Sex Res Soc Policy (2015) 12:137–144

DOI 10.1007/s13178-015-0183-z

Comparing Sexuality Communication Among Offspring

of Teen Parents and Adult Parents: a Different Role for Extended
Family
Jennifer M. Grossman & Allison J. Tracy &
Amanda M. Richer & Sumru Erkut

Published online: 19 February 2015

# Springer Science+Business Media New York 2015

Abstract This brief report examined teenagers’ sexuality com- conversations about sex (DiIorio et al. 2003). However, stud-
munication with their parents and extended families. It compared ies suggest that many adolescents view other family members
who teens of early parents (those who had children when they as resources for information about sex, particularly among
were adolescents) and teens of later parents (those who were urban youth (Harper et al. 2012; Teitelman et al. 2009). This
adults when they had children) talk to about sex. Eighth grade communication may be especially relevant for adolescents,
students (N=1281) in 24 schools completed survey items about whose increasing autonomy and agency and growing social
their communication about sex. Structural equation modeling networks (Zimmer-Gembeck and Collins 2003; Zimmerman
was used to predict communication profiles, while adjusting and Cleary 2006) can provide new opportunities to engage
for the nesting of students within schools. After controlling for with extended family networks. These networks often include
teens’ age, gender, race/ethnicity, grades, parent/guardian close- relatives such as grandparents, aunts, uncles, as well as god-
ness, and social desirability of survey responses, as well as fam- parents and neighbors (Levitt et al. 2005; Pernice-Duca 2010).
ily status and median family income, results showed that teens of Consistent with prior findings, our use of the term Bextended
early (teen) parents were more likely than teens of later (adult) family^ encompasses the larger family ecology of family
parents to talk with both parents and extended family about sex members and non-family adults who participants identify as
and less likely than later parents to talk only with parents. These key sources of sexual communication. Extended family sexu-
findings indicate that realities of teen sexuality communication ality communication may be particularly prevalent among
for teens of early parents may extend beyond a parent-teen mod- families of teen parents (referred to as Bearly parents^ in this
el to include extended family. Extended family involvement in paper), who often have high levels of extended family in-
educational outreach is a potential untapped resource to support volvement and support for child-rearing (Gordon et al. 2004;
sexual health for teens of early parents. Pinzon and Jones 2012). High rates of teen sex and pregnancy
among adolescents of early parents as compared with teens
whose parents were older when they had children (referred to
Keywords Sexuality communication . Extended family .
as Blater parents^ in this paper) (Johnson and Tyler 2007;
Teen parent . Urban youth . Adolescence . Extended family
Sipsma et al. 2010) also provide impetus to explore the poten-
tial of extended family sexuality communication as a resource
Family sexuality communication can reduce teen sexual risk- to support sexual health for this group. While many systemic
taking behavior (Murry et al. 2014; Trejos-Castillo and issues, such as poverty and lack of educational and employ-
Vazsonyi 2009). Most research on family sexuality communi- ment opportunities, contribute to intergenerational patterns of
cation has focused on parents as the primary partners for teen teen parenthood (Kearney and Levine 2011), understanding
family sexuality communication may serve as a potential pro-
tective resource to support the sexual health of teens with early
J. M. Grossman (*) : A. J. Tracy : A. M. Richer : S. Erkut
parents.
Cheever House/Wellesley Centers for Women, Wellesley College,
106 Central Street, Wellesley, MA 02481, USA Despite the high risk of early sex and pregnancy for teens of
e-mail: [email protected] early parents and high levels of extended family involvement
138 Sex Res Soc Policy (2015) 12:137–144

