Important Non-parental Adults and Positive Youth Development Across Mid- to Late-Adolescence: The Moderating Effect of Parenting Profiles

Abstract

Both parents and important non-parental adults have influential roles in promoting positive youth development (PYD). Little research, however, has examined the simultaneous effects of both parents and important non-parental adults for PYD. We assessed the relationships among youth-reported parenting profiles and important non-parental adult relationships in predicting the Five Cs of PYD (competence, confidence, connection, character, and caring) in four cross-sectional waves of data from the 4-H Study of PYD (Grade 9: N = 975, 61.1 % female; Grade 10: N = 1,855, 63.4 % female; Grade 11: N = 983, 67.9 % female; Grade 12: N = 703, 69.3 % female). The results indicated the existence of latent profiles of youth-reported parenting styles based on maternal warmth, parental school involvement, and parental monitoring that were consistent with previously identified profiles (authoritative, authoritarian, permissive, and uninvolved) as well as reflecting several novel profiles (highly involved, integrative, school-focused, controlling). Parenting profile membership predicted mean differences in the Five Cs at each wave, and also moderated the relationships between the presence of an important non-parental adult and the Five Cs. In general, authoritative and highly involved parenting predicted higher levels of PYD and a higher likelihood of being connected to an important non-parental adult. We discuss the implications of these findings for future research on adult influences of youth development and for programs that involve adults in attempts to promote PYD.

Full text

EMPIRICAL RESEARCH
Important Non-parental Adults and Positive Youth Development
Across Mid- to Late-Adolescence: The Moderating Effect
of Parenting Profiles
Edmond P. Bowers •Sara K. Johnson •Mary H. Buckingham •Santiago Gasca •
Daniel J. A. Warren •Jacqueline V. Lerner •Richard M. Lerner
Received: 12 September 2013 / Accepted: 15 January 2014 / Published online: 21 February 2014
ÓSpringer Science+Business Media New York 2014
Abstract Both parents and important non-parental adults
have influential roles in promoting positive youth devel-
opment (PYD). Little research, however, has examined the
simultaneous effects of both parents and important non-
parental adults for PYD. We assessed the relationships
among youth-reported parenting profiles and important
non-parental adult relationships in predicting the Five Cs of
PYD (competence, confidence, connection, character, and
caring) in four cross-sectional waves of data from the 4-H
Study of PYD (Grade 9: N =975, 61.1 % female; Grade
10: N =1,855, 63.4 % female; Grade 11: N =983,
67.9 % female; Grade 12: N =703, 69.3 % female). The
results indicated the existence of latent profiles of youth-
reported parenting styles based on maternal warmth,
parental school involvement, and parental monitoring that
were consistent with previously identified profiles
(authoritative, authoritarian, permissive, and uninvolved)
as well as reflecting several novel profiles (highly involved,
integrative, school-focused, controlling). Parenting profile
membership predicted mean differences in the Five Cs at
each wave, and also moderated the relationships between
the presence of an important non-parental adult and the
Five Cs. In general, authoritative and highly involved
parenting predicted higher levels of PYD and a higher
likelihood of being connected to an important non-parental
adult. We discuss the implications of these findings for
future research on adult influences of youth development
and for programs that involve adults in attempts to promote
PYD.
Keywords Parents Non-parental adults Positive youth
development Developmental systems Latent profile
analysis
Introduction
The contemporary perspectives about adolescence devel-
opment are framed by relational developmental systems
models (Lerner et al. 2013, in press; Overton 2013), which
posit that development is defined by dynamic and mutually
influential, bidirectional person/?context relationships.
Derived from relational developmental systems models, the
Five Cs model of positive youth development (PYD)
hypothesizes that when the strengths of youth are aligned
with resources in their contexts (‘‘developmental assets;’’
Benson et al. 2011), youth thriving (as defined by the Five
Cs of competence, confidence, connection, character, and
caring) is promoted. The developmental assets included in
this model have been associated with several contexts
(families, schools, out-of-school-time activities), but recent
evidence suggests that relationships with committed, caring
adults in adolescents’ lives within these contexts are the
most important assets for predicting higher levels of PYD
and lower levels of risk behaviors (Li and Julian 2012;
Theokas and Lerner 2006).
Whereas parents and parent–youth relationships are
important developmental assets in the positive develop-
ment of youth (e.g., Flouri and Buchanan 2004; Laursen
and Collins 2009; Lewin-Bizan et al. 2010), support from
important non-parental adults, such as extended family,
teachers, mentors, coaches, and neighbors also enhances
E. P. Bowers S. K. Johnson M. H. Buckingham S. Gasca 
D. J. A. Warren R. M. Lerner (&)
Institute for Applied Research in Youth Development, Tufts
University, Medford, MA, USA
e-mail: iaryd.pubs@gmail.com
J. V. Lerner
Boston College, Chestnut Hill, MA, USA
123
J Youth Adolescence (2014) 43:897–918
DOI 10.1007/s10964-014-0095-x

PYD levels (Bowers et al. 2012; Erickson et al. 2009;
Kogan and Brody 2010; Rhodes and Lowe 2009). How-
ever, most studies examining the influence of the adult
milieu on youth development have focused on the impact
of parents and parent–youth relationships and have not
considered the potential effects of important non-parental
adults.
This emphasis is not unexpected, as parents represent
the developmental asset that accounts for the most variance
in PYD-related outcomes (Bowers et al. 2011b; Theokas
and Lerner 2006). Moreover, the common indicators of
parent–adolescent relationship quality, such as perceived
maternal warmth, parental monitoring, and parental school
involvement are related to many youth outcomes
(e.g., Baumrind 1991; Bowers et al. 2011a; Dishion and
McMahon 1998; Flouri and Buchanan 2002; Smetana et al.
2006). However, relational developmental systems-derived
models, such as the Five Cs model of PYD (Lerner et al.
2013, in press), emphasize that development is embodied
within an ecological system involving cross-context rela-
tionships. Therefore, it is important to consider the non-
parental factors that can influence youth development.
Important non-parental adults are often the instructors,
advocates, and role models in contexts salient to youth,
such as sports, hobbies, and other out-of-school-time
activities; therefore, youth relationships with such adults
may serve as ‘‘developmental assets’’ that promote PYD
outcomes (Benson et al. 2006; Bowers et al. 2012). Indeed,
the presence and characteristics of youth relationships with
important non-parental adults have been linked to a range
of psychological, socioemotional, and behavioral outcomes
across adolescence (e.g., Bowers et al. 2012; Dubois and
Silverthorn 2005a,b; Greenberger et al. 1998; Zimmerman
et al. 2002). The influence of these important non-parental
adults may be especially salient in adolescence, as
youth build identities outside the home (Co
ˆte
´2009; Marcia
1980).
Little research, however, has appraised the simultaneous
effects of resources within both familial and non-familial
relationships for indices of healthy and positive develop-
ment in adolescence. Accordingly, this study examined
individual thriving (as indexed by the Five Cs of PYD)
across mid- to late-adolescence, as it may be moderated by
youth relationships with parents (as indexed by school
involvement, warmth, and monitoring) and important non-
parental adults.
Youth Relationships with Parents and PYD
Across more than a half century, the history of theory and
research about parenting parallels the classic debate in the
study of personality about traits versus types (Nesselroade
1988) in many ways and, as well, the more recent interest
in variable-centered versus person-centered approaches to
data analysis (Bergmann and Magnusson 1997). For
instance, reflecting a trait-type and variable centered
approach, Sears et al. (1957) discussed the social learning-
based behaviors that created variable clusters, or patterns,
of child rearing. In turn, reflecting a typological and per-
son-centered approach, Baumrind (1978) described three
types of parenting: authoritative, authoritarian, and per-
missive. Steinberg et al. (1991) expanded this typology by
differentiating between permissiveness that reflected
indulgent parenting versus permissiveness that reflected
negligent parenting.
Today, however, using the lens of relational develop-
mental systems conceptions (e.g., Overton 2013;Overton
and Mu
¨ller 2013), both the trait-like and the type-like
approaches to parenting may be argued to fall short of being
maximally useful. The extant trait-like concepts fail to
integrate the behavioral, cognitive, and emotional compo-
nents of the engagement that parents have with their children
and, as well, fail to appreciate the plasticity across ontogeny
of these attributes of parenting (Lerner and Benson 2013). In
turn, typologies also fall short in regard to the plasticity of
human behavior (cognition or emotion). Scholars using
either the Baumrind (1978) or the Steinberg et al. (1991)
formulations pay little heed to the idea that the sets of
characteristics associated with any typology are unlikely to
characterize parents’ functioning across time (e.g., consider
the demands on a parent by unruly, disobedient behaviors
from a 2-year old vs from a 14-year old) and across place
(consider a child’s loud and disruptive behaviors in the
family living room vs a service in a place of worship).
Therefore, research derived from relational develop-
mental models should include measures of parenting that
index youth–parent relationships across several key con-
texts such as family, peers, and school and at different time
points. Indeed, the measures of youth–parent relationship
quality in each of these contexts, such as perceived parental
warmth (behaviors indicating a parent’s acceptance, nur-
turance, and support), parental monitoring or knowledge
(the extent to which parents keep track of their child’s
behavior and acquaintances), and parental school involve-
ment (the extent to which parents take an active role in
their child’s education) are related to many youth outcomes
(e.g., Baumrind 1991; Bowers et al. 2011a; Dishion and
McMahon 1998; Flouri and Buchanan 2004; Lewin-Bizan
et al. 2010), and their influence may differ based on youth
age (Eccles and Harold 1996; McElhaney et al. 2009) and
socioeconomic and cultural background (e.g., Brody and
Flor 1998). However, few studies have considered these
measures jointly in indexing parenting or in predicting
youth outcomes across several years.
The majority of empirical work that has linked youth–
parent relationship quality to adolescent outcomes has
898 J Youth Adolescence (2014) 43:897–918
123

focused on parenting style (typologies based on parental
warmth and control or analogous dimensions of parenting)
as a key predictor of positive adolescent adjustment
(Baumrind 1978). The studies of parenting style consis-
tently report that ‘‘authoritative parenting,’’ which is
marked by parental rule induction, firmness, and support, is
related to indices of PYD (Baumrind 1978,1991). The
combination of parental responsiveness and demanding-
ness is consistently related to adolescent adjustment, school
performance, and psychosocial maturity (Steinberg 2001).
In addition, disaggregating the dimensions of parenting
style, Gray and Steinberg (1999) also found that warmth
predicted higher academic competence, lower levels of
behavior problems, and better psychosocial development.
However, as noted, youth socioeconomic status (SES) or
culture and, as well, ethnic/racial background, may mod-
erate the effect of parenting on youth outcomes. For youth
of color or youth from low SES backgrounds, authoritarian
parenting may not be as harmful, and may actually be
beneficial (Brody and Flor 1998; Furstenberg et al. 1999;
Hill et al. 2003). For example, traditional Mexican–
American mothers living from disadvantaged areas who
employed warm, yet strict and sometimes harsh control,
had children with fewer conduct problems (Hill et al.
2003). Similarly, when African-American families live in
poverty stricken, high crime neighborhoods with little
community support, strict control has been linked to less
antisocial behavior, better self-regulation, and less chance
of victimization (Brody and Flor 1998). Therefore,
research on parenting and youth outcomes should consider
the potential moderating effect of youth and family
demographics.
Parental monitoring or knowledge is also regarded as a
crucial intra-familial process affecting adolescent behavior
across multiple domains (Crouter et al. 1990; Kerr and
Stattin 2000; Kerr et al. 2010; Maccoby and Martin 1983).
Research has linked the presence of parental monitoring
with the absence of adolescent antisocial and delinquent
behaviors (Crouter and Head 2002; Kerr and Stattin, 2000).
However, some studies have found that extreme vigilance
in tracking and surveillance may be linked to various forms
of poor adjustment (Peterson et al. 1988; Rodin 1990;
Syme 1990). This research suggests that high levels of
parental monitoring may impede adolescents’ personal
control, sense of self-efficacy, and independence. In addi-
tion, the relationship between parental monitoring and
youth outcomes may be moderated by youth age, as
younger adolescents are more likely to disclose more
information to their parents (and are expected to disclose
more) than older adolescents (Finkenauer et al. 2002). As
youth develop autonomy over adolescence, parents are less
vigilant in monitoring their activities (McElhaney et al.
2009). Nevertheless, studies that have examined specific
monitoring behaviors have generally supported the notion
that appropriate rules and supervision are related to lower
problem behaviors (Hayes et al. 2004).
Finally, it has been suggested that the influence of
parental school involvement wanes as adolescence com-
mences, and parents become less involved as school sub-
jects become more difficult and youth seek autonomy
(Eccles and Harold 1996). However, across adolescence
increased parental school involvement is related to
increased intentional self regulation (Bowers et al. 2011a),
future aspirations (Hill et al. 2004), academic competence
(Li et al. 2010), and student achievement (Hill et al. 2004;
Hill and Tyson 2009). Therefore, more research across
several years of adolescence is needed to clarify these
findings.
Research on parental warmth, monitoring, and
involvement is limited when viewed through the perspec-
tive of contemporary relational developmental systems
models of adolescent development. First, as noted above,
few studies include indices of parenting across several
contexts: family (warmth), peers (monitoring), and school
(school involvement) and across several years of youth
development. Second, relational development systems
models stress dynamic cross-contextual influences on
human development. A majority of the studies on youth–
parent relationships examine adolescent development
within the immediate family sphere, with few assessing
resources provided by extra-familial relationships. Recent
work, however, has indicated that there is a substantial
influence of non-parental adults on adolescent development
(e.g., Dubois and Silverthorn 2005a,b). Therefore, the
present study used latent profile analysis to explore patterns
of parenting characteristics (regarding warmth, monitoring,
and school involvement), and to examine how these par-
enting profiles may moderate the relationships among
important non-parental adult relationships and PYD in
4 years of adolescence.
Youth Relationships with Important Non-parental
Adults and PYD
Relationships with important non-parental adults have also
been shown to help enhance positive development among
youth (Bowers et al. 2012; Dubois and Silverthorn 2005a,
b; Greenberger et al. 1998; Haddad et al. 2011; Zimmer-
man et al. 2002). Important non-parental adult relationships
can occur in a variety of forms, from relationships with
teachers, coaches and community members, to relation-
ships with older siblings, aunts, uncles, and other family
members. These adults have been labeled in the literature
as natural mentors (e.g., Dubois and Silverthorn 2005a,b;
Hurd et al. 2013), informal mentors (Kogan and Brody
2010), or as very important nonparental adults (VIPs; e.g.,
J Youth Adolescence (2014) 43:897–918 899
123

