SPECIAL SECTION ARTICLE
A 10-year follow-up twin and sibling study of marital conflict,
monitoring, siblings, and peers
JENAE M. NEIDERHISER,aKRISTINE MARCEAU,aAND DAVID REISSb
aPennsylvania State University; andbYale University Child Study Center
This study examined genetic and environmental influences on associations among marital conflict about the child, parental monitoring, sibling relationship
from same-sex sibling pairs and parents when the siblings were 10–18 years old (M ¼ 14.5 and 12.9 years for Child 1 and Child 2, respectively) and
20–35 years old (M ¼ 26.8 and 25.5 years for Child 1 and Child 2, respectively). Findings indicate four factors that explain the initiation of illegal drug use:
two shaped by genetic influences and two shaped byenvironments shared bysiblings. The two geneticallyshaped factors probably have distinct mechanisms:
one a child-initiated coercive process in the family and the other parent and peer processes shaped by the child’s disclosure. The environmentally
influenced factors seem distinctively shaped by poor parental monitoring of both sibs and the effects of siblings on each other’s deviancy.
The links between family climate, peer characteristics, and
drug use have been the focus of much research over the
past several decades. Parenting, especially parental knowl-
edge about their children’s friends and whereabouts (i.e.,
monitoring), is the family measure most thoroughly exam-
ined as a predictor of adolescent initiation and continued
use of drugs and related behaviors. The handful of studies
that have considered other aspects of the family like marital
conflict and hostility in sibling relationships have also found
evidence for links with problematic adolescent outcomes in-
cluding drug use. Outside of the family, peer characteristics
(i.e., delinquency) have well-established links with adoles-
cent behavior problems and drug use. However, there have
been fewer studies that have considered how these different
interpersonal relationships may work together in influencing
drug use, and there have not been satisfactory studies of
whether these influences reflect exogenous pressures acting
on the child or whether characteristics of the child initiate
and shape some or all of these apparent environmental risks.
To help clarify how these influences are balanced, and the
relative role of the child and exogenous social influences, it is
fortunate that there is a rapidly growing literature on the role
of genetic factors on interpersonal relationships and on the
covariation of interpersonal relationships and the adjustment
of family members. When children’s genes influence the ob-
jective features of their social environment or theirown expe-
rience of it, genotype–environment correlation is indicated.
The links between interpersonal relationships and adolescent
adolescent on his or her family and peers as well as on his or
her likelihood ofusingdrugs. Thecurrentreportwill examine
genetic and environmental influences on the covariation
among marital conflict, parental knowledge, sibling conflict,
and deviant peer group characteristics during adolescence
and initiation of drug use by young adulthood. The interper-
sonal relationships are considered in order of proximal rele-
vance to the adolescent’s likelihood of initiation of drug
use by young adulthood, with the most distal to the child
accounted for first (marital conflict) and the most proximal
to the child considered last (peer group characteristics). In or-
sample of twins and siblings varying in degree of genetic re-
latedness was used. Thus, in this report we will be able to
identify genotype–environment correlations among interper-
sonal relationships on the initiation of drug use.
Family Relationships and Drug Initiation
There is evidence from developmental and family process lit-
erature that marital conflict is linked with adolescent adjust-
Address correspondence and reprint requests to: Jenae M. Neiderhiser,
ing, University Park, PA 16802; E-mail: firstname.lastname@example.org.
The Nonshared Environment in Adolescent Development project was sup-
ported by National Institute of Mental Health Grant R01MH43373 (Time 1,
David Reiss, Principal Investigator [PI]) and Grant R01MH59014 (Time 3,
Jenae M. Neiderhiser, PI) and by the William T. Grant Foundation (Time
1, David Reiss, PI).
Development and Psychopathology 25 (2013), 133–149
#Cambridge University Press 2013
ment problems. Conflict between parents, especially conflict
that children experience directly, is associated with increased
internalizing, externalizing, and a host of other undesirable
outcomes in adolescents (e.g., Amato, 1991; Cummings,
2004; Harold & Conger, 1997; Rhoades et al., 2011). There
are, however, few studies that have examined the impact of
marital conflict on drug use in adolescents, with most studies
focusing on more general adjustment problems like external-
izing. There is evidence that impulse control, aggression, and
other externalizing behaviors predict initiation of drug use in
tween marital conflict and externalizing behaviors also sug-
gest an increased risk of initiation of drug use.
A number of possible mechanisms through which marital
conflict may be associated with child functioning have been
identified. There may be a direct effect of marital conflict
on the child, as described above. It is also likely that marital
conflict disrupts or impairs parenting, thus indirectly influ-
encing child adjustment through parental behaviors by, for
example, decreasing effective monitoring (e.g., Cummings,
1994; Gable, Belsky, & Crnic, 1992; Peterson & Zill, 1986).
A few studies have attempted to systematically examine the
ordering of effects among marital conflict, parenting, and
child adjustment. One such found that the effects of marital
by multiple types of parenting behaviors, including monitor-
ing, with some differences for mothers and fathers (Buehler,
Benson, & Gerard, 2006). Overall, it is clear that marital con-
flictis associatedwith child adjustment andthatparenting be-
haviors, including monitoring, at least partially mediate this
effect on children. Because adolescent drug initiation is an
adjustment problem highly comorbid with externalizing
problems(e.g., Disney,Elkins,McGue, &Iacono,1999;Hel-
strom, Bryan, Hutchison, Riggs, & Blechman, 2004; Molina
initiation via similar mechanisms, operating in part through
There are a plethora of studies that have found that more
effective parental monitoring is related to a reduced likeli-
hood to initiate drug use and related behaviors during adoles-
cence (Dishion & Tipsord, 2011; Lac & Crano, 2009). Stud-
ies examining initiation and continued drug use have found
parental monitoring to be one of the most predictive factors
in adolescent drug use (e.g., Chilcoat & Anthony, 1996;
Dishion, Capaldi, Spracklen, & Li, 1995). In general, these
are with, where they are, and when they will return (parental
izing behaviors, associate with fewer delinquent peers, and
are less likely to initiate use of drugs. This pattern of findings
has been shown for initiation of commonly used substances
like tobacco, alcohol, and marijuana (Crano, Siegel, Alvaro,
Lac, & Hemovich, 2008; Dishion et al., 1995; Dishion &
Loeber, 1985; Dorius, Bahr, Hoffmann, & Harmon, 2004;
Duncan, Duncan, Biglan, & Ary, 1998; Lac & Crano,
2009; Piko & Kovacs, 2010) as well as for use of other less
commonly used drugs, like ecstasy (Martins, Storr, Alex-
andre, & Chilcoat, 2008; Wu, Liu, & Fan, 2010) and inhal-
ants (Nonnemaker, Crankshaw, Shive, Hussin, & Farrelly,
2011). Although other parenting behaviors (i.e., hostility,
warmth, involvement, and coercive discipline) have also
been shown to predict adolescent substance use (e.g., Cheng
& Lo, 2010; Fletcher, Steinberg, & Williams-Wheeler, 2004;
Siebenbruner, Englund, Egeland, & Hudson, 2006), gener-
allyparental monitoring andknowledgecontinuestobeasig-
nificant predictor above and beyond other parenting behav-
iors (see Chen, Storr, & Anthony, 2005 for an exception to
this).Together,thisevidence highlightstheimportance ofpa-
rental monitoring in particularon the likelihood of children to
initiate a number of different substances. More effective pa-
rental monitoring is associated with a decreased likelihood
to initiate drugs.
What is not clear from the studies reported above is the di-
rection of effects: do parents who effectively monitor their
adolescent have more control, thus taking an active part in
preventing drug initiation, or is effective monitoring an ado-
lescent-driven construct with adolescents who are less likely
to initiate drug use more likely to tell their parents about their
whereabouts? This latter explanation is consistent with the
and that it is adolescent disclosure that predicts the likelihood
of delinquent behaviors, not parental monitoring or attempts
at control (Kerr & Stattin, 2000; Kerr, Stattin, & Burk, 2010;
Stattin & Kerr, 2000). Findings from genetically informed
designs indicate that there are genetic influences on parental
monitoring (e.g., Neiderhiser et al., 2004; Neiderhiser, Reiss,
Lichtenstein, Spotts, & Ganiban, 2007), thus supporting an
adolescent-driven mechanism of effect.
Siblings are another source of potential influence on ado-
lescent initiation of drug use. There have been a number of
studies that have examined deviancy training in sibling pairs
with clear evidence that having a deviant sibling increases an
adolescent’s risk of deviant behavior, including initiation of
substance use (e.g., Avenevoli & Merikangas, 2003; Comp-
ton, Snyder, Schrepferman, Bank, & Shortt, 2003; Natsuaki,
Ge, Reiss, & Neiderhiser, 2009; Rowe & Gulley, 1992). The
quality of the sibling relationship appears to be an important
factor for the similarityof siblings for several problematic be-
haviors. For example, siblings with close relationships were
more likely to have similar attitudes and experiences with
sexual behavior in a large sample of adolescent siblings
(McHale, Bissell, &Kim,2009).Atleastonestudyhasfound
no association between sibling relationship quality and sub-
stance use when sibling deviancy was accounted for (Storm-
shak, Comeau, & Shepard, 2004), suggesting that deviancy
training is a likely mechanism for associations between
sibling deviancy and initiation of drug use in adolescents.
In a different study not directly assessing drug use, hostile
sibling relationships predicted delinquency in boys and girls,
while warm sibling relationships predicted delinquency in
boys only (Slomkowski, Rende, Conger, Simons, & Conger,
J. M. Neiderhiser, K. Marceau, and D. Reiss
2001). These studies clearly indicate that siblings can exert
influence on the likelihood of adolescentsto initiate drug use.
ies that have found that the risk of drug use increases with the
number of peers who use (Brook, Kessler, & Cohen, 1999;
Ennett et al., 2006) and that the perceptions of peer use is a
strong predictor of adolescent use (Agrawal & Lynskey,
2007). Peers have also been implicated in deviancy training
such that friendships with deviant peers increase delin-
quency, substance use, violence, and adult maladjustment
(Dishion, McCord, & Poulin, 1999; Dishion & Owen, 2002).