for teens in this group, little research has explored extended delaying sexual activity and increasing correct and consistent
family sexuality communication for these teens. In one of few use of protection among teens who are sexually active. This
studies of this topic, qualitative research on urban parents found evaluation involved 24 schools in the greater Boston area.
that early parents reported greater extended family sexuality Schools were eligible for study participation if they were lo-
communication with their teens than later parents (Grossman cated within a 25-mile radius of Boston, included sixth-, sev-
et al. 2013). These parents identified extended family as a re- enth-, and eighth grade classrooms, and had between 2 and 10
source to support their teens’ healthy development and rein- 6th grade classrooms. For the 6th grade survey, 19 schools
force messages about delaying sex. These parents may be opted to use passive parental consent, while 5 schools chose
drawing on extended family members as a way to increase their active consent. By 8th grade, all schools opted to use passive
teens’ access to supportive resources for sex education to re- parental consent. Consent forms were sent home to parents in
duce the likelihood of early parenthood in the next generation. multiple languages along with a project description in each
A conceptual model for family sexuality communication year of the study. Approximately 5 % of parents either
has been formulated by Jaccard and colleagues to understand did not return consent forms (in schools choosing active
how parents talk with their teens about sex and relationships consent) or opted out (in schools choosing passive con-
(Jaccard et al. 2002). This model identifies key elements to sent). Less than 1 % of students chose not to take the
understand how family sexuality communication can influ- survey and were taken to another classroom when sur-
ence adolescents’ sexual risk behavior. The theory draws from veys were administered. The survey took about 45 min
five basic elements of how individuals communicate with one to complete. Survey data were collected from fall 2009
another: the source of communication (who talks to the teen), to spring 2012 by evaluation team members who intro-
the message communicated, the channel through which the duced the study and were available to answer questions
message is transmitted, the recipient of communication (i.e., and collect completed surveys. Identifying information
the teen), and the context of communication (social or cultural was separated from students’ survey responses and kept
settings of communication, such as household structure). in a locked cabinet. The research was approved by
While this communication model has been used in prior re- Wellesley College’s Institutional Review Board and by
search on parent-teen sexuality communication (e.g., relevant review bodies for participating schools. See
Guilamo-Ramos et al. 2006), the current study extends explo- Erkut and colleagues for a full description of recruit-
ration of two aspects of this model by investigating who in the ment and data collection procedures (Erkut et al. 2013).
family teens talk to about sex (the source of communication) The analysis for the current study used the 8th grade survey
among teens with early and later parents (context of data from the larger study because extended family sexuality
communication). communication questions were only asked in the final year of
We posit that early vs. later parent status serves as a critical the survey.
context of teens’ backgrounds, which shapes whether they
talk about sex with parents, extended family members, both
parents and extended family members, or no one in their fam- Measures
ily. The goal of this brief report is to assess whether teens of
early parents differ from teens of later parents in their parent Age Age was calculated using date of birth.
and extended family sexuality communication among a di-
verse group of 8th grade students. Based on the importance Race/Ethnicity Students reporting a single race/ethnicity were
of extended family among families of early parents, we hy- categorized in their self-identified group. Those reporting
pothesize that teens of early parents are more likely to report multiple groups were identified as biracial. Following
talking about sex with both parents and extended family and Census definitions that Hispanics can be of any race (Ennis
with extended family only, and would be less likely to report et al. 2010), students who self-identified as Latino with or
talking with parents only than teens of later parents. without another racial identification were coded as Latino.

Family Status Students were asked BWho do you live with?^

Methods Responses included Btwo parents in one place,^ Bone parent,^
Btwo parents in different places,^ Bgrandparents or other fam-
Survey Participants ily members,^ or Bother.^ We used a Btwo-parent vs all others^
dichotomy as a covariate in the analysis.
This study used secondary data from a randomized control
longitudinal evaluation of a sex education intervention. The Median Household Income Family income level was obtain-
long-term goal of the intervention, Get Real: Comprehensive ed for the Census tract associated with student reported home
Sex Education that Works, is to promote sexual health through address.
Sex Res Soc Policy (2015) 12:137–144 139

