Content uploaded by Alessio Vieno
Author content
All content in this area was uploaded by Alessio Vieno
Content may be subject to copyright.
Parenting and Antisocial Behavior: A Model of the Relationship Between
Adolescent Self-Disclosure, Parental Closeness, Parental Control, and
Adolescent Antisocial Behavior
Alessio Vieno
University of Padova
Maury Nation
Vanderbilt University
Massimiliano Pastore and Massimo Santinello
University of Padova
This study used data collected from a sample of 840 Italian adolescents (418 boys; Mage ⫽12.58) and
their parents (657 mothers; Mage ⫽43.78) to explore the relations between parenting, adolescent
self-disclosure, and antisocial behavior. In the hypothesized model, parenting practices (e.g., parental
monitoring and control) have direct effects on parental knowledge and antisocial behavior. Parenting
style (e.g., parent– child closeness), on the other hand, is directly related to adolescent self-disclosure,
which in turn is positively related to parental knowledge and negatively related to adolescents’ antisocial
behavior. A structural equation model, which incorporated data from parents and adolescents, largely
supported the hypothesized model. Gender-specific models also found some gender differences among
adolescents and parents, as the hypothesized model adequately fit the subsample of mothers but not
fathers. Mothers’ closeness to girls predicted their knowledge of their daughters’ behavior; mothers’
control predicted boys’ antisocial behavior.
Keywords: parenting, monitoring, self-disclosure, deviant behavior, early adolescence
Numerous studies have examined the role parenting plays in the
cause and prevention of adolescent antisocial behavior. These
studies suggest that some aspects of parents’ behavior are associ-
ated with adolescent problem behavior (e.g., L. G. Simons &
Conger, 2007). These studies, along with Patterson’s studies of
juvenile delinquency (Patterson, 1982; Patterson, Reid, & Dishion,
1992), indicate that parental knowledge of adolescents’ where-
abouts and activities is an important predictor of antisocial behav-
ior. Parental knowledge has often been conceptualized as primarily
a function of parenting practices, such as parental monitoring (e.g.,
soliciting information from an adolescent) and parental control
(e.g., requiring information prior to granting permission; for a
review see Crouter & Head, 2002). That is, active parent involve-
ment, through solicitation and control, increases parental knowl-
edge, which ultimately acts as a protective factor against poor
psychosocial outcomes for adolescents. Despite the numerous stud-
ies linking parental knowledge and adolescent outcomes (e.g., Dar-
ling, Cumsille, Caldwell, & Dowdy, 2006; Dishion & McMahon,
1998; Fletcher, Steinberg, & Williams-Wheeler, 2004; Patterson &
Stouthamer-Loeber, 1984; Soenens, Vansteenkiste, Luyckx, &
Goossens, 2006), questions remain about how these constructs are
connected.
Recent studies have questioned the assumption that parental
knowledge is primarily a byproduct of parental practices. Indeed,
parents may receive information about their adolescent’s activities
through (a) asking their adolescent, (b) limiting or controlling their
adolescent’s activities, or (c) the adolescent’s self-disclosure (Kerr
& Stattin, 2000). Kerr and Stattin (2000) suggested that parental
knowledge may be more related to individual differences in ado-
lescent self-disclosure than the result of parental practices (see also
Stattin & Kerr, 2000). One implication of their research is that
parental control and monitoring may both be positively associated
with parental knowledge, yet these are clearly different constructs.
In fact, they have suggested that specific parenting behaviors, such
as soliciting information from the adolescent, may be relatively
unimportant in determining levels of parental knowledge. A sec-
ond implication is that parenting style, as indicated by the emo-
tional climate of the relationship, may be more important in
establishing a relational context in which adolescents feel com-
fortable sharing information, thereby increasing parental knowl-
edge. The purpose of this study is to examine the relationship
between parenting practices, parenting style, and adolescent self-
disclosure, using a sample of Italian middle school students.
Subsequent studies (Darling et al., 2006; Fletcher et al., 2004;
Laird, Pettit, Bates, & Dodge, 2003; Smetana, Crean, & Daddis,
2002; Soenens et al., 2006) have continued to explore Kerr and
Stattin’s (2000; Stattin & Kerr, 2000) hypotheses. For example,
Alessio Vieno, Massimiliano Pastore, and Massimo Santinello, Depart-
ment of Developmental and Social Psychology, University of Padova,
Padova, Italy; Maury Nation, Center for Community Studies, Vanderbilt
University.
The research for this article was funded by the Italian Ministry of
University and Research (MIUR) in 2003 (Protocol 2003111478).
Correspondence concerning this article should be addressed to Alessio
Vieno, Department of Developmental and Social Psychology, University of
Padova, via Belzoni, 80 -35131- Padova, Italy. E-mail: alessio.vieno@
unipd.it
Developmental Psychology © 2009 American Psychological Association
2009, Vol. 45, No. 6, 1509–1519 0012-1649/09/$12.00 DOI: 10.1037/a0016929
1509
Soenens et al. (2006) argued that the relation between parenting
behaviors and adolescent self-disclosure is more complex because
the relation between parenting practices and adolescent disclosure
has to be considered in the broader context of the quality of the
parent– child relationship. Through structural equation modeling,
Soenens et al. demonstrated that self-disclosure mediated the re-
lations between parenting practices (monitoring and control) and
parental knowledge, which in turn was related to antisocial behav-
ior and affiliation with peers engaging in problem behavior.
Reviews of Kerr and Stattin (2000; Stattin & Kerr, 2000), along
with subsequent studies, suggest that their conclusions may be
limited by the data sources, analysis strategies, and sample char-
acteristics. Specifically, some of these studies used single data
sources (either the parent or the adolescent). Fletcher et al. (2004),
for example, used only adolescent self-reports of these measures.
In studies where information was available from both parents and
children, such as Soenens et al. (2006), the information was
analyzed separately. Thus, the association between variables might
be partly a methodological artifact. Another limitation is that the
majority of related studies have focused on middle and late ado-
lescence (from 14 to 21 years of age). However, early adolescence
is a critical period in the development of many risk behaviors
(Moffitt, Caspi, Dickson, Silva, & Stanton, 1996). Further, al-
though peers may become a more important reference group in
shaping adolescent behaviors (Brown, Mounts, Lamborn, & Stein-
berg, 1993; Meeus & Dekovic, 1995), numerous studies show that
the parent– child relationship remains important for the psychoso-
cial adjustment of young adolescents (e.g., M. B. Simons, Chen,
Abroms, & Haynie, 2004; R. L. Simons, Chao, Conger, & Elder,
2001; Snyder & Huntley, 1990). Some have even suggested that
early adolescence is the key period in which a trajectory is set for
future behavior problems (Pettit, Bates, Dodge, & Meece, 1999).
In addition, a majority of these studies did not address whether
the relationships between parental monitoring, self-disclosure, and
antisocial behavior differ according to a child’s gender. For ex-
ample, Gorman-Smith and Loeber (2005) found that parental mon-
itoring was an important predictor of delinquency for both adoles-
cent boys and girls. Others (e.g., Scaramella, Conger, & Simons,
1999; Spoth, Neppl, Goldberg-Lillehoj, Jung, & Ramisetty-
Mikler, 2006) have found that a positive relationship with parents
is associated with fewer social and behavioral problems for both
genders. The emerging literature on the relationship between child
self-disclosure and parental knowledge is inconclusive with re-
spect to the impact of gender; some studies found no gender
differences (Soenens et al., 2006), and others reported different
patterns for males and females (Crouter & Head, 2002). In general,
studies suggest that male adolescents tend to self-disclose less
frequently to parents about risky behaviors (e.g., sexual behavior;
Consedine, Sabag-Cohen, & Krivoshekova, 2007), raising the
question of whether Kerr and Stattin’s (2000; Stattin & Kerr, 2000)
model and subsequent models that include self-disclosure are
gender specific.