There has been some debate about the direction of effects of
these associations. Is it that deviant peers train friends in de-
viance or do individuals at risk of drug use select deviant
peers? Studies that have attempted to disentangle the direction
tionship. For example, longitudinal studies suggest that both
peer drug use and participant drug use are highly stable over
time, with somewhat more evidence that peers influence use
of substances over time (Sieving, Perry, & Williams, 2000).
A study using a short-term longitudinal design found that
adolescents who were already using substances tended to se-
lect friends who also used and that these friends contributed to
changes in adolescent substance use over time (Poulin, Kies-
ner, Pedersen, & Dishion, 2011). An earlier study found that
both selection and socialization occurs in friendships and
marijuana use over time (Kandel, 1978). Thus, although
peers are clearly implicated in the initiation of drug use,
stable over time.
Integrating Across Marital Conflict, Parental
Monitoring, Sibling Relationships, and Peers
Although marital conflict, parental monitoring, sibling rela-
tionships, and peer characteristics have all been shown to
have a direct association with initiation of drug use and related
behaviors, as reviewed above, it is unlikely that any of these
constructs are operating to influence the adolescent in isola-
tilityonadolescentexternalizing problems is mediated in part
by parenting behaviors, including less effective monitoring
(e.g., Buehler et al., 2006; Cummings, 1994; Gable et al.,
1992; Peterson & Zill, 1986). However, no studies were iden-
tified linking marital quality and parenting to substance use.
drug use have found some evidence that parents have an
important influence on the types of friends with whom their
adolescent(s) associate. For example, ineffective parental
monitoring predicted involvement with a deviant peer net-
work that in turn was associated with a rise in substance
use in the transition to high school (Dishion et al., 1995).
There is also some evidence that associating with deviant
peers may mediate the effect of parenting on substance use
though at least one study found important differences for the
with parental monitoring moderating both associations (Kies-
ner et al., 2010). In other words, the impact of parenting be-
haviors, including monitoring, influences adolescent drug
use indirectly via peer characteristics. However, the likeli-
hood that an adolescent who uses drugs with his friends
will also usewhen alone is decreased when parental monitor-
ing is effective. Other studies of the links between parenting
related to peer group characteristics indirectly via adolescent
drug use and self-reliance (Brown, Mounts, Lamborn, &
Steinberg, 1993) and that poor parenting practices, including
ineffective monitoring, increase the influence of deviant
peers on adolescent drug use (Duncan et al., 1998; Kung &
Fewer studies have examined associations among sibling
relationships and peer characteristics. Peer and sibling sub-
stance use are better predictors of adolescent substance use
than parental alcohol use, and sibling substance use also pre-
dicts peer substance use (Windle, 2000). In other words, the
transmission of substance use within a family appears to be
lateral (via sibling influences) rather than vertical (via paren-
tal influences), and sibling factors predict and probably pre-
cede peer factors. In another study both sibling and peer
deviance directly predicted adolescent substance use, al-
though when both were included as predictors of change in
substance use over time, only sibling deviance remained sig-
nificant (Stormshaket al.,2004).The findings fromthis latter
study suggest that sibling factors may be more important in
predicting adolescent drug use. Both of these studies suggest
that siblings play an important role in influencing adolescent
drug use, preceding the influence of drug using peers and
possibly overriding the influence of peers.
We found no studies that considered marital conflict, pa-
rental monitoring, sibling relationships, and peer characteris-
tics influences on initiation of drug use or related behaviors.
The studies that have considered more than one of these con-
structs in the same model have found evidence for richer pat-
terns of association than indicated by examining any one of
these constructs alone. Both parental behavior and sibling re-
mediation and moderation of associations between interper-
sonal relationships and adolescent substance use is indicated
in some studies. There is a clear need to consider how these
different relationships may work together to influence the in-
itiation of drug use in order to better understand the processes
that may be involved. The findings reviewed above suggest
that the effects of family factors appear to have a more pro-
found effect when they are more proximal to the adolescent,
supporting an approach that progresses from more distal to
tion of drug use. Specifically, we propose an ordering from
marital conflict about the child, parental monitoring, sibling
relationship, peer characteristics, to initiation of drug use.
Genetic and Environmental Influences on
Interpersonal Relationships and Drug Use
Strong causal inferences about the role of social factors indrug
useand abuse cannot be drawnfrom studiesthat do not usege-
netically informed designs for three reasons. First, there is am-
ple evidence that substance use itself is influenced by genetic
cally accounting for more than half of the total variance (e.g.,
Kendler, Meyers, & Prescott, 2007). These same genetic fac-
tors may influence adolescent relationships with parents and
peers, but this second effect of genetic factors might simply
be a noncausal epiphenomenon. Second, adolescent relation-
ships may mediate genetic influences on substance use. Ge-
netic factors related to substance use may first influence rela-
tionships and, as a consequence of that influence, go on to
influence substance use. Third, interpersonal relationships
may have theirownmaineffects on substanceuse independent
use. These possibilities are essential in constructing any causal
model of substance use. They are based on a well-established
literature substantiating the role of genetic influences of the
adolescent on measures of family relationships like parenting
Doyle, 1997; Feinberg, Reiss, Neiderhiser, & Hetherington,
2005; Rasbash, Jenkins, O’Connor, Tackett, & Reiss, 2011)
and even for marital conflict abouteach child (Reiss, Neiderhi-
ser, Hetherington, & Plomin, 2000). Genetic influences have
istics (Beaver et al., 2009; Bullock, Deater-Deckard, & Leve,
2006; Cleveland, Wiebe, & Rowe, 2005; Iervolino et al.,
2002; Kendler, Jacobson, et al., 2007; Pike & Atzaba-Poria,
2003). In general, these studies have found that parenting is in-
fluenced by genetic, shared environmental, and nonshared
environmental influences; peer relationships primarily by ge-
netic and nonshared environmental influences; and sibling re-
lationships primarily by genetic and shared environmental in-
fluences (Kendler & Baker, 2007; Ulbricht & Neiderhiser,
2010). In other words, parenting may reflect evoked responses
of parentsto genetically influenced characteristics of the child,
while peer characteristics may be the result of adolescents’ se-
lecting peers or by being selected on the basis of the adoles-
cent’s genetically influenced characteristics. In both cases, ge-
notype–environment correlation isindicated,although the type
of genotype–environment correlation is unclear. At least one
study has reported that genetic influences explain the covaria-
tion between parental discipline and quality of peer relation-
tent with the univariate findings reported above. Marital
conflict about the child, a marital variable that is distinct for
each twin and/or sibling, shows substantial and significant ge-
netic influence, with the remaining variance due to primarily
nonshared environmental influences (Reiss et al., 2000). This
can also be interpreted as evocative effects of the child on his
or her parents’ marital conflict because the similarity in levels
nessof the children. Insum,there is ample evidence of genetic
influence on measures of interpersonal relationships with sys-
tematic variation based on which relationship is examined.
Genetic and environmental influences on substance use
Substance abuse disorders are highly heritable, with genetic
iance (e.g., Kendler, Myers, et al., 2007; Lynskey, Agrawal,
& Heath, 2010; Rhee et al., 2003). Different patterns of ge-
netic and environmental influences have been found, how-
ever, based on the developmental progression of substance
use. In other words, initiation of use is largely due to shared
environmental and some genetic influences, while continued
use and dependence is primarily due to genetic influences
(Agrawal & Lynskey, 2008; Derringer, Krueger, McGue, &
Iacono, 2008; Hopfer, Crowley, & Hewitt, 2003; Kendler,
influences on initiation of drug use are typically described as
having to do with availability of substances and sibling ef-
fects rather than common rearing environment factors. These
findings suggest that although individuals are more likely to
progress from experimentation to problem use for reasons to
do with their genes, the opportunity to experiment is largely
attributable to the environment that they share with their sib-
lings. Thus, minimizing the opportunities to initiate drug use
may be a very effective intervention strategy even in indi-
viduals who may be at greater genetic “risk” of becoming de-
There are also fairly clear findings from genetically in-
formed studiesthat the bulkof genetic influence on substance
use is not specific to a particular substance. A review of twin
and adoption studies examining genetic and environmental
influences on adolescent substance use found evidence for
common genetic influence across substances (Hopfer et al.,
2003). More recent reports support this conclusion. For ex-
ample, common genetic influences have been found across
multiple classes of illicit drugs in male twins (Kendler,
Jacobson, Prescott, & Neale, 2003), although a subsequent
analysis of male and female twins found that illicit and legal
drug dependence loaded on different genetic factors, indicat-
ing that there are distinct genetic influences for the two cate-
gories of drug dependence (Kendler, Myers, et al., 2007). In
both studies, although there are genetic influences specific to
each drug, the bulk of genetic variance was shared across the
drugs. These data suggest that a single index of drug use in
early adulthood, one that combines all substances of abuse,
adequately reflects the cumulative genetic and environmental
influences on substance use in adulthood. We use this strat-
egy in the current study.
Genetic and environmental influences on associations
among interpersonal relationships and outcomes
A powerful way to understand how interpersonal relation-
ships may impact outcomes, including initiation of drug
J. M. Neiderhiser, K. Marceau, and D. Reiss
use, is to examine genetic and environmental influences on
the associations among relationships and outcomes. By ex-
amining genetic and environmental influences on these asso-
ciations, we can understand whether evocative adolescent ef-
fects contribute to both interpersonal relationships and drug
initiation, or whether interpersonal relationships are potential
environment correlation) predicting adolescent drug initia-
tion. Few studies have investigated the role of adolescents’
flict and adolescent adjustment problems. There is some evi-
dence that the association between marital conflict about the
child and adolescent antisocial behavior is explained in near
equal parts by genetic and shared environmental influences
(Reiss et al., 2000). A children of twins approach found
that marital instability was linked to young adult alcohol,
drug, and behavioral problems for environmental reasons ra-
ther than through passive genotype–environment correlation
(D’Onofrio et al., 2005). Another study using the children
oftwins designfound that geneticinfluences mediatedtheas-
sociation between marital conflict frequency and conduct
problems in offspring (Harden et al., 2007). Together, these
few studies suggest both genetic and environmental media-
tion of associations between marital relationships and off-
spring adjustment, indicating that marital conflict likely im-
pacts offspring adjustment via both evocative and passive
genotype–environment correlation processes.