Grades Students reported their Btypical^ grades ranging from education intervention (which included parent activities, but
BMostly As^ to BMostly Ds and Fs.^ did not include an extended family component) did not affect
study associations, we completed regression analyses with
Parent/Guardian Closeness Students were asked how close and without intervention vs. comparison group status as a
they felt to their parent/guardians (scale 1–5, not close at all to covariate. Intervention group status did not change the model,
very close). so was left out of the final analysis.
Model parameters and sample descriptives were used to
Teen Parent Students were asked to report whether their calculate predicted probabilities of a teen having a particular
mother or father became a parent when they were a teen. sexuality communication profile. We conducted post hoc tests
Students answering Byes^ were categorized as having a teen of these predicted probabilities and their parametric standard
parent. errors, conditioned on having an early parent. Each pair of
predicted probabilities was statistically compared using a
Family Sexuality Communication Students were asked Wald test of equality.
BWhich adults in your family have you talked to about
sex? Check all that apply.^ To include kinship networks
(such as godparent, family friend), we also included a Results
parallel question about whether teens had talked with a
Bfriend of the family^ about sex. Family options included Analyses included all students who completed survey items
mother, father, grandmother, grandfather, aunts, uncles, regarding communication about sex and having an early par-
older sister, older brother, and cousin. These items were ent (N=1281, 79 % of the 1626 students who took the 8th
used to create family sexuality communication profiles grade survey: 162 with early parents and 1119 with later par-
described below. ents). The sample was 52 % female; 35 % Latino, 31 % Black,
17 % White, 15 % Biracial, and 3 % Asian; and had a mean
Social Desirability An abbreviated version of Baxter and col- age of 13.97 years (SD=0.70). Average grades received in 8th
leagues’ (2004) social desirability index was used. Item scores grade were mostly Bs or mostly Bs and Cs. Most teens report-
were summed, with higher scores representing higher social ed feeling close to their parents/guardians and just under half
desirability. (46 %) reported living in a two-parent family. The median
household income was $49,562 (SD =$21,259). Seventy-
Analysis five percent of teens with early parents and 58 % of teens with
later parents reported talking with some extended family about
Logistic regression analyses were conducted in a structural sex.
equation modeling framework using the Mplus statistical Table 1 shows the results of the multiple group logistic
package (Muthén and Muthén 1998-2012) in order to adjust regression models. Examination of covariate effects show that
for the nesting of students within schools, which we treated as girls were more likely than boys to talk to parents only (B=
a design characteristic. This model was a multiple groups (2 0.60, SE=0.17, p<0.01, odds ratio=1.83), extended family
groups: adolescent has/does not have an early parent) multi- only, (B=0.66, SE=0.19, p<0.01, odds ratio=1.94), and both
nomial regression model predicting the relative likelihood of parents and extended family members (B=0.65, SE=0.16,
an adolescent belonging to each family communication pro- p<0.001, odds ratio=1.91) rather than neither parents nor
file: Bparents only,^ Bextended family only,^ and Bboth par- extended family. Latino and Black teens were more likely than
ents and extended family.^ BNeither parents nor family^ was White teens (the comparison group for race/ethnicity) to talk
the fourth group, which served in these models as the compar- with extended family only as opposed to talking with neither
ison category. Analyses controlled for teens’ age, gender, parents nor extended family (Latino: B = 0.96, SE =0.42,
race/ethnicity, grades, parent/guardian closeness, and social p < 0.05, odds ratio = 2.61; Black: B = 1.01, SE = 0.42,
desirability of survey responses, as well as family status and p<0.05, odds ratio=2.75). Latino teens were also more likely
median family income. Teens’ grades were included as a co- than White teens to talk with both parents and extended family
variate because poor academic achievement is associated with (B=0.90, SE=0.34, p<0.01, odds ratio=1.46). Teens living in
early sexual activity (Frisco 2008), which may predict higher areas with a higher median income were more likely to talk
levels of extended family sexuality communication with parents only (B=0.12, SE=0.03, p<0.01, odds ratio=
(Grossman et al. 2014a, b). 1.12) or to both parents and extended family rather than nei-
Due to limitations in the sample size for teens of early ther (B=0.10, SE=0.04, p<0.05, odds ratio=1.11). Teens liv-
parents, race/ethnicity was only included as covariate, rather ing with both parents were less likely to talk with both parents
than exploring interaction effects between race/ethnicity and and extended family than with neither (B=−0.46, SE=0.20,
early parent status. To ensure that participation in the sex p<0.05, odds ratio=0.63). Teens with poorer grades were
140 Sex Res Soc Policy (2015) 12:137–144

Table 1 Profiles of communication about sex; omnibus test and post hoc tests of differences

Parent(s) only vs. neither Extended family vs. neither Both vs. neither

B SE Odds ratio B SE Odds ratio B SE Odds ratio

Threshold (no early parent) −0.20 0.11 −0.21 0.12 0.62 0.09***
Threshold (have early parent) −0.34 0.27 0.25 0.24 1.22 0.22***
Age (8th grade) 0.04 0.12 1.04 0.24 0.13 1.27 0.20 0.11 1.22
Female 0.60 0.17** 1.83 0.66 0.19** 1.94 0.65 0.16*** 1.91
Biracial 0.21 0.36 1.24 0.65 0.42 1.92 0.66 0.34 1.94
Latino 0.61 0.41 1.84 0.96 0.42* 2.61 0.90 0.34** 2.46
Black 0.23 0.45 1.26 1.01 0.42* 2.75 0.36 0.34 1.43
Two-parent family −0.16 0.23 0.85 −0.18 0.20 0.83 −0.46 0.20* 0.63
Median income 0.12 0.03** 1.12 0.03 0.06 1.03 0.10 0.04* 1.11
Social desirability −0.10 0.15 0.91 −0.31 0.24 0.73 −0.72 0.16*** 0.49
Poor grades −0.08 0.06 0.93 0.14 0.04*** 1.15 0.00 0.05 1.00
Parent/guardian Closeness 0.65 0.10*** 1.91 −0.10 0.06 0.91 0.31 0.08*** 1.37

