department of Health
Our results support the protective effects of modifiable and Human Services: Centers for Disease Control and Prevention
June 4, 2010.
family function variables, specifically disapproving parental [3] Lammers C, Ireland M, Resnick M, Blum R. Influences on
attitudes, connectedness to family, and adolescents’ adolescents’ decision to postpone onset of sexual intercourse: a
communication with their mothers. Overall, these results survival analysis of virginity among youths aged 13 to 18 years. J
indicate that health professionals who work with adolescents Adolesc Health 2000; 26(1): 42-8.
in the community should be alert to adolescents’ potentially [4] Eaton DK, Kann L, Kinchen S, et al. Youth risk behavior
surveillance—United States, 2007. USA: Department of Health and
overestimated confidence, which may manifest itself in the Human Services: Centers for Disease Control and Prevention 2008.
areas of sexuality and alcohol use. [5] Laflin MT, Wang J, Barry M. A longitudinal study of adolescent
transition from virgin to nonvirgin status. J Adolesc Health 2008;
Our findings suggest that while mother–child discussion 42(3): 228-36.
about sex might be important, it is statistically significant [6] Center for Disease Control Prevention. National Youth Risk
that adolescents who feel connected and have a stronger Behavior Survey Overview. Atlanta, GA: US Department of Health
relationship with their parents will be more likely to have and Human Services, Division of Adolescent and School Health
2009.
later sexual onset and fewer sex partners. Thus, to prevent [7] Centers for Disease Control and Prevention. Youth Risk Behavior
adolescents from risky behaviors, it is imperative for health Surveillance — United States, 2009. USA: Department of Health
care professionals to encourage strengthening of and Human Services Centers for Disease Control and Prevention.
communication and boundaries between adolescents and June 4, 2010.
their parents and families. Specifically, health care providers [8] Whitbeck LB, Yoder KA, Hoyt DR. Early adolescent sexual
activity: a developmental study. J Marriage Fam 1999; 61(4): 934-
and others involved in the health of adolescents can work to 46.
increase modifiable family function factors, including [9] O’Donnell L, Stueve A, Duran R, et al. Parenting practices,
disapproving parental attitudes, increased parental parents’ underestimation of daughters’ risks, and alcohol and
monitoring, father and family connectedness, and general sexual behaviors of urban girls. J Adolesc Health 2008; 42(5): 496-
502.
adolescent–mother communication. It has been shown that [10] Resnick MD, Bearman PS, Blum RW, et al. Protecting adolescents
parental behavior can be changed and has a positive effect on from harm. Findings from the National Longitudinal Study on
adolescent health outcomes. According Hawkins, Catalano, Adolescent Health. JAMA 1997; 278(10): 823-32.
and Miller [37], parenting skills training and parental [11] Baumrind D. Effects of authoritative parental control on child
involvement have been shown to produce protective effects behavior. Child Dev1966; 37: 887-907.
[12] Alexander CS, Astone NM, Bandeen-Roche K, Ensminger ME,
on adolescent health behavior. Health care providers should Mekos D, Roche KM. Parenting influences on early sex initiation
encourage parents to provide sufficient guidance and among adolescents: how neighborhood matters. J Fam Issues 2005;
monitoring to steer adolescents away from risky situations. 26(1): 32-54.
As adolescents try to negotiate personal and social pressures, [13] Dancy BL. African-American adolescent females’ predictors of
having sex. J Natl Black Nurses Assoc 2006; 17(2): 30-8.
parents need to provide opportunities and support for their [14] Dancy BL, Crittenden KS, Ning H. African-American adolescent
adolescents to learn and develop self-regulation efficacy. girls’ initiation of sexual activity: survival analysis. Womens
Health care professionals can play an integral role in this Health Issues (WHI) 2010; 20(2): 146-55.
dynamic learning process by providing parents continued [15] DeHaan L, Boljevac T. Alcohol use among rural middle school
training and support. students: adolescents, parents, teachers, and community leaders’
perceptions. J Sch Health 2009; 79(2): 58-66.
CONFLICT OF INTEREST [16] Lohman BJ, Billings A. Protective and risk factors associated with
adolescent boys’ early sexual debut and risky sexual behaviors. J
The authors confirm that this article content has no Youth Adolesc 2008; 37(6):723.
conflict of interest. [17] Riesch SK, Anderson LS, Krueger HA. Parent–child
communication processes: preventing children's health-risk
ACKNOWLEDGEMENTS behavior. J Spec Pediatr Nurs 2006; 11(1): 41-56.
[18] Ennett ST, Bauman KE, Foshee VA. Parent–child communication
This research uses data from the National Longitudinal about adolescent tobacco and alcohol use: what do parents say and
Study of Adolescent Health, a project directed by Kathleen does it affect youth behavior? J Marriage Fam 2001;63(1):48-62.
[19] Rodgers KB. Parenting processes related to sexual risk-taking
Mullan Harris and designed by J. Richard Udry, Peter S. behaviors of adolescent males and females. J Marriage Fam 1999;
Bearman, and Kathleen Mullan Harris at the University of 61(1): 99-109.
North Carolina at Chapel Hill, and funded by grant P01- [20] Kao TA, Loveland-Cherry C, Guthrie B. Maternal influences on
HD31921 from the Eunice Kennedy Shriver National Asian American Pacific Islander adolescents' perceived maternal
sexual expectations and their sexual initiation. J Fam Issues 2010;
Institute of Child Health and Human Development with 31(3):381-406.
additional funding from 23 other federal agencies and [21] Resnick MD, Bearman PS, Blum RW, et al. Protecting adolescents
foundations. Special acknowledgment is given to Ronald R. from harm: findings from the national longitudinal study on
Rindfuss and Barbara Entwisle for assistance in the original adolescent health. JAMA 1997; 278(10): 823-32.
18 The Open Family Studies Journal, 2013, Volume 5 Kao and Carter

