Child Development, July/August 2000, Volume 71, Number 4, Pages 1072–1085
Parental Monitoring: A Reinterpretation
Håkan Stattin and Margaret Kerr
Monitoring (tracking and surveillance) of children’s behavior is considered an essential parenting skill. Numerous
studies show that well-monitored youths are less involved in delinquency and other normbreaking behaviors,
and scholars conclude that parents should track their children more carefully. This study questions
that conclusion. We point out that monitoring measures typically assess parents’ knowledge but not its source,
and parents could get knowledge from their children’s free disclosure of information as well as their own active
surveillance efforts. In our study of 703 14-year-olds in central Sweden and their parents, parental knowledge
came mainly from child disclosure, and child disclosure was the source of knowledge that was most
closely linked to broad and narrow measures of delinquency (normbreaking and police contact). These results
held for both children’s and parents’ reports, for both sexes, and were independent of whether the children
were exhibiting problem behavior or not. We conclude that tracking and surveillance is not the best prescription
for parental behavior and that a new prescription must rest on an understanding of the factors that determine
child disclosure.
INTRODUCTION
Parents often comment on the temptations that teenagers
today face, and they want to know what they can
do to minimize the chances that their child will fall
prey to bad influences. What advice can developmental
psychology offer them? One clear answer seems to
be that they can keep a close eye on what their children
are doing, where they are going, and whom they are
with—they can monitor their children’s behavior.
Parental monitoring is conceptualized as “a set of
correlated parenting behaviors involving attention to
and tracking of the child’s whereabouts, activities,
and adaptations” (Dishion & McMahon, 1998, p. 61).
This definition of monitoring as parental action is similar
to the dictionary definition, in which the verb “to
monitor” means “to keep watch over or check as a
means of control” (Read et al., 1995, p. 822). Parental
monitoring figures prominently in theoretical models
of the development of antisocial behavior. Theoretical
models make parenting practices, including inadequate
monitoring, recurring links in a causal chain
that starts with disruptive behavior, leads to hanging
out with deviant peers, and results in antisocial behavior
(Reid & Patterson, 1989; Snyder & Patterson,
1987). Indeed, cross-sectional and longitudinal
studies show that poorly monitored adolescents tend
to be antisocial, delinquent, or criminal (for a review
of early work, see Patterson & Stouthamer-Loeber,
1984; for empirical examples, see Cernkovich & Giordano,
1987; Crouter, MacDermid, McHale, & PerryJenkins,
1990; McCord, 1986, Sampson & Laub, 1994;
Weintraub & Gold, 1991). Poorly monitored youths
also tend to use illegal substances (Flannery, Vazsonyi,
Torquati, & Fridrich, 1994) and associate with
peers who approve of drug use (Chassin, Pillow, Curran,
Molina, & Barrera, 1993). They are more likely to begin
using tobacco (Biglan, Duncan, Ary, & Smolkowski,
1995) and to increase their drug use over time (Fletcher,
Darling, & Steinberg, 1995), and the poor monitoring
displayed by alcoholic parents seems to explain their
children’s use of illegal substances (Chassin et al.,
1993). Poorly monitored youths do worse in school
(Crouter et al., 1990; White & Kaufman, 1997) and engage
in more risky sexual activity (Metzler, Noell,
Biglan, Ary, & Smolkowski, 1994; Romer et al., 1994).
Evidence also suggests that poorly monitored youths
have deviant friends (Dishion, Capaldi, Spracklen, &
Li, 1995) and that they may become delinquent because
of peer pressure (Fridrich & Flannery, 1995). Furthermore,
longitudinal evidence suggests that children’s
delinquency and drug use are related to poor parental
monitoring later on, and that poor monitoring is related
to later delinquency (Aseltine, 1995; Barber, 1996).
The obvious prescription for parental behavior is
better monitoring—active surveillance or tracking of
children’s behavior. Snyder and Patterson (1987) suggest
that parents must give children a set of rules
about where they may go, with whom they may associate,
and when they must be home and then “‘check
up’ or track compliance with those rules, and take effective
disciplinary action when the rules are violated”
(p. 226). Joint monitoring has been suggested
as a final strategy for reducing risky sexual behavior:
“. . . encourage parents of children’s friends to join together
to monitor the behavior of their children”
(Romer et al., 1994, p. 985). Monitoring has also been
© 2000 by the Society for Research in Child Development, Inc.
All rights reserved. 0009-3920/2000/7104-0023
Stattin and Kerr 1073
suggested for preventing drug use: “The overall lesson
of this study appears to be that parental monitoring is
an appropriate strategy for parents attempting to deter
adolescents from engaging in substance use. Strong
parental monitoring helps to deter adolescents from
using alcohol and drugs themselves and, as a consequence,
prevents nonusing adolescents from associating
with drug-using peers. Since it is these drugusing
peers who are likely to pressure teenagers to
initiate or elevate substance use, strongly monitored
adolescents are, in essence, doubly protected from
substance use involvement . . .” (Fletcher et al., 1995,
pp. 269–270). Clearly, the prescription is for parents
to use a firm hand and actively control their children’s
behavior and associations.
Recommending active control and surveillance
might be reasonable, on the basis of a cursory reading
of the findings just reviewed, but it might be wrong.