The multiple thresholds reported above reflect the freely estimated regression equations for the Bno early parent^ and Bhave early parent^ groups predicting
the multinomial dependent variable representing communication group (parents only, extended family only, and both, with neither as the comparison
category). Threshold values are scaled such that lower thresholds translate to higher probability values
*p<0.05; **p<0.01; ***p<0.001

more likely to talk with extended family members only than odds ratio=0.91), although this reached the level of a trend
with neither (B=0.14, SE=0.04, p<0.001, odds ratio=1.15). (p<0.10). Given the importance of teen-parent quality of re-
Teens who tended to give more socially desirable responses lationships to teen sexual risk and prevention (Perrino et al.
were less likely to report talking with both parents and extend- 2000) and the inconsistent findings for this variable in our
ed family than neither parents nor extended family (B=−0.72, analyses, we ran post hoc analyses to further explore these
SE=0.16, p<0.001, odds ratio=0.49). associations. Post hoc tests revealed that the effects of parent
Teens who feel closer to their parent(s) were more likely to closeness differ significantly across communication profiles
talk to parents only about sex (B=0.65, SE=0.10, p<0.001, (0.65 vs. −0.10: Wald=64.02, df=1, p<0.001; 0.65 vs. 0.31:
odds ratio=1.91) or to both extended family and parents (B= Wald=14.15, df=1, p<0.001; −0.10 vs. 0.31; Wald=40.01,
0.31, SE=0.08, p<0.001, odds ratio=1.37) than to neither. df=1, p<0.001).
Teens who feel closer to their parent(s) were just as likely to Figure 1 shows that the omnibus test for differences in the
talk to an extended family member only as compared to nei- predicted probabilities of belonging to the communication
ther parents nor extended family (B=−0.10, SE=0.06, ns, profiles across teens with early parents and those with later

100%
Later Parent (n=1119)
90%
Predicted Probabilities

Early Parent (n=162)

80%
70%
60% 53%
50%
41%
40%
30% 22%
18% 18% 20% 16%
20%
11%
10%
0%
Parent(s) only Extended family Both parent(s) and Neither parent(s)
only extended family nor extended
family
Fig. 1 Whom teens of early and later parents talk to about sex. These across each pair. Omnibus: Wald=16.81, df=3, p<0.001; parents only:
probabilities were calculated using the MODEL CONSTRAINT Wald=5.96, df=1, p<0.05; extended family only: Wald=0.30, df=1, ns;
command in Mplus to specify the probability equations and the both extended family and parent(s): Wald=11.41, df=1, p<0.001; neither
MODEL TEST command to statistically compare the resulting values extended family nor parent(s): Wald=4.76, df, p<0.05
Sex Res Soc Policy (2015) 12:137–144 141