Even fewer studies have examined the relation between parental
monitoring and self-disclosure based on parents’ gender. However,
several studies suggest that there are differences in the parenting
styles of mothers and fathers and that some of these differences
may be evidenced in their children’s delinquent behavior. For
example, maternal and paternal differences have been observed in
aspects of child rearing such as emotional sensitivity and structur-
ing of children’s play (Lovas, 2005). In relation to parenting style,
L. G. Simons and Conger (2007) observed that fathers were less
likely to practice authoritative parenting. Also, studies of single
parents show that parents’ gender is a significant predictor of
adolescents’ involvement in alcohol and drug use (Hoffman &
Johnson, 1998). In fact, Demuth and Brown (2004) found that
single fathers had higher family incomes but were less involved
and provided less supervision and monitoring; these differences
were associated with more antisocial behavior. Although their
study was conducted on a U.S. sample, the mother–father differ-
ences are consistent with the parental roles in many Italian families
(Claes et al., 2005), where fathers are more likely to be more
involved in managing the economic wellbeing of the family than in
child rearing (Ramella & Sindoni, 1997).
Most studies examining adolescent self-disclosure have not ex-
amined the differences between parents’ gender. Those that have
examined parental gender differences have provided mixed results.
Soenens et al. (2006) examined parent reports and child reports in
separate models. They found that their model fit equally well for
maternal and paternal behavior based on both parent and adoles-
cent reports. On the other hand, Smetana, Metzger, Gettman, and
Campione-Barr (2006) found that adolescents were more willing
to disclose to mothers than to fathers, which suggests that mothers
may be more likely to have knowledge of their adolescents’
behavior.
The Present Study
The results across previous studies and the emerging questions
related to parental knowledge and adolescent behavior indicate the
need for additional studies. Indeed, if researchers are to understand
why these constructs are associated with poor adolescent out-
comes, it is essential to understand the relations between these
constructs. The principal aim of the present study is to test and
extend the integrated model of parenting and adolescent behavior
problems determined by Soenens et al. (2006). We do this by
exploring the direct and indirect associations between parental
control, closeness of parent– child relationship, adolescents’ will-
ingness to self-disclose, and their collective relation with parental
knowledge and antisocial behavior (see Figure 1). Furthermore, we
examine whether the association between these constructs differs
as a function of both parent and child gender. Finally, in contrast
to Soenens et al., we analyze both parent and child data in a single
model, as opposed to separate models for child and parent data.
When models are run separately (for child and parents), significant
predictions are likely. However, the association between variables
may be partially due to a methodological error. Researchers (e.g.,
Bartels et al., 2004; Cook & Goldstein, 1993) have suggested that
structural equation modeling could be used to address this limita-
tion. Thus, we address this issue by combining parent and adoles-
cent information when considering parental control and parent–
child closeness to create a latent construct.
According to our model, adolescents who have close relation-
ships with their parents and are exposed to more parental control
should be more likely to disclose, in part because of synergy
between parenting style and parenting practices. Kerr and Stattin
(2000; Stattin & Kerr, 2000) implicitly conceptualized parental
knowledge as a byproduct of parental practices or specific behav-
iors designed to accomplish the goal of increasing parental knowl-
1510 VIENO, NATION, PASTORE, AND SANTINELLO
edge. However, Darling and Steinberg (1993) suggested that so-
cializing adolescents into parental goals and values is a byproduct
of parenting practices and parenting style, where parenting style is
a more general set of behaviors that establishes an attitude toward
the child and sets the emotional climate or closeness of the rela-
tionship. Depending on the context established by the parental
style, identical parental practices, such as those related to parental
control, may be experienced very differently by an adolescent as
either caring concern or an intrusion.
This line of thinking suggests that if parents’ styles are such that
they are involved and cultivate close relationships with their ado-
lescents, the adolescents may in turn feel that their parents are
entitled to more information about their unsupervised activities;
therefore, these adolescents may spontaneously offer information
and respond more favorably to parental control. Thus, parental
control can have a direct effect on problem behavior by limiting
the opportunities for adolescents to engage in problem behaviors
and indirect effects depending on whether adolescents are com-
fortable disclosing their whereabouts and activities to parents. This
is consistent with Kerns, Aspelmeier, Gentzler, and Grabill (2001),
who found that when it comes to adolescent self-disclosure, pa-
rental warmth, responsiveness, and availability predict higher lev-
els of parental knowledge than parental control.
Our model extends previous models by exploring an early
adolescent sample and by including both parent and child ratings
of the closeness of their relationship. More importantly, we exam-
ine a model that proposes the relation between parenting and
adolescent engagement in antisocial behavior is direct or indirect,
depending on whether the focus is on the practices associated with
parental control or on parental style as indicated by the closeness
of the parent– child relationship. As illustrated in Figure 1, we
hypothesize that the effect of parent– child closeness on antisocial
behavior is primarily indirect and is mediated through adolescent
self-disclosure. Parental control, on the other hand, has direct
effects on adolescent antisocial behavior. However, this direct
relation is complemented by an indirect relation through adoles-
cent self-disclosure. That is, the overall effect of parental control
on adolescent problem behavior is enhanced when adolescents
disclose their behavior. Although we may find gender differences
in the amount of self-disclosure, we believe that the impact of
self-disclosure will be similar regardless of gender (similar to
Gorman-Smith & Loeber, 2005).
A significant implication of this model is that the connection
between parental knowledge and adolescent problem behaviors
may be spurious. More specifically, we suggest that parental
knowledge is an indicator of the effectiveness of parental style and
practices. We further suggest that there is no inherent power in
parents knowing what their adolescents are doing. If both adoles-
cent self-disclosure and parental knowledge reflect the quality of
the parent– child relationship, parental knowledge may be less
important or appear unimportant in predicting adolescent behavior
problems in models that contain both constructs.
Finally, we extend previous research by examining these ques-
tions using an Italian sample. Claes, Lacourse, Bouchard, and
Perucchini (2003) reported that, compared with North American
and North European youth (where the majority of research has
been conducted), Italian youth reported higher levels of conflicts
but also warmer relationships with their parents. Moreover, North
American and North European adolescents reported that their
parents more often adopted an interactive style (discussion) when
rules were not respected, whereas Italian youths more frequently
named punitive reactions from their parents in such situations.
Because parental practices are often shaped by culture (Goodnow,
1985), an Italian sample provides an important opportunity to
Disclosure
Closeness
Antisocial
Behavior
Mother
Students
Students
Mother
Control
Students
Mother
Knowledge
Students
Students
.29
.27
.54
.16
.15
.64
.16
ns
.50
-.23
.13
-.16
-.36
.76*
.57*
.21
.39
.50
Figure 1. Standardized path coefficients for the model proposed by Soenens et al. (2006) for mothers.
2
(13) ⫽
53.11, p⬍.01, comparative fit index ⫽.97, normed fit index ⫽.96, non-normed fit index ⫽.93, root-mean-
square error of approximation ⫽.06 (.05–.08).
ⴱ
We assigned the fixed value .15 times the variance of yto the
corresponding parameter theta-epsilon (Jo¨reskog & So¨ rbom, 1996a, p. 196).
1511
PARENTING, DISCLOSURE, AND ANTISOCIAL BEHAVIOR
examine the relationship between these constructs in a culturally
specific context.
Method
Sample
The present study was conducted in Padova, a midsize city in
the northeast of Italy. A total of 1,196 students were enrolled in the
sample schools; 35 of them did not regularly attend school. There-
fore, the total possible sample was 1,161. Parental permission to
participate was obtained for 1,157 students. Of those, 1,147 early
adolescents (588 boys and 559 girls) participated in the study.