Greenberger et al. 1998). Whatever their label, youth
reports of important non-parental adults in their lives have
been linked to adolescent outcomes, such as educational
accomplishments, lower risk and problem behaviors, and
lower depressive symptoms (Dubois and Silverthorn
2005a; Greenberger et al. 1998; Zimmerman et al. 2002).
In addition, specific characteristics of these relationships,
including warmth, acceptance, and closeness, have been
related to these youth outcomes (Bowers et al. 2012;
DuBois and Silverthorn 2005b; Haddad et al. 2011; Kogan
et al. 2011; Rhodes et al. 1992).
For example, using the Add Health data set, Dubois and
Silverthorn (2005a) found that youth that reported having
natural mentors were more likely to complete high school
and attend college, were more likely to report higher levels
of self-esteem, and were more likely to be physically
active. In addition, these youth with natural mentors were
less likely to be part of negative outcomes, such as being
involved in gangs and risk taking behaviors. Similar find-
ings have been seen in studies that focus on rural African-
American youth (Kogan and Brody 2010), urban African-
American youth (Hurd et al. 2011), and European-Ameri-
can youth (Bowers et al. 2012).
These types of relationships with important non-parental
adults appear to have protective qualities for youth, and
they may also help enhance developmental outcomes, but,
as noted, they have not been studied extensively. As such,
it may be that these reported positive effects of important
non-parental adult relationships only hold for young people
with poor relationships with their parents; that is, important
non-parental adults serve as ‘‘compensatory resources’’
(Erickson et al. 2009). On the other hand important non-
parental adult relationships may only promote PYD when
youth also have good familial relationships; that is they
serve as ‘‘complementary resources’’ (Erickson et al.
2009). By examining youth relationships with both parents
and non-parental adults, the present study aims to address
such issues by extending prior work on the integrative
influence of parents and important non-parental adults on
adolescent development.
The Integrative Influence of Youth Relationships
with Parents and Important Non-parental Adults
for PYD
Although there has been a long-standing interest within
developmental science in the ecology of human develop-
ment as involving variables linked to intraindividual change
trajectories (Bronfenbrenner 1979; Bronfenbrenner and
Morris 2006), most research studying youth/?adult rela-
tionships has nevertheless focused on within-context influ-
ences (e.g., parents or teachers or mentors) on individual
development. However, consistent with Bronfenbrenner’s
(1979; Bronfenbrenner and Morris 2006) ideas, relational
developmental systems models (Overton 2013) emphasize
that development within a context is embodied within a
system involving cross-context relationships (e.g., parents
and teachers and mentors). There are, however, only a few
studies that have assessed the relationships among parents,
important non-parental adults, and youth development from
an embodied approach (e.g., Erickson et al. 2009; Hurd et al.
2013; Kogan and Brody 2010). These studies provide some
indications of what relationships we might expect when
examining the integrative effects of adults on youth.
Erickson et al. (2009) used data from the Add Health
and Add Health Academic Achievement studies (AHAA)
to examine the role of informal mentors in predicting
educational outcomes within the broader ecological con-
text. In particular they examined whether the presence of
resources within the family, school, and peer contexts
influenced (1) the probability of having an informal men-
toring relationship and (2) the effectiveness of that infor-
mal mentoring relationship for youth educational
outcomes. The findings indicated that, while youth with
many contextual resources are also more likely to report
having an informal mentor, the effect of having an informal
mentor (especially if that informal mentor was a teacher)
was actually greater for youth with few contextual
resources.
Similarly, Kogan and Brody (2010) studied individual
and contextual correlates of depressive symptoms among a
sample of rural African-American youth. They found that
support from informal mentors had a greater impact on
youth hope, perceived life chances, and self regulation for
youth in familial contexts in which parents practiced less
autonomy-promotive parenting behaviors. Whereas higher
autonomy-promoting parenting in general predicted greater
self-regulation and lower levels of depressive symptoms,
higher levels of supportive mentoring buffered the negative
effect of low autonomy-promoting parenting on youth self-
regulation.
Finally, Hurd et al. (2013) investigated the relationships
among involved-vigilant parenting, natural mentoring
relationships, youth social skills, and psychological well-
being among a sample of socioeconomically diverse,
African-American youth. Hurd and colleagues found that
mentoring was a complementary resource as youth with
natural mentoring relationships reported greater levels of
involved-vigilant parenting than youth without natural
mentors. Youth with more connected natural mentoring
relationships had higher social skills and psychological
well-being than youth with less connected or no natural
mentoring relationships. The type of natural mentoring
relationship youth had also moderated the effect of
involved-vigilant parenting on youth outcomes. Whereas
involved-vigilant parenting was related positively to youth
900 J Youth Adolescence (2014) 43:897–918
123

social skills and psychological well-being among the entire
sample of youth, the strength of these relationships was
weaker for youth with more connected natural mentoring
relationships. Hurd and colleagues suggested that this
finding may indicate that parents and natural mentors share
socializing responsibilities for young people, whereas those
youth with less connected, or with no non-parental, rela-
tionships may be more strongly influenced by their parents.
In general, these studies suggest that parenting may
moderate the influence of important non-parental adult
relationships on youth PYD. The nature of these interac-
tions may depend on the outcome of interest (e.g., com-
petence vs character) as well as on the profiles of parental
relationships that are identified. Nevertheless, despite past
findings regarding cross-contextual influences on adoles-
cent development, more research is needed about such
influences in order to elucidate the importance for adoles-
cent development of broadly, ecologically embodied,
mutually-influential individual-context relationships more
fully (Overton 2013). These prior studies have either
examined the role of youth–adult relationships in specific
samples of youth (African-American urban or African-
American rural youth) or relied on retrospective reports of
important non-parental adult relationships (Add Health).
Using data from the 4-H Study of PYD (Lerner et al. 2005,
2009,2010,2011), the present study extends this work by
examining the role of both parental and non-parental adult
relationships to predict youth development in adolescence.
Study Aims and Hypotheses
This study examined how the characteristics of youth
relationships with both parents and important non-parental
adults integratively influenced the Five Cs of PYD across
mid- to late-adolescence. Using data from youth in Waves
5 through 8 (approximately Grades 9–12) of the 4-H Study
of PYD (Lerner et al. 2005,2009,2010,2011), we first
investigated whether latent profiles based on participants’
perceived relationships with parents existed (regarding
monitoring, school involvement, and warmth). We also
tested whether the Five Cs of PYD differed according to
parenting profile membership. Next, we investigated whe-
ther profile membership was related to participants’ reports
of relationships with important non-parental adults.
Finally, we examined whether parenting profile member-
ship moderated the relationship between participants’
reported important non-parental adult relationships and the
Five Cs of PYD.
The plasticity of individual functioning constitutes a
strength of the person, enabling adjustment of behavior
(cognitions or emotions) to fit the adaptive demands of
time and place, including demands placed on the person by
the behavior of others (Lerner 1984). As such, because it
would afford the maximum probability for plasticity and
hence adaptability across grades, we expected that the
predominant latent class involved in perceived parenting
would correspond to a grouping showing at least moderate
levels of functioning in regard to all three assessed char-
acteristics. We term this predicted latent class ‘‘integrative
parenting,’’ and predicted that at all four grade levels we
assessed, integrative parenting should be the most fre-
quently observed latent class and the profile linked to the
highest levels of PYD and the lowest levels of risk
behaviors.
Although we would not dispute that there may be latent
classes of parenting that correspond to those configurations
discussed by Baumrind (1978) or Steinberg, et al. (1991),
that is, authoritative, authoritarian, and permissive parent-
ing styles, we do not think that such groupings should be as
large, ubiquitous across grade levels, or linked consistently
with indicators of PYD as the ‘‘integrative group.’’ Because
such groups would reflect a less plastic repertoire of par-
enting attributes, these latent classes would involve levels
of parenting characteristics neither affording the most
adaptive parent functioning or the most positive linkage
with youth behaviors. We hypothesized that these latent
class profiles would be differentially related to the Five Cs
of PYD. For example, authoritative parenting would be
associated with higher levels of the Cs, whereas
authoritarian parenting would be associated with lower
levels of the Cs. Prior research has indicated that the
relationship between parenting and youth outcomes may
vary according to SES, ethnicity/race, and other demo-
graphics (e.g., Furstenberg et al. 1999; Hill et al. 2003);
however, as we discuss more in the Discussion, we were
not able to examine these differences due to the complexity
of the models needed to test our primary hypotheses.
We also expected these parenting profiles would be
differentially related to the probability of youth reporting
whether they had an important non-parental adult. Based
on prior work (Erickson et al. 2009; Hurd et al. 2013), we
expected that integrative parenting would be most likely to
be linked to the presence of important non-parental adults
in the lives of youth. Because integrative parents are more
likely to be engaged with adaptive developmental regula-
tions (Brandtsta
¨dter 1998) than would be the case with
other latent classes of parenting, such individuals would be
most likely to model for their children (and to be sup-
portive of) engagement by their children with the positive-
development supportive features of their context.
Finally, we did not hypothesize the direction of the
moderating effect of parenting profile membership on the
relationship between important non-parental adult rela-
tionships and the Five Cs of PYD. Prior findings have been
mixed in regard to these questions. These parent-important
J Youth Adolescence (2014) 43:897–918 901
123