parents was significant (Wald=16.81, df=3, p<0.001). Post prior research. Girls showed higher levels of all types of sex-
hoc tests of differences within each communication profile uality communication as compared to boys, which fits with
showed that adolescents of early parents were significantly research findings that girls talk with their parents about sex
less likely than teens of later parents to talk with parents only more than boys (Omar et al. 2003; Swain et al. 2006). The
(Wald=5.96, df=1, p<0.05) and to talk with neither parents current findings suggest that this may hold true for extended
nor extended family (Wald=4.76, df=1, p<0.05) but were family as well. Findings that Black and Latino teens showed
significantly more likely to talk with both parents and extend- more extended family sexuality communication than White
ed family (Wald=11.41, df=1, p<0.001). Teens with early teens are consistent with research identifying the importance
parents were equally as likely to talk with extended family of extended family relationships in Black and Latino families
only as teens with later parents (Wald=0.30, df=1, ns). (Jones et al. 2007; Stanton-Salazar 2001). There is little prior
research regarding associations of other demographic vari-
ables, such as family structure, neighborhood income, and
Discussion students’ grades with extended family sexuality communica-
tion. The finding that teens who tended to give more socially
While urban teens show high rates of extended family com- desirable responses were less likely to report talking with par-
munication overall (Grossman et al. 2014a, b), extended fam- ents and extended family than with neither suggests that teens
ily communication may be particularly prevalent for teens of who put a high value on how they are perceived by others may
early parents. This study’s findings suggest that a dyadic mod- feel less comfortable discussing sexual issues with family
el for sexuality communication, that is restricted to parents members.
and teens, may not reflect the realities of many families. Findings that greater closeness with parents predicts higher
Across this urban sample, a combination of parent and extend- likelihood of talking with parents only or both parents and
ed family sexuality communication was the most common extended family, but not extended family only suggest that
profile. Consistent with Jaccard and his colleagues’ theoretical teens who are close with their parents may be more comfort-
model for sexuality communication (Jaccard et al. 2002), the able talking with parents about sex. However, one might hope
findings support the importance of examining the life contexts that teens who feel less close to their parents would view
of teens (in this case, having an early parent) and teens’ extended family as an alternative resource for talking about
sources of family sexuality communication (parents and ex- sex. This would fit with research showing that extended fam-
tended family). Further, profile analyses suggest that commu- ily members can fill in gaps as primary caregivers when a
nication differences are more nuanced than whether teens talk parent is unable to take on traditional caretaking roles
to extended family and need to be understood within a larger (Bertera and Crewe 2013; Harris 2013). This was not the case
context of parent and extended family sexuality communica- here, although associations for less parent closeness and more
tion. Specifically, study findings showed support for hypoth- extended family only sexuality communication reached a
eses that teens of early parents would be more likely to talk trend level. Post hoc analyses showed that parent closeness
about sex with parents and extended family and less likely to operates differently for each communication profile, with the
talk with parents only than teens of later parents. These find- strongest association with parent only sexuality communica-
ings suggest that family sexuality communication for teens of tion, followed by both parents and extended family, and the
early parents is likely to be a group effort, with multiple family weakest association with extended family only communica-
members (including parents) involved. This may in part reflect tion. It may be that a measure of closeness with extended
early parents’ reliance on extended family support for family is needed to determine whether extended family can
childrearing in general, which has been well documented provide an alternative to parents in this domain.
(Smith 2000). The results also fit with qualitative findings
showing that early parents see extended family as a resource Limitations and Future Directions
to support their teens’ healthy development and actively
engage family members in sexuality communication with This study is limited by the use of secondary analysis, which
their teens (Grossman et al. 2013). However, the hypothesis entailed that the major constructs of interest needed to be
that teens of early parents would be more likely to talk to measured by variables that were predetermined and only pres-
extended family only was not supported. It may be that the ent in the final wave of survey data. Specifically, the data
unique role of extended family for teens of early parents is lacked information on when, why, and which sexuality-
primarily a collaborative one, rather than one in which extend- related topics teens discuss with extended family. The small
ed family members take on primary responsibility for sexuality sample size of teens with early parents for this study limits the
communication. potential to explore interaction effects of race/ethnicity with
Covariate associations of gender and racial/ethnic group early parent status and associations between family commu-
with sexuality communication profiles were consistent with nication profiles and teen sexual behavior. In addition, due to a
142 Sex Res Soc Policy (2015) 12:137–144

lack of gender information about extended family members sex may help teens identify potential partners for
(e.g., whether a cousin or sibling is male or female), data was assigned family activities (Grossman et al. 2014a, b).
insufficient to explore interactions between the gender of the Including extended family members as recognized re-
teen and the gender of the family member with communica- sources for family assignments could also expand stu-
tion profiles. Findings from the current study and prior re- dents’ participation in family activities, as teens may be
search indicate the importance of gender in family sexuality reluctant to talk with parents about sex, especially when
communication (e.g., Teitelman et al. 2009; Sneed 2008) and they become sexually active (Crohn 2010). Pediatricians
suggest this as an area for future study. also serve as resources for adolescents’ sex education
To more fully understand the health implications of and may be able to share guidance for family sexuality
extended family sexuality communication, particularly communication with teens and their family members.
for teens of early parents, future work should include Given that they identify talking about sex with their
longitudinal assessment of urban teens’ parent and ex- patients as a high priority need and an area of minimal
tended family sexuality communication, early or later expertise (Miller et al. 2008), they may be open to
parent status, and teen sexual behavior. Prior findings opportunities to learn how to support family sexuality
that parent-teen sexuality communication can protect communication. Trusted adults outside of the family,
teens from risky sexual behavior (Murry et al. 2014; such as teachers and sexuality educators, have also been
Usher-Seriki et al. 2008) suggest that talking with other identified as resources for teen sexuality communication
family members about sex may have a protective influ- (Namisi et al. 2009; Teitelman et al. 2009) and may be
ence. This may be particularly true for teens of early valuable sources of sexuality communication.
parents, where a broader family ecology, rather than While state policies vary in their parameters for sex
solely parental relationships, may provide a base of education, many states identify parental involvement as
family support. Future study of multiple family roles a key aspect of effective sex education programs
in sexuality communication also raises complexities re- (National Conference of State Legislatures 2014). This
garding whether messages about sex differ across family study’s findings suggest that policies that support paren-
members, as their protective impact may rest on the tal involvement can benefit from an expanded under-
health-promoting and consistent nature of the communi- standing of Bparental^ to include extended family in
cation. Longitudinal work that extends over teens’ tran- sex education. This change in policy could provide op-
sition to sex is also needed given findings that some portunities for educators and health providers to en-
adolescents seek extended family as a more comfortable hance health-promoting extended family sexuality com-
alternative to parents as they begin to explore sex and munication and to leverage existing communication in
relationships (Crohn 2010; Teitelman et al. 2009). While support of teens’ sexual health. This may be especially
extended family played a role in sexuality communica- true for offspring of early parents, who have high vul-
tion for many early adolescents in this study, it may be nerability for sexual risk behavior (Johnson and Tyler
even more prevalent during middle and late adolescence 2007; Sipsma et al. 2010), but also have a potential
as a greater number of teens become sexually active. resource given their unique engagement with extended
family sexuality communication.
Implications for Policy and Practice