Parents received one questionnaire. We invited parents to decide
if the mother or father would complete the survey. Eighty-four
percent (N⫽959) of parents agreed to participate. Of the 959
parent participants (84% of the total sample), 742 (77.3%) were
mothers, 210 (21.9%) were fathers, and 7 (0.8%) were the adoles-
cents’ grandparents or adult siblings (Mage ⫽42.78, SD ⫽5.19).
Due to their small numbers, surveys completed by grandparents or
adult siblings were removed from the analyses.
Analysis Sample
Because some parents did not respond to the questionnaire (N⫽
188) and some of the children and parents (N⫽119) were missing
data required by the imputation procedure (see Statistical Analy-
ses), the theoretical model was tested on a final sample of 840
adolescents and their parents. Ages of students (418 boys and 422
girls) ranged from 11 to 15, with a mean of 12.58 (SD ⫽.93).
Frequency distributions by grade showed that a total of 310 stu-
dents were in the sixth grade (36.9%), 249 in the seventh grade
(29.5%), and 281 in the eighth grade (33.6%). Participants were
predominantly born in Italy (95.3%), with small percentages of
participants from Eastern Europe (2.5%) and North Africa (1%),
and the remainder coming from Asia, the United States, and
Western Europe. In relation to the family structure, 91.4% of the
students came from a two-parent family (with parents married and
living together), 4.1% lived in a stepfamily, and 4.5% lived with
one parent (mainly the mother).
Of the 840 parents (Mage ⫽43.78), 657 were mothers. Moth-
ers’ mean age was 42.07 years (SD ⫽4.97). On an 8-point rating
scale, their mean educational level was 4.27 (SD ⫽1.47), indicat-
ing that they had at least 11 years of education on average. Fathers’
mean age was 45.49 years (SD ⫽4.99). Their mean educational
level was 4.55 (SD ⫽1.47), indicating that they had 13 years of
education on average. Compared to the excluded sample, the
analysis sample was equally distributed in terms of students’
gender,
2
(1) ⫽1.81, ns, and age, t(1142) ⫽0.69, ns.
Given the way mothers and fathers were recruited into the study,
we also compared the percentage of boys (78.6%) and girls
(76.7%) whose mother participated. No differences were found,
2
(1) ⫽1.09, ns. Also, we found no differences related to mothers’
and fathers’ education, t(920) ⫽2.13, ns, or to the youths’ age,
t(920) ⫽1.43, ns.
Finally, because we required parental participation and active
parental consent, it is possible that the youth included in the
analysis may differ systematically from those who were excluded
(Weinberger, Tublin, Ford, & Feldman, 1990). For this reason, we
used a ttest to compare participants to students who were excluded
because their parents did not participate in the study. Overall, we
did not find any differences in terms of parental control, t(1153) ⫽
1.10, ns; mothers’ closeness, t(1076) ⫽0.12, ns; fathers’ close-
ness, t(1067) ⫽0.04, ns; self-disclosure to mother, t(1134) ⫽2.72,
ns; self-disclosure to father, t(1109) ⫽0.67, ns; mothers’ knowl-
edge, t(1129) ⫽3.80, ns; fathers’ knowledge, t(1105) ⫽1.66, ns;
and adolescents’ antisocial behavior, t(1153) ⫽0.77, ns.
Procedure
The present study was approved by the institutional review
committee at the University of Padova. Parents of all sixth, sev-
enth, and eighth grade students (from five public middle schools)
were asked for their consent to participate and for their permission
to allow their children to participate in the study.
Students filled out their portion of the questionnaire during a
single class period. Data were collected during a 4-week period
and were proctored by research assistants. Students were given
approximately 50 min to complete the questionnaire.
The parent’s questionnaire was given to the student in a sealed
envelope. In addition to the questionnaire, the envelope contained
basic information and instructions for completing and returning the
survey in a sealed envelope. Each child returned a sealed envelope
to the teacher, containing either the completed survey or the
parents’ decision not to participate.
Measures
Questionnaires for both youths and parents were composed of
measures of parenting (closeness and control) and antisocial be-
havior taken from international literature and translated, piloted,
and validated in Italian. Using a procedure similar to Stattin and
Kerr (2000), we investigated self-disclosure and parental knowl-
edge via the questionnaire administered to students.
Parent– child closeness. The Closeness to Parents Scale
(Buchanan, Maccoby, & Dornbusch, 1991; Vieno, 2006) was used
by adolescent participants to rate their closeness to both their
mother and father. The scale was composed of nine items to rate
each parent (e.g., “How often does your [mother/father] express
affection or liking for you?”). The participants responded on a
scale ranging from 1 (not at all)to5(often). The Cronbach’s alpha
reliability for the nine-item scale was .86 for mothers (bootstrap
95% CI [.83, .88]) and .90 for fathers (bootstrap 95% CI [.88,
.92]). Responses were averaged for the measures of mother– and
father– child closeness (child report).
Parents responded regarding their own behavior in terms of their
willingness to listen to and share feelings and experiences with
their children (Child Rearing Practices Report [CRPR]; Rickel &
Biasatti, 1982). The questionnaire was composed of 10 items (e.g.,
“I joke and play with my child”). Parents responded on a scale
ranging from 1 (not at all descriptive of me)to5(highly descrip-
tive of me). The alpha for the 10-item scale was .87 (bootstrap 95%
CI [.84, .90]). Responses were averaged for the measure of parent–
child closeness (parent report).
Parental control. Parental control was measured by a proxy,
conceptualized in accordance with Fletcher et al. (2004) to assess
the extent to which decisions regarding key areas of adolescents’
lives were made by parents, rather than by adolescents themselves.
1512 VIENO, NATION, PASTORE, AND SANTINELLO
Ten items were drawn from a scale developed by Steinberg (1987)
to assess the level of decision-making autonomy and were then
used to approximate parental control over different activities and
behaviors (e.g., “Which friend I spent time with” and “How I
spend my money”). The participants responded on a 5-point scale:
(a) “I decide this without discussing it with parents”; (b) “I make
the final decision after discussing it with my parents”; (c) “My
parents and I make the decision together”; (4) “My parents make
the final decision after discussing it with me”; and (d) “My parents
decide this without discussing it with me.” The 10 items were
averaged to yield a single scale with an alpha of .76 (Child Report;
bootstrap 95% CI [.72, .79]).
Parents rated the same items with respect to their parenting
behavior. The Cronbach’s alpha for the 10-item scale was .73
(bootstrap 95% CI [.69, .77]). Responses were averaged for the
measure of parental control (Parent Report).
Adolescents’ self-disclosure. Adolescents answered four items
relating to their voluntary self-disclosure (two for father and two
for mother). These items composed the self-disclosure subscale of
the monitoring scale proposed by Small, Silverberg, and Kerns
(1993) to distinguish between monitoring, knowledge, and self-
disclosure. The four items were as follows: “I spontaneously tell
my [mother/father] which friends I hang out with before I go out”
and “I spontaneously tell my [mother/father] how I hang out with
my friends before I go out.” The participants responded on a scale
ranging from 1 (never)to4(always). The correlation for the
two-item scale was r⫽.45 for mothers (bootstrap 95% CI [.43, .48])
and r⫽.61 for fathers (bootstrap 95% CI [.56, .66]). Responses were
averaged for the measures of adolescents’ self-disclosure to their
mothers and fathers (Child Report).
It should be noted that this operationalization of self-disclosure
only partially overlaps with the measure used by Stattin and Kerr
(2000). This decision was made for two reasons. First, in the initial
stage of the study (Vieno, 2006), we found no variability on the
two items concerning a night out. This is likely because of the
early stage of adolescence considered in the present study (M
age ⫽12.58). On the other hand, we found problems with the
reverse coded item, “Do you keep a lot of secrets from your
parents about what you do during your free time?” For this reason,
in the second stage of the study we decided to assess disclosure
using the two items mentioned, as they are essentially the same
subscale presented by Small et al. (1993).