Why? Because parents who score high on monitoring
might not be exercising control or practicing surveillance
at all. The most often-used monitoring measures
ask about parents’ knowledge of their children’s activities,
but they seldom ask about active tracking and
checking (the definition of monitoring). For example,
items such as the following ask adolescents to rate
their parents’ knowledge: “How much do your parents
REALLY know . . . Who your friends are? Where
you go at night? How you spend your money? What
you do with your free time? Where you are most afternoons
after school?” (Fletcher et al., 1995, p. 262);
“Do your parents know where you are when you are
away from home? Do your parents know who you are
with when you are away from home?” (Weintraub &
Gold, 1991, p. 272); and “In my free time away from
home, my parents know who I’m with and where I am.”
(Cernkovich & Giordano, 1987, p. 303). These measures
do not ask how parents came to know these things.
Other measures assess knowledge by asking parents
and their children the same set of questions about the
child’s activities and then assessing agreement between
the two sets of answers (Crouter, Manke, & McHale,
1995; Crouter et al., 1990; Patterson & StouthamerLoeber,
1984). But again, there is no telling how the information
was gained. Even though the term monitoring
implies that the measures represent parents’ tracking
and surveillance efforts, they actually represent an end
product: parents’ knowledge. [Hereafter, we use quotation
marks to distinguish parental knowledge measures
(“monitoring”) from the construct (monitoring)].
In fact, parents could get knowledge of their children’s
activities in at least three conceivable ways.
First, the children could tell them spontaneously,
without any prompting (child disclosure). Second,
parents could ask their children and their children’s
friends for the information (parental solicitation).
Third, parents could impose rules and restrictions on
their children’s activities and associations, thereby
controlling the amount of freedom children have to
do things without telling them (parental control).
Which source of information is really behind the
parental knowledge measures that have been called
monitoring? Some preexisting findings suggest that it
might be child disclosure. When several aspects of
parenting are examined together, the findings often
suggest that parent–child communication is more
beneficial than surveillance and control. One study
tested the idea that attachment to parents lowers the
likelihood of delinquency (Cernkovich & Giordano,
1987). The researchers used several indirect measures
of attachment, some of which dealt with parent–child
communication and the closeness of the relationship.
The results suggested that delinquents had poor communication
with their parents. They were lower than
nondelinquents on caring and trust (intimacy in the
parent–child relationship), identity support (parents’
respect, acceptance, and support), and instrumental
communication (discussion of future plans). Many of
the measures that were related to delinquency, then,
dealt directly with parent–child communication. Furthermore,
although delinquents were lower on “control
and supervision,” the measure included knowledge
of the child’s activities rather than active control
or supervision efforts, so it might actually have been a
measure of child disclosure. This study, then, suggests
that communication is at least as important as
control. Others suggest that it is much more important.
In one study of parental involvement and school
performance, parent–child agreement and parent–
child discussion predicted higher grade-point averages
and achievement test scores, but measures of
surveillance did not (Otto & Atkinson, 1997). In fact,
parental monitoring of school work predicted lower,
not higher, grades and test scores. In other words,
communication was linked to good performance; surveillance
was linked to bad performance. Further, in an
intervention study that attempted to reduce adolescents’
substance use, parents were trained and encouraged
to exert more active efforts to control their children’s
associations with drug- and alcohol-using peers
and their access to alcohol (Cohen & Rice, 1995). The intervention
had no effect on adolescents’ substance use.
Nonetheless, parental knowledge of the child’s whereabouts,
good parent–child rapport, and a respectful
parent–child relationship were all associated with less
substance use. Again, parent–child communication
was beneficial; surveillance and control were not.
The present study addresses two major questions
about parental monitoring. The first concerns whether
1074 Child Development
parents’ knowledge of their children’s whereabouts
and activities (“monitoring”) actually comes from their
own active efforts, as the term monitoring implies.
We look at three potential sources of information—
child disclosure, parental solicitation, and parental
control—and ask which of these explains the largest
portion of the variance in “monitoring” (parents’
knowledge). The second question concerns the negative
association between monitoring and normbreaking
behavior that so many studies have reported.
These studies most often conclude that parents’ control
and surveillance efforts prevent adolescents from
getting into trouble. But is this a valid conclusion? We
answer this by looking at which of the three potential
sources of parents’ knowledge is most strongly related
to normbreaking behavior. For each of these
questions, we test for possible moderating effects of
gender and the child’s misbehavior to determine
whether our findings are generalizable or apply only
to children with certain characteristics.
METHOD
Participants
Participants were 14-year-old youths from seven
mid-Sweden communities, and their parents. The registered
crime rate for youngsters in these communities
is higher than the national average but somewhat
lower than for the metropolitan areas in Sweden. The
participants represent the whole range of socioeconomic
backgrounds in the communities. As is the case
for the whole country, 91% of the fathers and 80% of
the mothers had at least part time employment (6%
of the fathers and 7% of the mothers were unemployed).
Seventy-six percent of the youngsters lived
with both biological parents.
Students in all 32 8th-grade classes in these communities
were asked to join the study (N ϭ 763). They
took part in the study unless their parents returned a
form stating that they did not want their child to participate
(10 parents returned this form). Neither parents
nor children were paid for their participation. Of
763 students, 703 (92%) were present on the day of the
data collection and answered the questionnaires.
Another questionnaire was sent home to each
child’s biological parent or legal guardian in the
home where the child lived during the school week.