Manystudies haveexaminedgenetic and environmental in-
fluences underlying the association between parenting and
adolescent outcomes. Most of the covariation between parent-
ing and externalizing behaviors can be explained by genetic
influences (e.g., Evrony, Ulbricht, & Neiderhiser, 2010;
Ulbricht & Neiderhiser, 2010). In other words, the genetic in-
fluences of the child on parenting behaviors also account for
variation in externalizing behaviors. One of the first studies
to take this approach examined parental negativityand adoles-
& Plomin, 1996). Although the focus of that report was on
finding nonshared environmental influences that contributed
to covariation between family relationships and adolescent ad-
tion between parental negativity and adolescent antisocial be-
havior. Many studies have found a similar pattern of findings
usingdifferent samples,withmost concludingthat geneticfac-
tors are significant influences on the covariation between par-
enting and adolescent antisocial and related behaviors (Burt,
Krueger, McGue, & Iacono, 2003; Button, Lau, Maughan,
& Eley, 2008; Narusyte, Andershed, Neiderhiser, & Lichten-
stein, 2007). These findings, similar to those of genetic influ-
ences on parenting alone, indicate that genotype–environment
correlation is likely to be operating, although it provides little
information on whether these effects are evocative or passive
(because parents and children share genes and environments).
A numberof studies have used different designs or taken a
longitudinal approach to examining associations between
parentingandadolescentexternalizing behaviorstobetter un-
derstand howand why these constructs are related. Forexam-
ple, two adoption studies found that birth parent psychopa-
thology is linked to adoptive parents’ harsh parenting via the
adopted child’s behavior, indicating evocative genotype–
environment correlation (Ge et al., 1996; Riggins-Caspers,
Cadoret, Knutson, & Langbehn, 2003), and a different adop-
tion study found a direct association between birth parent an-
tisocial behavior and negative parenting (O’Connor, Deater-
design examined parenting behaviorsand later delinquent be-
haviors in adolescents and found that the associations were
best explained by shared environmental and genetic influ-
have been at least two longitudinal studies of parenting and
adolescent outcomes. In both, parenting was associated
with change in adolescent antisocial behavior, and a substan-
tial portion of this association was due to genetic influences
(Burt, McGue, Krueger, & Iacono, 2005; Neiderhiser, Reiss,
Hetherington, & Plomin, 1999). In sum, a variety of studies
using different samples, designs, and measures find evidence
for genetic influences on the covariation of parenting and ex-
ternalizing and related outcomes in offspring.
A handful of studies have examined genetic and environ-
mental influences on links between other sibling and peer
characteristics and relationships, and substance use and re-
lated outcomes. A series of studies using a large representa-
tive sample of twins and siblings found that sibling relation-
ship was linked with smoking and drinking and that these
associations are shared environmental, not genetic, in nature
(Rende, Slomkowski, Lloyd-Richardson, & Niaura, 2005;
Slomkowski, Rende, Novak, Lloyd-Richardson, & Niaura,
2005). In addition, sibling negativity was linked with adoles-
cent externalizing behaviors for mostly shared environmental
reasons (Reiss et al., 2000). Shared environmental influences
on family functioning and adolescent substance usewere also
found in a large sibling adoption study, with the shared envi-
ronmental influences due to sibling effects, not parent effects
(McGue, Sharma, & Benson, 1996). A report from the same
group using a sample of 14-year-old twins found that only
shared environmental influences explained the covariation
among peer deviance, parent–child relationship problems,
and early substance use (Walden, McGue, Iacono, Burt, &
Elkins, 2004). A study of a large sample of adult twin men
who rated their peer characteristics and conduct problems
during adolescence retrospectively via the Life History Cal-
endar method found that genetic influences on conduct prob-
lems were linked to later peer deviance and did not change
over time, while shared environmental influences explained
the covariance between peer deviance and conduct problems
2008). A subsequent report using the same sample found that
genetic and environmental influences on cannabis use were
correlated with peer deviance and that the direction of effects
was from cannabis use to peer deviance, not vice versa (Gil-
lespie, Neale, Jacobson, & Kendler, 2009). A similar pattern
of findings was found for adolescent peer delinquency and
smoking during young adulthood in a longitudinal sample
of twins and siblings, with the association explained by
only genetic and nonshared environmental influences (Hara-
keh et al., 2008), and between peer and adolescent delin-
quency in a different sample of twins and siblings (Beaver,
DeLisi, Wright, & Vaughn, 2009). In sum, sibling relation-
ships have been found to be linked to externalizing and re-
lated behaviors because of shared environmental influences,
which is most likely a sibling deviancy training process,
whereas findings from studies examining peer delinquency
and adolescent and young adult delinquency and substance
use consistently show that genetic influences were important
genotype–environment correlation in these associations.
Therefore, although deviancy training is a proposed
ships on adolescent drug use, findings from behavioral genet-
ics studies suggest that the evocative effects of the drug-using
adolescent are particularly important in the deviancy training
of his/her own peers, while shared environmental influences
contribute to similarities between siblings. It is important to
note here, however, that the shared environmental influences
from siblings are conceptualized differently than those from
parents (although there are no differences in the modeling).
As noted earlier, shared environmental influences due to sib-
lings arelikelytheresult ofreciprocalprocesses, while shared
environmental influence due to rearing environment from
parents are more commonly considered the result of the rear-
ing environment acting on the child (Rowe & Rogers, 1997).
The current report seeks to build on and extend the existing
work examining genetic and environmental influences on in-
terpersonal relationships and the initiation of illegal drug use
by young adulthood. Although a number of studies have ex-
amined bivariate associations among these relationships and
adolescent or young adult behaviors, some longitudinally,
few have considered multiple relationships. The phenotypic
literature has indicated that marital conflict and parental mon-
itoring areimportant family factorsthat mayincrease thelike-
lihood of drug initiation and sibling conflict and peer delin-
quency as possible avenues for contagion or deviancy
training. We will consider all four interpersonal factors: mar-
ital conflict about each adolescent, parental monitoring,
sibling negativity,and peergroupdelinquencyduringadoles-
cence in relation to initiation of drug use byyoungadulthood,
using a genetically informed design. Clarifying the patterns
of genetic and environmental contributions to these asso-
ciations will help to advance our understanding of how ge-
notype–environment correlation and direct environmental
influences may be operating within the context of these
interpersonal relationships to influence the initiation of
shared Environment and Adolescent Development (NEAD)
project (Neiderhiser, Reiss, & Hetherington, 2007; Reiss
et al., 2000). The NEAD project assessed 720 same-sex
(48% female) twin and sibling pairs from families across
cent siblings no more than 4 years apart in age. NEAD com-
prises six sibling types in two family types: monozygotic
and full siblings in nondivorced families (FI, N ¼ 95 pairs);
and full siblings (FS, N ¼ 182 pairs), half-siblings (HS,
N ¼ 109 pairs), and stepsiblings (US, N ¼ 130 pairs) in step-
dren in stepfamilies were required to be married at least 5
years prior to data collection to ensure that none of the step-
families were in the unstable early phase of family formation.
The nondivorced families with full siblings were recruited
through random-digit dialing of 10,000 telephone numbers
throughout the United States. Except for a small subsample
dom-digit dialing procedures, most of the other sibling types
were recruited through a national market survey of 675,000
Three cross-sectional assessments occurred in NEAD, the
with adolescents still living at home at least 50% of the time
and who participated in Time 1 were eligible for the Time 2
assessment 2–3 years later (N ¼ 408). All families who par-
ticipated in the Time 1 assessment were eligible to participate
in the Time 3 assessment as young adults. In order to maxi-
mize the analytic sample size, data from the Time 1 and
Time 3 assessments were used in the current report. At
Time 1 siblings were 10–18 years old (average age was
14.5 years for Sibling 1 and 12.9 for Sibling 2) and were
ily income was $25,000–$35,000. The Time 3 assessment
took place 7–13 years later. At Time 3 siblings were 20–35
years old (average age ¼ 26.8 for Sibling 1 and 25.5 for Sib-
The median family income for parents was higher than at
the Time 1 assessment ($60,000–$69,000). At Time 3 the
median income was $40,000–$49,000 for Sibling 1 and
$30,000–$39,000 for Sibling 2. Of the original 720 families,
516 families could be recontacted, and some data were col-
lected on 413 families. Families who did not participate in
the Time 3 assessment were generally younger, F (340) ¼
1.44, p , .05, and had lower family income, F (361) ¼
1.27, p , .05, but did not differ from families who did parti-
cipate in the Time 3 assessment on other demographic and
study variable characteristics (e.g., gender of siblings, sibling
J. M. Neiderhiser, K. Marceau, and D. Reiss
antisocial behavior, peer college orientation, peer delinquen-
cy and substance use, parents’ age or education, and family
income). Demographic information for the Time 1 and
Time 3 assessments (used in the present report) is presented
in Table 1.
At Time 1 most families were visited twice by two interview-
ers. Both parents and the two adolescents completed ques-
tionnaires and were videotaped during the visit in dyadic, tri-
adic, and tetradic family interactions. Additional questionnaire
data were obtained from questionnaires that were mailed
cified areas of disagreement that were then discussed and vi-
deotaped in the 10-min interactions. A global coding system
on a 5-point Likert scale was used to rate the videotaped in-
teractions (Hetherington & Clingempeel, 1992). In the pres-
ent study we focused on the dyadic interactions between the
Twins were rated for physical similarity (e.g., eye and hair
color) by the interviewer, the parents, and self-report using a
questionnaire designed for adolescents (Nichols & Bilbro,
1966). If any differences in physical characteristics were re-
ported or if respondents reported that people never were con-
fused about the identity of the twins, the twin pair was clas-
sified as DZ. Twelve of the twin pairs could not be
classified as MZ or DZ and were excluded from these analy-
ses (,2% of the twin pairs). The accuracy of this method for
assigning zygosity has been shown to be more than 95% ac-
curate when compared with DNA tests (Nichols & Bilbro,
1966; Spitz, Moutier, Reed, Busnel, & Marchaland, 1996).
Our analysis sample reflects the 708 families for which we
have zygosity data. Missing data were accommodated using
full information maximum likelihood during analysis.
Measurement included parent and adolescent reports and vi-
deotaped dyadic interactions.