Parental knowledge. Adolescents answered four items (two
for father and two for mother) concerning the level of knowledge
they offered to parents about their free- time activities. The items
used are the parental knowledge subscale of the monitoring scale
proposed by Small et al. (1993). The four items were as follows:
“My [mother/father] knows where I spend my free time after
school” and “My [mother/father] knows where I go when I go
out.” The participants responded on a scale ranging from 1 (never)
to4(always). The correlation for the two-item scale was r⫽.46
for mothers (bootstrap 95% CI [.44, .49]) and r⫽.62 for fathers
(bootstrap 95% CI [.55, .65]). Responses were averaged for the
measures of knowledge among mothers and fathers (Child Report).
It should be noted that the implementation of this measure was
modified for the same problems encountered for the self-disclosure
measure.
Adolescents’ antisocial behavior. An adapted version of the
self-report measure of antisocial behavior (Kiesner, 2002; Vieno,
2006) was used to measure antisocial behavior among youths. The
scale was composed of 11 items (e.g., “How many times did you
take money from your parents without their permission?”). Partic-
ipants were asked to respond by considering their actions within
the past 30 days. The answer options were based on a 6-point
ordinal scale, ranging from 1 (never)to6(more than 20 times).
The Cronbach’s alpha for this scale was .78 (bootstrap 95% CI
[.74, .81]). Responses were averaged for the measure of antisocial
behavior among adolescents (Child Report).
Parents rated 11 items from the Problem Checklist (PLST;
Oregon Social Learning Center, 1997) with respect to their child’s
misbehavior. Parents responded on a scale ranging from 1 (never)
to6(always/almost always). The alpha for the 11-item scale was
.83 (bootstrap 95% CI [.81, .84]). Responses were averaged for the
measure of antisocial behavior among adolescents (Parent Report).
Statistical Analyses
Before running the analyses, we performed data normalization
and imputation of missing values for all variables. Imputation of
missing values and computation of normal scores were performed
using PRELIS 2, based on a scheme described by Jo¨reskog and
So¨rbom (1996b, pp. 153–156) and a formula reported by Jo¨reskog,
So¨rbom, du Toit, and du Toit (2000, pp. 161–162). The missing
values were imputed based upon values observed in other cases
that had a similar response pattern over a set of matching variables.
Consequently, if there were also missing values for the matching
variables, the value could not be imputed. Thus, the cases excluded
were eliminated because so many data for these cases were miss-
ing that a value could not be imputed.
Structural equation modeling (Jo¨reskog & So¨rbom, 1996a), imple-
mented by the program LISREL (Version 8.50), was used to test
model fit separately for mothers and fathers. To address some of the
limitations in previous research (e.g., Soenens et al., 2006), we used
both parent and adolescent informants to create a latent construct for
parent–adolescent closeness and parental control, as well as antisocial
behavior; we used only information from student informants to create
latent constructs for self-disclosure and parental knowledge.
We considered a variety of indices as indicators of the model’s
overall goodness of fit: Chi-square (
2
), for example, was used as
a test of the null hypothesis that the model fit the data. However,
reliance on chi-square has been criticized, especially in the case of
large samples (more than 200; Jo¨reskog & So¨rbom, 1996a; Saris,
1982). For that reason, we also used the comparative fit index
(CFI) and non-normed fit index (NNFI), with values ranging from
0 (a poor fit) to 1 (a perfect fit). We also used the root-mean-square
error of approximation (RMSEA), which is considered a measure
of a good fit when lower than .06 (Hu & Bentler, 1999), and 90%
CI for RMSEA. Finally, we examined the squared multiple corre-
lations for the structural equations.
To evaluate child gender differences in the model, a multigroup
approach was used (Jo¨reskog & So¨rbom, 1996a; see, e.g., Byrne,
1989). This approach allows estimation of the fit of the model and
the parameters simultaneously on different subgroups. In particu-
lar, the hypothesis of the invariance of the covariance matrix (⌺)
and the hypothesis of the form invariance (same dimensions and
same patterns of fixed, free, and constrained values in all matrices;
k) on different groups tested the fit and parameters of the model
comparing boys and girls.
1513
PARENTING, DISCLOSURE, AND ANTISOCIAL BEHAVIOR
Results
Descriptive Statistics
Table 1 illustrates descriptive statistics for the subsample of
fathers and mothers on which we tested the models. The table
shows bivariate correlations among variables for the two sub-
samples and separates means and standard deviations by students’
gender. Similar to previous studies (see Kerr & Stattin, 2000), boys
scored higher on antisocial behavior on the basis of self- and
parental reports. Compared with girls, boys perceived more control
from their fathers; girls perceived more control from their mothers.
Boys scored higher on closeness to mother and on mother’s control
(both child and parent report). Girls scored higher on self-
disclosure to mothers and mothers’ knowledge. In the case of
fathers, in general the correlations were small but in the expected
direction. Although the correlation between all the indicators of the
same constructs measured on the children and parents was signif-
icant, the magnitude was relatively low, ranging from .18 to .27 for
fathers and from .24 to 34 for mothers. These results are generally
consistent with the results obtained by Soenens et al. (2006) and
Stattin and Kerr (2000).
Testing the Theoretical Model for Mothers
Analyses began by testing the proposed model (Figure 1) in
which all paths among the variables were assessed. One path
coefficient, from closeness to antisocial behavior, was not signif-
icant. Figure 1 represents the tested model with estimated stan-
dardized parameters. Nonsignificant paths are indicated by ns.
The resulting model for mothers produced these fit indices:
2
(13) ⫽53.11, p⬍.01, CFI ⫽.97, NNFI ⫽.93, RMSEA ⫽.06
(.05–.08). Observing the indices, it is possible to conclude that the
model produces an adequate fit. Moreover, the squared multiple
correlations for a structural equation are: R
1
2
⫽.41, R
2
2
⫽.44,
R
3
2
⫽.46, where 1 is disclosure, 2 is knowledge, and 3is
antisocial behavior. Thus, the model accounts for 41% of the
variance in adolescent disclosure, 44% of parental knowledge, and
46% of variance in antisocial behavior. Moreover, the global R
2
for the model was .47.
To more adequately evaluate how well the theoretical model fit
the data, we conducted a series of incremental fit tests to compare
the model obtained with other alternative models within a nested
model comparison framework (Bentler & Bonnet, 1980; Bollen,
1989; Ge, Conger, Lorenz, & Simons, 1994). The following four
theoretically meaningful models were compared: baseline, bivari-
ate (two gammas), no beta (four gammas), and the hypothesized
model:
1. The baseline is a null-gamma and beta model in which
only the measurement model is estimated and all the
structural coefficients are assumed to be zero (that is, we
assume that all the constructs are orthogonal).
2. A bivariate model (two gammas) is a model that is
traditionally employed in research on parenting and ad-
olescent antisocial behavior: Adolescents’ antisocial be-
havior was predicted by parental control and mother–
child closeness, but adolescents’ disclosure and mothers’
knowledge were isolated.