non-parental adult interactions may have either a com-
pensatory or a complementary effect on youth outcomes
(Erickson et al. 2009; Hurd et al. 2013). These effects
could be due to the outcome of interest (which C of PYD),
the parenting profile, or the age of the youth.
Method
Full details of the 4-H Study of PYD have been presented
elsewhere (Lerner et al. 2005,2009,2010,2011). There-
fore, we present here only the features of the methods
relevant to the present research, which includes data from
Waves 5 through 8. A discussion of the overall method of
the 4-H Study is provided in the introductory article of this
special issue (Bowers et al., in press).
Participants
Participants for the present study were gathered from
Waves 5 through 8 (Grades 9 through 12) of the 4-H Study
of PYD. While there are a large proportion of longitudinal
participants across these waves of the 4-H Study, we
treated each wave as a separate cross-sectional sample.
This decision was due to a large proportion of participants
missing from wave to wave, and the complexity of the final
models that were selected to test our hypotheses.
The Wave 5 sample was comprised of 975 ninth grade
youth (61.1 % female; mean age =14.93, SD =1.10).
The majority of youth reported their ethnicity as European
American (59.5 %), but African-American (8.2 %), Latino
American (9.3 %), and Asian American (2.8 %) youth
were also represented, among others. In Wave 5, 4.9 % of
participants’ mothers reported maternal education of less
than 12 years, 13.1 % reported 12 years, 26.9 % reported
between 13 and 14 years, 10.7 % reported 16 years, 7.7 %
reported more than 16 years of education, and 36.6 % did
not report their level of education.
The Wave 6 sample was comprised of 1855 tenth grade
youth (63.4 % female; mean age =15.71, SD =1.37).
The majority of youth reported their ethnicity as European
American (75.9 %), but African-American (5.6 %), Latino
American (6.5 %), and Asian American (1.6 %) youth
were also represented. In Wave 6, 1.2 % of participants’
mothers reported maternal education of less than 12 years,
5.9 % reported 12 years, 12.8 % reported between 13 and
14 years, 10.6 % reported 16 years, 5.0 % reported more
than 16 years of education, and 64.7 % did not report their
level of education.
The Wave 7 sample was comprised of 983 eleventh
grade youth (67.9 % female; mean age =16.50,
SD =1.45). The majority of youth reported their ethnicity
as European American (78.5 %), but African-American
(5.2 %), Latino American (2.3 %), and Asian American
(2.4 %) youth were also represented. In Wave 7, 1.2 % of
participants’ mothers reported maternal education of less
than 12 years, 6.5 % reported 12 years, 18.0 % reported
between 13 and 14 years, 13.1 % reported 16 years, 6.8 %
reported more than 16 years of education, and 54.4 % did
not report their level of education.
The Wave 8 sample was comprised of 703 twelfth grade
youth (69.3 % female; mean age =17.61, SD =1.51).
The majority of youth reported their ethnicity as European
American (75.2 %), but African-American (3.7 %), Latino
American (4.8 %), and Asian American (3.7 %) youth
were also represented. In Wave 8, 2.3 % of participants’
mothers reported maternal education of less than 12 years,
7.7 % reported 12 years, 17.0 % reported between 13 and
14 years, 15.0 % reported 16 years, 7.9 % reported more
than 16 years of education, and 50.1 % did not report their
level of education.
Measures
Maternal Warmth
The eight-item maternal warmth subscale of the Child’s
Report of Parenting Behaviors Inventory (CRPBI; Sch-
ludermann and Schludermann 1970) was used to assess
maternal warmth. The CRPBI is a widely used self-report
measure of children’s assessment of parenting practices.
Maternal warmth was conceptualized as behaviors that
indicate acceptance, nurturance, support, and a feeling of
being loved and wanted by the parent (Gray and Steinberg
1999). Examples of maternal warmth items include ‘‘My
mother speaks to me in a warm and friendly way’’ and ‘‘My
mother cheers me up when I am upset.’’ The response
format ranges from 0 =almost never to 4 =almost
always. Higher scores indicate higher warmth and nurtur-
ance. Overall, the CRPBI has good reliability (Cronbach’s
alpha =.80; Schludermann and Schludermann 1970). In
regard to validity, factor analyses have indicated that
Warmth is a replicable factor (Schwarz et al. 1985). There
is evidence for adequate convergent and discriminant
validity, for example, correlations between ratings by sib-
lings were r=.50 (p\.01) for maternal warmth; corre-
lations between adolescents and parents for warmth were in
the .4 range (Schwarz et al. 1985). In the present data set,
the Cronbach’s alphas for the maternal warmth measure
were .96 at Grades 9 through 12.
Parental School Involvement
We used three items from the Search Institute’s Profiles of
Student Life—Attitudes and Behaviors (PSL-AB) ques-
tionnaire (Leffert et al. 1998) to create the parental school
902 J Youth Adolescence (2014) 43:897–918
123

involvement scale (Theokas et al. 2005). An example item
is ‘‘How often does one of your parents ask about your
homework?’’ Each item is measured using a five point
Likert-type scale ranging from 0 =never to 4 =very often
with a higher score reflecting greater parental involvement.
The scale is computed by taking the mean of at least three
of the four items. Cronbach’s alphas for this scale were
0.84 for Grade 9; 0.80 for Grades 10 and 11, and 0.81 for
Grade 12.
Parental Monitoring
We used the eight-item Parental Monitoring Scale (PMS;
Small and Kerns 1993) to measure parental monitoring.
This self-report instrument assesses the adolescent’s per-
ception of parental monitoring or the extent to which their
parents keep track of their behavior and acquaintances.
Example items are ‘‘My parent(s) know where I am after
school,’’ and ‘‘My parent(s) know who my friends are.’’
The response format ranges from 1 =never to 5 =always,
with higher scores indicating higher parental monitoring.
The scale is scored by calculating the mean of all the items.
The PMS scale was developed for a Small and Kerns
(1993) study, based on interview research of Patterson and
Stouthamer-Loeber (1984). The PMS scale has been
reported to have reliability (Cronbach’s alpha =.87) and
predictive validity (Small and Kerns 1993). Cronbach’s
alpha was 0.92 for Grades 9–11; and 0.93 for Grade 12.
Important Non-parental Adult Relationships
To index the presence of an important non-parental adult,
we used one item derived from the Monitoring the Future
(MTF 2000) study that asks participants the following
question, ‘‘Other than your parents, is there at least one
other adult you would feel able to talk to if you were
having problems in your life?’’ The response options are
1=yes, for most or all problems,2=yes, for at least
some of my problems, and 3 =no. We did not ask the
youth to identify (name) this adult.
Positive Youth Development
The approach to PYD used by Lerner et al. (2005)employed
several measures to index PYD, which is operationalized
through the assessment of the Five Cs—competence, con-
fidence, character, connection, and caring. Each ‘‘C’’ com-
prises a number of well-validated scales designed to assess
the essential elements of the definition of the construct.
General information regarding the measurement of each of
the Cs is presented below. Full details about these measures,
their construction, and validity and reliability can be found
in Lerner et al. (2005) and Bowers et al. (2010).
The Five Cs comprising the PYD construct are opera-
tionalized as follows:
Competence is a positive view of one’s action in
domain-specific areas including the social and academic
domains (11 items for Grades 9–12). Cronbach’s alpha was
0.83 for Grades 9, 11, and 12, and 0.84 for Grade 10.
Confidence is an internal sense of overall positive self-
worth, identity, and feelings about one’s physical appear-
ance (16 items for Grades 9–12). Cronbach’s alpha was
0.92 for Grades 9–12.
Character involves respect for societal and cultural
rules, possession of standards for correct behaviors, a sense
of right and wrong, and integrity (20 items for Grades
9–12). Cronbach’s alpha was 0.89 for Grades 9 and 11, and
0.90 for Grades 10 and 12.
Connection involves a positive bond with people and
institutions that are reflected in healthy, bidirectional
exchanges between the individual and peers, family,
school, and community in which both parties contribute to
the relationship (22 items for Grades 9–12). Cronbach’s
alpha was 0.90 for Grades 9 and 12, and 0.89 for Grades 10
and 11.
Caring is the degree of sympathy and empathy, that is,
the degree to which participants feel sorry for the distress
of others (9 items for Waves 5–8). Cronbach’s alpha was
0.84 for Grades 9 and 10, 0.82 for Grade 11, and 0.83 for
Grade 12.
Procedure
In Waves 1 through 3 (Grades 5 through 7) of the 4-H
Study, data collection from youth was conducted by trained
study staff or, at more distant locations, hired assistants. A
detailed protocol was used to ensure that data collection
was administered uniformly and to ensure the return of all
study materials. After Wave 1, youth who were absent on
the day of the survey or were from schools or programs that
did not allow on-site testing were contacted by e-mail,
mail, or phone, and were asked to complete and return the
survey to us. Beginning in Wave 5 (Grade 9), youth
completed the survey online unless they requested a paper
survey. Parents completed online or paper surveys. Paper
surveys were delivered to their homes by their children or
through the mail (in the latter case, return postage was
provided).
Data Analysis Plan
Data analysis involved two major steps. First, we under-
took analyses to identify profiles. Second, we conducted
analyses relating profiles to both the Five Cs and to
important non-parental adults.
J Youth Adolescence (2014) 43:897–918 903
123

Identifying Profiles
To investigate our initial question regarding the existence of
profiles of participants’ perceived relationships with parents,
we conducted latent profile analyses (LPA) at each wave.
LPA is a mixture modeling technique analogous to latent
class analysis (LCA; Collins and Lanza 2010), except the
indicators of the latent profiles are continuous in LPA rather
than categorical as in LCA. As with LCA, the aim of an
LPA is to identify subgroups of individuals who are similar
to each other on a specific group of variables (i.e., indicators
of perceived parenting) and different from individuals in
other subgroups (Muthe
´n and Muthe
´n2000). These sub-
groups are not directly observable but must be inferred from
relationships among the observed variables (i.e., partici-
pants’ responses to survey items about their relationships
with their parents). In the current study, we used maternal
warmth, parental school involvement, and parental moni-
toring as indicators of the latent profile variable.
A primary consideration in specifying LPAs is whether
to uphold the assumption of conditional (or local) inde-
pendence. In conventional LCA models, conditional inde-
pendence means that variables must be uncorrelated within
class or profile (Collins and Lanza 2010). Another way of
stating this assumption is correlations between observed
variables are assumed to exist because of the underlying
latent classes/profiles, and once the classes/profiles are
modeled these associations should no longer be necessary.
This assumption, however, is often less realistic for latent
profile models (i.e., those with continuous indicators), and
it is recommended that continuous indicators be allowed to
correlate within profiles (i.e., a multivariate normal mixture
model, McLachlan and Peel 2000). We tested models both
with and without conditional independence. In all cases,
the multivariate normal mixture models provided a better
fit to the data, so we report only those results.
The procedure for conducting LPA involves testing
models with varying numbers of profiles and comparing fit
indices, as well as theoretical interpretability, to decide on
the number of profiles that provide the best fit to the data.
In terms of statistical model fit indices, the most commonly
used indices include information criteria [e.g., the Bayesian
Information Criterion (BIC); Schwarz 1978], the bootstrap
likelihood ratio test (BLRT; McLachlan and Peel 2000),
and the Lo–Mendell–Rubin likelihood test (LMR; Lo et al.
2001). We examined all of these indices but gave special
weight to the BIC and BLRT because in prior simulation
work (Nylund et al. 2007), these two tests were the most
accurate in suggesting the appropriate number of classes.
We also examined the interpretability of each model,
including the prevalence of the profiles, the response pat-
terns of each profile, and the profile’s correspondence with
theoretical expectations.
Relating the Parenting Profiles to the Five Cs of PYD
and Relationships with Important Non-parental Adults
To investigate the relationship between the latent profiles
of perceived parenting and the auxiliary observed variables
of the Five Cs of PYD and participants’ reports of rela-
tionships with important non-parental adults, we used the
three-step procedure available in MPlus (Asparouhov and
Muthe
´n2013). This procedure, newly available in Version
7, allows researchers to examine the relationship between
the latent profile variable and the other variables of interest
independently (i.e., without including these variables in the
estimation of the latent profile model itself) while still
incorporating the classification uncertainty (i.e., measure-
ment error) associated with latent profile models. We used
an extension of this procedure to test whether profiles of
perceived parenting moderated the relationship between
the Five Cs of PYD and participants’ reported relationships
with important non-parental adults. In this procedure,
individuals’ most likely latent class membership was used
as a latent class indicator variable with uncertainty rates
(i.e., measurement error) prefixed at the class membership
probabilities obtained in the original latent profile analysis.
We then specified the regression model of interest and
tested whether the relationships in that model differed
within the various latent profiles (i.e., we tested whether
latent profile membership was a moderator). This analysis
enabled us to test a separate regression model while
retaining uncertainly in profile membership.
Results
Profiles of Perceived Parenting
Table 1shows model fit statistics at each wave for models
of perceived parenting with two through five, six, or seven
profiles. For each wave, we decided which number of
profiles to choose based on several criteria. First, the pro-
files with the lowest values for AIC and BIC, along with
the highest entropy, were preferred. In addition, the pvalue
of the LMR and BLRT tests showed whether a certain
number of profiles (e.g., three) provided a significant
improvement in model fit over a model with one fewer
profile (e.g., two). In many cases, however, the AIC and
BIC continued to decrease, and the BLRT test did not
produce a non-significant pvalue (in other words, adding
profiles continued to improve model fit, as judged by those
criteria, up to six or even seven profiles). This effect
commonly happens in latent profile analyses, however, and
does not guarantee that the largest number of profiles
always provides the most theoretically meaningful or
substantively interpretable solution. In such situations, we
904 J Youth Adolescence (2014) 43:897–918
123