Parents were asked to return the completed questionnaire
by mail, and 76% did so. In 71.4% of cases,
mothers filled out the questionnaire alone, in 14.4% of
cases fathers filled it out alone, in 12.9% of cases,
mothers and fathers worked together, and in 1.3% of
cases, a guardian other than a parent filled out the
questionnaire. Recent studies suggest that mothers
and fathers can have different levels of knowledge
under certain conditions (e.g., Crouter, Helms-Erikson,
Updegraff, & McHale, 1999). This is an important
issue, but we cannot address it in this study because
we cannot compare mothers and fathers in the same
families. In our study, according to both parents’ and
children’s reports, parental knowledge did not depend
upon the sex of the parent who responded.
In one of these communities, 14-year-olds from a
different cohort (N ϭ 36) served as a pilot sample for
a subsequent study. They answered the questions that
were used in this study on two occasions, 2 months
apart, and their responses were used to calculate the
test–retest reliabilities that we report for these measures.
Even though the response rate among parents was
high, those who responded might have been a biased
sample. To examine a possible selection effect, we
compared the children whose parents returned the
questionnaire (n ϭ 539) with those whose parents did
not (n ϭ 164) on all child-reported parental “monitoring”
measures and measures of normbreaking and police
contact (see following description). The children
whose parents returned the questionnaire did not differ
from the other children on normbreaking, police
contact, parental solicitation, or child disclosure. They
did, however, report somewhat lower parental “monitoring,”
p Ͻ .05, and parental control, p Ͻ .01.
Measures
Parental “monitoring.” In keeping with the monitoring
literature, we have operationalized “monitoring”
as parents’ knowledge of the child’s whereabouts,
activities, and associations. Using 5-point
Likert scales, children answered nine questions about
their parents’ knowledge. The questions were, “Do
your parents: know what you do during your free
time? know who you have as friends during your
free time? usually know what type of homework you
have? know what you spend your money on? usually
know when you have an exam or paper due at school?
know how you do in different subjects at school? know
where you go when you are out with friends at night?
normally know where you go and what you do after
school?” and “In the last month, have your parents
ever had no idea of where you were at night?” Parents
answered the same questions, with only minor
changes in wording where necessary (e.g., “Do you:
know what your child does during his or her free
time? know who your child has as friends during his
or her free time? . . .”). Means were calculated for the
child-report items (␣ reliability ϭ .86) and parentreport
items (␣ reliability ϭ .89). The test–retest reli-
Stattin and Kerr 1075
ability for child-reported “monitoring” was substantial,
r(36) ϭ .83.
Child disclosure. Our child disclosure measure
comprised five items. The children’s questions were,
“Do you spontaneously tell your parents about your
friends (which friends you hang out with and how they
think and feel about various things)?” “How often do
you usually want to tell your parents about school
(how each subject is going; your relationships with
teachers)?” “Do you keep a lot of secrets from your
parents about what you do during your free time?”
“Do you hide a lot from your parents about what you
do during nights and weekends?” and “Do you like
to tell your parents about what you did and where
you went during the evening?” Parents answered the
same questions, with only minor changes in wording
where necessary. The ␣ reliabilities were .84 for parents’
reports and .81 for children’s reports. Childreported
disclosure was highly reliable, according to
the 2-month test–retest correlation, r(34) ϭ .87.
Parental solicitation. Five items were averaged to
form the parental solicitation measure. The children’s
items were, “How often do your parents talk with
your friends when they come over to your house?”
“How often do your parents ask you about what happened
during your free time?” “During the past
month, how often have your parents initiated a conversation
with you about your free time?” “When did
your parents last have extra time to sit down and listen
to you when you talk about what happened during
your free time?” and “How often do your parents
ask you to sit and tell them what happened at school
on a regular school day?” Parents answered the same
questions, with slight changes in wording where necessary.
The ␣ reliabilities were .77 and .75 for youthreported
and parent-reported solicitation, respectively.
Child-reported solicitation was highly reliable, according
to the 2-month test–retest correlation, r(35) ϭ .82.
Parental control. Our parental control measure
comprised six items. Youths answered: “Must you
have your parents’ permission before you go out during
the weeknights?” “If you go out on a Saturday
evening, must you inform your parents beforehand
about who will be along as well as where you will be
going?” “If you have been out past curfew, do your
parents require that you explain why and tell who
you were with?” “Do your parents demand that they
know where you are in the evenings, who you are
going to be with, and what you are going to do?”
“Must you ask your parents before you can make
plans with friends about what you will do on a Saturday
night?” and “Do your parents require that you
tell them how you spend your money?” Parents answered
the same questions, with minor changes in
wording. The ␣ reliabilities were .82 and .77 for
youths’ reports and parents’ reports, respectively. The
2-month test–retest reliability for child-reported parental
control was high, r(33) ϭ .86.
Normbreaking. Our normbreaking measure comprised
nine items. Youths answered these questions
about their behavior over the past year: “Have you
drunk beer, liquor, or wine to the point of feeling
drunk?” “Have you tried hashish, cannabis or marijuana?”
“Have you pilfered from school?” “Have you
purposely vandalized or taken part in vandalizing
something that did not belong to you such as a window
display, car, telephone booth, bank, or garden?” “Have
you taken items from a mall, store, or newsstand
without paying?” “Have you taken money from
home?” “Have you bullied someone or together with
others mobbed or bullied other students (e.g., ignored,
made fun of, or teased)?” “Have you been a
part of a physical fight?” “Have you been caught by
the police?” Also, parents reported whether they
thought their child had done any of these things. The
␣ reliabilities for normbreaking were .79 and .62 for
youths’ and parents’ reports, respectively. For some
of the analyses, the single item asking whether they
had been caught by the police over the past year was
used as a narrower measure of normbreaking.