Table 1
Means, Standard Deviations, t Values for Gender, and Correlations Between Variables Separated for Mother and Father (Mother Above the Diagonal)
Variable 12345678
Girls
M(SD)
Boys
M(SD)
tvalues
(df ⫽656)
1. Parent–child closeness (CR) — .34
ⴱⴱ
.19
ⴱⴱ
.10
ⴱ
.47
ⴱⴱ
.38
ⴱⴱ
⫺.31
ⴱⴱ
⫺.22
ⴱⴱ
4.16 (0.71) 4.15 (0.67) 0.05
2. Parent–child closeness (PR) .18
ⴱ
— .12
ⴱⴱ
.10
ⴱ
.16
ⴱⴱ
.13
ⴱⴱ
⫺.10
ⴱ
⫺.26
ⴱⴱ
4.45 (0.53) 4.57 (0.48) 9.75
ⴱⴱ
3. Parental control (CR) .15
ⴱ
.04 — .32
ⴱⴱ
.37
ⴱⴱ
.35
ⴱⴱ
⫺.30
ⴱⴱ
⫺.04 2.30 (0.52) 2.38 (0.65) 3.35
ⴱ
4. Parental control (PR) .05 ⫺.13 .23
ⴱⴱ
— .08
ⴱ
.09
ⴱ
⫺.09
ⴱ
.03 2.58 (0.45) 2.67 (0.48) 5.90
ⴱ
5. Adol. self-disclosure (CR) .51
ⴱⴱ
.25
ⴱⴱ
.26
ⴱⴱ
.07 — .55
ⴱⴱ
⫺.44
ⴱⴱ
⫺.16
ⴱⴱ
2.97 (0.77) 2.80 (0.86) 7.33
ⴱⴱ
6. Parental knowledge (CR) .41
ⴱⴱ
.11 .27
ⴱⴱ
.40 .68
ⴱⴱ
—⫺.48
ⴱⴱ
⫺.16
ⴱⴱ
3.65 (0.56) 3.47 (0.65) 13.43
ⴱⴱ
7. Adol. antisocial behavior (CR) ⫺.24
ⴱⴱ
⫺.14 ⫺.27
ⴱⴱ
.09 ⫺.38
ⴱⴱ
⫺.29
ⴱⴱ
— .24
ⴱⴱ
1.30 (0.39) 1.50 (0.53) 28.61
ⴱⴱ
8. Adol. antisocial behavior (PR) ⫺.17
ⴱⴱ
⫺.27
ⴱⴱ
⫺.05 .06 ⫺.20
ⴱⴱ
⫺.28
ⴱⴱ
.27
ⴱⴱ
— 2.11 (0.64) 2.30 (0.64) 13.22
ⴱⴱ
M(SD)M(SD)M(SD)M(SD)M(SD)M(SD)M(SD)M(SD)
Females 3.92 (0.75) 4.32 (0.61) 2.28 (0.51) 2.58 (0.47) 2.66 (1.00) 3.17 (0.86) 1.25 (0.27) 1.98 (0.57)
Males 4.07 (0.76) 4.25 (0.70) 2.32 (0.56) 2.75 (0.45) 2.66 (0.92) 3.06 (0.92) 1.43 (0.42) 2.14 (0.64)
tvalues (df ⫽182) 1.68 0.56 0.24 6.19
ⴱⴱ
0.01 0.72 12.10
ⴱⴱ
2.92
ⴱ
Note. CR ⫽child report; PR ⫽parent report; adol. ⫽adolescent.
ⴱ
p⬍.05.
ⴱⴱ
p⬍.01.
1514 VIENO, NATION, PASTORE, AND SANTINELLO
3. A no-beta model (four gammas): Adolescents’ antisocial
behavior and disclosure and mothers’ knowledge were
predicted by both parental control and mother– child
closeness, but no betas are included.
4. Our hypothesized model, as presented in Figure 1, pos-
tulates a mediating process linking parenting (control and
closeness), adolescents’ disclosure and mothers’ knowl-
edge, and adolescent antisocial behavior.
The results of the comparisons across these models are presented
in Table 2.
First, the bivariate model (parenting predicting adolescent anti-
social behavior) provided a better fit with the data than the baseline
model. The coefficients in Table 2 show a significant reduction in
2
with three degrees of freedom (⌬
2
⫽299.52). By adding adolescent
disclosure and mothers’ knowledge constructs into our estimation
(just as a predictor and not as a mediating factor), the model
produced a significant improvement over the bivariate model
(⌬
2
⫽605.77). Finally, over the no-beta (four gammas) model, a
significant improvement was found by adding the connection
(betas) between adolescents’ disclosure and mothers’ knowledge
and adolescent antisocial behavior (⌬
2
⫽35.94). In observing the
remaining fit indices, both NNFI and CFI seem to show a gradient
in which there is a substantial and continuous fitness improvement
from the baseline to the hypothesized model.
Testing the Theoretical Model for Fathers
Analyses began by testing the proposed model (Figure 1) in
which all paths among the variables were assessed. The resulting
model for fathers did not produce an adequate fit:
2
(13) ⫽43.48,
p⬍.01, CFI ⫽.89, NNFI ⫽.80, RMSEA ⫽.11 (.08 –.15).
To examine where the problem occurred with the model for
fathers, we conducted a series of incremental fit tests to compare
the model obtained with other alternative models within a nested
model comparison framework. Again, four theoretically meaning-
ful models were compared in the tests: baseline, bivariate (two
gammas), no beta (four gammas), and the hypothesized model.
The results of the comparisons across these models are presented
in Table 3.
First, the bivariate model (parenting predicting adolescent anti-
social behavior) provided a better fit with the data than the baseline
model. The coefficients in Table 3 show a significant reduction in
2
with three degrees of freedom (⌬
2
⫽62.01). By adding
adolescent disclosure and fathers’ knowledge constructs into our
estimation (just as a predictor and not as a mediating factor), the
model produced a dramatically significant improvement over the
bivariate model (⌬
2
⫽189.96). Finally, over the no-beta (four
gammas) model, no significant improvement was found by adding
the connection (betas) between adolescents’ disclosure and fa-
thers’ knowledge and adolescent antisocial behavior (⌬
2
⫽4.83).
Both the NNFI and the CFI show a gradient in which we see a
substantial and continuous improvement until the no-beta (four
gammas) model. Thus, results indicate that adolescents’ antisocial
behavior and disclosure and father’s knowledge were directly
predicted by both father control and father– child closeness.
Testing the Theoretical Model for Mothers Based on
Child’s Gender
After evaluating the overall fit of the model for mothers, mul-
tigroup comparisons were used to examine the extent to which this
model is consistent, in terms of covariance matrices (⌺) and forms
(dimensions, and patterns of fixed, free, and constrained values; k)
across students’ gender. All the fit indices presented indicate
significant statistical differences in the covariance matrices,
2
(36) ⫽506.83, p⬍.01, CFI ⫽.63, NNFI ⫽.42, RMSEA ⫽.20
(.18 –.21), and forms,
2
(29) ⫽144.02, p⬍.01, CFI ⫽.93,
NNFI ⫽.87, RMSEA ⫽.11 (.09 –.12), between boys and girls. It
was therefore necessary to analyze and compare the structural
parameters of the model for the different subgroups (see Figure 2).
Almost all the paths are the same as the original model. How-
ever, the difference is that for boys there are significant paths
between control and antisocial behavior as well as disclosure and
antisocial behavior (but not for girls). For girls there is a significant
path between closeness and parental knowledge (but not for boys).
Moreover, the squared multiple correlations for structural equation
for boys are R
1
2
⫽.40, R
2
2
⫽.46, R
3
2
⫽.37; for girls they are
R
1
2
⫽.57, R
2
2
⫽.45, R
3
2
⫽.72. Differences in the models
suggest that as it relates to antisocial behavior, the benefits of
parental control and parental style differ according to the child’s
gender, with boys benefiting more from parental control and girls
benefiting from more closeness in the parent– child relationship.
Discussion
The present study employed a multi-informant design to exam-
ine the adolescent–parent processes associated with antisocial be-
havior in a sample of Italian early adolescents. The results pro-
vided general support for our proposed model for mothers and
partially conflicted with the results obtained by Soenens et al.