Sex education programs often focus on parents as the

primary sex educators of their children (Kirby and Conclusions
Miller 2002). While parents play a critical role in
talking with their teens about sex (Albert 2010), the This study suggests that family sexuality communication
findings from this study suggest the need for sex edu- may differ for teens of early and later parents, with
cation programs to target a broader family system, es- policy implications for teen sexual health and preven-
pecially for programs that include children of early par- tion. Parents serve a critical role in talking with teens
ents. Sex education programs that include family com- about sex (Albert 2010) and existing sex education pro-
ponents have shown effectiveness in delaying teen sex grams primarily emphasize parent-teen communication
(Grossman, Tracy, Charmaraman, Ceder, & Erkut, 2014; (Kirby and Miller 2002). However, this study’s findings
Tortolero et al. 2010), but we know of no evaluations of suggest the importance of broadening the focus to in-
programs that include extended family involvement. clude extended family, particularly for teens of early
While programs have limited resources to reach out to parents. For them especially, greater involvement of ex-
extended family members, asking teens to identify tended family in educational outreach policy and prac-
which family members they trust to talk with about tice may be a benefit in preventing teen risky behavior.
Sex Res Soc Policy (2015) 12:137–144 143

Acknowledgments This manuscript was supported by the National Gordon, R. A., Chase-Lansdale, P. L., & Brooks-Gunn, J. (2004).
Institute of Child Health and Human Development of the National Insti- Extended households and the life course of young mothers: under-
tutes of Health under grant number R03 HD073381-01. The content is standing the associations using a sample of mothers with premature,
solely the responsibility of the authors and does not necessarily represent low birth weight babies. Child Development, 75(4), 1013–1038. doi:
the official views of the National Institutes of Health. Everyone who 10.1111/j.1467-8624.2004.00723.x.
contributed significantly to this work is listed here. We thank Ineke Ceder Grossman J. M., Charmaraman, L., & Erkut, S. (2013). Do as I say, not as
for her edits and support. I did: how parents talk with early adolescents about sex. Journal of
Family Issues.
Conflict of Interest The authors declare that they have no conflict of Grossman, J. M., Tracy, A. J., Charmaraman, L., Ceder, I., & Erkut, S.
interest. (2014a). Protective effects of middle school comprehensive sex ed-
ucation with family involvement. Journal of School Health, 84,
Funding This manuscript was supported by the National Institute of 739–747.
Child Health and Human Development of the National Institutes of Grossman, J. M., Tracy, A. J., Richer, A., & Erkut, S. (2014b). The role of
Health under grant number R03 HD073381-01. The content is solely extended family in teen sexual health. Journal of Adolescent
the responsibility of the authors and does not necessarily represent the Research.
official views of the National Institutes of Health. Guilamo-Ramos, V., Jaccard, J., Dittus, P., & Bouris, A. M. (2006).
Parental expertise, trustworthiness, and accessibility: parent-
Ethical Approval All procedures performed in studies involving hu- adolescent communication and adolescent risk behavior. Journal
man participants were in accordance with the ethical standards of the of Marriage and Family, 68(5), 1229–1246.
institutional and/or national research committee and with the 1964 Harris, D. M. (2013). Grandmas’ hands rocked the cradle. Children and
Helsinki declaration and its later amendments or comparable ethical Youth Services Review, 35(12), 2072–2079. doi:10.1016/j.
standards. childyouth.2013.09.022.
Harper, G. W., Timmons, A., Motley, D. N., Tyler, D. D., Catania, J. A.,
Compliance with Ethical Standards The authors declare that they Boyer, C. B., & Dolcini, M. M. (2012). BIt takes a village:^ familial
have no conflict of interest. All procedures performed in studies involving messages regarding dating among African American adolescents.
human participants were in accordance with the ethical standards of the Research in Human Development, 9, 29–53.