Parent–child relationships. Children answered eight
questions about the quality of their relationships with
their mothers: “How often do you feel disappointed
with your mother?” “How well do you and your
mother understand each other?” “Do you wish that
your mother was different?” “Do you and your mother
quarrel and fight with each other?” “How often do
you feel proud of your mother?” “Do you accept your
mother the way she is?” “How often do you feel angry
or irritated by your mother?” and “Does your
mother support and encourage you?” They answered
the same questions about their fathers. The scales
were reflected, when necessary, so that higher scores
indicated more positive relationships. The variables
were aggregated by parent first and then a total
parent–child relationship variable was formed by
taking the mean of the mother–child and father–child
relationship variables. For children of single parents,
the variable for the one parent was used. The ␣ reliability
calculated on mother- and father-relations items,
combined, was .89. The test–retest correlation was
moderately high, r(33) ϭ .75.
Family closeness. Parents completed the Swedish
Family Climate Scale (Hanson, 1989). They were presented
with 85 adjectives, which they marked as
either appropriate or inappropriate for describing their
own family climate. We used the “closeness” scale,
which consists of 18 adjectives: happy, warm, stable,
1076 Child Development
easy, harmonious, gentle, loving, natural, safe, kind,
meaningful, secure, considerate, pleasant, friendly,
calm, praising, humble. The psychometric properties
of this measure have been reported previously (Hanson,
1989).
Skewness of some measures. Note that the distributions
of parent-reported “monitoring” and childreported
normbreaking were somewhat skewed,
skewness ϭ Ϫ2.31 and 2.02 for monitoring and
normbreaking, respectively. Our analyses involving
these measures have been done with the basic, untransformed
data. However, we have also transformed
these variables by taking the natural logarithms.
Analyses using the transformed variables
yielded results that were almost identical to the original
analyses. (Small differences appeared only on the
third decimal place.) This result was probably due to
the large sample size.
Procedure
The children filled out the questionnaires during
regular school hours, and they were assured of the
confidentiality of their answers. They were informed
that their parents would answer similar questions.
Research assistants administered the questionnaires
at the schools. Teachers were not present. Parents responded
by filling out and mailing in a questionnaire.
They were informed that their children had answered
similar questions at school.
RESULTS
Gender Differences
Mean level differences between boys and girls on
some of the variables appear in Table 1 along with
t-tests of the differences. As shown in the table, there
are no gender differences in parental “monitoring” as
reported by either children or parents. According to
both children’s and parents’ reports, however, girls
freely disclose more than boys to their parents. Parents
solicit more information from girls than boys, according
to children but not parents, and although
girls report being controlled more than boys, parents
report controlling boys more than girls. But the actual
differences in control are small. Boys report having
better relationships than girls do with their parents.
Finally, boys and girls are similar on normbreaking
and police contact, but gender differences did appear
on some of the individual normbreaking items. Girls
were higher than boys on some and boys were higher
than girls on others. This probably accounts for the
gender similarity on the scale as a whole. In particular,
girls were more likely than boys to report taking
money from home, t(673.06) ϭ 2.1, p ϭ .04, and they
were marginally more likely to report drinking alcohol,
t(686) ϭ 1.7, p ϭ .08. Boys, on the other hand,
were more likely to report vandalizing property,
t(520.47) ϭ Ϫ5.1, p Ͻ .001, and fighting, t(596.77) ϭ
Ϫ4.0, p Ͻ .001.
For this study, however, the pertinent question is
not whether parents monitor, control, or solicit information
from girls more than boys, or whether girls
disclose more than boys. Rather, the question is
whether gender moderates the relations among these
variables such that our reinterpretation of monitoring
would apply to one sex more than the other. Therefore,
for each of our major findings, we include tests
for gender interactions to determine whether the
findings apply to one sex more than the other.
What Does “Monitoring” Represent?
To discover what “monitoring” really represents,
we first examined the correlations linking childreported
and parent-reported “monitoring” (parental
knowledge) with the three hypothesized sources of
information: child disclosure, parental solicitation,
and parental control. These relations are reported in
Table 2 along with the intercorrelations among the
sources of information. According to children’s reports,
each of the three sources correlated significantly with
“monitoring,” r(692) ϭ .66, p Ͻ .001, r(688) ϭ .42, p Ͻ
.001, and r(684) ϭ .41, p Ͻ .001 for child disclosure, parental
solicitation, and parental control, respectively,
but the correlation between child disclosure and
Table 1 Mean Scores as a Function of Gender For All Variables
Used in the Study
Girls Boys t
Children’s reports
Parental monitoring .04 (.70) Ϫ.04 (.67) 1.62
Child disclosure .10 (.76) Ϫ.10 (.72) 3.60***
Parental solicitation .15 (.71) Ϫ.15 (.71) 5.45***
Parental control .08 (.74) Ϫ.08 (.70) 3.00**
Parent–child relationship Ϫ.09 (.68) .10 (.54) Ϫ3.92***
Normbreaking Ϫ.03 (.62) .03 (.67) Ϫ1.31
Police contact 1.09 (.35) 1.12 (.38) Ϫ.76
Parent’s reports
Parental monitoring .01 (.79) Ϫ.01 (.67) .25
Child disclosure .13 (.75) Ϫ.11 (.78) 3.74***
Parental solicitation .03 (.69) Ϫ.02 (.72) .73
Parental control Ϫ.04 (.67) .08 (.67) Ϫ2.11*
Normbreaking Ϫ.02 (.49) .01 (.54) Ϫ.73
Note: Standardized scores, except for police contact, which ranged
from 1– 4. Standard deviations are given in parentheses.