(2006). In support of our proposed model, we found that maternal
Table 2
Incremental Fit Comparisons for Nested Model (Mother)
Model
2
df p NNFI CFI
Change in
2
⌬
2
df p
Baseline (no gammas, no betas) 994.34 22 ⬍.001 .28 .44
Bivariate (two gammas) 694.82 19 ⬍.001 .31 .53 299.52 3 ⬍.001
No-beta (four gammas) 89.05 16 ⬍.001 .91 .95 605.77 3 ⬍.001
Hypothesized (four gammas, two betas) 53.11 13 ⬍.001 .93 .97 35.94 3 ⬍.001
Note. NNFI ⫽non-normed fit index; CFI ⫽comparative fit index.
1515
PARENTING, DISCLOSURE, AND ANTISOCIAL BEHAVIOR
control is positively related to early adolescents’ self-disclosure
and to mothers’ knowledge of their child’s behavior. On the other
hand, mother– child closeness is positively related to early adoles-
cents’ self-disclosure and to mothers’ knowledge and adolescents’
antisocial behavior. Our results indicate that the extent to which
mothers provided high levels of control over children’s behavior
and established a close relationship promoted self-disclosure and
decreased the probability that their child would engage in antiso-
cial behavior. This result supports Darling and Steinberg’s (1993)
argument that parenting practices and parenting style may have
distinct roles in regulating children’s behavior, with parenting style
promoting a positive family climate, which in turn promotes
parent– child emotional closeness and fosters the disclosure of
personal information from the child.
Also, our analysis suggests that mothers’ control was not di-
rectly associated with lower levels of adolescents’ involvement in
antisocial behavior; that is, most of the effects of parental control
on antisocial behavior were due to the increased disposition of the
child to disclose to their parents. According to Soenens et al.
(2006), this suggests that the direct effects of parental control on
problem behavior might be mediated mostly by adolescent disclo-
sure; it also suggests that parents might balance active surveillance
and firm standards for behavior (Gray & Steinberg, 1999) with a
focus on promoting a positive relationship with their child (Stattin
& Kerr, 2000).
The amount of variance in adolescent self-disclosure that was
explained by mother– child closeness suggests that there are other
important factors that may impact an adolescent’s willingness to
self-disclose. The importance of self-disclosure in predicting an-
tisocial behavior suggests that further investigation is needed to
determine child and environmental factors that may promote self-
disclosure. Biological factors, such as temperament, may affect
adolescents’ willingness to self-disclose or may affect how they
experience the parent– child relationship. Also, the general mea-
sure of disclosure used in this study does not allow us to determine
how much the nature of the information they are choosing to
Table 3
Incremental Fit Comparisons for Nested Model (Father)
Model
2
df p NNFI CFI
Change in
2
⌬
2
df p
Baseline (no gammas, no betas) 300.28 24
a
⬍.001 .21 .33
Bivariate (two gammas) 238.27 20
a
⬍.001 .17 .41 62.01 4 ⬍.001
No-beta (four gammas) 48.32 16 ⬍.001 .81 .89 189.96 4 ⬍.001
Hypothesized (four gammas, two betas) 43.48 13 ⬍.001 .80 .89 4.83 3 ns
Note. NNFI ⫽non-normed fit index; CFI ⫽comparative fit index.
a
Difference in terms of degrees of freedom from the mother model is attributable to the number of fixed parameters.
Disclosure
Closeness
Antisocial
Behavior
Mother
Students
Students
Mother
Control
Students
Mother
Knowledge
Students
Students
.33
(.16)
.22
(.37)
.59
(.52)
.16
(.16)
.12
(.23)
.61
(.56)
.15
(.18)
-.14
(ns)
.47
(.68)
-.12
(ns)
ns
(.36) -.18
(-.40)
-.27
(-.33)
.79*
(.72*)
.59*
(.53*)
.16
(.23)
.43
(.29)
.56
(.16)
Figure 2. Standardized path coefficients for the mother model, separately for student gender (girls).
2
(12) ⫽39.57,
p⬍.01, comparative fit index (CFI) ⫽.96, NFI ⫽.94, non-normed fit index (NNFI) ⫽.93, root-mean-square error
of approximation (RMSEA) ⫽.06 (.04 –.09);
2
(12) ⫽21.50, p⬍.05, CFI ⫽.99, NFI ⫽.97, NNFI ⫽.97,
RMSEA ⫽.05 (.01–.08).
ⴱ
We assigned the fixed value .15 times the variance of yto the corresponding parameter
theta-epsilon (Jo¨reskog & So¨rbom, 1996a, p. 196).
1516 VIENO, NATION, PASTORE, AND SANTINELLO
disclose may affect adolescents’ willingness to disclose. Recently,
Smetana et al. (2006) reported that adolescents felt more obligated
to disclose information concerning school work than information
related to personal and moral issues. Future research would be
helpful to examine whether the patterns we observed hold true
across different content or types of behaviors.
The results of the study indicate that the model did not ade-
quately fit for fathers. Although the research related to gender
differences in parenting styles and practices is sparse, this result
suggests that mothers and fathers may be distinct in the ways in
which they seek and receive information. Because disclosure and
parental knowledge are central constructs in this study, results
confirm (Smetana et al., 2006) that adolescents are generally more
willing to disclose to mothers than to fathers, and for this reason
mothers may have more influence on the behavior of their child.
Also, the lack of fit of the model may be an indicator of gender
differences in parental roles in Italian families. That is, Italian
fathers are much more focused on taking care of the economic
needs of the family and are less involved in the management of
children’s behavior.
Many of the differences found in the analysis based on the
child’s gender are consistent with previous research on gender
differences. It seems that girls’ closeness to their mothers in-
creased their mothers’ knowledge. Also, closeness was not related
to mothers’ knowledge of boys’ behavior. However, boys overall
reported closer relationships with their mothers than did girls. This
finding is in contrast to previous research (e.g., Claes et al., 2003)
and suggests several possible interpretations. First, it may indicate
that mothers may cater to boys more because of the greater value
placed on the wellbeing of boys. However, no existing studies
suggest that the privileging of boys is more pronounced in Italy
than in other western cultures that do not share these results.
Second, it may reflect the greater effort mothers may exert to know
what their male children are doing, because boys are less likely to
disclose and parents are likely to have less knowledge of their
behavior. If parents expect closeness to increase their knowledge
(as our data indicate for girls), they may escalate their attempts to
connect with boys to achieve a similar level of knowledge. Finally,
it may reflect mothers’ attempts to counteract the greater likeli-
hood that their male children may become involved in antisocial
behavior. Consistent with previous studies, our data indicated that
overall, boys were more likely to participate in antisocial behavior.
Consequently, Italian parents may put more effort into trying to
mitigate the greater risk by favoring their male children. More
research examining the cultural and racial dynamics of mother–
adolescent closeness is needed to substantiate any of these hypoth-
eses.
In contrast to the findings for mothers, there were no gender
differences observed in the children’s relationship to their fathers.
This result is consistent with the general idea that boys are less
likely to self-disclose to parents than girls, and thus parents are less
informed about their male child’s activities (Crouter & Head,
2002). It seems that this gender effect is independent of the level
of closeness to the parents. This may also be explained in terms of
the tendency for boys to be more secretive about their behavior
(Smetana et al., 2006). Moreover, in considering this result we
must take into account that some studies suggest that during early
adolescence, girls tend to be more emotionally involved in their
families than boys (see Butler & Nolen-Hoeksema, 1994).
Overall, these results do contribute to the growing evidence that
adolescents play an important and active role in regulating the
amount of information that parents have about their behavior and
in allowing their parents to monitor their behavior. At the same
time, this also indicates that a parent’s ability to cultivate an open
and positive relationship may be as important as any specific
parenting practice in determining the amount of information they
have about their adolescents’ behavior and in influencing their
child’s behavior. For this reason, parenting interventions that pro-
mote skills designed to improve the overall relationships and
family dynamic may be particularly powerful in decreasing ado-
lescent behavior problems (Kumpfer & Alvarado, 2003).
Limitations and Conclusions
The present data have several notable limitations. The principal
limitation is the way in which parents were selected to participate.