Marital conflict about the child. At Time 1 marital conflict
about the child was assessed using an average of mother
and father reports of conflict with spouse about each sibling
on the Child Rearing Issues-Self and Spouse scale (Hether-
ington & Clingempeel, 1992). Mothers and fathers reported
on how often, on a scale of 0 (never) to 7 (more than once
a day), they had arguments about Sibling 1 and Sibling 2
(separately) in the past month (e.g., children’s behaviors to-
ward self and siblings, children’s grades, chores, or allow-
ances). Mother and father reports were moderately correlated
(Sibling 1: r ¼ .31, p , .05; Sibling 2: r ¼ .25, p , .05).
Parental monitoring. At Time 1 parental monitoring was
computed as the sum of mother and father reports of their
knowledge about the adolescent’s friends and whereabouts
(Hetherington & Clingempeel, 1992; as . 0.90). Mothers
and fathers were asked 13 items about how much they
know about Sibling 1’s and Sibling 2’s lives in several areas
(e.g., choice of friends, who they are, and what they are like;
and child’s school life, such as teachers, homework, and
grades) on a scale of 1 (never) to 5 (always). Mother and fa-
ther reports were moderately correlated (Sibling 1: r ¼ .46,
p , .05; Sibling 2: r ¼ .40, p , .05).
Sibling negativity. Time 1 sibling conflict was assessed using
a composite of child, parent, and observer ratings (Furman &
Buhrmester, 1985; Hetherington et al., 1999; Schaefer &
Edgerton, 1981). Parents reported on the sibling rivalry,
aggression, and avoidance subscales of the Sibling Inventory
Siblings reported on the amount of criticism they felt from
their sibling using the criticism subscale of the Network
of Relationship Inventory (Hetherington & Clingempeel,
1992; as. 0.72), self-report on the sibling rivalry, aggres-
sion, and avoidance subscales of the Sibling Inventory of Be-
havior (Hetherington & Clingempeel, 1992; as .0.71), the
total score for the sibling interaction task (a questionnaire
where siblings rate the frequency of disagreement on a num-
ments; Hetherington & Clingempeel, 1992; as. 0.30), and
the symbolic and physical aggression subscales of the Con-
flict Tactics Inventory (Straus, 1979, 1987; as.0.81).
Observers rated anger, coercion, and transactional conflict
during the dyadic sibling interactions (intraclass correlations
[ICCs] . .62). Scores for parent, adolescent, and observer re-
ports were standardized and summed to form composite sib-
ling negativity scores (r ¼ .24–.60, ps , .05; see Reiss et al.,
Peer group delinquency. Combined motherand father reports
were used to index the adolescent’s delinquent peer group
Table 1. Sample demographic information
Time 1Time 3
CharacteristicSibling 1Sibling 2Sibling 1 Sibling 2
38.1 (5.2)41.0 (6.5)51.4 (4.6)54.2 (6.1)
13.8 (2.3) 13.9 (2.7)14.2 (2.7)14.6 (2.8)
Note: The values are means (standard deviations).
(Iervolino et al., 2002; Manke, McGuire, Reiss, Hethering-
ton, & Plomin, 1995). Mothers and fathers reported on Sib-
ling 1’s and Sibling 2’s peer group delinquency using 8 items
assessing how like the child’s friends are to adjectives (i.e.,
delinquent or rebellious) on a 1 (very much unlike) to 4
(very much like) scale. Scores were acceptably reliable (as.
0.76). Mother and father reports were moderately correlated
(Sibling 1: r ¼ .38, p , .05, Sibling 2: r ¼ .35, p , .05).
Initiation of illegal drugs by young adulthood. Substance use
was assessed at Time 3 using a construct indicating the num-
ber of illicit drugs young adults had initiated (i.e., marijuana,
pills, crack, cocaine, LSD, PCP, heroin, mushrooms, inhal-
ants/paint/glue, or meth; Elliott & Huizinga, 1983; Jessor,
Donovan, & Widmer, 1980; Jessor & Jessor, 1977). Across
siblings, 47% of the sample did not initiate any drugs, 28%
of the sample initiated one drug, and 10% of the sample
had initiated four or more drugs.
Data analyses proceeded in several steps. Child age, sex, the
interaction between age and sex, and age difference between
nontwin siblings were controlled for by regressing the var-
iance in each variable out of the study variables (McGue &
Bouchard, 1984). The continuous residual scores were used
in all analyses.
First, phenotypic correlations were computed to examine
flict about the child, parental monitoring, sibling negativity,
and peer group delinquency) and the number of illegal drugs
initiated by young adulthood, as well as the intercorrelations
among the interpersonal variables. We also regressed all four
interpersonal variables on drug use in a phenotypic multiple
regression to determine whether marital conflict about the
child, parental monitoring, sibling negativity, and peer group
delinquencyeach explained unique variance in the numberof
illegal drugs initiated by young adulthood.
means and standard deviations were examined across sibling
type (MZ twins, DZ twins, and FI in nondivorced families;
FS,HS,US instepfamilies), and sibling ICCs werecomputed
using double-entered data separately for each sibling type for
fluences are suggested if ICCs decrease according to decreas-
ing genetic similarity (MZ . DZ ¼ FI ¼ FS . HS . US).
Shared environmental influences are suggested if ICCs are
similar for genetically nonidentical siblings (DZ ¼ FI ¼ FS
. 0, HS . O, US . 0). Finally, MZ ICCs , 1 indicate non-
shared environmental influences.
Third, we conducted multivariate Cholesky biometrical
models of the raw data using Mx (Neale, 1994) to systemati-
cally estimate genetic and environmental influences on the
associations among the interpersonal relationship variables
and initiation of illegal drug use. Figure 1 depicts the model
for one sibling. The parameter estimates are set to be equal
tinctive pairs of latent factors representing genetic influences
(i.e., A1 for Sibling 1 and A1 for Sibling 2, A2 for Sibling 1
and A2 for Sibling 2, A3 for Sibling1 and A3 for Sibling 2,
A4 for Sibling 1 and A4 for Sibling 2, and A5 for Sibling 1
and A5 for Sibling 2) set according to the genetic similarity
of the siblings. Correlations between pairs of distinct shared
environmental latent variables (C1–C5) were each set to 1,
and nonshared environmental latent factors (E1–E5) were
uncorrelated with each other or any other latent factor.
The order in which variables are entered into the model is
important in multivariate Cholesky decompositions because
the genetic and environmental influences contributing to as-
sociations among the first and subsequent variables are con-
strained by the genetic and environmental influences on the
first variable entered into the model. As described earlier,
we derived the orderof our variables based on the patterns re-
viewed in the literature from predictors most distal to most
proximal to initiation of drug use: (a) marital conflict about
the child, (b) parental monitoring, (c) sibling negativity, (d)
peer group delinquency, and (e) the numberof drugs initiated
by young adulthood, the outcome.
Multivariate Cholesky decompositions first estimate the
genetic, shared, and nonshared environmental influences on
the first variable (i.e., marital conflict about the child).
Then, the association between the first and second variables
(i.e., the association between marital conflict about the child
and parental monitoring) is decomposed into genetic, shared,
and nonshared environmental influences, followed by the as-
sociation between the first and third variables, and so on.
Figure 1. An illustration of the multivariate Cholesky model. The latent fac-
tors A1, C1, and E1 refer to the respective genetic, shared, and nonshared
environmental influences on marital conflict. A2, C2, and E2 refer to the re-
spective unique genetic, shared, and nonshared environmental influences on
parental monitoring, or the genetic, shared, and nonshared environmental in-
fluences on parental monitoring that are not shared with marital conflict. A3,
conflict or parental monitoring. A4, C4, and E4 refer to the respective unique
genetic, shared, and nonshared environmental influences on peer group de-
linquency not shared with any of the previous constructs; and A5, C5, and
E5 represent the respective unique genetic, shared, and nonshared environ-
mental influences on number of drugs initiated by young adulthood not in
common with any of the predictors. All paths are estimated in the fully satu-
J. M. Neiderhiser, K. Marceau, and D. Reiss
Unique genetic, shared, and nonshared environmental influ-
ences are then estimated for the second variable (parental
and third variables, second and fourth variables, and so forth.
This is repeated for each subsequent variable entered into the
model, culminating in estimating the unique genetic, shared,
and nonshared environmental influences on the outcome
variable (i.e., numberof drugs initiated by young adulthood).
We tested a saturated model estimating all genetic, shared,
and nonshared environmental paths possible, and then sys-
tematically dropped pathsthat were not significantly different
from zero (as judged by 95% confidence interval [CI]).
Model differences were explored using a nested model ap-
proach that compared the constrained models to the saturated
model. Differences in chi-square estimates between models
tested whether the constrained models resulted in a signifi-
cant decrement of model fit. If a model removing paths did
not result in a significant decrement in fit from the saturated
model, we concluded the path was no different from zero
model) was the better fitting and more parsimonious model.
We first dropped each nonsignificant path individually. If
there was a significant decrement in model fit (despite non-
nificant and included during the next step in model fitting,
when nonsignificant paths were dropped in pairs. We also
tested dropped paths in larger groups until we arrived at the
best fitting model. Results from both the full model and
the best fitting model are presented.
Model assumptions. These types of behavioral genetic mod-
els assume that shared and nonshared environmental effects
are the same across sibling types and that assortative mating
(i.e., nonrandom mating that could result in suppressed ge-
netic influences because DZs and full siblings would share
more than 50% of their genes) is not occurring. These as-
sumptions have previously been tested and have been gener-
ington, & Reiss, 1996). Further, there were no systematic
differences for the validity of equal twin and sibling environ-
ments in NEAD (Reiss et al., 2000), indicating that the equal
environments assumption is reasonable for the current study.
We also are assuming that the orderof influence of predictors
flows in the direction previously specified.
Behavioral genetic models are also particularly sensitive to
skewness (Purcell, 2002). Therefore, we used log transforma-
tions to obtain a normal distribution for both marital conflict
about the child and peer group delinquency. We also used
SAS PROC RANK to normalize data by substituting values
variations within the range of normalityof skewness and kur-
tosis in the distributions of each variable would not affect
modeling (Eaves et al., 1997). The data were double entered
prior to control for within-family confounds before perform-
ing any ICCs and phenotypic associations.