evaluated the point at which the additional profiles were
either too small to be meaningful (e.g., a profile with one
member) or were very similar to existing profiles.
We chose a four-profile solution at Grade 9, a five-
profile solution at Grade 10, a five-profile solution at Grade
11, and a six-profile solution at Grade 12. Figures 1,2,3,4
show profile-specific means on the three parenting indica-
tors for the chosen solutions. As described above, the
model fit indices suggested which number of profiles would
fit best. It is important to note that the profiles may not all
differ from each other in regard to statistical significance
involving the item response probabilities and means of
each of the included variables. We holistically evaluated
the profiles to identify patterns of responses rather than to
note specific areas of difference.
Grade 9
We identified four profiles of perceived parenting at Grade
9. The first profile, which we termed Integrative, comprised
26 % of the sample (n=214) and was characterized by
relatively moderate and similar levels of all three predic-
tors; however, it was not the largest parenting profile as we
expected. The second profile, which we labeled School-
Focused, was made up of 4.9 % of the sample (n=45).
This profile was characterized by lower levels of warmth
and monitoring but moderate levels of involvement in
school. The third profile, 62.7 % of participants (n=582),
we termed Authoritative because it was characterized by
higher levels of warmth and monitoring but lower levels of
involvement (Since this sample is mostly ninth graders,
high levels of parental school involvement may not be
expected from authoritative parents). The final profile,
6.5 % of the sample (n=60), we labeled Authoritarian;
this profile was characterized by relatively moderate levels
of involvement, lower levels of warmth, and higher levels
of monitoring (Fig. 1).
Grade 10
At Grade 10, we identified five profiles of perceived par-
enting. Three profiles had been previously identified at
Grade 9: Integrative (n=240, 13.5 %), which contrary to
our prediction was not the largest profile; Authoritative
profile (n =524, 29.5 %), and Authoritarian (n=52,
3.0 %). We also identified two new profiles. The Unin-
volved profile comprised 3.8 % of the sample (n=68) and
was characterized by relatively lower levels of all three
parenting indicators, whereas the Highly Involved profile
consisted of half of the sample (n=895) and was char-
acterized by higher levels of monitoring, warmth, and
involvement (Fig. 2).
Grade 11
We also identified five profiles at Grade 11, four of which
were similar to those identified in Grade 10. The Unin-
volved profile (n=18, 1.9 %) was again characterized by
Table 1 Fit statistics for latent profile analyses for Grades 9 through
10
Grade Number
of profiles
BIC AIC Entropy LMR
pvalue
BLRT
pvalue
9 2 6,961 .78 .000 .000
3 6,912 .83 .000 .000
4 6,884 .80 .1369 .000
5 6,860 .82 .3960 .000
10 2 12,871 .85 .000 .000
3 12,705 .85 .000 .000
4 12,570 .85 .001 .000
5 12,516 .84 .002 .000
6 12,474 .86 .021 .000
11 2 6,772 .79 .064 .000
3 6,690 .82 .002 .000
4 6,629 .85 .104 .000
5 6,615 .84 .269 .000
6 6,592 .87 .009 .000
12 2 4,952 .82 .024 .000
3 4,904 .83 .003 .000
4 4,870 .86 .004 .000
5 4,853 .84 .533 .000
6 4,842 .82 .254 .000
7 4,843 .81 .098 .000
Bold values indicate profile solution selected at each Grade
Fig. 1 Four-profile model of
youth-reported parenting at
Grade 9 of the 4-H Study of
PYD. Note Integrative
(n =214); School-focused
(n =45); Authoritative
(n =582); and Authoritarian
(n =60)
J Youth Adolescence (2014) 43:897–918 905
123

lower levels of all three parenting indicators, the Integra-
tive profile (n=284, 29.4 %) was, again counter to pre-
diction, not the largest profile; a Highly Involved profile
(n=593, 61.4 %) was marked by higher warmth and
monitoring with relatively moderate involvement; and an
Authoritarian profile (n=55, 5.7 %) was characterized by
higher monitoring, relatively moderate involvement, and
lower warmth. The Permissive profile, (n=16, 1.6 %)
newly identified at this wave, was characterized by a pat-
tern opposite of the Authoritarian profile. This profile was
marked by relatively higher warmth and relatively low-to-
moderate involvement and monitoring (Fig. 3).
Grade 12
At Grade 12, we identified six profiles, five of which were
similar to those identified in prior waves: Integrative
(n=130, 19.1 %), which once more was not the largest
profile; Authoritarian (n=9, 1.3 %); Highly Involved
(n=413, 60.1 %); Uninvolved (n =14, 2.1 %); and
Permissive (n=18, 2.6 %). The sixth profile, Controlling,
(n=97, 14.2 %) was marked by relatively lower maternal
warmth but relatively higher parental school involvement
and parental monitoring (Fig. 4).
Mean Differences on the Five Cs Across Profiles
We next tested for mean differences in the Five Cs
according to profile membership using the 3-step procedure
implemented in MPlus. In order to account for the multiple
comparisons conducted at each wave, we applied a Bon-
ferroni correction to control for a family wise error rate of
a=.10; therefore corrected as ranged from a=0 .016 at
Grade 9 to a=0.006 at Grade 12. Table 2shows the
results of these analyses.
At Grade 9, participants classified into the Authoritative
Parenting profile had significantly higher mean scores on
all Five of the Cs compared to participants in the other
three profiles of perceived parenting (all p’s \.016). Par-
ticipants in the School-Focused parenting profile showed
consistently low scores on the Five Cs, particularly con-
nection. Participants in the School-Focused profile reported
significantly lower connection that youth in the
Authoritarian, and Integrative profile as well. There were
Fig. 2 Five-profile model of
youth-reported parenting at
Grade 10 of the 4-H Study of
PYD. Note Integrative
(n =240); Authoritarian
(n =52); Highly-involved
(n =895); Authoritative
(n =524); and Uninvolved
(n =68)
Fig. 3 Five-profile model of
youth-reported parenting at
Grade 11 of the 4-H Study of
PYD. Note Uninvolved
(n =18); Permissive (n =16);
Integrative (n =284); Highly-
Involved (n =593); and
Authoritarian (n =55)
Fig. 4 Six-profile model of
youth-reported parenting at
Grade 12 of the 4-H Study of
PYD. Note Integrative
(n =130); Authoritarian
(n =9); Controlling (n =97);
Highly Involved (n =413);
Uninvolved (n =14); and
Permissive (n =18)
906 J Youth Adolescence (2014) 43:897–918
123

no other significant mean differences in confidence, char-
acter, caring and competence for youth in the Integrative,
Authoritarian, or School-Focused profiles.
At Grade 10, youth who perceived their parents as
Highly Involved had significantly higher mean scores in all
five Cs than all other profiles (all p’s \.01). Those in the
Authoritative profile had the next highest set of scores
(with the exception of confidence for which they were
similar to Integrative and Uninvolved parenting youth).
Integrative and Uninvolved participants had similar scores
on all 5 Cs with the exception of connection, on which
Integrative participants had higher scores. Youth who
perceived their parents as Authoritarian had consistently
low scores on all Five Cs (aside from connection, in which
youth with Uninvolved parents were lower), and their
scores at Grade 10 were lower than those of youth in this
profile at Grade 9.
At Grade 11, participants in the Highly Involved par-
enting profile again showed the highest levels of the Five
Cs, with the exception of competence (for which their
score was slightly lower than, but not different from, youth
in the Permissive profile), but the advantage of Highly
Involved parenting became less clear. Those youth with
Highly Involved parents reported significantly higher
means on all Cs as compared to the Integrative profile (all
p’s \.01), but they were only significantly better than
Authoritarian profile youth in confidence, connection, and
competence; the Uninvolved parenting profile youth in
caring, connection, and competence; and the Permissive
parenting profile youth in character. Similar to Grade 10,
the scores of participants in the Integrative and Uninvolved
parenting profiles were similar on confidence, character,
caring, and competence—those in the Integrative parenting
profile reported higher scores for connection. Those in the
Permissive parenting profile showed the lowest levels of
character and caring, whereas those in the Uninvolved
profile had the lowest score on connection.
At Grade 12, participants in the Highly Involved and
Permissive parenting profiles generally had the pattern of
highest scores on confidence, character, connection, and
competence. Youth in the Authoritarian profile reported the
second highest caring score, but no comparisons to the
Authoritarian profile were significant due to the few youth in
this profile (n =9). Uninvolved participants again had the
most problematic profiles as they reported significantly lower
confidence and connection than youth in the Integrative,
Highly Involved, and Permissive profiles (all p’s\.006).
The newly identified group, youth with Controlling parent-
ing, reported significantly lower levels of the Cs (except for
caring) as compared to Highly Involved parenting.
Table 2 Mean scores on the Five Cs of PYD by latent profile of perceived parenting
Grade Profile Confidence Character Caring Connection Competence
9 Integrative 60.71
c
59.51
c
60.57
c
55.55
b,c
63.32
c
School-focused 53.74
c
53.85
c
54.66
c
36.74
a,c,d
56.58
c
Authoritative 73.58
a,b,d
77.88
a,b,d
77.78
a,b,d
74.86
a,b,d
79.59
a,b,d
Authoritarian 53.20
c
62.92
c
59.80
c
52.92
b,c
70.59
c
10 Integrative 57.85
b,c,d
60.67
b,c,d
65.16
c,d
55.10
c,d,e
61.44
c,d
Authoritarian 40.90
a,c,d,e
49.21
a,c,d
48.11
c,d
47.91
c,d
52.53
c,d
Highly involved 74.06
a,b,d,e
78.35
a,b,d,e
80.18
a,b,d,e
77.59
a,b,d,e
79.22
a,b,d,e
Authoritative 62.59
a,b,c
70.46
a,b,c,e
73.29
a,b,c,e
66.45
a,b,c,e
70.96
a,b,c,e
Uninvolved 59.14
b,c
59.32
c,d
59.78
c,d
45.50
a,c,d
61.08
c,d
11 Uninvolved 62.51 66.73 62.97
d
42.67
b,c,d
64.78
b,d
Permissive 67.45 62.51
d
61.07 68.10
a
80.70
a,e
Integrative 62.27
d
68.65
d
73.58
d
63.42
a,d,e
72.20
d
Highly involved 72.09
c,e
78.16
b,c
80.12
a,c
75.91
a,c,e
79.22
a,c,e
Authoritarian 57.62
d
71.26 71.04 55.14
c,d
66.05
b,d
12 Integrative 61.08
d,e,f
64.92
d,f
68.83
d
57.33
d,e,f
67.62
d
Authoritarian 52.73 73.34 81.38 60.23 69.62
Controlling 57.30
d,f
72.02
d,f
76.92 60.79
d,e
72.16
d
Highly involved 74.83
a,c,e
80.92
a,c
81.47
a
77.33
a,c,e
81.18
a,c
Uninvolved 49.06
a,d,f
61.85 63.78 43.97
a,c,d,f
54.97
Permissive 77.04
a,c,e
80.45
a,c
77.75 70.03
a,e
79.65
Within the same wave and column,
a
different from Group 1;
b
different from Group 2;
c
different from Group 3;
d
different from Group 4;
e
different from Group 5;
f
different from Group 6. p\.016 at Grade 5; p\.01 at Grades 10 and 11; p\.006 at Grade 12
J Youth Adolescence (2014) 43:897–918 907
123

Profiles Differences in Participants’ Reports
of Relationships with Important Non-parental Adults
The third step in our analysis was to investigate whether
participants were more or less likely to report parenting
profiles based on their relationship with an important non-
parental adult. Table 3provides the proportion of youth
who reported having an important non-parental adult to
talk to about none, some, or all of their problems across the
parenting profiles at each Grade. Table 4shows the results
of a set of multinomial logistic regressions predicting
parenting profile membership from important non-parental
adult relationship status. A specific parenting profile was
chosen as a reference group for each multinomial regres-
sion equation, and the impact of important non-parental
adult relationship status on parenting profile membership
was computed. Each model tested whether youth reporting
that they had a least one adult to talk to about all of their
problems or some of their problems predicted parenting
profile membership in comparison to a selected reference
parenting profile. The multinomial logit model provides
raw logistic regression coefficients which are the logs of
the ratios of the two probabilities. These coefficients are
often exponentiated and interpreted as odds ratios. Stan-
dard interpretation of the odds ratio is for a unit of change
in the predictor variable, the odds of being in a specific
profile relative to the referent group are expected to change
by a factor of the respective parameter estimate given the
other variables in the model are held constant. An odds
ratio greater (less) than 1 indicates an increase (decrease) in
the odds of a youth belonging to a particular parenting
profile as compared to the reference parenting profile.
Table 4presents the odds ratios for these comparisons
based on youth relationships with an important non-
parental adult.
We found several significant differences in parenting
profile membership based on important non-parental adult
relationships in regard to whether the participant had an
adult to talk to about none, some, or all of his or her
problems. At Grade 9, youth of Authoritative parents were
more likely to report having someone they could talk to
about their problems compared to participants with
Authoritarian, Integrative, or School-Focused parents. For
example, youth who reported having an adult to talk to
about some of their problems had 5.81 times larger odds
than youth with no important non-parental adult to be in the
Authoritative parenting profile as compared to the
Authoritarian profile (33.1 vs 3.0 % of the sample), and
youth who reported having an adult to talk to about all of
their problems had 7.09 times larger odds than youth with
Table 3 Proportion (%) of All youth responses to the question ‘‘other than your parents, is there at least one other adult you would feel able to
talk to if you were having problems in your life?’’
Grade 9 Latent profile of perceived parenting
Integrative School-focused Authoritative Authoritarian
No adult 5.1 1.8 5.9 1.8
Some problems 12.9 2.1 33.1 3.0
All problems 7.4 0.6 24.4 1.9
Grade 10 Integrative Authoritarian Highly involved Authoritative Uninvolved
No adult 3.6 0.5 3.5 2.5 1.2
Some problems 6.2 1.5 21.8 16.0 1.4
All problems 3.7 1.0 24.9 11.0 1.2
Grade 11 Uninvolved Permissive Integrative Highly involved Authoritarian
No adult 0.7 0.3 3.0 3.3 1.2
Some problems 0.9 0.2 15.9 29.6 1.9
All problems 0.2 0.9 10.3 28.8 2.7
Grade 12 Integrative Authoritarian Controlling Highly involved Uninvolved Permissive
No adult 2.1 0.0 1.7 2.3 0.8 0.2
Some problems 8.6 0.8 7.7 27.6 0.5 0.9
All problems 8.1 0.5 5.1 31.1 0.8 1.7
Percentages are based on youth who were placed in a parenting profile and responded to the item concerning their relationship to an important
non-parental adult. Absolute numbers do not match the percentages of youth in each profile reported in the profiles of perceived parenting results
908 J Youth Adolescence (2014) 43:897–918
123