By allowing parents to decide which parent participated in the
study, we ended up with a significantly smaller sample of fathers
than mothers. The selection process raises the possibility that
fathers who did participate may differ from fathers who did not
participate on important attributes such as their level of involve-
ment or their closeness to their child. Furthermore, the smaller
number of fathers reduced the statistical power of the parent-
specific analysis, thus making it more difficult to detect some of
the more subtle relations that may exist in the father-specific
model. Despite the smaller numbers, the sample of fathers was
large enough to test our model based on the ratio between the
sample size and the number of parameters (Bentler, 1995).
The threats to external and statistical conclusion validity suggest
the need for caution in our conclusions about fathers’ effects on
children’s antisocial behavior. Although our model for fathers did
not produce an adequate fit, our results do suggest that fathers’
closeness and control can play a role in predicting adolescent
self-disclosure and antisocial behavior. It is important for future
studies to include a more systematic sample of fathers (i.e., by
random selection of the participating parent or by having both
parents provide information) to determine the magnitude and re-
liability of this effect. A second limitation is the low correlation
between the indicators of self-disclosure and parental knowledge,
together with their implementation, which only partially overlaps
with the measure used by Stattin and Kerr (2000). Therefore, it is
possible that some of the observed differences may be due to
instrumentation. Further, the parental control variable emphasizes
parental contributions to the actual decision-making as opposed to
parental monitoring and discipline (e.g., Spoth et al., 2006). Al-
though we believe that this implementation taps into core aspects
of the construct, it is important to qualify our inferences related to
parental control because of the differences in measurement strat-
egies.
Also, the present cross-sectional design does not allow us to
determine the stability of the effects or to study how the relations
between these variables changes over time. Longitudinal (includ-
ing experimental and other panel) studies of parenting, adolescent
self-disclosure, and parental knowledge are needed to determine
the causal relations with antisocial behavior. In any case, the poor
fit obtained from the test of the alternative model indicates that, at
least in early adolescence, parenting variables are more likely to
1517
PARENTING, DISCLOSURE, AND ANTISOCIAL BEHAVIOR
predict self-disclosure than whether or not the child is engaging in
antisocial behavior.
Another issue of note is the low correlation between parent and
youth reports. This is an effect that has been observed in previous
studies (see Tein, Roosa, & Michaels, 1994). We combined these
responses into a latent construct. By doing this we reduce the
artifact (source error effects) that increases the probability of
finding significant relations among variables when using one in-
formant (see Bartels et al., 2004; Cook & Goldstein, 1993). A final
issue is that the sample was drawn from a region in northeastern
Italy and may not be generalizable to early adolescents in other
parts of Italy, where parenting styles and other cultural factors may
be very different (see Claes et al., 2003).
Despite these limitations, the present study does provide impor-
tant insight into the complementary role that parents and adoles-
cents play in promoting positive adolescent outcomes. It suggests
that parents can still play an important role in terms of influencing
their children’s behavior during early adolescence by cultivating a
close relationship and exercising some control. However, early
adolescence marks a special developmental stage for both adoles-
cents and parents. For adolescents, it is the period in which they
begin to exercise some independence and make their own choices
regarding their behavior and how much information they share
with their parents. For parents, it marks a period in which they
begin to develop a more adult relationship with their child. This
study suggests that for mothers in particular, maintaining some of
the boundaries and limits that are consistent with earlier develop-
mental stages can reduce adolescent antisocial behavior. However,
it is important that they recognize the limits of their control and
promote open and close relationships in which their adolescents
are comfortable sharing information about their activities.
References
Bartels, M., Boomsma, D. I., Hudziak, J. J., Rietveld, M. J. H., van
Beijsterveldt, T. C. E. M., & van den Oord, E. J. C. G. (2004).
Disentangling genetic, environmental, and rater effects on internalizing
and externalizing problem behavior in 10-year-old twins. Twin Re-
search, 7, 162–175.
Bentler, P. M. (1995). EQS structural equations program manual. Encino,
CA: Multivariate Software.
Bentler, P. M., & Bonnet, D. G. (1980). Significance tests and goodness of
fit in the analysis of covariance structures. Psychological Bulletin, 88,
588 – 606.
Bollen, K. A. (1989). A new incremental fit index for general structural
equation models. Sociological Methods and Research, 17, 303–316.
Brown, B. B., Mounts, N., Lamborn, S. D., & Steinberg, L. (1993).
Parenting practices and peer group affiliation in adolescence. Child
Development, 64, 467– 482.
Buchanan, C. M., Maccoby, E. E., & Dornbusch, S. M. (1991). Caught
between parents: Adolescents’ experience in divorced homes. Child
Development, 62, 1008 –1029.
Butler, L. D., & Nolen-Hoeksema, S. (1994). Gender differences in re-
sponses to depressed mood in a college sample. Sex Roles, 30, 331–346.
Byrne, B. M. (1989). Multigroup comparison and the assumption of
equivalent construct validity across groups: Methodological and sub-
stantive issues. Multivariate Behavioral Research, 24, 503–523.
Claes, M., Lacourse, E., Bouchard, C., & Perucchini, P. (2003). Parental
practices in late adolescence, a comparison of three countries: Canada,
France and Italy. Journal of Adolescence, 26, 387–399.
Claes, M., Lacourse, E., Ercolani, A. P., Pierro, A., Leone, L., & Presaghi,
F. (2005). Parenting, peer orientation, drug use, and antisocial behavior
in late adolescence: A cross-national study. Journal of Youth and Ado-
lescence, 34, 401– 411.
Consedine, N. S., Sabag-Cohen, S., & Krivoshekova, Y. S. (2007). Ethnic,
gender, and socioeconomic differences in young adults’ self-disclosure:
Who discloses what and to whom? Cultural Diversity & Ethnic Minority
Psychology, 13, 254 –263.
Cook, W. L., & Goldstein, M. J. (1993). Multiple perspectives on family
relationships: A latent variables model. Child Development, 64, 1377–
1388.
Crouter, A. C., & Head, M. R. (2002). Parental monitoring and knowledge
of children. In M. H. Bornstein (Ed.), Handbook of parenting: Vol. 3.
Being and becoming a parent (pp. 461– 483). Mahwah, NJ: Erlbaum.
Darling, N., Cumsille, P., Caldwell, L. L., & Dowdy, B. (2006). Predictors
of adolescents’ disclosure to parents and perceived parental knowledge:
Between- and within-person differences. Journal of Youth and Adoles-
cence, 35, 667– 678.
Darling, N., & Steinberg, L. (1993). Parenting style as context: An inte-
grative model. Psychological Bulletin, 113, 487– 496.
Demuth, S., & Brown, S. L. (2004). Family structure, family processes, and
adolescent delinquency: The significance of parental absence versus
parental gender. Journal of Research in Crime and Delinquency, 41,
58 – 81.
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,
61–75.
Fletcher, A. C., Steinberg, L., & Williams-Wheeler, M. (2004). Parental
influences on adolescent problem behavior: Revisiting Stattin and Kerr.
Child Development, 75, 781–796.
Ge, X., Conger, R. D., Lorenz, F. O., & Simons, R. L. (1994). Parents’
stressful life events and adolescents’ depressed mood. Journal of Health
and Social Behavior, 35, 28 – 44.
Goodnow, J. J. (1985). Change and variation in ideas about childhood and
parenting. In I. E. Sigel (Ed.), Parental belief systems: The psychological
consequences for children (pp. 235–270).Hillsdale, NJ: Erlbaum.
Gorman-Smith, D., & Loeber, R. (2005). Are developmental pathways in
disruptive behaviors the same for girls and boys? Journal of Clinical
Child and Adolescent Psychiatry, 14, 15–27.