Genetic and environmental influences on drug use
The intraclass twin/sibling correlations showed a slightly de-
creasing pattern, suggesting substantial shared environment,
but also genetic and nonshared environmental influences on
drug use (MZ ¼ 0.71, DZ ¼ 0.70, FI ¼ 0.40, FS ¼ 0.56,
HS ¼ 0.31, US ¼ 0.41). Univariate biometric analyses sup-
ported this pattern of findings. Results showed significant ge-
netic (32%; 95% CI ¼ 0.06–0.57), shared environmental
(41%; 95% CI ¼ 0.19–0.59), and nonshared environmental
influences (27%; 95% CI ¼ 0.18–0.41) for the number of
drugs initiated by young adulthood with the variance nearly
evenly split between the three.
Phenotypic correlations are presented in Table 2. Overall,
each interpersonal predictor was positively associated with
the number of drugs initiated in young adulthood (r . .13,
ps , .05), except parental monitoring, which was related to
fewer drugs initiated, as expected (r ¼ –.17, p , .05). All
of the interpersonal predictors were correlated with one an-
other (rs . .06, ps , .05).
to estimate the extent to which each interpersonal variable
predicted unique variance in the outcome, drug initiation.
Marital conflict, parental monitoring, sibling negativity, and
tors, with drug initiation as the dependent variable. The over-
all model was significant, F (4, 472) ¼ 12.74, p , .001, and
predicted about 10% of the variance in drug initiation (R2¼
.097). Parental monitoring (b ¼ 0.08, t ¼ 2.05, p , .05), sib-
ling negativity (b ¼ 0.09, t ¼ 2.21, p , .05), and peer delin-
quency (b ¼ 0.19, t ¼ 4.49, p , .001) each contributed sig-
Table 2. Correlations among study variables
N ¼ 657
N ¼ 667
N ¼ 695
N ¼ 257
N ¼ 646
N ¼ 658
N ¼ 246
N ¼ 669
N ¼ 250
N ¼ 259
*p , .05.
Table 3. Model fitting results for the fully saturated model
A1 C1E1 A2 C2E2 A3C3 E3A4 C4E4 A5C5 E5
Drug use0 0.550.51
Note: Unstandardizedbetaweightsare presented(95% confidenceintervals). Thelatent factors A1,C1,and E1 refer to therespective genetic, shared, andnonshared environmental influenceson maritalconflict.A2,
C2, and E2 refer to the respective unique genetic, shared, and nonshared environmental influences on parental monitoring or the genetic, shared, and nonshared environmental influences on parental monitoring that
are not shared with marital conflict. A3, C3, and E3 refer to the respective unique genetic, shared, and nonshared environmental influences on sibling negativity not in common with either marital conflict or parental
monitoring. A4, C4, and E4 refer to the respective unique genetic, shared, and nonshared environmental influences on peer group delinquency not shared with any of the previous constructs; and A5, C5, and E5
represent the respective unique genetic, shared, and nonshared environmental influences on number of drugs initiated by young adulthood not in common with any of the predictors.
Table 4. Model fitting results for the best-fitting model
A1C1 E1A2C2E2 A3C3E3A4C4 E4A5 C5 E5
Marital conflict 0.49
Drug Use0000 0.11
Note: Unstandardized betaweights are presented (95% confidence intervals). Parameterestimates of 0 indicate that the parameter was not included in the best-fitting model. The latent factors A1, C1, and E1 refer to
the respective genetic, shared, and nonshared environmental influences on marital conflict. A2, C2, and E2 refer to the respective unique genetic, shared, and nonshared environmental influences on parental mon-
itoring or the genetic, shared, and nonshared environmental influences on parental monitoring that are not shared with marital conflict. A3, C3, and E3 refer to the respective unique genetic, shared, and nonshared
environmental influences on sibling negativity not in common with either marital conflict or parental monitoring. A4, C4, and E4 refer to the respective unique genetic, shared, and nonshared environmental in-
fluences on peer group delinquency not shared with any of the previous constructs; and A5, C5, and E5 represent the respective unique genetic, shared, and nonshared environmental influences on number of drugs
initiated by young adulthood not in common with any of the predictors.
aA parameter estimate is not significantly different from zero but is required to be in the model (or the model fails).
nificant unique variance in drug initiation, controlling for the
other interpersonal variables. Marital conflict contributed
trolling for the other interpersonal variable (b ¼ 0.08, t ¼
1.85,p ¼ .065). Thus, we continued with the proposedmodel
including all four interpersonal variables and drug initiation
for the behavioral genetic analyses.
Unstandardized parameter estimates and 95% CIs for the
(fully saturated), Table 4 (best fitting), and Figure 2 (best
fitting). The constrained model (dropping nonsignificant
paths) fit the data better than the fully saturated model, fully
saturated model: –2 log likelihood (–2LL) (3088) ¼ 7326.5,
Akaike information criterion (AIC) ¼ 1150.3; constrained
model: –2LL (5971) ¼ 14173.6, AIC ¼ 2231.6; x2diff
(22) ¼ 30.3, p ¼ .11, AIC diff ¼ –13.7.
The best fitting model (Figure 2) revealed a genetic factor
explaining variance common to marital conflict about the
itiation (A1). Therewas also a discrete genetic factorexplain-
ing variance in parental monitoring, peer group delinquency,
and drug initiation that was distinct from the genetic influ-
ences shared with marital conflict (A2). A final genetic factor
explained unique variance in sibling negativity (A3). Peer
group delinquency and drug initiation showed no unique ge-
netic influences after accounting for the genetic variance
shared with marital conflict and parental monitoring. There
was a shared environmental factor explaining variance in
marital conflict about the child and sibling negativity (C1).
A separate shared environmental factor explained variance
shared by parental monitoring, peer group delinquency, and
drug initiation that was distinct from variance shared with
marital conflict (C2). A third distinct shared environmental
factor explained shared environmental influences on sibling
environmental influences on marital conflict and parental
monitoring (C3). Finally, there were unique shared environ-
mental factors influencing peer group delinquency (C4) and
drug initiation (C5). Nonshared environmental factors did
not explain any of the covariation among interpersonal vari-
ables and drug use. Instead, there were discrete nonshared
environmental influences for each construct (E1–E5).
This study sought to clarify the associations among interper-
sonal relationships within the family and peer group charac-
teristics during adolescence influencing initiation of drug
use by young adulthood. Findings indicate that interpersonal
relationships during adolescence are associated with initia-
tion of drug use for both genetic and environmental reasons
and that these associations cluster in meaningful ways, sug-
gesting four distinct factors for the initiation of drug use. In
other words, marital conflict, sibling negativity, and peer
group delinquency are all associated with initiation of drug
use and with one another for shared genetic, but not environ-
mental, reasons. Parental monitoring and peer group delin-
quency are associated with initiation of drug use through
ditional shared environmental factor is common to only sib-
ling negativity and initiation of drug use. It is interesting
that even afterall of the variance shared with interpersonal re-
lationships during middle adolescence had been accounted
for there were still sizable and significant shared and non-
shared environmental influences on initiation of drug use
by young adulthood but no remaining genetic influences.
These findings and their implications are discussed.
The first genetic pathway identified in this report reflects
adolescent characteristics eliciting conflict in their relation-
ships and also increasing their likelihood of associating
with deviant peers and initiating drug use. This pathway is
consistent with the coercive family cycle well documented
by Patterson (1982) and others (e.g., Compton et al., 2003;
Conger, Ge, Elder, Lorenz, & Simons, 1994; Dishion, Dun-
can, Eddy, Fagot, & Fetrow, 1994; Loeber & Dishion,
1984). In his initial work on the coercive cycle, Patterson fo-
cused on families of younger aggressive boys. He was clear
that cycles of aggression between parent and child were initi-
ated by the child, a finding that would fit with the importance
of the child’s genetic influences shown here. However, our
study arrived on the developmental scene later than Patter-
son’s.Heredeviantchild behaviors maybehavingawideref-
Figure 2. Significant paths for the best fitting model. The latent factors A1,
C1, and E1 refer to the respective genetic, shared, and nonshared environ-
mental influences on marital conflict. A2, C2, and E2 refer to the respective
unique genetic, shared, and nonshared environmental influences on parental
monitoring, or the genetic, shared, and nonshared environmental influences
on parental monitoring that are not shared with marital conflict. A3, C3, and
E3 refer to the respective unique genetic, shared, and nonshared environ-
mental influences on sibling negativity, not in common with either marital
conflict or parental monitoring. C4 and E4 refer to the respective unique
shared and nonshared environmental influences on peer group delinquency
not shared with any of the previous constructs, and C5 and E5 represent
the respective unique shared and nonshared environmental influences on
number of drugs initiated by young adulthood not in common with any of
the predictors. Significant paths are shown with a solid line. The nonsignifi-
cant path estimate is depicted with a hashed line.
J. M. Neiderhiser, K. Marceau, and D. Reiss
fect, on sibling and on marital conflict about the adolescent.
Patterson tracked the links between the coercive cycles in the
family and the child’s entering delinquent peer groups with
the link occurring through the school. Aggressive boys
were not only aggressive at home but also aggressive at
school where, rejected by nonaggressive and/or nondeviant
peers, they sought the company of other deviant children.
By adolescence, the child’s genetically influenced character-
istics seem, from our findings, to have had a detrimental im-
pact on multiple relationships and thereby have increased
their risk for adult substance use.
The second genetically influenced pathway centers on in-
adequate parental monitoring, peer group delinquency, and
initiation of illegal drug use and may reflect lack of adoles-
cent disclosure of their whereabouts and activities. In other
words, the genetic influences of the child that influence the
likelihood that they will elicit monitoring from their parents
are correlated with their association with deviant peers and
their likelihood of initiating illegal drug use by young adult-
hood. This pattern of findings is consistent with the adoles-
cent disclosure explanation of parental monitoring and
knowledge first proposed by Stattin and Kerr a decade ago
(Kerr & Stattin, 2000; Stattin & Kerr, 2000). According to
this explanation, the effects of parental monitoring on adoles-
cent delinquent behavior is not the result of parents’ active ef-
forts to find out who their children are associating with and
where and what they are doing but is instead due to their
adolescent’s disclosure of their activities to their parents.
We also identified two factors shaped by environments
siblings share. Recent studies have drawn attention to this do-
main of influences, especially for family relationships, exter-
nalizing behaviors, and drug use. For example, as behavioral
genetic studies include more direct measures of the environ-
ment like parenting behaviors, shared environmental influ-
ences have been found to account for substantial proportions
of the total variance (e.g., Elkins, McGue, & Iacono, 1997;
Plomin, Reiss, Hetherington, & Howe, 1994). Shared envi-
ronmental influences have consistently been indicated for
adolescent delinquency and antisocial behaviors (Rhee &
Waldman, 2002), and a recent meta-analysis found evidence
for significant and meaningful shared environmental influ-
ences on behaviors within the externalizing and internalizing
spectrums (attention-deficit/hyperactivitydisorder was anex-
ception with no shared environmental influence; Burt, 2009).