institutional and/or national research committee and with the 1964 Jaccard, J., Dodge, T., & Dittus, P. (2002). Parent-adolescent communi-
Helsinki declaration and its later amendments or comparable ethical stan- cation about sex and birth control: a conceptual framework. New
dards. Informed consent was obtained from all individual participants Directions for Child and Adolescent Development, 97, 9–42.
included in the study. Johnson, K. A., & Tyler, K. A. (2007). Adolescent sexual onset: an
intergenerational analysis. Journal of Youth and Adolescence,
Informed Consent Informed consent was obtained from all individual 36(7), 939–949. doi:10.1007/s10964-006-9165-z.
participants included in the study. Jones, D. J., Zalot, A. A., Foster, S. E., Sterrett, E., & Chester, C. (2007).
A review of childrearing in African American single mother fami-
lies: the relevance of a coparenting framework. Journal of Child and
Family Studies, 16, 671–683.
References Kearney, M. S., & Levine, P. B. (2011). Income inequality and early non-
marital childbearing: An economic exploration of the Bculture of
despair.^ (NBER Working Paper 17157). Retrieved from the
Albert, B. (2010). With One Voice 2010: America’s adults and teens National Bureau of Economic Research website http://www.nber.
sound off about teen pregnancy. Washington: The National org/papers/w17157.
Campaign to Prevent Teen and Unplanned Pregnancy. Kirby, D., & Miller, B. C. (2002). Interventions designed to promote
Baxter, S. D., Smith, A. F., Litaker, M. S., Baglio, M. L., Guinn, C. H., & parent-teen communication about sexuality. In S. Feldman & D.
Shaffer, N. M. (2004). Children’s social desirability and dietary re- Rosenthal (Eds.), Talking sexuality: parent-adolescent
ports. Journal of Nutrition, Education, and Behavior, 36, 84–89. communication (pp. 93–100). San Francisco, CA: Jossey-Bass.
Bertera, E. M., & Crewe, S. E. (2013). Parenthood in the twenty-first Levitt, M. J., Levitt, J., Bustos, G. L., Crooks, N. A., Santos, J. D., Telan,
century: African American grandparents as surrogate parents. P., Hodgetts-Barber, J., & Milevsky, A. (2005). Patterns of social
Journal of Human Behavior in the Social Environment, 23(2), support in the middle childhood to early adolescent transition: im-
178–192. doi:10.1080/10911359.2013.747348. plications for adjustment. Social Development, 14, 398–420.
Crohn, H. M. (2010). Communication about sexuality with mothers and Miller, K. S., Wyckoff, S. C., Lin, C. Y., Whitaker, D. J., Sukalac, T., &
stepmothers from the perspective of young adult daughters. Journal Fowler, M. G. (2008). Pediatricians’ role and practices regarding
of Divorce & Remarriage, 51(6), 348–365. provision of guidance about sexual risk reduction to parents. The
DiIorio, C., Pluhar, E., & Belcher, L. (2003). Parent-child communication Journal of Primary Prevention, 29(3), 279–291. doi:10.1007/
about sexuality: a review of the literature from 1980–2002. Journal s10935-008-0137-9.
of HIV/AIDS Prevention and Education for Adolescents and Murry, V. M., McNair, L. D., Myers, S. S., Chen, Y.-f., & Brody, G. H.
Children, 5, 7–32. (2014). Intervention induced changes in perceptions of parenting
Ennis, S. R., Rios-Vargas, M., & Albert, N. G. (2010). The Hispanic and risk opportunities among rural African American. Journal of
population, 2010. U.S. Census. Retrieved from http://www.census. Child and Family Studies, 23(2), 422–436. doi:10.1007/s10826-
gov/prod/cen2010/briefs/c2010br-04.pdf 013-9714-5.
Erkut, S., Grossman, J. M., Frye, A., Ceder, I., Charmaraman, L., & Muthén, L. K., & Muthén, B. O. (1998). Mplus User’s Guide. Seventh
Tracy, A. (2013). Can sex education delay early sexual debut? Edition. Los Angeles, CA: Muthén & Muthén.
Journal of Early Adolescence, 33, 479–494. Namisi, F. S., Flisher, A. J., Overland, S., Bastien, S., Onya, H., Kaaya,
Frisco, M. L. (2008). Adolescents’ sexual behavior and academic attain- S., & Aarø, L. E. (2009). Sociodemographic variations in commu-
ment. Sociology of Education, 81(3), 284–311. doi:10.1177/ nication on sexuality and HIV/AIDS with parents, family members
003804070808100304. and teachers among in-school adolescents: a multi-site study in
144 Sex Res Soc Policy (2015) 12:137–144