no important non-parental adult to be in the Authoritative
parenting profile as compared to the Authoritarian profile
(24.4 vs 1.9 % of the sample). Consistent with our pre-
diction, however, youth who reported having someone to
talk to about all of their problems were more likely to be in
the Integrative parenting profile (7.4 % of the sample) than
in the School-Focused profile (0.6 % of the sample; 7.10
times larger odds).
Table 4 Odds ratios from logistic regression predicting profile membership from relationships with important non-parental adults
Other than your parents, is there at least one other adult you would feel able to talk to if you were having problems in your life? (reference
category = no)
Latent profile of perceived parenting
Grade 9 Integrative School-focused Authoritative Authoritarian
Some problems 2.25 .92 5.81** Reference
All problems 2.06 .29 7.09** Reference
Some problems 0.38** 0.15*** Reference
All problems 0.29*** 0.04** Reference
Some problems 2.45 Reference
All problems 7.10* Reference
Grade 10 Integrative Authoritarian Highly involved Authoritative Uninvolved
Some problems 1.45 2.27 5.64*** 6.42** Reference
All problems 0.94 0.96 7.46*** 4.62*** Reference
Some problems 0.22*** 0.35 0.87 Reference
All problems 0.21*** 0.21 1.61 Reference
Some problems 0.25*** 0.40 Reference
All problems 0.04*** 0.12* Reference
Some problems 0.63 Reference
All problems 0.98 Reference
Grade 11 Uninvolved Permissive Integrative Highly involved Authoritarian
Some problems 1.31 0
a
5.21** 11.02*** Reference
All problems 0.11 2.10 1.68 5.92*** Reference
Some problems 0.12*** 0
a
0.47* Reference
All problems 0
a
0.35 0.29** Reference
Some problems 0.25* 0
a
Reference
All problems 0.06* 1.24 Reference
Some problems 15.00
a
Reference
All problems 0.05 Reference
Grade 12 Integrative Authoritarian Controlling Highly involved Uninvolved Permissive
Some problems 0.56 15.00
a
0.63 2.22 0.06 Reference
All problems 0.27 15.00
a
0.17 1.32 0.07 Reference
Some problems 8.75** 15.00
a
9.97* 15.00
a
Reference
All problems 3.89 15.00
a
2.56 15.00
a
Reference
Some problems 0.25* 15.00
a
0.28* Reference
All problems 0.20* 15.00
a
0.13** Reference
Some problems 0.85 15.00
a
Reference
All problems 1.52 15.00
a
Reference
Some problems 0
a
Reference
All problems 0
a
Reference
*p\.05; ** p\.01; *** p\.001
a
Indicates threshold set at extreme value (15 or 0) by Mplus during estimation
J Youth Adolescence (2014) 43:897–918 909
123

We found a similar pattern at Grade 10; participants
with Highly Involved and Authoritative parents being
much more likely to report having someone to talk about
some or all their problems (46.7 and 27.0 %, respectively)
compared to participants with Uninvolved and Integrative
parents (2.6 and 9.9 %, respectively). Youth with Highly
Involved parents were also more likely to report having an
important non-parental adult to talk to about all of their
Table 5 Beta coefficients for regression of Five Cs on relationships with important non-parental adults
Other than your parents, is there at least one other adult you would feel able to talk to if you were having problems in your life? (reference
category = no; first line represents coefficient for ‘‘some of my problems’’ and second is coefficient for ‘‘all of my problems’’)
Grade Profile Confidence Character Caring Connection Competence
9 Integrative 7.02 4.15 4.58 3.35 -0.55
6.80 7.57 8.22 6.56* 6.35
School-focused -20.46* 15.09** 0.91 11.55 27.37
-13.19 6.20 7.45 7.66 -13.58
Authoritative 8.78* 1.83 0.97 5.44** 2.36
15.66*** 5.78** 4.06 10.33** 6.18
Authoritarian 12.32 16.01* 15.73* 20.55*** 15.10*
7.23 15.78* 16.92 18.06** 14.23**
10 Integrative 4.25 1.40 5.72 6.73 4.80
7.61* 2.46 13.11* 15.92*** 7.01
Authoritarian 19.29* 29.13** 34.75 11.53 11.34
14.05 33.39** 24.92 -1.00 2.30
Highly involved -0.55 5.92 5.10 3.19 2.78
4.91 9.95** 9.53** 8.08** 6.34*
Authoritative 4.22 3.54 0.55 6.72** 8.24**
6.89 5.14 3.47 10.00*** 9.04**
Uninvolved 6.17 11.23 20.39 29.60 9.33
11.23 -0.06 14.54 16.30 11.36
11 Uninvolved -7.89 37.56*** 38.21** 8.66 3.15
-39.41* 12.11 34.76 4.21 -27.22***
Permissive -21.67 -35.14* 27.20** -17.05 -4.23
23.87 -0.28 7.06 -2.34 26.99***
Integrative 3.29 2.03 -0.23 12.08*** 1.91
10.48 7.26 3.06 18.69*** 4.90
Highly involved 9.50 5.64 3.06 6.85* 10.53**
13.71** 10.33** 5.61 13.02*** 13.84***
Authoritarian -5.65 -3.40 5.73 0.78 -2.04
8.27 4.16 10.29 9.71 8.59
12 Integrative -1.64 4.53 0.83 10.99* 2.09
1.76 4.74 12.10* 16.62** 8.86*
Authoritarian -22.54 18.91** 35.33*** 8.52 9.73
12.05 9.35** 24.58*** 43.52*** 27.36***
Controlling 4.88 0.43 0.88 5.73 -0.97
8.04 9.06* 8.01 13.93** 6.46
Highly involved 11.93* 5.48 4.47 7.82* 6.55
18.15*** 11.14** 5.23 14.03*** 9.88
Uninvolved -2.00 1.05 2.34 18.30* 21.01*
13.45 37.55*** 37.77*** 8.14 19.18**
Permissive -22.56** -13.61*** -6.24 0.07 -12.20
-16.36** -20.01*** -17.29** 10.05 -8.98*
*p\.05; ** p\.01; *** p\.001
910 J Youth Adolescence (2014) 43:897–918
123

problems than youth in the Authoritarian parenting profile
(24.9 vs 1.0 %, respectively).
At Grade 11, participants with Uninvolved or
Authoritarian parents were less likely to report having an
important non-parental adult to talk to about some or all of
their problems (1.1 and 4.6 %, respectively) than partici-
pants with either Integrative or Highly Involved parents
(26.2 and 58.4 %, respectively). Participants with Integra-
tive parents, in turn, were less likely to report having an
important non-parental adult to talk about some or all
problems with than were participants with Highly Involved
parents (25.3 vs 58.4 %, respectively).
At Grade 12, a similar pattern emerged as youth with
Uninvolved parents were again less likely to report having
an important non-parental adult to talk about some of their
problems with than youth with Integrative or Controlling
parents (0.5 vs 8.6 and 7.7 %, respectively). In turn, youth
in the Highly Involved parenting profile were more likely
to report having an important non-parental adult to talk to
about some or all of their problems (58.7 %) than youth in
the Integrative (16.7 %) and Controlling (12.8) parenting
profiles.
Profile Moderation of the Relationship Between
Important Non-parental Adults and the Five Cs
Our final analyses involved testing whether parenting
profile membership moderated the relationship between
participants’ reports of their important non-parental adult
relationships and the Five Cs of PYD. Full results of these
analyses are presented in Table 5.
At Grade 9, we found moderation for all Five Cs.
Having an adult to talk to about some or all of their
problems was associated with increased connection
for participants in the Integrative, Authoritative, and
Authoritarian parenting profiles; increased confidence for
participants of Authoritative parents; increased character
for youth who perceived their parents to be School-
Focused, Authoritative, or Authoritarian; and increased
caring and competence for youth with Authoritarian par-
ents. In contrast, having an adult to talk to about some
problems (but not all) was associated with a decrease in
confidence among youth in the School-Focused profile.
At Grade 10, having an adult to talk to about some or all
of their problems was associated with increased connection
for participants in the Integrative, Highly Involved, and
Authoritative parenting profiles. In turn, having an impor-
tant non-parental adult to talk to about some of all of their
problems was associated with increased confidence for
participants of Integrative parents and increased character
for youth who perceived their parents to be Authoritarian.
Finally, having an important non-parental adult to talk to
about some of all of their problems was associated with
increased caring for youth in the Integrative and Highly
Involved profiles, and increased competence for youth in
the Highly Involved and Authoritative profiles.
At Grade 11, having an adult to talk to about some or all
of their problems was associated with increased connection
for participants in the Integrative and Highly Involved
parenting profiles. In addition, having an important non-
parental adult to talk to was associated with increased
confidence for participants of Highly Involved parents and
increased character for youth who perceived their parents
to be Uninvolved or Highly Involved. The degree of
important non-parental adult availability to talk was also
associated with increased caring for youth in the Unin-
volved and Permissive profiles and increased competence
for youth in the Permissive profile. In contrast, having an
adult to talk to about all problems was associated with
decreased confidence and competence for those in the
Uninvolved profile and having an adult to talk to about
some problems was associated with decreased character for
participants who perceived their parents to be Permissive.
At Grade 12, having an adult to talk to about some or all
of their problems was associated with increased connection
for participants in all parenting profiles except Permissive
parenting and increased confidence for participants of
Highly Involved parents. In addition, having an important
non-parental adult to talk was also associated with increased
character for youth who perceived their parents to be Highly
Involved, Authoritarian or Uninvolved and increased caring
for youth in the Integrative, Authoritarian and Uninvolved
profiles. Having an important non-parental adult to talk to
about some or all of their problems was also associated with
increased competence for youth in the Integrative,
Authoritarian, and Uninvolved profiles. In contrast, for
participants in the Permissive profile having an adult to talk
to about some or all problems was associated with decreased
confidence, character, caring, and competence.
Discussion
Relational developmental systems models emphasize that
development is embodied within a system involving cross-
context relationships (e.g., Bronfenbrenner and Morris
2006; Overton 2013). Little research, however, has
appraised the simultaneous effects of multiple contexts for
indices of healthy and positive development in adolescence.
In the present study, we used relational developmental sys-
tems thinking to examine youth-reported relationships with
both parents and important non-parental adults as they
coalescence within four years of high school to predict
positive youth outcomes as indexed by the Five Cs of PYD
(Bowers et al. 2010; Lerner et al. 2005). We tested several
hypotheses to describe and explain the relationships among
J Youth Adolescence (2014) 43:897–918 911
123