Gray, M. R., & Steinberg, L. (1999). Unpacking authoritative parenting:
Reassessing a multidimensional construct. Journal of Marriage and the
Family, 61, 574 –587.
Hoffman, J. P., & Johnson, R. A. (1998). A national portrait of family
structure and adolescent drug use. Journal of Marriage and the Family,
41, 392– 407.
Hu, L., & Bentler, P. M. (1999). Cut-off criteria for fit indexes in covari-
ance structure analysis: Conventional criteria versus new alternatives.
Structural Equation Modelling, 6, 1–55.
Jo¨reskog, K. G., & So¨ rbom, D. (1996a). LISREL 8: User’s reference guide.
Chicago, IL: Scientific Software International.
Jo¨reskog, K., & So¨ rbom, D. (1996b). PRELIS 2: User’s reference guide.
Lincolnwood, IL: Scientific Software International.
Jo¨reskog, K., So¨ rbom, D., du Toit, S., & du Toit, M. (2000). LISREL 8:
New statistical features. Chicago, IL: Scientific Software.
Kerns, K. A., Aspelmeier, J. E., Gentzler, A. L., & Grabill, C. M. (2001).
Parent– child attachment and monitoring in middle childhood. Journal of
Family Psychology, 15, 69 – 81.
Kerr, M., & Stattin, H. (2000). What parents know, how they know it, and
several forms of adolescent adjustment: Further support for a reinter-
pretation of monitoring. Developmental Psychology, 36, 366 –380.
Kiesner, J. (2002). Depressive symptoms in early adolescence: Their
relations with classroom problem behavior and peer status. Journal of
Research on Adolescence, 12, 463– 478.
Kumpfer, K. L., & Alvarado, R. (2003). Family-strengthening approaches
1518 VIENO, NATION, PASTORE, AND SANTINELLO
for the prevention of youth problem behaviors. American Psychologist,
58, 457– 465.
Laird, R. D., Pettit, G. S., Bates, J. E., & Dodge, K. A. (2003). Parents’
monitoring-relevant knowledge and adolescents’ delinquent behavior:
Evidence of correlated developmental changes and reciprocal influences.
Child Development, 74, 752–768.
Lovas, G. S. (2005). Gender and patterns of emotional availability in
mother–toddler and father–toddler dyads. Infant Mental Health Journal,
26, 327–353.
Meeus, W., & Dekovic, M. (1995). Identity development, parental and peer
support in adolescence: Results of a national Dutch survey. Adolescence,
30, 931–944.
Moffitt, T. E., Caspi, A., Dickson, N., Silva, P., & Stanton, W. (1996).
Childhood-onset versus adolescent-onset antisocial conduct problems in
males: Natural history from ages 3 to 18 years. Development and
Psychopathology, 8, 399 – 424.
Oregon Social Learning Center (1997). Problem checklist (PLST). Unpub-
lished instrument. Eugene, OR: Author.
Patterson, G. R. (1982). A social learning approach: Vol. 3. Coercive
family process. Eugene, OR: Castalia.
Patterson, G. R., Reid, J. B., & Dishion, T. (1992). A social interactional
approach: Vol. 4. Antisocial boys. Eugene, OR: Castalia.
Patterson, G. R., & Stouthamer-Loeber, M. (1984). The correlation of
family management practices and delinquency. Child Development, 55,
1299 –1307.
Pettit, G. S., Bates, J. E., Dodge, K. A., & Meece, D. W. (1999). The
impact of after-school peer contact on early adolescent externalizing
problems is moderated by parental monitoring, perceived neighborhood
safety, and prior adjustment. Child Development, 70, 768 –778.
Ramella, E., & Sindoni, G. (1997). La rappresentazione delle relazioni
familiari nell’infanzia e nella preadolescenza [Representation of family
relationships during infancy and preadolescence]. Neuropsicologia In-
fantile Psicopedagogia Riabilitazione, 23, 89 –97.
Rickel, A. U., & Biasatti, L. L. (1982). Modification of the Block Child
Rearing Practices Report. Journal of Clinical Psychology, 38, 129 –134.
Saris, W. E. (1982). Different questions, different variables. In C. Fornell
(Ed.), A second generation of multivariate analysis: Vol. 2. Measure-
ment and evaluation (pp. 78–96).New York, New York: Praeger.
Scaramella, L. V., Conger, R. D., & Simons, R. L. (1999). Parental
protective influences and gender-specific increases in adolescent inter-
nalizing and externalizing problems. Journal of Research on Adoles-
cence, 9, 111–141.
Simons, L. G., & Conger, R. D. (2007). Linking mother–father differences
in parenting to a typology of family parenting styles and adolescent
outcomes. Journal of Family Issues, 28, 212–241.
Simons, M. B., Chen, R., Abroms, L., & Haynie, D. L. (2004). Latent
growth curve analyses of peer and parent influences on smoking pro-
gression among early adolescents. Health Psychology, 23, 612– 621.
Simons, R. L., Chao, W., Conger, R. D., & Elder, G. H. (2001). Quality of
parenting as mediator of the effect of childhood defiance on adolescent
friendship choices and delinquency: A growth curve analysis. Journal of
Marriage and the Family, 63, 63–79.
Small, S. A., Silverberg, S. B., & Kerns, D. (1993). Adolescents’ percep-
tions of the costs and benefits of engaging in health-compromising
behaviors. Journal of Youth and Adolescence, 22, 73– 87.
Smetana, J. G., Crean, H. F., & Daddis, C. (2002). Family processes and
problem behaviors in middle-class African American adolescents. Jour-
nal of Research on Adolescence, 12, 275–304.
Smetana, J. G., Metzger, A., Gettman, D. C., & Campione-Barr, N. (2006).
Disclosure and secrecy in adolescent–parent relationships. Child Devel-
opment, 77, 201–217.
Snyder, J., & Huntley, D. (1990). Troubled families and troubled youth:
The development of antisocial behavior and depression in children. In
P. E. Leone (Ed.), Understanding troubled and troubling youth (pp.
194 –225). Thousand Oaks, CA: Sage.
Soenens, B., Vansteenkiste, M., Luyckx, K., & Goossens, L. (2006).
Parenting and adolescent problem behavior: An integrated model with
adolescent self-disclosure and perceived parental knowledge as inter-
vening variables. Developmental Psychology, 42, 305–318.
Spoth, R., Neppl, T., Goldberg-Lillehoj, C., Jung, T., & Ramisetty-Mikler,
S. (2006). Gender related quality of parent– child interactions and early
adolescent problem behaviors. Journal of Family Issues, 27, 826 – 849.
Stattin, H., & Kerr, M. (2000). Parental monitoring: A reinterpretation.
Child Development, 71, 1072–1085.
Steinberg, L. (1987). Single parents, stepparents, and the susceptibility of
adolescents to antisocial peer pressure. Child Development, 58, 269 –
275.
Tein, J.-Y., Roosa, M. W., & Michaels, M. (1994). Agreement between
parent and child reports on parental behaviors. Journal of Marriage and
the Family, 56, 341–355.
Vieno, A. (2006). Cosa pensano di sapere e cosa realmente sanno i genitori
e i comportamenti antisociali dei figli durante la pre-adolescenza [What
parents think they know, what they really know, and the antisocial
behavior of early adolescents]. Italian Journal of Developmental Psy-
chology, 84, 24 –36.
Weinberger, D. A., Tublin, S. K., Ford, M. E., & Feldman, S. S. (1990).
Preadolescents’ social– emotional adjustment and selective attrition in
family research. Child Development, 61, 1374 –1386.
Received May 25, 2007
Revision received June 4, 2009
Accepted June 22, 2009 䡲
1519
PARENTING, DISCLOSURE, AND ANTISOCIAL BEHAVIOR
A preview of this full-text is provided by American Psychological Association.
Content available from Developmental Psychology
This content is subject to copyright. Terms and conditions apply.