Given the findings that shared environmental influences are
important for both interpersonal relationships and measures
of adjustment, identifying two shared environmental factors
that explain variation of initiation of illegal drug use is espe-
The first shared environmental factor reflects a different
type of parental monitoring effect: one that is parent driven
rather than child driven. In this case, inadequate parental
monitoring, peer group delinquency, and initiation of drug
members similar. Thus, consistent with the literature on pa-
rental monitoring and drug use (e.g., Dishion & Tipsord,
2011), parents who are more effective monitors have children
who are less likely to associate with deviant peers and to in-
itiate drug use and these associations. A parsimonious expla-
initiates a sequence of events that pivots around subsequent
peer group delinquency and risk for substance abuse. The de-
velopmental sequence of parental monitoring and peer group
delinquency is not established in this study but, as we have
reviewed, has been documented elsewhere.
The second pathway shaped by shared environment reflects
what may be a sibling deviancy training (e.g., Compton et al.,
to the covariation of sibling negativity and initiation of illegal
drug use. These findings may provide one of the clearest exam-
ples of sibling deviancy training to date because this construct
includes only the shared environmental influences that are un-
conflict about the child) and are uncorrelated with peer group
tor emerged that seems to represent a general family climate of
conflict and negativity forassociations between marital conflict
environmental factor, however, is uncorrelated with the sibling
deviancy shared environmental factor.
Limitations and future directions
There are a number of limitations of the current report. We
cannot clearly distinguish among the different types of geno-
likely that many different types of genotype–environment
correlation are operating simultaneously. However, we can
say that we find clear evidence for genotype–environment
correlation in that genetic influences of the child explain as-
sociations among interpersonal relationships and initiation
of drug use. Another limitation of this report is the cross-
sectionalnatureofthe design. Although we assessed thesam-
Nonetheless, all subjects who participated during the young
adult assessment were beyond the age at which initiation of
drug use typically occurs. The outcome variable, initiation
of any illegal drug, could be seen as a shortcoming of the cur-
rent report. Based on previous work that has found clear evi-
dence for a common genetic liability for use of illegal drugs
Neale, 1999; Kendler, Karkowski, Neale, & Prescott, 2000;
Kendler, Karkowski, & Prescott, 1999), we feel that this ag-
gregate variable is appropriate. Finally, the use of parent re-
ports of adolescent peer group characteristics limits our con-
fidence that this is an index of the actual characteristics of the
adolescent’s peer group characteristics. Prior research using
the NEAD sample using this measure and the child report
measure (collected only at Time 2) found different patterns
of genetic and environmental influences (Iervolino et al.,
2002; Manke et al., 1995). Parent reports showed more ge-
netic influences, while child self-reports showed primarily
nonshared environmental and modest shared environmental
influences. Because the child self-report version was not ad-
preted with some caution.
Despite these limitations, this study has clear implications
for our understanding of the pathogenesis and prevention of
drug use in earlyadulthood. First, it clarifies that, for samples
similar in demographic characteristics to the one used here,
environmental causes of substance abuse are important. Our
data emphasize a pattern of recent findings that siblings are
an important subsystem in the family and are a potential
breeding ground for deviant behavior and substance abuse
(e.g., Compton et al., 2003; Rende et al., 2005; Slomkowski
et al., 2005). Preventionists often overlook siblings in their
exclusive attention on other aspects of the family.
Second, our findings provide some support for both sides
of the current debate about the pivotal role of parental moni-
toring in the initiation of drug use. Stattin and Kerr (2000)
reflects problems of the child: they do not disclose what they
do and hence the parents cannot act to prevent or reduce de-
tified, the one that pivots around parental monitoring, is not
consistent with Stattin and Kerr’s argument. Taken together,
our findings support the position that inadequate parental
monitoring has two sources: a child problem, perhaps inade-
quate disclosure, and a parent problem, inadequate surveil-
lance. Moreover, these are noncontingent and separable fam-
ily problems. These findings suggest some families may
show one type of problem and not the other, thus assessing
families based on this critical distinction and tailoring inter-
ventions to improve child disclosure, where warranted, or to
embolden parental surveillance, where warranted, or both is
one way that these findings could be used to reduce the like-
lihood of the initiation of drug use.
Third, the lack of genetic influences on initiation of drug
use by young adulthood that are not correlated with interper-
sonal relationships during adolescence is a novel finding that
could have important implications. Our model accounts for
all the genetic influences on initiation of drug use. It weights
heavily a general hypothesis that the mechanisms by which
genetic influences on drug use are expressed are mediated
via their impact on family and peer relationships. If their in-
fluence on these relationships could be blunted, then the ad-
verse effect of these genetic influences could be partially or
entirely prevented. We need, of course, more rigorous studies
here toestablish, with certainty, that the genetic influences on
relationship systems are not just epiphenomena. The surest
way to do this is to conduct preventive interventions within
the context of a twin or adoption design. Interventions aimed
at marital conflict and sibling deviancy training should, in
such a design, eliminate entirely the heritability of substance
use. The data we report here not only suggest the importance
a strong caution about the current passion to use gene identi-
fication studies to identify “druggable targets” (Uhl et al.,
2008) for the prevention or treatment of substance abuse.
Family and peer relationships now clearly qualify as inter-
mediate phenotypes and, if genetically informed randomized
controlled trials support their causal role in expression of ge-
netic influences on drug abuse, they become legitimate ther-
apeutic targets not for drugs, with all their unwanted side ef-
fects, but for psychosocial interventionsthat may be saferand
Agrawal, A., & Lynskey, M. T. (2007). Does gender contribute to heteroge-
neity in criteriaforcannabis abuse and dependence? Results from the na-
tional epidemiological survey on alcohol and related conditions. Drug
and Alcohol Dependence, 88, 300–307.
Agrawal, A., & Lynskey, M. T. (2008). Are there genetic influences on ad-
diction: Evidence from family, adoption and twin studies. Addiction,
Amato, P. R. (1991). Psychological distress and the recall of childhood
family characteristics. Journal of Marriage and the Family, 53, 1011–
Avenevoli, S., & Merikangas, K. R. (2003). Familial influences on adoles-
cent smoking. Addiction, 98, 1–20.
Beaver, K. M., DeLisi, M., Wright, J. P., & Vaughn, M. G. (2009).
Gene–environment interplay and delinquent involvement: Evidence of
direct, indirect, and interactive effects. Journal of Adolescent Research,
Beaver, K. M., Shutt, J. E., Boutwell, B. B., Ratchford, M., Roberts, K., &
Barnes, J. C. (2009). Genetic and environmental influences on levels of
self-control and delinquent peer affiliation: Results from a longitudinal
sample of adolescent twins. Criminal Justice and Behavior, 36, 8.
Brook, J. S., Kessler, R. C., & Cohen, P. (1999). The onset of marijuana use
from preadolescence and early adolescence to young adulthood. Devel-
opment and Psychopathology, 11, 901–914.
practices and peer group affiliation in adolescence. Child Development,
of parenting. Journal of Research on Adolescence, 16, 265–291.
Bullock, B. M., Deater-Deckard, K., & Leve, L. D. (2006). Deviant peer af-
filiationandproblembehavior: Atestofgeneticand environmentalinflu-
ences. Journal of Abnormal Child Psychology, 34, 29–41.
Burt, S. A. (2009). Rethinking environmental contributionsto child and ado-
lescentpsychopathology:A meta-analysis ofsharedenvironmentalinflu-
ences. Psychological Bulletin, 135, 6.
Burt, S. A., Krueger, R. F., McGue, M., & Iacono, W. (2003). Parent–child
conflict and the comorbidity among childhood externalizing disorders.
Archives of General Psychiatry, 60, 505–513.
Burt, S. A., McGue, M., Krueger, R. F., & Iacono, W. G. (2005). How are
parent–child conflict and childhood externalizing symptoms related
velopment and Psychopathology, 17, 145–165.
Burt, S. A., McGue, M., Krueger, R. F., & Iacono, W. G. (2007). Environ-
environment. Journal of Abnormal Child Psychology, 35, 787–800.
Bussell, D. A., Neiderhiser, J. M., Pike, A., Plomin, R., Simmens, S., Howe,
G. W., et al.(1999).Adolescents’ relationshipsto siblings and mothers:A
Button, T. M. M., Lau, J. Y. F., Maughan, B., & Eley, T. C. (2008). Parental
punitive discipline, negative life events and gene–environment interplay
in the development of externalizing behavior. Psychological Medicine,
J. M. Neiderhiser, K. Marceau, and D. Reiss
Chen, C. Y., Storr, C. L., & Anthony, J. C. (2005). Influences of parenting
practices on the risk of having a chance to try cannabis. Pediatrics,
Cheng, T. C., & Lo, C. C. (2010). The roles of parenting and child welfare
services in alcohol use by adolescents. Children and Youth Services Re-
view, 32, 38–43.
Chilcoat, H. D., & Anthony, J. C. (1996). Impact of parent monitoring on in-
itiation of drug use through late childhood. Journal of the American
Academy of Child & Adolescent Psychiatry, 35, 91–100.
smoking and drinking friends among adolescents: A behavioral–genetic
evaluation. Journal of Genetic Psychology, 166, 153–169.
Compton, K., Snyder, J., Schrepferman, L., Bank, L., & Shortt, J. W. (2003).
The contribution of parents and siblings to antisocial and depressive be-
havior in adolescents: A double jeopardy coercion model. Development
and Psychopathology, 15, 163–182.
Conger, R. D., Ge, X., Elder, G. H., Lorenz, F. O., & Simons, R. L. (1994).
Economic stress, coercive family process, and developmental problems
of adolescents. Child Development, 65, 541–561.
Crano, W. D., Siegel, J. T., Alvaro, E. M., Lac, A., & Hemovich, V. (2008).
The at-risk adolescent marijuana nonuser: Expanding the standard dis-
tinction. Prevention Science, 9, 129–137.
Cummings, E. (1994). Marital conflict and children’s functioning. Social
Development, 3, 16–36.