these contextual resources for youth PYD outcomes with
the aim to extend earlier work on parenting styles (e.g.,
Baumrind 1991; Steinberg et al. 1991; Gray and Steinberg
1999; Maccoby and Martin 1983) and important non-
parental adults (Bowers et al. 2012; DuBois and Silverthorn
2005a,b).
As we expected, several profiles of perceived parenting
that were similar to parenting ‘‘styles’’ identified previ-
ously were identified (e.g., Baumrind 1991; Maccoby and
Martin 1983). At all four waves, we identified parenting
profiles analogous to authoritarian and authoritative/highly
involved types. In addition, we identified an uninvolved
parenting profile at three grades and a permissive parenting
profile at two grades. The appearance of these two profiles
at later grades is consistent with research indicating that
parents generally decrease their monitoring of adolescents
as they mature (McElhaney et al. 2009). In turn, we also
identified several parenting profiles in the current study that
have not been described in earlier studies. These novel
profiles are most likely due to our inclusion of parenting
characteristics concerning the home (warmth), school
(involvement), and out-of-school time (monitoring) con-
texts. Most parenting typology work (e.g., Baumrind 1991;
Maccoby and Martin 1983) includes indices of behavioral
control and warmth, but not educational involvement.
At all four waves, we identified an Integrative parenting
profile (relatively moderate parental school involvement,
monitoring, and warmth) as we hypothesized. We also
identified a School-Focused profile (Higher parental school
involvement, lower warmth and monitoring) at Grade 9 and
a Controlling parenting profile (relatively high parental
school involvement and monitoring, but lower warmth) at
Grade 12. Furthermore, analyses of the latent profile fre-
quencies indicated that the majority of youth in our sample
perceived their parents to exhibit a relatively positive
parenting style/profile. At each grade, many youth were
members of the Integrative profile. However, although we
expected that most youth would be associated with this
profile, most youth were actually members of the Author-
itative or Highly Involved parenting profiles. In turn, there
were theoretically expected differences in the Five Cs of
PYD according to profile membership; however, Integra-
tive parenting was not associated with higher PYD. In
general, youth who perceived their parents as being
Authoritative or Highly Involved consistently had higher
levels of PYD than the Integrative, Uninvolved, and
Authoritarian profiles at each wave.
These findings were consistent with prior work (e.g.,
Baumrind 1991; Maccoby and Martin 1983; Steinberg
2001), but not consistent with our hypotheses. As noted, we
expected that an Integrated profile, involving moderate
levels of each of the three assessed parenting attributes,
would be most frequently seen. However, this expectation
was not the case for any of the four grades. In turn, we
based this expectation on the view that such a profile would
reflect maximum plasticity, affording parents the capacity
to modulate their parenting behaviors to fit the time and
place wherein they were interacting with their children. We
expected, therefore, that youth having Integrative parents
would show the highest levels of PYD and also would be
most likely to report having an important non-parental
adult in their lives. Only partial support for these predic-
tions was found.
Of course, the methodology of this study could not
directly assess (1) whether this Integrative profile actually
involved more plastic attributes than other parenting profile
or, relatedly (2) whether parents whom we have labeled
integrated actually modulated their parenting across time
(the age of their children) and place (the situation wherein
parent–youth interaction occurred). Future longitudinal
research, involving direct behavioral assessments of par-
enting behaviors, will be needed to test the ideas about the
plasticity and enactment across time and place of the
profile we have termed integrative parenting.
The effect of parenting on PYD outcomes across the
waves was most evident for connection. Youth in parenting
profiles characterized by low perceived warmth from par-
ents (Uninvolved, Authoritarian) reported lower feelings of
connection to family, peers, and community. Parents who
are engaged in their children’s lives in a warm manner
seem to provide youth with the skills and motivation to
develop healthy relationships outside of the home.
School-Focused youth at Grade 9 had particularly low
PYD scores, but this profile was not present at any of the
other grades. Perhaps parents who were previously school-
focused as their children entered ninth grade also began to
monitor their maturing adolescents in other domains as
these youth sought autonomy in the form of new activities
and friends in later years of high school. The increased
monitoring practices of these school-focused parents may
have increased the likelihood that their parenting might be
deemed authoritarian or controlling at later grades. How-
ever, we were not able to test this hypothesis directly in our
work due to treating the sample as embedded within cross-
sectional grades.
The low C scores reported by the School-Focused youth
were also reflected in the low C scores of participants of
Authoritarian parents at Grades 10 and 11. Youth in the
Authoritarian parenting profile were actually doing worse
than youth in the Uninvolved parenting profile at Grade 10.
However in Grades 11 and 12, children of Uninvolved
parents were generally reporting the lowest C scores,
especially in terms of connection. Consistent with prior
research on parenting (Gray and Steinberg 1999), our
results suggest that (1) parents who monitor their children
at a relatively high level without also expressing a
912 J Youth Adolescence (2014) 43:897–918
123

commensurate level of warmth; and (2) parents who are
uninvolved or neglectful of their children, are most likely
to have children reporting lower PYD-related outcomes.
Therefore, the influence of parental school involvement did
not wane over time (Eccles and Harold 1996), but it was an
asset for positive development in adolescence (Hill et al.
2004), provided that parents also expressed warmth and
acceptance for their children.
In turn, youth who perceived their parents as permissive
were doing relatively well and were not significantly dif-
ferent on any of the Cs from youth with highly involved
parents at Grade 12. Permissive youth were significantly
better than Integrative, Controlling, and Uninvolved par-
ents on several PYD outcomes at Grade 12. These findings
suggest that as youth progress through adolescence, par-
enting marked by relatively high warmth and relatively low
involvement and monitoring is just as likely to be associ-
ated with positive outcomes as parenting marked by rela-
tively high warmth, involvement, and monitoring. These
findings echo the mixed findings associated with the
parental monitoring literature (Crouter et al. 1990; Peterson
et al. 1988), but are consistent with work showing parents
tend to decrease their oversight of adolescents as they
develop (McElhaney et al. 2009). Perhaps as adolescents
exhibit appropriate prosocial behaviors, parents are less
likely to monitor their behavior closely or be involved in
their schoolwork. Warmth seems to be the primary pre-
dictor of positive outcomes (Gray and Steinberg 1999), but
from a relational systems perspective, whether highly
involved parenting versus permissive parenting is benefi-
cial for PYD may depend on individual characteristics
(e.g., age, self-regulatory skills) or factors in the larger
ecological context (e.g., other social support). More work
is needed on what individual and contextual factors might
moderate the impact of these two very different parenting
styles on adolescent development; that is, for what youth in
what settings are these parenting styles most favorable? We
discuss findings below about the role of important non-
parental adults in moderating the effect of parenting on
PYD that may shed some light on possible contextual
factors that need to be considered.
Parenting profiles were also related to the likelihood of
having an important non-parental adult. Consistent with
prior work (Erickson et al. 2009; Hurd et al. 2013), positive
parenting profiles were associated with the likelihood of
reporting the presence of an important non-parental adult.
However, youth with Authoritative or Highly Involved
parenting were most likely to report an important non-
parental adult relationship, not youth with Integrative
parenting as we expected. Youth with higher contextual
resources in the parental domain (those youth in the
Authoritative and Highly Involved Parenting profiles) are
more likely to report having someone other than their
parents to talk to about some or all of their problems (Er-
ickson et al. 2009). Conversely, youth in School-Focused
(Grade 9), Authoritarian (Grade 10 and 11), and Unin-
volved parenting profiles (Grades 10 through 12) were
significantly less likely to report having an important non-
parental adult to talk to if they were having problems even
when compared to youth of Integrative parenting. Perhaps
youth in the Authoritative and Highly Involved profiles are
able to learn from their parents the knowledge, schema, and
skills needed to develop relationships with, and recruit
resources from, other adults around them. However, the
small profile sizes (\20 for some profiles) generated very
high odds ratios (e.g., for the Authoritarian profile in Grade
12), so these results should be interpreted with caution.
Finally, although youth with less optimal parenting were
less likely to report having an important non-parental adult,
we did find that parenting profile did moderate the rela-
tionship between having an important non-parental adult
and the Five Cs of PYD, such that having an important
non-parental adult seemed to matter more for youth with
certain profiles of parenting. These findings are consistent
with prior work that considered familial or parental and
important non-parental adult resources simultaneously
(e.g., Erickson et al. 2009; Hurd et al. 2013; Kogan and
Brody 2010). The moderating effect of parenting on the
relationship between important non-parental adults and the
Five Cs of PYD depended on the PYD outcome of interest
and the type of parenting youth reported. Therefore, the
interaction between parenting and important non-parental
adult relationships could have had a complementary
(combined positive influence of resources on an outcome);
compensatory (a positive relationship buffered the effect of
a poor relationship); or detrimental (combined negative
influence of resources) effect.
In general, there was a positive effect of an important
non-parental adult on connection for most youth, except for
those youth in the Permissive parenting profile. This gen-
eral finding is consistent with Rhodes (2005) contention
that positive relationships with important non-parental
adults may help youth learn how to have more positive
interactions with others. When we considered youth reports
of having an important non-parental adult to talk to about
some or all of their problems, we also found that were also
stronger results for Character and somewhat for Compe-
tence and Confidence. Youth from authoritative and highly
involved parenting profiles benefitted from having an
important non-parental adult they could talk to about their
problems. When these youth reported having an important
non-parental adult to talk to about all of their problems,
they reported significantly higher Character at all four
grades and Confidence in Grades 9, 11, and 12.
Having an important non-parental adult to talk to about
their problems also served as a compensatory resource for
J Youth Adolescence (2014) 43:897–918 913
123

many youth. Youth who were identified as having rela-
tively problematic parenting profiles—School-focused,
Authoritarian, Uninvolved, and Controlling—reported
higher levels of Character when they had an important non-
parental adult to talk to about their problems. These results
are consistent with prior work indicating that positive non-
parental adult involvement is especially critical for youth
who often do not report having these non-parental supports
(e.g., Erickson et al. 2009). These results also suggest that
when important non-parental adults are there to listen to
youth who do not see their parents as warm or involved in
their lives, youth are more likely to have integrity, be
honest, follow rules, value diversity, and be socially con-
scious. These results are consistent with earlier work
linking the quantity of important non-parental adult rela-
tionships and the emotional closeness of important non-
parental adult relationships to character (Bowers et al.
2012). Future work should disentangle the effects of
youth–adult relationships on specific aspects of character
as these studies have both used a composite index of
character.
Finally, these analyses also indicated there were several
‘‘contrary’’ results, that is, results in which the moderating
effects were negative.When youth reported having an
important non-parental adult to talk to about their prob-
lems, they also reported decreased scores on several of the
Five Cs. These patterns were only found among less
favorable parenting profiles—School-Focused at Grade 9,
Uninvolved at Grade 11, and Permissive at Grades 11 and
12—which may indicate that the characteristics of the non-
parental adults themselves are particularly important for
these youth. Of particular note were the effects of having
an important non-parental adult on the Confidence, Char-
acter and Caring of youth with Permissive parents. The
results from Table 2indicated that youth in the Permissive
profile were generally doing well in terms of the Five Cs.
However, youth outcomes differed when one considered
the larger adult milieu. Permissive parents might not have
any influence over who their child selects as an important
non-parental adult, and therefore, these youth might be
vulnerable to the effects of ‘‘poor’’ important non-parental
adults or role models who may be themselves depressed or
engage in problem behaviors (Haddad et al. 2011). Another
possibility is that since these adolescents have been granted
freedom by their parents generally seen as normative
(McElhaney et al. 2009), having a non-parental adult
available may negatively affect youth autonomy (Peterson
et al. 1988). Again there were many analyses where the
coefficients were large but non-significant, most likely due
to small sample sizes within many of the profiles. The
negative interactive effects may also result from these
small sample sizes and associated decreased variation. The
parameter estimates are much more stable in the large
groups, so those patterns of results should be interpreted
with caution.
Although the present study provided insight into the
nuanced relationship between parental and non-parental
resources in predicting adolescent outcomes, there were
several limitations to consider when interpreting our
results. First, the results were based on four separate cross-
sectional waves of the 4-H Study of PYD rather than
considering the joint-effect of cross-contextual resources
on the positive and healthy development of youth. The
decision to consider the relationships among parenting,
important non-parental adults, and PYD concurrently
rather than longitudinally was due to the complexity of the
models considered in conjunction with the low proportion
of overlap in participation across waves. We were not able
to conduct latent transitional analyses as they were often
not enough youth in consecutive waves to examine whether
or not they remained in the same parenting profile from
wave to wave. Future research should investigate these
patterns longitudinally to determine how the coalescence of
familial and extra-familial relationships influences youth
development across adolescence. In addition, the small
sizes of some profiles within waves give us caution when
making interpretations.
The 4-H Study of PYD includes youth from a diversity
of demographic, economic, and social backgrounds; how-
ever, the participants generally report positive relationships
with family, friends, and community and are, overall, doing
relatively well in terms of healthy functioning. Future work
should examine parental and non-parental influences at the
extremes of parenting, for example in families marked by
overparenting or ‘‘helicopter’’ parenting (Padilla-Walker
and Nelson 2012) as well as those families marked by
neglect and abuse.
In addition, while the 4-H Study sample is diverse, the
sample is largely homogenous (over 60 % White/Cauca-
sian and over 60 % female). Youth SES and ethnic/racial
background may moderate the effect of parenting on youth
outcomes as authoritarian parenting may be beneficial for
minority and low-resourced youth (Brody and Flor 1998;
Furstenberg et al. 1999; Hill et al. 2003). However, model
complexity as well as the small profile numbers at some
grades limited our ability to test whether these relationships
differed in relationship to demographics. Future work
could test these models with larger and more diverse
samples.
The wording of the important non-parental adult item is
also an issue when interpreting the results. Without pro-
viding clearer criteria for what constituted an important
non-parental adult, or asking youth to identify their rela-
tionship to the important non-parental adult (e.g., coach,
aunt, teacher), the youth could have been referring to a
mentor whom they were assigned through a formal
914 J Youth Adolescence (2014) 43:897–918
123