[22] Rienks SL, Wadsworth ME, Markman HJ, Einhorn L, Moran Etter esteem and sense of familism. Marriage Fam Rev 2004; 36(1/2),
E. Father involvement in urban low income fathers: baseline 35-65.
associations and changes resulting from preventive intervention. [31] Chen Z, Dornbusch SM, Liu RX. Direct and indirect pathways
Fam Relations 2011; 60(2): 191-204. between parental constructive behavior and adolescent affiliation
[23] Amoran OE. Parental influence on reproductive health behaviour with achievement-oriented peers. Child Fam Stud 2007; 16(6):
of youths in Ibadan, Nigeria. Afr J Med Med Sci 2008; 37(1): 21-7. 837-58.
[24] Bandura A. Self-efficacy: toward a unifying theory of behavioral [32] Balodis IM, Potenza MN, Olmstead MC. Binge drinking in
change. Psychological Rev 1977;84(2):191-215. undergraduates: relationships with sex, drinking behaviors,
[25] McAlister, AL, Perry CL, Parcel GS. How individuals, impulsivity, and the perceived effects of alcohol. Behav Pharmacol
environments, and health behavior interact: social cognitive theory. 2009; 20(5-6): 518-26.
In: Glanz K, Rimer BK, Lewis FM, eds. Health behavior and health [33] Quinn PD, Fromme K. Self-regulation as a protective factor against
education: theory, research and practice. 4th ed. San Francisco, risky drinking and sexual behavior. Psychol Addict Behav 2010;
CA: Jossey-Bass 2008. 24(3): 376-85.
[26] Bandura A. Social cognitive theory in cultural context. Appl [34] Testa M, Hoffman JH, Livingston JA. Alcohol and sexual risk
Psychol-Int Rev 2002;51(2):269-90. behaviors as mediators of the sexual victimization-revictimization
[27] Bandura, A, Caprara, GV, Barbaranelli, C, Regalia, C, Scabini, E. relationship. J Consult Clin Psychol 2010; 78(2): 249-59.
Impact of family efficacy beliefs on quality of family functioning [35] Cox MJ, Paley B. Families as systems. Annu Rev Psychol 1997;
and satisfaction with family life. Appl Psychol-Int Rev 2011; 60: 48: 243-266.
421-48. [36] Bersamin M, Todd M, Fisher DA, Hill DL, Grube JW, Walker S.
[28] Bandura A. Adolescent development from an agentic perspective. Parenting practices and adolescent sexual behavior: a longitudinal
Greenwich, Connecticut: IAP information age publishing, 2006. study. J Marriage Fam 2008; 70(1): 97-112.
[29] National longitudinal study of adolescent health (Add Health): [37] Hawkins JD, Catalano RF, Miller JY. Risk and protective factors
Research Design, 2009. [cited: 29th Nov 2012]. Available from: for alcohol and other drug problems in adolescence and early
http://ww.cpc.unc.edu/projects/addhealth/design. adulthood: implications for substance abuse prevention. Psychol B
[30] Bush, KR, Supple, AJ, Lash, SB. Mexican adolescents’ perceptions 1992; 112: 64-105.
of parental behaviors and authority as predictors of their self-

Received: July 17, 2013 Revised: September 23, 2013 Accepted: September 24, 2013

© Kao and Carter; Licensee Bentham Open.

This is an open access article licensed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/3.0/)
which permits unrestricted, non-commercial use, distribution and reproduction in any medium, provided the work is properly cited.