Recent advances and emerging themes in process-oriented research.
Journal of Child Psychology and Psychiatry, 43, 31–63.
Derringer, J., Krueger, R. F., McGue, M., & Iacono, W. G. (2008). Genetic
and environmental contributions to the diversity of substances used in
adolescent twins: A longitudinal study of age and sex effects. Addiction,
male adolescent drug use. Development and Psychopathology, 7, 803–
Dishion, T. J., Duncan, T. E., Eddy, J. M., Fagot, B. I., & Fetrow, R. (1994).
adaptation. Social Development, 3, 255–268.
Dishion, T. J., & Loeber, R. (1985). Adolescent marijuana and alcohol use:
The role of parents and peers revisited. American Journal of Drug and
Alcohol Abuse, 11, 11–25.
Dishion, T. J., McCord, J., & Poulin, F. (1999). When interventions harm:
Peer groupsand problembehavior.AmericanPsychologist, 54, 755–764.
Dishion, T. J., & Owen, L. D. (2002). A longitudinal analysis of friendships
and substance use: Bidirectional influence from adolescence to adult-
hood. Developmental Psychology, 38, 480–491.
Dishion, T. J., & Tipsord, J. M. (2011). Peer contagion in child and adoles-
cent social and emotional development. Annual Review of Psychology,
Disney, E. R., Elkins, I. J., McGue, M., & Iacono, W. J. (1999). Effects of
ADHD, conduct disorder, and genderon substance use and abuse in ado-
lescence. American Journal of Psychiatry, 156, 1515–1521.
Madden, P. A., et al. (2005). A genetically informed study of marital in-
stability and its association with offspring psychopathology. Journal of
Abnormal Psychology, 114, 570–586.
Dorius,C. J., Bahr,S. J.,Hoffmann,J. P., & Harmon, E. L. (2004).Parenting
practices as moderators of the relationship between peers and adolescent
marijuana use. Journal of Marriage and Family, 66, 163–178.
Duncan, S. C., Duncan, T. E., Biglan, A., & Ary, D. (1998). Contributions of
tivariate latent growth modeling approach. Drug and Alcohol Depen-
dence, 50, 57–71.
Eaves,L. J.,Silberg,J. L.,Maes,H. H.,Simonoff, E., Pickles,A.,Rutter,M.,
et al. (1997). Genetics and developmental psychopathology: 2. The main
effects of genes and environment on behavioral problems in the Virginia
Twin Study of Adolescent Behavioral Development. Journal of Child
Psychology and Psychiatry and Allied Disciplines, 38, 965–980.
Elkins, I. J., Iacono, W. G., & Doyle, A. E. (1997). Characteristics associated
with the persistence of antisocial behavior: Results from recent longitu-
dinal research. Aggression and Violent Behavior, 2, 101–124.
influences on parent–son relationships: Evidence for increasing
genetic influence during adolescence. Developmental Psychology, 33,
Elliott, D. S., & Huizinga,D. (1983).Social class and delinquent-behavior in
a national youth panel—1976–1980. Criminology, 21, 149–177.
and child adjustment: Parent–child conflict and perceived attachments as
mediators, potentiators, and mitigators of risk. Development and Psycho-
pathology, 16, 631–648.
network analysis. Journal of Research on Adolescence, 16, 159–186.
Ernst, M., Luckenbaugh, D. A., Moolchan, E. T., Leff, M. K., Allen, R.,
Eshel, N., et al. (2006). Behavioral predictors of substance-use initiation
in adolescents with and without attention-deficit/hyperactivity disorder.
Pediatrics, 117, 2030–2039.
Evrony, H., Ulbricht, J., & Neiderhiser, J. M. (2010). Understanding gene,
environment, and Gene?Environment interaction effects: The example
of childhood externalizing disorders. In K. P. Tercyak (Ed.), Handbook
of genomics and the family: Psychosocial context for children and ado-
lescents (pp. 59–86). New York: Springer.
Feinberg,M. E., Reiss, D., Neiderhiser, J. M., & Hetherington, E. M. (2005).
Differential association of family subsystem negativity on siblings’ mal-
adjustment: Using behavior genetic methodsto test process theory. Jour-
nal of Family Psychology, 19, 601–610.
Fletcher, A. C., Steinberg, L., & Williams-Wheeler, M. (2004). Parental in-
fluences on adolescent problem behavior: Revisiting Stattin and Kerr.
Child Development, 75, 781–796.
relationships in their social networks. Developmental Psychology, 21,
Gable, S., Belsky, J., & Crnic, K. (1992). Marriage, parenting, and child de-
velopment: Progress and prospects. Journal of Family Psychology, 5,
Ge, X., Conger, R. D., Cadoret, R. J., Neiderhiser, J. M., Yates, W., &
Troughton, E. (1996). The developmental interface between nature and
behaviors. Developmental Psychology, 32, 574–589.
Gillespie, N. A., Neale, M. C., Jacobson, K., & Kendler, K. S. (2009). Mod-
eling the genetic and environmental association between peer group de-
viance and cannabis use in male twins. Addiction, 104, 420–429.
Harakeh, Z., Neiderhiser, J. M., Spotts, E. L., Engels, R., Scholte,R. H. J., &
Reiss, D. (2008). Genetic factors contribute to the association between
peers and young adults smoking: Univariate and multivariate behavioral
genetic analyses. Addictive Behaviors, 33, 1113–1122.
Harden, K. P., Turkheimer, E., Emery, R. E., D’Onofrio, B. M., Slutske,
W. S., Heath, A. C., et al. (2007). Marital conflict and conduct problems
in children of twins. Child Development, 78, 1–18.
The role of adolescent awareness. Child Development, 68, 333–350.
Helstrom, A., Bryan, A., Hutchison, K. E., Riggs, P. D., & Blechman, E. A.
(2004). Tobacco and alcohol use as an explanation for the association be-
tween externalizing behavior and illicit drug use among delinquent ado-
lescents. Prevention Science, 5, 267–277.
sitions: A family systems perspective. Monographs of the Society for Re-
search in Child Development, 57(2–3, Serial No. 227), 1–238.
Hetherington, E. M., Henderson, S. H., O’Connor, T. G., Insabella, G. M.,
Taylor, L., Anderson, E. R., et al. (1999). Adolescent siblings in step-
families: Family functioning and adolescent adjustment. Monographs
of the Society for Research in Child Development, 64(4, Serial No.
Hopfer, C. J., Crowley, T. J., & Hewitt, J. K. (2003). Review of twin and
adoption studies of adolescent substance use. Journal of the American
Academy of Child & Adolescent Psychiatry, 42, 710–719.
Iervolino, A. C., Pike, A., Manke, B., Reiss, D., Hetherington, E., & Plomin,
R. (2002). Genetic and environmental influences in adolescent peer so-
cialization: Evidence from two genetically sensitive designs. Child De-
velopment, 73, 162–174.
Jessor, R., Donovan, J. E., & Widmer, K. (1980). Psychosocial factors in
adolescent alcohol and drug use: The 1978 National Sample Study
and the 1974–78 Panel Study. Boulder, CO: University of Colorado, In-
stitute of Behavioral Science.
Jessor, R., & Jessor, S. L. (1977). Problem behaviorand psychosocial devel-
opment. New York: Academic Press.
ence, 181, 1067–1070.
Kandel, D. B. (1978). Homophily, selection, and socialization in adolescent
friendships. American Journal of Sociology, 84, 427–436.
Kendler, K. S. (2001). Twin studies of psychiatric illness: An update. Ar-
chives of General Psychiatry, 58, 1005–1014.
Kendler, K. S., & Baker, J. H. (2007). Genetic influences on measures of the
environment: A systematic review. Psychological Medicine, 37, 615–626.
Prescott, C. A. (2007). Creating a social world—A developmental
twin study of peer-group deviance. Archives of General Psychiatry, 64,
der and peer deviance in males. Psychological Medicine, 38, 1001–1011.
Kendler, K. S., Jacobson, K. C., Prescott, C. A., & Neale, M. C. (2003).
Specificity of genetic and environmental risk factors for use and abuse/
dependence of cannabis, cocaine, hallucinogens, sedatives, stimulants,
and opiates in male twins. American Journal of Psychiatry, 160,
Kendler, K. S., Karkowski, L. M., Corey, L. A., Prescott, C. A., & Neale, M.
C. (1999). Genetic and environmental risk factors in the aetiology of il-
licit drug initiation and subsequent misuse in women. British Journal
of Psychiatry, 175, 351–356.
Kendler, K. S., Karkowski, L. M., Neale, M. C., & Prescott, C. A. (2000).
Illicit psychoactive substance use, heavy use, abuse, and dependence in
a US population-based sample of male twins. Archives of General Psy-
chiatry, 57, 261–269.
Kendler, K. S., Karkowski, L., & Prescott, C. A. (1999). Hallucinogen, opi-
ate, sedative and stimulantuse and abuse in a population-basedsampleof
female twins. Acta Psychiatrica Scandinavica, 99, 368–376.
Kendler,K. S., Myers, J., & Prescott, C. A. (2007).Specificityof geneticand
environmental risk factors for symptoms of cannabis, cocaine, alcohol,
caffeine, and nicotine dependence. Archives of General Psychiatry, 64,
Kendler, K. S., Schmitt, E., Aggen, S. H., & Prescott, C. A. (2008). Genetic
use from early adolescence to middle adulthood. Archives of General
Psychiatry, 65, 674–682.
Kerr, M., & Stattin, H. K. (2000). What parents know, how they know it, and
several forms of adolescent adjustment: Further support for a reinterpre-
tation of monitoring. Developmental Psychology, 36, 366–380.
Kerr, M., Stattin, H., & Burk, W. J. (2010). A reinterpretation of parental
monitoring in longitudinal perspective. Journal of Research on Adoles-
cence, 20, 39–64.
friendsmoderating and mediating effects of parental monitoring and peer
activity contexts. Merrill–Palmer Quarterly, 56, 529–556.
Kung, E. M., & Farrell, A. D. (2000). The role of parents and peers in early
adolescent substance use: An examination of mediating and moderating
effects. Journal of Child and Family Studies, 9, 509–528.
Lac, A., & Crano, W. D. (2009). Monitoring matters: Meta-analytic review
reveals the reliable linkage of parental monitoring with adolescent mari-
juana use. Perspectives on Psychological Science, 4, 578–586.