* p Ͻ .05; ** p Ͻ .01; *** p Ͻ .001.
Stattin and Kerr 1077
“monitoring” was significantly stronger than the next
largest correlation, z ϭ 6.4, p Ͻ .001. Using parentreported
measures, the results were almost identical.
Again, all of the correlations were statistically significant,
r(574) ϭ .63, p Ͻ .001, r(575) ϭ .43, p Ͻ .001, and
r(561) ϭ .28, p Ͻ .001 for child disclosure, parental solicitation,
and parental control, respectively, but the
correlation between child disclosure and “monitoring”
was significantly stronger than the next largest correlation,
z ϭ 4.8, p Ͻ .001. Note, further, that these differential
links to monitoring are not due to differences in
skewness. All measures of disclosure, solicitation, and
control had distributions that were close to normal,
skewness statistics ϭ Ϫ.23, .18, and .20 for children’s reports
and Ϫ.89, Ϫ.59, and Ϫ.59 for parent’s reports of
disclosure, solicitation, and control, respectively.
As shown in Table 2, in addition to the fact that
they are all related to “monitoring,” the information
sources are all significantly, positively intercorrelated.
Within parents’ and children’s reports, child disclosure
and parental solicitation were the most highly intercorrelated
and child disclosure and parental control were
the least. In addition, note that parents and adolescents
agreed substantially about these aspects of their
relationships. All correlations between parents’ and
adolescents’ reports were significant at or below the
.001 level, as shown on the diagonal in Table 2. Parents
and adolescents agreed most on child disclosure,
r(537) ϭ .45, less on “monitoring,” r(537) ϭ .38, and
parental solicitation, r(526) ϭ .33, and least on parental
control, r(508) ϭ .30. Parents and children agree,
then, that parents’ information comes from spontaneous
child disclosure and parents’ active efforts but
that more of it comes from child disclosure. However,
because these sources of information are all intercorrelated,
the question is what they each contribute
independently.
To determine how much of parents’ knowledge
comes from each of these sources, independent of the
others, we used a multiple regression approach. We
entered the three potential information sources simultaneously
in a regression model predicting “monitoring.”
Because we know from Table 1 that girls disclose
more information than boys do to their parents,
we also included sex in this regression model along
with Sex ϫ Disclosure, Sex ϫ Solicitation, and Sex ϫ
Control interaction terms. The results appear in the
upper portion of Table 3.
Which source of information is most predictive of
parents’ knowledge (“monitoring”)? The answer is
child disclosure. Using children’s reports of all the
variables, child disclosure and parental control both
independently predict the amount of information
parents have, but child disclosure is the more important
of the two. With these variables controlled, parental
solicitation is unrelated to “monitoring.” Using
parents’ reports of all the variables, all three independently
predict “monitoring”; however, as with the
children’s reports, child disclosure appears to be
more important than the variables that measure parents’
active efforts. Parents get most of their information
about their children’s activities from their children’s
willing disclosure, rather than from their active
surveillance and control efforts. Additionally, as shown
in the table, neither sex nor any of the interaction terms
approach significance for children’s or parents’ reports.
(We also performed these analyses by entering
sex before entering all other variables. It was nonsignificant
even when entered alone, as one would expect
from the fact that there was no significant sex difference
on “monitoring” in Table 1.) Despite the meanlevel
sex differences in child disclosure, solicitation,
and control, the relations between these variables and
“monitoring” are the same for boys and girls.
Table 2 Intercorrelations Among “Monitoring” (Parents’ Knowledge) and Three Hypothesized
Sources of Knowledge
Child
Disclosure
Parental
Solicitation
Parental
Control
Parental
Monitoring
Child disclosure [.45***] .55*** .36*** .66***
(537) (685) (681) (692)
Parental solicitation .46*** [.33***] .43*** .42***
(571) (526) (682) (688)
Parental control .12** .20*** [.30***] .41***
(588) (560) (508) (684)
Parental monitoring .63*** .43*** .28*** [.38***]
(574) (575) (561) (537)
Note: Degrees of freedom are in parentheses. Children’s reports are above and parent’s reports below
the diagonal. Correlations between parents’ and children’s reports are on the diagonal in brackets.
** p Ͻ .01; *** p Ͻ .001.
1078 Child Development
Once children have freely disclosed information
about their whereabouts and activities, how much additional
information will parents get from their own
active efforts? To answer this, we entered child disclosure
first in a regression model predicting “monitoring”
and then added parental solicitation and control
to the model. The results appear in the lower portion
of Table 3. By itself, child disclosure explains a substantial
proportion of the variance in “monitoring” (44%
from the children’s points of view and 38% from the parents’
points of view). Adding the other two sources to
the model produces a statistically significant, but small,
increase in the variance explained (3% for children’s reports
and 5% for parents’ reports). These results suggest
that children freely tell their parents a large proportion
of what their parents know about their whereabouts
and activities. Parents’ efforts to gain information reliably
produce more, but only a small amount more. In
summary, then, when “monitoring” is recognized as
knowledge and we ask where parents have gotten that
knowledge, we find that tracking or surveillance efforts,
which are implied by the term monitoring and its previous
interpretations, are less important than children’s
spontaneous sharing of information.