Tanzania and South Africa. Scandinavian Journal of Public Health, Swain, C. R., Ackerman, L. K., & Ackerman, M. A. (2006). The influ-
37(suppl 2), 65–74. ence of individual characteristics and contraceptive beliefs on
National Conference of State Legislatures. (2014). State policies on sex parent-teen sexual communications: a structural model. Journal of
education in schools. Retrieved from http://www.ncsl.org/research/ Adolescent Health, 38(6), 753.e759–753.e718. doi:10.1016/j.
health/state-policies-onsex-education-in-schools.aspx. jadohealth.2005.08.015.
Omar, H., McElderry, D., & Zakharia, R. (2003). Educating adolescents Teitelman, A. M., Bohinski, J. M., & Boente, A. (2009). The social
about puberty: what are we missing? International Journal of context of sexual health and sexual risk for urban adolescent girls
Adolescent Medicine and Health, 15(1), 79–83. doi:10.1515/ in the United States. Issues in Mental Health Nursing, 30(7), 460–
ijamh.2003.15.1.79. 469. doi:10.1080/01612840802641735.
Pernice-Duca, F. M. (2010). An examination of family and social support Tortolero, S. R., Markham, C. M., Peskin, M. F., Shegog, R., Addy, R. C.,
networks as a function of ethnicity and gender: a descriptive study of Escobar-Chaves, S. L., & Baumler, E. R. (2010). It’s your game:
youths from three ethnic reference groups. Journal of Youth Studies, Keep it real: delaying sexual behavior with an effective middle
13(3), 391–402. school program. Journal of Adolescent Health, 46(2), 169–179.
Perrino, T., González-Soldevilla, A., Pantin, H., & Szapocznik, J. (2000). doi:10.1016/j.jadohealth.2009.06.008.
The role of families in adolescent HIV prevention: a review. Clinical Trejos-Castillo, E., & Vazsonyi, A. T. (2009). Risky sexual behaviors in
Child and Family Psychology Review, 3(2), 81–96. first and second generation Hispanic immigrant youth. Journal of
Pinzon, J. L., & Jones, V. F. (2012). Care of adolescent parents and their Youth and Adolescence, 38(5), 719–731. doi:10.1007/s10964-008-
children. Pediatrics, 130(6), e1743–e1756. doi:10.1542/peds. 2012- 9369-5.
2879. Usher-Seriki, K. K., Bynum, M. S., & Callands, T. A. (2008). Mother-
Sipsma, H., Biello, K. B., Cole-Lewis, H., & Kershaw, T. (2010). Like daughter communication about sex and sexual intercourse among
father, like son: the intergenerational cycle of adolescent fatherhood. middle- to upper-class African American girls. Journal of Family
American Journal of Public Health, 100(3), 517–524. Issues, 29(7), 901–917. doi:10.1177/0192513x07311951.
Smith, K. (2000). Who’s minding the kids? Child care arrangements: Fall
Zimmer-Gembeck, M. J., & Collins, W. A. (2003). Autonomy develop-
1995. (Current Population Reports P70-70.). Washington: U.S.
ment during adolescence. In G. R. Adams & M. D. Berzonsky
Census Bureau.
(Eds.), Blackwell handbook of adolescence (pp. 175–204). Oxford,
Sneed, C. D. (2008). Parent-adolescent communication about sex: the
England: Blackwell Publishing.
impact of content and comfort on adolescent sexual behavior.
Journal of HIV/AIDS Prevention in Children & Youth, 9(1), 70– Zimmerman, B. J., & Cleary, T. J. (2006). Adolescents’ development of
83. doi:10.1080/10698370802126477. personal agency: the role of self-efficacy beliefs and self-regulatory
Stanton-Salazar, R. D. (2001). Manufacturing hope and despair: the skill. In F. Pajares & T. Urdan (Eds.), Self-efficacy beliefs of
school and kin support networks of U.S.-Mexican youth. New adolescents (pp. 45–70). Greenwich, CT: Information Age
York, NY: Teachers College Press. Publishing.
Reproduced with permission of the copyright owner. Further reproduction prohibited without
permission.