Loeber, R., & Dishion, T. J. (1984). Boys who fight at home and school:
Family conditions influencing cross-setting consistency. Journal of Con-
sulting and Clinical Psychology, 52, 759–768.
research on adolescent substance use: Methods,findings,and challenges.
Journal of the American Academy of Child & Adolescent Psychiatry, 49,
Manke, B., McGuire, S., Reiss, D., Hetherington, E. M., & Plomin, R. (1995).
Genetic contributions to adolescents’ extrafamilial social interactions:
Teachers, best friends, and peers. Social Development, 4, 238–256.
Martins, S. S., Storr, C. L., Alexandre, P. K., & Chilcoat, H. D. (2008). Ado-
lescent ecstasy and other drug use in the National Survey of Parents and
Youth: The role of sensation-seeking, parental monitoring and peer’s
drug use. Addictive Behaviors, 33, 919–933.
McGue, M., & Bouchard, T. J., Jr. (1984). Adjustment of twin data for the
effects of age and sex. Behavior Genetics, 14, 325–343.
McGue,M., Sharma,A., & Benson,P. (1996). Theeffect of commonrearing
mental Psychology, 32, 604–613.
McHale, S. M., Bissell, J., & Kim, J. Y. (2009). Sibling relationship, family,
and genetic factors in sibling similarity in sexual risk. Journal of Family
Psychology, 23, 562–572.
Molina, B. S. G., & Pelham, W. E. (2003). Childhood predictors of adoles-
cent substance use in a longitudinal study of children with ADHD. Jour-
nal of Abnormal Psychology, 112, 497–507.
Aggression as a mediator of genetic contributions to the association be-
tween negative parent–child relationships and adolescent antisocial be-
havior. European Child and Adolescent Psychiatry, 16, 128–137.
Natsuaki, M. N., Ge, X., Reiss, D., & Neiderhiser, J. M. (2009). Aggressive
behavior between siblings and the development of externalizing prob-
lems: Evidence from a genetically sensitive study. Developmental Psy-
chology, 45, 11.
ginia Commonwealth University, Department of Psychiatry.
Neiderhiser, J. M., Reiss, D., & Hetherington, E. M. (2007). The Nonshared
Environment in Adolescent Development (NEAD) Project: A longitu-
dinalfamilystudyoftwinsand siblingsfromadolescenceto youngadult-
hood. Twin Research and Human Genetics, 10, 74–83.
Neiderhiser, J.M., Reiss, D., Hetherington,E.,& Plomin,R.(1999). Relation-
ships between parentingand adolescent adjustment overtime: Genetic and
environmental contributions. Developmental Psychology, 35, 680–692.
Neiderhiser, J. A., Reiss, D., Lichtenstein, P., Spotts, E. L., & Ganiban, J.
(2007). Father–adolescent relationships and the role of genotype–envi-
ronment correlation. Journal of Family Psychology, 21, 560–571.
Neiderhiser, J. M., Reiss, D., Pedersen, N. L., Lichtenstein, P., Spotts, E. L.,
& Hansson, K. (2004).Genetic and environmental influences on mother-
ing of adolescents: A comparison of two samples. Developmental Psy-
chology, 40, 335–351.
Nichols, R. C., & Bilbro, W. C. (1966). The diagnosis of twin zygosity. Acta
Geneticae et Statistica Medica, 16, 265–275.
Nonnemaker, J. M., Crankshaw, E. C., Shive, D. R., Hussin, A. H., & Far-
relly, M. C. (2011). Inhalant use initiation among U.S. adolescents: Evi-
dence from the National Surveyof Parents and Youth using discrete-time
survival analysis. Addictive Behaviors, 36, 878–881.
O’Connor, T. G., Deater-Deckard, K., Fulker, D., Rutter, M., & Plomin, R.
(1998). Genotype–environment correlations in late childhood and early
adolescence: Antisocial behavioral problemsand coercive parenting. De-
velopmental Psychology, 34, 970–981.
Patterson, G. (1982). Coercive family process: A social learning approach.
Eugene, OR: Castalia.
Peterson, J. L., & Zill, N. (1986). Marital disruption, parent–child relation-
ships and behavior problems in children. Journal of Marriage and the
Family, 48, 295–307.
ogy and Psychiatry, 44, 598–611.
Pike, A., & Eley, T. C. (2009). Links between parenting and extra-familial
relationships: Nature or nurture? Journal of Adolescence, 32, 519–533.
Pike, A., McGuire, S., Hetherington, E., & Reiss, D. (1996). Familyenviron-
ment and adolescent depressive symptoms and antisocial behavior: A
multivariate genetic analysis. Developmental Psychology, 32, 590–604.
Pike, A., McGuire, S., Hetherington, E. M., Reiss, D., & Plomin, R. (1996).
Family environment and adolescent depressive symptoms and antisocial
behavior: A multivariate genetic analysis. Developmental Psychology,
Piko, B. F., & Kovacs, E. (2010). Do parents and school matter? Protective
factors for adolescent substance use. Addictive Behaviors, 35, 53–56.
Plomin, R., Reiss, D., Hetherington, E., & Howe, G. W. (1994). Nature and
nurture: Genetic contributions to measures of the family environment.
Developmental Psychology, 30, 32–43.
Poulin, F., Kiesner, J., Pedersen, S., & Dishion, T. J. (2011). A short-term
longitudinal analysis of friendship selection on early adolescent sub-
stance use. Journal of Adolescence, 34, 249–256.
action in twin analysis. Twin Research, 5, 554–571.
Rasbash, J., Jenkins, J., O’Connor, T. G., Tackett, J., & Reiss, D. (2011). A
genetically informative sample. Journal of Personality and Social Psy-
chology, 100, 474–491.
Reiss, D., Neiderhiser, J. M., Hetherington, E. M., & Plomin, R. (2000). The
relationship code: Deciphering genetic and social influences on adoles-
cent development. Cambridge, MA: Harvard University Press.
Rende, R., Slomkowski, C., Lloyd-Richardson, E., & Niaura, R. (2005). Sib-
ling effects on substance use in adolescence: Social contagion and ge-
netic relatedness. Journal of Family Psychology, 19, 611–618.
J. M. Neiderhiser, K. Marceau, and D. Reiss
Rhee, S. H., Hewitt, J. K., Young, S. E., Corley, R. P., Crowley, T. J., & Stal- Download full-text
lings, M. C. (2003). Genetic and environmental influences on substance
initiation, use, and problem use in adolescents. Archives of General Psy-
chiatry, 60, 1256–1264.
Rhee,S. H.,& Waldman,I. D. (2002).Geneticandenvironmentalinfluences
on antisocial behavior: A meta-analysis of twin and adoption studies.
Psychological Bulletin, 128, 490–529.
Reiss, D. (2011). Longitudinal pathways from marital hostility to child
anger during toddlerhood: Genetic susceptibility and indirect effects
via harsh parenting. Journal of Family Psychology, 25, 282–291.
Biology–environmentinteractionand evocative biology–environmentcor-
problem adolescent behaviors. Behavior Genetics, 33, 205–220.
Rowe, D. C., & Gulley, B. L. (1992). Sibling effects on substance use and
delinquency. Criminology, 30, 217–233.
Rowe, D. C., & Rogers, J. L. (1997). Behavioral genetics, adolescent de-
viance, and “d”: Contributions and issues. In G. R. Adams (Ed.), Ad-
vances in adolescent psychology. Hillsdale, NJ: Erlbaum.
Schaefer,E., & Edgerton,M.(1981). Thesiblinginventoryof behavior. Cha-
pel Hill, NC: University of North Carolina Press.
Siebenbruner,J.,Englund, M.M.,Egeland,B., & Hudson, K.(2006).Devel-
opmental antecedents of late adolescence substance use patterns. Devel-
opment and Psychopathology, 18, 551–571.
Sieving,R. E., Perry, C. L., & Williams, C. L. (2000). Dofriendships change
ence in young adolescents’ alcohol use. Journal of Adolescent Health,
Slomkowski, C., Rende, R., Conger, K. J., Simons, R. L., & Conger, R. D.
(2001). Sisters, brothers, and delinquency: Evaluating social influence
during early and middle adolescence. Child Development, 72, 271–283.
Slomkowski, C., Rende, R., Novak, S., Lloyd-Richardson, E., & Niaura, R.
(2005). Sibling effects on smoking in adolescence: Evidence for social in-
fluence from a genetically informative design. Addiction, 100, 430–438.
Spitz, E., Moutier, R., Reed, T., Busnel, M. C., & Marchaland, C. (1996).
Comparative diagnoses of twin zygosity by SSLP variant analysis, ques-
tionnaire, and dermatoglyphic analysis. Behavior Genetics, 26, 55–63.
Stattin, H., & Kerr, M. (2000). Parental monitoring: A reinterpretation. Child
Development, 71, 1072–1085.
Stormshak,E. A., Comeau, C. A., & Shepard, S. A. (2004). The relative con-
tribution of sibling deviance and peer deviance in the prediction of sub-
stance use across middle childhood. Journal of Abnormal Child Psychol-
ogy, 32, 635–649.
Straus, M. A. (1979). Measuring intrafamily conflict and violence: The Con-
Straus, M. A. (1987). Conflict Tactics Scales. In N. Freedman & R. Sherman
(Eds.), Handbook of measurements for marriage and family therapy
(pp. 159–161). New York: Brunner/Mazel.
Uhl, G. R., Drgon, T., Johnson, C., Li, C. Y., Contoreggi, C., Hess, J., et al.
Genome-wide association approaches identify “connectivity constella-
tion” and drug target genes with pleiotropic effects. Annals of the
New York Academy of Sciences, 1141, 318–381.
Ulbricht, J. A., & Neiderhiser, J. M. (2010). Genotype–environment correla-
tion and family relationships. Unpublished manuscript.
tifying shared environmental contributions to early substance use: The
respective roles of peers and parents. Journal of Abnormal Psychology,
Windle, M. (2000). Parental, sibling, and peer influences on adolescent sub-
stance use and alcohol problems. Applied Developmental Science, 4, 98–
Wu, P., Liu, X. L., & Fan, B. (2010). Factors associated with initiation of ec-
stasy use among US adolescents: Findings from a national survey. Drug
and Alcohol Dependence, 106, 193–198.