This finding could, conceivably, appear because
disclosure is highly important among children who
have nothing to hide from their parents, even though
it may not be among those who have a lot to hide. In
other words, the child’s behavior might interact with
these sources to predict parent’s knowledge (“monitoring”).
Further analyses, however, suggest not. We
tested for interactions between the information
sources and dichotomized measures of children’s
problem behavior: (1) hanging out on the streets in
the evening (seldom ϭ once a month or less, n ϭ 518;
often ϭ once a week or more, n ϭ 168) and (2) normbreaking
(low ϭ lower than .5 SD on the normbreaking
behavior scale, n ϭ 553; high ϭ .5 SD or higher on
the normbreaking scale, n ϭ 143). As shown in Table
4, each of these possible moderators are independently
linked to parents’ knowledge (“monitoring”).
However, there is little evidence that they moderate
the relations between child disclosure and “monitoring.”
Most of the interaction-term slopes are near zero
and nonsignificant, and child disclosure is the strongest
predictor in every model, which indicates that
whether the child is misbehaving or not, parents get
most of their knowledge from the child’s free disclosure.
There is one weakly significant interaction—the
Disclosure ϫ Hanging Out interaction for childreported
measures. We plotted the interaction by solving
the regression equation for high and low child
disclosure among those who were hanging out often
and seldom. This plot revealed that for children who
often hung out on the streets in the evening, low disclosure
of information was linked to particularly low
levels of parental knowledge (“monitoring”). But
high disclosure was still linked to high levels of parental
knowledge (“monitoring”). Thus, among children who
often hang out on the streets, child disclosure is an even
more important source of parental knowledge.
Monitoring and Normbreaking
“Monitoring” has been linked in the literature to
lower incidences of delinquency, smoking, drug use,
and other normbreaking behaviors, and it is in this
study, as well, r(693) ϭ Ϫ.50, p Ͻ .001 and r(534) ϭ
Ϫ.34, p Ͻ .001 for bivariate correlations linking childreported
normbreaking with child-reported and parentreported
“monitoring,” respectively. Additionally, each
of the child-reported sources of information correlated
significantly with normbreaking, r(688) ϭ Ϫ.39,
p Ͻ .001, r(685) ϭ Ϫ.13, p Ͻ .001, and r(680) ϭ Ϫ.24,
p Ͻ .001 for disclosure, solicitation, and control, respectively.
Of the parent-reported sources, only child
disclosure correlated significantly with normbreaking,
r(530) ϭ Ϫ.27, p Ͻ .001, r(531) ϭ Ϫ.01, ns, and
r(517) ϭ Ϫ.06, ns, for disclosure, solicitation, and control,
respectively.
The question, however, is which source of information
is most important in predicting normbreaking,
independent of the others. As shown in Table 5, the
answer, again, is child disclosure. Higher levels of
child disclosure correspond to lower levels of normTable
3 Simultaneous and Heirarchical Regression Analyses
Predicting “Monitoring” (Parental Knowledge) from Potential
Information Sources and Child Sex
Child Report Parent Report
␤ ⌬R2 ␤ ⌬R2
Simultaneous inclusion
Disclosure .64*** .59***
Solicitation .03 .15**
Control .22*** .19***
Sex .04 .05
Disclosure ϫ Sex Ϫ.07 Ϫ.07
Solicitation ϫ Sex Ϫ.01 .00
Control ϫ Sex Ϫ.02 .49*** Ϫ.02 .44***
Stepwise inclusion
Model 1
Disclosure .66*** .44*** .62*** .38***
Model 2
Solicitation &
Control added .03*** .05***
** p Ͻ .01; *** p Ͻ .001.
Stattin and Kerr 1079
breaking, independent of parental solicitation and
control. The standardized slopes for disclosure in
both child- and parent-report models are more than
twice as large as the next largest slopes. For control
and solicitation, on the other hand, the independent
connections to normbreaking are less clear and consistent.
Independent of disclosure and solicitation,
child-reported parental control is linked to lower
normbreaking but parent-reported control is not. The
findings concerning parental solicitation run counter
to expectations. With child disclosure and parental
control held constant, parental solicitation is linked to
higher, not lower, normbreaking. The more parents
ask their children about their activities, the more
normbreaking their children tend to do. Note, further,
that these results hold when parent-reported sources
are used to predict parent-reported normbreaking,
which is highly skewed because most parents report
no normbreaking, ␤ ϭ Ϫ.37, p Ͻ .001, ␤ ϭ .18, p Ͻ .01,
and ␤ ϭ .00, ns for disclosure, solicitation, and control,
respectively. Because of a skewed distribution of parental
reports of children’s normbreaking, we dichotomized
the measure of child normbreaking into two
groups: one in which parents reported no norm violations
or just minor ones (75.1%) and another in which
parents reported several norm violations (24.9%). A
logistic regression analysis revealed that child disclosure
predicted negatively, b ϭ Ϫ.81, Wald statistic ϭ
31.45, p Ͻ .001, parental solicitation predicted positively,
b ϭ .41, Wald statistic ϭ 5.85, p Ͻ .05, and parental
control was nonsignificant, b ϭ Ϫ.15, Wald statistic
ϭ 0.99, p ϭ ns.