mentoring program; formal mentors are not traditionally
considered important non-parental adults. Taking these
limitations into account, these findings should be inter-
preted with caution. Who a youth is designating as an
important non-parental adult might explain a great deal
about the magnitude of the impact of this person on PYD.
Future research should allow participants to note the
identity of the important non-parental adult to whom they
are responding. These considerations will allow researchers
to disentangle the effects of each type of relationship.
Finally, future work should include additional measures to
index the specific quality of the important non-parental
adult relationship. The present paper only included a gen-
eral item measuring important non-parental adult presence.
However, prior work has indicated that specific qualities of
the important non-parental adult relationship (Bowers et al.
2012; Dubois and Silverthorn 2005b) and characteristics of
the adults themselves (e.g., Haddad et al. 2011) are influ-
ential in predicting youth outcomes.
Conclusion
The present findings could be helpful to both parents and
practitioners. The findings are consistent with the wisdom
that parents should work to develop relationships with their
children that are marked by warmth and acceptance, but
also with the knowledge of where their children go and
with whom they associate. The results also indicate that
being involved with young people’s academic life and
school is only detrimental to youth outcomes when that
involvement is parents’ only concern. Engagement in the
multiple domains of a young person’s life is associated
with general youth well-being.
The results suggest that a key factor in promoting young
people’s positive development may be to increase their
connections to, and experiences with, important non-
parental adults. Not only do we want to connect youth to
non-parental adults, but we also want those adults to build
the skills to develop a deeper relationship with young
people. Youth need to see important non-parental adults as
a problem-solving resource or as an ‘‘open ear.’’ Therefore,
out-of-school-time and youth development programs
should include training and support for youth-serving
professionals so that they learn the best ways to develop
positive relationships with young people that enable them
to have these characteristics.
While important non-parental adults were generally
beneficial to all youth regardless of parenting, identifying
important non-parental adults is especially important for
youth whose parents may not be fully engaged in their
lives, as these youth were more likely to report not having
an important non-parental adult. In turn, when these youth
had an important non-parental adult in their lives, the
effects were greater than for youth from more resource-
advantaged backgrounds (see too Erickson et al. 2009).
The results also suggest that character education programs
that are developed for ‘‘these at-risk’’ youth should include
a mentoring component or should work to engage other
adults in the young person’s life as important non-parental
adults generally benefitted these young people’s character.
Important non-parental adult relationships were most
consistently related to youth connection. Therefore,
important non-parental adults appear to open up the doors
for improved youth relationships with families, peers, their
school, and their community (Rhodes 2005). The social
capital (Coleman 1988) provided by these adults may
provide youth the assets they need to develop the mutually
beneficial person/?context relationships that define
positive development.
Finally, this research recognizes but cannot examine
whether the characteristics of children can elicit different
behaviors in their parents and other caregivers (Bell 1968;
Lewis and Rosenblum 1974; Thomas et al. 1963). By
affecting those who are seeking to affect them, children are
agents (producers) of their own development (Lerner 1982;
Lerner and Busch-Rossnagel 1981). Parent or adult rearing,
as well as child rearing, exists (Lerner 2004). Future work
framed from relational developmental systems models
should include both youth and parent characteristics to
explain how youth–adult relationships help to define
development more fully.
Acknowledgments This research was supported in part by Grants
from the National 4-H Council, the Altria Corporation, the Thrive
Foundation for Youth, and the John Templeton Foundation.
Authors’ Contributions E.B. conceived of the study, drafted the
manuscript, and performed interpretation of the data; S.J. performed
the statistical analysis and interpretation of the data; M.B., S.G., and
D.W. helped to draft the manuscript. J.L. and R.L. conceived of the
study, and participated in its design and coordination, and helped to
draft the manuscript. All authors read and approved the final
manuscript.
References
Asparouhov, T., & Muthe
´n, B. (2013). Auxiliary variables in mixture
modeling: A 3-step approach using MPlus. MPlus Web Notes:
No.14. Retrieved from https://www.statmodel.com/examples/
webnotes/AuxMixture_submitted_corrected_webnote.pdf.
Baumrind, D. (1978). Parental disciplinary patterns and social
competence in children. Youth & Society, 9(3), 239–276.
Baumrind, D. (1991). Parenting styles and adolescent development.
In R. M. Lerner, A. Petersen, & J. Brooks-Gunn (Eds.),
Encyclopedia of adolescence (pp. 746–758). Garland Publishing,
Inc.
Bell, R. Q. (1968). A reinterpretation of the direction of effects in
studies of socialization. Psychological Review, 75, 81–95.
J Youth Adolescence (2014) 43:897–918 915
123

Benson, P. L., Scales, P. C., Hamilton, S. F., & Sesma, A. (2006).
Positive youth development: Theory, research, and applications.
In W. Damon & R. Lerner (Eds.), Handbook of child psychol-
ogy: Theoretical models of human development (Vol. 1,
pp. 894–941). Hoboken, NJ: Wiley.
Benson, P. L., Scales, P. C., & Syvertsen, A. K. (2011). The
contribution of the developmental assets framework to positive
youth development theory and practice. In R. M. Lerner, J.
V. Lerner, & J. B. Benson (Eds.), Advances in Child Develop-
ment and Behavior, 41, 195–228.
Bergman, L. R., & Magnusson, D. (1997). A person-oriented
approach in research on developmental psychopathology. Devel-
opment and Psychopathology, 9, 291–319.
Bowers, E. P., Geldhof, G. J., Johnson, S. K., Lerner, J. V., & Lerner,
R. M. (in press). Thriving across the adolescent years: A view of
the issues. Journal of Youth and Adolescence.
Bowers, E. P., Geldhof, G., Schmid, K. L., Napolitano, C. M., Minor,
K., & Lerner, J. V. (2012). Relationships with important
nonparental adults and positive youth development: An exam-
ination of youth self-regulatory strengths as mediators. Research
in Human Development, 9(4), 298–316.
Bowers, E. P., Gestsdo
´ttir, S., Geldhof, J., Nikitin, J., von Eye, A., &
Lerner, R. M. (2011a). Developmental trajectories of intentional
self regulation in adolescence: The role of parenting and
implications for positive and problematic outcomes among
diverse youth. Journal of Adolescence, 34(6), 1193–1206.
Bowers, E. P., Li, Y., Kiely, M. K., Brittian, A., Lerner, J. V., &
Lerner, R. M. (2010). The five Cs model of positive youth
development: A longitudinal analysis of confirmatory factor
structure and measurement invariance. Journal of Youth and
Adolescence, 39(7), 720–735.
Bowers, E. P., von Eye, A., Lerner, J. V., Arbeit, M. R., Weiner, M.
B., Chase, P., et al. (2011b). The role of ecological assets in
positive and problematic developmental trajectories. Journal of
Adolescence, 34(6), 1151–1166.
Brandtsta
¨dter, J. (1998). Action perspectives on human development.
In W. Damon & R. M. Lerner (Eds.) & R. M. Lerner (Vol. Ed.),
Handbook of child psychology: Theoretical models of human
development (Vol. 1, pp. 807–863). New York: Wiley.
Brody, G. H., & Flor, D. L. (1998). Maternal resources, parenting
practices, and child competence in rural, single-parent African
American families. Child Development,69(3), 803–816.
Bronfenbrenner, U. (1979). Contexts of child rearing: Problems and
prospects. American Psychologist, 34(10), 844–850.
Bronfenbrenner, U., & Morris, P. A. (2006). The bioecological model
of human development. In R. M. Lerner & W. Damon (Eds.),
Handbook of child psychology, Theoretical models of human
development (Vol. 1, pp. 793–828). Hoboken, NJ: Wiley.
Coleman, J. S. (1988). Social capital in the creation of human capital.
American Journal of Sociology, 94, S95–S120.
Collins, L. M., & Lanza, S. T. (2010). Latent class and latent
transition analysis: With applications in the social, behavioral,
and health sciences. Hoboken, NJ: Wiley.
Co
ˆte
´, J. E. (2009). Identity formation and self-development in
adolescence. In R. M. Lerner & L. Steinberg (Eds.), Handbook of
adolescent psychology: Individual bases of adolescent develop-
ment (Vol. 1, pp. 266–304). Hoboken, NJ US: John Wiley &
Sons Inc.
Crouter, A. C., & Head, M. R. (2002). Parental monitoring and
knowledge of children. In M. Bornstein (Ed.), Handbook of
parenting: Becoming and being a parent (Vol. 3, pp. 461–483).
Mahwah, NJ: Erlbaum.
Crouter, A. C., MacDermid, S. M., McHale, S. M., & Perry-Jenkins,
M. (1990). Parental monitoring and perceptions of children’s
school performance and conduct in dual-and single-earner
families. Developmental Psychology, 26(4), 649.
Dishion, T. J., & McMahon, R. J. (1998). Parental monitoring and the
prevention of child and adolescent problem behavior: A
conceptual and empirical formulation. Clinical Child and Family
Psychology Review, 1(1), 61–75.
DuBois, D. L., & Silverthorn, N. (2005a). Characteristics of natural
mentoring relationships and adolescent adjustment: Evidence
from a national study. Journal of Primary Prevention, 26, 69–92.
DuBois, D. L., & Silverthorn, N. (2005b). Natural mentoring
relationships and adolescent health: Evidence from a national
study. American Journal of Public Health, 95, 518–524.
Eccles, J. S., & Harold, R. D. (1996). Family involvement in
children’s and adolescents’ schooling. In A. Booth & J. F. Dunn
(Eds.), Family–school links: How do they affect educational
outcomes? (pp. 3–34). Hillsdale, NJ England: Lawrence Erlbaum
Associates Inc.
Erickson, L. D., McDonald, S., & Elder, G. H. (2009). Informal
mentors and education: Complementary or compensatory
resources? Sociology of Education, 82(4), 344–367.
Finkenauer, C., Engels, R. C. M. E., & Meeus, W. (2002). Keeping
secrets from parents: Advantages and disadvantages of secrecy
in adolescence. Journal of Youth and Adolescence, 31, 123–136.
Flouri, E., & Buchanan, A. (2002). What predicts good relationships
with parents in adolescence and partners in adult life: Findings
from the 1958 British birth cohort. Journal of Family Psychol-
ogy, 16(2), 186.
Flouri, E., & Buchanan, A. (2004). Early father’s and mother’s
involvement and child’s later educational outcomes. British
Journal of Educational Psychology, 74(2), 141–153.
Furstenberg, F. F., Jr., Cook, T. D., Eccles, J., Elder, G. H., Jr., &
Sameroff, A. (1999). Managing to make it: Urban families and
adolescent success. Chicago: The University of Chicago Press.
Gray, M., & Steinberg, L. (1999). Unpacking authoritative parenting:
Reassessing a multidimensional construct. Journal of Marriage
and the Family, 61(3), 574–587.
Greenberger, E., Chen, C., & Beam, M. R. (1998). The role of ‘‘very
important’’ nonparental adults in adolescent development. Jour-
nal of Youth and Adolescence, 27(3), 321–343.
Haddad, E., Chen, C., & Greenberger, E. (2011). The role of
important non-parental adults (VIPs) in the lives of older
adolescents: A comparison of three ethnic groups. Journal of
Youth and Adolescence, 40(3), 310–319.
Hayes, L., Hudson, A., & Matthews, J. (2004). Parental monitoring
behaviors: A model of rules, supervision, and conflict. Behavior
Therapy, 35, 587–604.
Hill, N. E., Bush, K. R., & Roosa, M. W. (2003). Parenting and family
socialization strategies and children’s mental health: Low-
income Mexican-American and Euro-American mothers and
children. Child Development,74, 189–204.
Hill, N. E., Castellino, D. R., Lansford, J. E., Nowlin, P., Dodge, K.
A., Bates, J. E., et al. (2004). Parent academic involvement as
related to school behavior, achievement, and aspirations:
Demographic variations across adolescence. Child Development,
75(5), 1491–1509.
Hill, N. E., & Tyson, D. F. (2009). Parental involvement in middle
school: A meta-analytic assessment of the strategies that
promote achievement. Developmental Psychology, 45(3), 740.
Hurd, N. M., Varner, F. A., & Rowley, S. J. (2013). Involved-vigilant
parenting and socio-emotional well-being among black youth:
The moderating influence of natural mentoring relationships.
Journal of Youth and Adolescence, 42, 1583–1595.
Hurd, N. M., Zimmerman, M. A., & Reischl, T. M. (2011). Role
model behavior and youth violence: A study of positive and
negative effects. The Journal of Early Adolescence, 31(2),
323–354.
Kerr, M., & Stattin, H. (2000). What parents know, how they know it,
and several forms of adolescent adjustment: Further support for a
916 J Youth Adolescence (2014) 43:897–918
123