The lower portion of Table 5 reveals, further, that
these findings apply to both boys and girls. Sex does
not interact with any of the sources of parental information
in predicting normbreaking.
Again, the child’s behavior might moderate these
relations. Perhaps child disclosure is not the most important
link to lower normbreaking among children
who are hanging out on the streets in the evening.
Perhaps, for those children, solicitation or control becomes
more important than child disclosure as predictors
of lower normbreaking. To test this, we included
the dichotomized variable for the child’s hanging out
on the streets in the evening and its interactions with
the sources of information. These results appear in
Table 6. As expected, hanging out on the streets in the
evening is strongly linked to normbreaking. According
to children’s reports of disclosure, solicitation,
and control, shown in the left-hand column of Table 6,
child disclosure is strongly linked to lower normbreaking,
and that link does not depend on whether
the child is hanging out on the streets in the evening.
In this model, there is also a weak Control ϫ Hanging
Out interaction. A plot of four points predicted by the
regression equation (high and low control; hanging
out often and seldom) revealed that low control was
mainly associated with higher normbreaking among
Table 4 Simultaneous Regression Analyses Predicting Monitoring
(Parental Knowledge) from Potential Information
Sources and Children’s Behavior
Child Report Parent Report
␤ R2 ␤ R2
Hanging out on the
streets in the eveninga
Disclosure .49*** .45***
Solicitation .07 .16***
Control .19*** .13***
Hanging out Ϫ.14*** Ϫ.15***
Disclosure ϫ Hanging Out .09* .01
Solicitation ϫ Hanging Out Ϫ.05 .06
Control ϫ Hanging Out Ϫ.00 .49*** .08 .45***
Child’s normbreakingb
Disclosure .50*** .43***
Solicitation .04 .19***
Control .15*** .15***
Normbreaking Ϫ.23*** Ϫ.17***
Disclosure ϫ Normbreaking .00 .04
Solicitation ϫ Normbreaking .02 .00
Control ϫ Normbreaking .04 .53*** .06 .45***
a Seldom ϭ once a month or less (n ϭ 518); often ϭ once a week or
more (n ϭ 168).
b Low ϭ lower than .5 SD on the normbreaking scale (n ϭ 553);
high ϭ .5 SD or higher on the normbreaking scale (n ϭ 143).
*p Ͻ .05; ** p Ͻ .01; *** p Ͻ .001.
Table 5 Simultaneous Regression Analyses Predicting Children’s
Self-Reported Normbreaking from Child’s Sex and
Sources of Parents’ Information
Child-Reported
Sources
Parent-Reported
Sources
␤ R2 ␤ R2
Predictors of normbreaking
Disclosure Ϫ.42*** Ϫ.38***
Solicitation .18*** .19***
Control Ϫ.17*** .19*** Ϫ.01 .12***
Test of moderating effects
of gender
Disclosure Ϫ.39*** Ϫ.43***
Solicitation .14* .20**
Control Ϫ.19*** Ϫ.00
Sex .01 Ϫ.02
Disclosure ϫ Sex Ϫ.05 .06
Solicitation ϫ Sex .06 .00
Control ϫ Sex .02 .19*** Ϫ.02 .13***
* p Ͻ .05; ** p Ͻ .01; *** p Ͻ .001.
1080 Child Development
children who often hung out on the streets in the
evening. For those who seldom did, parental control
was less linked to normbreaking. However, even
though parental control is associated with lower normbreaking
under some circumstances, the link between
this control interaction and normbreaking is only a
third as large as the main effect of child disclosure.
According to parents’ reports, shown in the righthand
column of Table 6, child disclosure is the source
of parental information that is most closely linked to
normbreaking, but the interaction shows that the association
is moderated by the child’s behavior. A plot
of the interaction revealed that child disclosure was
even more strongly connected to normbreaking among
children who often hung out on the streets in the
evening than among those who seldom did. For children
who hung out often, the less they disclosed to
their parents, the more likely they were to be breaking
norms, and their normbreaking was more extreme
than that of those children who seldom hung out on
the streets. In addition, according to parents’ reports
in Table 6, solicitation interacts with hanging out to
predict normbreaking. A plot of this interaction revealed
that higher levels of solicitation were linked to
higher levels of normbreaking, but this was much
more true when children often hung out on the streets
in the evening. Among parents of those children, solicitation
is probably a reaction to normbreaking
rather than an antecedent (i.e., parents’ attempts to
get information about what their children are doing
when they are often out in the evenings and getting
into trouble).
These findings suggest that the link between
“monitoring” and antisocial behavior exists not because
surveillance reduces antisocial behavior as has
often been claimed but rather because child disclosure
is heavily represented in “monitoring,” and children
who talk openly with their parents tend to commit
fewer antisocial acts. However, because studies
that find links between “monitoring” and antisocial
behavior often conclude that surveillance is the important
parental activity, we asked, finally, whether
parents’ active efforts added anything substantial to
the prediction of normbreaking, over and above child
disclosure. To answer this, we regressed normbreaking
on child disclosure first and then added parental solicitation
and control to the model. The results suggested
that parents’ active efforts have a reliable but
small effect. Child disclosure, by itself, significantly
predicted less normbreaking, R2 ϭ .15, p Ͻ .01. Adding
parental solicitation and control to the model produced
a statistically significant but small increase in
the variance explained, R2 change ϭ .03, p Ͻ .01.