reinterpretation of monitoring. Developmental Psychology,
36(3), 366–380.
Kerr, M., Stattin, H., & Burk, W. J. (2010). A reinterpretation of
parental monitoring in longitudinal perspective. Journal of
Research on Adolescence, 20(1), 39–64.
Kogan, S. M., & Brody, G. H. (2010). Linking parenting and informal
mentor processes to depressive symptoms among rural African
American young adult men. Cultural Diversity & Ethnic
Minority Psychology, 16(3), 299–306.
Kogan, S. M., Brody, G. H., & Chen, Y. (2011). Natural mentor
processes deter externalizing problems among rural African
American emerging adults: A prospective analysis. American
Journal of Community Psychology, 48, 272–283.
Laursen, B., & Collins, W. (2009). Parent–child relationships during
adolescence. In R. M. Lerner & L. Steinberg (Eds.), Handbook of
adolescent psychology: Contextual influences on adolescent devel-
opment (Vol. 2, pp. 3–42). Hoboken, NJ US: John Wiley & Sons Inc.
Leffert, N., Benson, P. L., Scales, P. C., Sharma, A. U., Drake, D. R.,
& Blyth, D. A. (1998). Developmental assets: Measurement and
prediction of risk behaviors among adolescents. Applied Devel-
opmental Science, 2(4), 209–230.
Lerner, R. M. (1982). Children and adolescents as producers of their
own development. Developmental Review, 2, 342–370.
Lerner, R. M. (1984). On the nature of human plasticity. New York:
Cambridge University Press.
Lerner, R. M. (2004). Liberty: Thriving and civic engagement among
America’s youth. Thousand Oaks, CA: Sage Publications.
Lerner, R. M., & Benson, J. B. (2013). Introduction: Embodiment and
epigenesis: A view of the issues. In R. M. Lerner & J. B. Benson
(Eds.), Advances in Child Development and Behavior: Embodi-
ment and epigenesis: Theoretical and methodological issues in
understanding the role of biology within the relational develop-
mental system (pp. 1–20). London, UK: Elsevier.
Lerner, J. V., Bowers, E. P., Minor, K., Boyd, M. J., Mueller, M. K.,
Schmid, K. L., Napolitano, C. M., Lewin-Bizan, S., & Lerner, R.
M. (2013). Positive youth development: Processes, philosophies,
and programs. In R. M. Lerner, M. A. Easterbrooks, & J. Mistry
(Eds.); I. B. Weiner (Editor-in-Chief). Handbook of psychology:
Developmental psychology (Vol. 6, pp. 365–392). Hoboken, NJ:
Wiley.
Lerner, R. M., & Busch-Rossnagel, N. A. (Eds.). (1981). Individuals
as producers of their development: A life-span perspective. New
York: Academic Press.
Lerner, R. M., Lerner, J. V., Almerigi, J., Theokas, C., Phelps, E.,
Gestsdottir, S., et al. (2005). Positive youth development,
participation in community youth development programs, and
community contributions of fifth grade adolescents: Findings
from the first wave of the 4-H Study of Positive Youth
Development. Journal of Early Adolescence, 25(1), 17–71.
Lerner, R. M., Lerner, J. V., Bowers, E. P., & Geldhof, G. J. (in
press). Positive youth development: A relational developmental
systems model. In Overton, W. F., & Molenaar, P. C. M. (Eds.);
R. M. Lerner (Editor-in-Chief). Handbook of child psychology
and developmental science, 7
th
edition. Volume 1: Theory and
method. New York: Wiley.
Lerner, R. M., Lerner, J. V., Bowers, E. P., Lewin-Bizan, S., & von
Eye, A. (Eds.) (2011). Individual and contextual bases of
thriving in adolescence: Findings from the 4-H Study of Positive
Youth Development. Journal of Adolescence, 34(6), 1107–1227.
Lerner, R. M., von Eye, A., Lerner, J. V., & Lewin-Bizan, S. (2009).
Exploring the foundations and functions of adolescent thriving within
the 4-H study of positive youth development: A view of the issues.
Journal of Applied Developmental Psychology, 30(5), 567–647.
Lerner, R. M., von Eye, A., Lerner, J. V., Lewin-Bizan, S., & Bowers,
E. P. (Eds.). (2010). The meaning and measurement of thriving
in adolescence: Findings from the 4-H Study of Positive Youth
Development. Journal of Youth and Adolescence, 39(7),
707–842.
Lewin-Bizan, S., Bowers, E., & Lerner, R. M. (2010). One good thing
leads to another: Cascades of positive youth development among
American adolescents. Development and Psychopathology, 22,
759–770.
Lewis, M., & Rosenblum, L. A. (1974). The effect of the infant on its
caregiver. New York: Wiley.
Li, J., & Julian, M. (2012). Developing relationships as the active
ingredient: A unifying working hypothesis of ‘‘what works’’
across intervention settings. American Journal of Orthopsychi-
atry, 82(2), 157–166.
Li, Y., Lerner, J., & Lerner, R. (2010). Personal and ecological assets
and academic competence in early adolescence: The mediating
role of school engagement. Journal of Youth and Adolescence,
39(7), 801–815.
Lo, Y., Mendell, N. R., & Rubin, D. B. (2001). Testing the number of
components in a normal mixture. Biometrika, 88, 767–778.
Maccoby, E. E., & Martin, J. A. (1983). Socialization in the context of
the family: Parent–child interaction. In P. H. Mussen (Ed.),
Handbook of child psychology (Vol. 4, pp. 1–101)., Socializa-
tion, personality, and social development New York: Wiley.
Marcia, J. E. (1980). Identity in adolescence. In J. Adelson (Ed.),
Handbook of adolescent psychology (pp. 159–187). New York:
Wiley & Sons.
McElhaney, K., Allen, J., Stephenson, C., & Hare, A. (2009).
Attachment and autonomy during adolescence. In R. M. Lerner
& L. Steinberg (Eds.), Handbook of adolescent psychology (Vol.
1, pp. 358–403). New York, NY: Plenum Press.
McLachlan, G. J., & Peel, D. (2000). Finite mixture models. New
York: Wiley.
Monitoring the Future. (2000). National survey on drug use,
1975–2000. Bethesda, MD: National Institute on Drug Abuse.
Muthe
´n, B. O., & Muthe
´n, L. K. (2000). Integrating person-centered
and variable-centered analyses: Growth mixture modeling with
latent trajectory classes. Alcoholism, Clinical and Experimental
Research, 24(6), 882–891.
Nesselroade, J. R. (1988). Some implications of the trait-state
distinction for the study of development across the life span:
The case of personality research. In P. B. Baltes, D. L. Feath-
erman, & R. M. Lerner (Eds.), Life-span development and
behavior (Vol. 8, pp. 163–189). Hillsdale, NJ: Lawrence
Erlbaum Associates Inc.
Nylund, K. L., Asparouhov, T., & Muthe
´n, B. O. (2007). Deciding on
the number of classes in latent class analysis and growth mixture
modeling: A Monte Carlo simulation study. Structural Equation
Modeling, 14(4), 535–569.
Overton, W. F. (2013). Relationism and relational-developmental-
systems: A paradigm for developmental science in the post-
Cartesian era. In R. M. Lerner & J. B. Benson (Eds.), Advances
in Child Development and Behavior: Embodiment and epigen-
esis: Theoretical and methodological issues in understanding the
role of biology within the relational developmental system (pp.
21–64). London, UK: Elsevier.
Overton, W. F., & Mu
¨ller, U. (2013). Development across the life
span: Philosophy, concepts, theory. In R. M. Lerner, M.
A. Easterbrooks, & J. Mistry (Eds.); I. B. Weiner (Editor-in-
Chief). Handbook of psychology: Developmental psychology
(Vol. 6, pp. 19–58). New York, NY: Wiley.
Padilla-Walker, L., & Nelson, L. (2012). Black hawk down?
Establishing helicopter parenting as a distinct construct from
other forms of parental control during emerging adulthood.
Journal of Adolescence, 35(5), 1177–1190.
Patterson, G., & Stouthamer-Loeber, M. (1984). The correlation of
family management practices and delinquency. Child Develop-
ment, 55(4), 1299–1307.
J Youth Adolescence (2014) 43:897–918 917
123

Peterson, C., Seligman, M. E., & Vaillant, G. E. (1988). Pessimistic
explanatory style is a risk factor for physical illness: A thirty-
five-year longitudinal study. Journal of Personality and Social
Psychology, 55(1), 23–27.
Rhodes, J. E. (2005). A model of youth mentoring. In D. L. DuBois &
M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 30–43).
Thousand Oaks, CA: Sage.
Rhodes, J. E., Ebert, L., & Fischer, K. (1992). Natural mentors: An
overlooked resource in the social networks of young, African
American mothers. American Journal of Community Psychol-
ogy, 20(4), 445–461.
Rhodes, J. E., & Lowe, S. R. (2009). Mentoring in adolescence. In R.
M. Lerner & L. Steinberg (Eds.), Handbook of adolescent
psychology: Contextual influences on adolescent development
(Vol. 2, pp. 152–190). Hoboken, NJ: Wiley.
Rodin, J. (1990). Control by any other name: Definitions, concepts,
and processes. In J. Rodin, C. Schooler, & K. Schaie (Eds.), Self-
directedness: Cause and effects throughout the life course (pp.
1–17). Hillsdale, NJ England: Lawrence Erlbaum Associates Inc.
Schludermann, E., & Schludermann, S. (1970). Replicability of
factors in children’s report of parent behavior (CRPBI). The
Journal of Psychology, 76(2), 239–249.
Schwarz, G. (1978). Estimating the dimension of a model. The Annals
of Statistics, 6(2), 461–464.
Schwarz, J., Barton-Henry, M. L., & Pruzinsky, T. (1985). Assessing
child-rearing behaviors: A comparison of ratings made by
mother, father, child, and sibling on the CRPBI. Child Devel-
opment, 56(2), 462–479.
Sears, R. R., Maccoby, E. E., & Levin, H. (1957). Patterns of child
rearing. Evanston, IL: Row, Peterson.
Small, S. A., & Kerns, D. (1993). Unwanted sexual activity among
peers during early and middle adolescence: Incidence and risk
factors. Journal of Marriage and the Family, 55(4), 941–952.
Smetana, J. G., Campione-Barr, N., & Metzger, A. (2006). Adoles-
cent development in interpersonal and societal contexts. Annual
Review of Psychology, 57, 255–284.
Steinberg, L. (2001). We know some things: Parent–adolescent
relationships in retrospect and prospect. Journal of Research on
Adolescence, 11(1), 1–19.
Steinberg, L., Mounts, N., Lamborn, S., & Dornbusch, S. (1991).
Authoritative parenting and adolescent adjustment across various
ecological niches. Journal of Research on Adolescence, 1,
19–36.
Syme, S. (1990). Control and health: An epidemiological perspective.
In J. Rodin, C. Schooler, & K. Schaie (Eds.), Self-directedness:
Cause and effects throughout the life course (pp. 213–229).
Hillsdale, NJ England: Lawrence Erlbaum Associates.
Theokas, C., Almerigi, J. B., Dowling, E. M., Benson, P. L., Scales, P.
C., Von Eye, A., et al. (2005). Conceptualizing and modeling
individual and ecological asset components of thriving in early
adolescence. Journal of Early Adolescence, 25(1), 113–143.
Theokas, C., & Lerner, R. M. (2006). Observed ecological assets in
families, schools, and neighborhoods: Conceptualization, mea-
surement, and relations with positive and negative developmen-
tal outcomes. Applied Developmental Science, 10(2), 61–74.
Thomas, A., Chess, S., Birch, H. G., Hertzig, M. E., & Korn, S.
(1963). Behavioral individuality in early childhood. New York:
New York University Press.
Zimmerman, M. A., Bingenheimer, J. B., & Notaro, P. C. (2002).
Natural mentors and adolescent resiliency: A study with urban
youth. American Journal of Community Psychology, 30(2),
221–243.
Edmond P. Bowers is a Research Assistant Professor at Tufts
University. He received his doctorate in Applied Developmental and
Educational Psychology from Boston College. His research interests
include a focus on important non-parental adults and the various ways
that these adults can promote positive development in young people.
Sara K. Johnson is a research assistant professor at the Institute for
Applied Research in Youth Development at Tufts University. Her
work integrates the study of positive youth development, civic
engagement, and identity development using mixed-methods
approaches.
Mary H. Buckingham is a doctoral student at the Institute for
Applied Research in Youth Development at Tufts University. Her
research interests involve the study of positive youth development
and parenting.
Santiago Gasca is a graduate student at the Institute for Applied
Research in Youth Development at Tufts University. His research
interests involve the study of positive youth development and
developmental psychopathology.
Daniel J. A. Warren is a doctoral student at the Institute for Applied
Research in Youth Development at Tufts University. His research
interests involve the study of positive youth development and
educational programs.
Jacqueline V. Lerner is a Professor of Applied Developmental and
Educational Psychology in the Department of Counseling, Applied
Developmental and Educational Psychology at Boston College. She
received her Ph.D. in Educational Psychology from the Pennsylvania
State University. Her research interests focus on the development of
children and adolescents in the contexts of family, school and
community, and the embedded relationships in these contexts that
contribute to positive development.
Richard M. Lerner is the Bergstrom Chair in Applied Develop-
mental Science and the Director of the Institute for Applied Research
in Youth Development at Tufts University. He received his Ph.D. in
developmental psychology from the City University of New York.
His work integrates the study of public policies and community-based
programs with the promotion of positive youth development and
youth contributions to civil society.
918 J Youth Adolescence (2014) 43:897–918
123