Note that the three hypothesized sources of information
together were approximately equal to “monitoring”
as predictors of normbreaking. For parents’
reports, “monitoring” accounted for 12% of the variance
in normbreaking; the three sources accounted
for 10%. The analogous figures for children’s reports
were 25% and 18%. The correlations within these
pairs were not significantly different from each other,
z ϭ 0.88 and 1.72, respectively.
Police contact. Until now, we have used a broad
measure of normbreaking behavior that included
norm violations at home, at school, and during leisure
time. We look, now, at a narrower measure: having
been in trouble with the police during the last
year. According to the self-reports of the 14-year-old
children, 8.8% had been caught by the police for some
offense during the past year, and according to the parents
and the children themselves, parents of these
children knew less about their daily activities than the
parents of the other children. Child-reported “monitoring”
was lower among the children who said they
had been caught by the police, t(692) ϭ 6.13, p Ͻ .001.
Mean z-scores on “monitoring” were Ϫ.50 (SD ϭ .71)
and .05 (SD ϭ .67) for those who had been caught by
the police and those who had not, respectively. Parent-reported
“monitoring” was also lower for those
who had had police contact, t(533) ϭ 2.83, p Ͻ .01.
Mean z-scores were Ϫ.27 (SD ϭ .48) and .05 (SD ϭ .73)
for those who had been caught by the police and
those who had not, respectively.
“Monitoring” (parents’ knowledge) is related to
police contact, but which source of parents’ knowledge
is most closely related? The answer is child disclosure,
just as it was for the broader normbreaking
measure. In a logistic regression analysis predicting
police contact from the child’s reported disclosure,
Table 6 Regression Analyses Predicting Child-Reported Normbreaking
from Sources of Parental Information and the Child’s
Behavior
Child-Reported
Sources
Parent-Reported
Sources
␤
Model
R2 ␤
Model
R2
Disclosure Ϫ.31*** Ϫ.21***
Solicitation .11* .08
Control Ϫ.07 Ϫ.03
Hanging outa .30*** .35***
Disclosure ϫ Hanging Out Ϫ.03 Ϫ.16**
Solicitation ϫ Hanging Out .06 .15**
Control ϫ Hanging Out Ϫ.11* .29*** .02 .26***
a Seldom ϭ once a month or less (n ϭ 518); often ϭ once a week or
more (n ϭ 168).
*p Ͻ .05; ** p Ͻ .01; *** p Ͻ .001.
Stattin and Kerr 1081
parental solicitation, and parental control, child disclosure
was most important, b ϭ Ϫ.62, Wald statistic
ϭ 8.25, p Ͻ .01. Parental control was also a significant
predictor, b ϭ Ϫ.50, Wald statistic ϭ 4.82, p Ͻ .05,
but parental solicitation was not, b ϭ .18, Wald statistic
ϭ 0.57, ns.
According to parents’ reports, 5.7 % of the children
had been caught by the police during the past year,
and again, the parents of these children knew less
about their activities, according to both sources.
Child-reported “monitoring” was lower among these
children, t(27.42) ϭ 3.23, p Ͻ .01. Mean z-scores on
“monitoring” were Ϫ.51 (SD ϭ .91) and .07 (SD ϭ .65)
for those who had been caught by the police and those
who had not, respectively. Parent-reported “monitoring”
was also lower, t(573) ϭ 2.40, p Ͻ .05. Mean
z-scores were Ϫ.30 (SD ϭ .81) and .02 (SD ϭ .72) for
those who had been caught by the police and those
who had not, respectively.
Furthermore, when the three sources of information
were used to predict parent-reported police contact,
child disclosure was the only significant predictor,
b ϭ Ϫ.80, Wald statistic ϭ 11.77, p Ͻ .01. In short,
independent of whether the outcome measure is
broad (normbreaking) or narrow (police contact) or
whether the children or the parents report on police
contact, child disclosure emerges as the most important
predictor.
Is child disclosure just a proxy for good parent–child relationships?
One could argue that the link between
child disclosure and normbreaking is primarily due
to the fact that children who confide in their parents
have close emotional bonds with their parents, and it
is the emotional bonds that makes them unlikely to be
involved in normbreaking. We used two different indicators
of parent–child relationships: one from the
children’s perspectives (parent–child relationships)
and one from the parents’ perspectives (family closeness).
The family closeness measure was taken on a
subset of the sample (n ϭ 154). These two measures
were positively correlated, r(154) ϭ .22, p Ͻ .01. In addition,
children’s reports of mother–child relationships
were positively correlated with their reports of
father-child relationships, r(662) ϭ .47, p Ͻ .001, and
the pattern of results was the same when mother– and
father–child relationships were examined separately.
Beginning with parent–child relationships, we
asked whether any of these variables would still appear
as predictors of normbreaking after controlling for
children’s reports of the parent–child relationship.
With normbreaking as the dependent variable in a regression
model, we forced parent–child relationships
in on the first step and then added child disclosure,
parental solicitation, and parental control on the second
step. Parent–child relationships was a significant
predictor of normbreaking when entered alone, ␤ ϭ
Ϫ.25, p Ͻ .001. However, each of the child-reported
sources still significantly predicted normbreaking independent
of our parent–child relationship measure,
␤ ϭ Ϫ.35, p Ͻ .001, ␤ ϭ Ϫ.09, p Ͻ .05, and ␤ ϭ Ϫ.25,
p Ͻ .001, for child disclosure, parental solicitation,
and parental control, respectively. Moreover, when all
of the variables were entered simultaneously to test
for their independent contributions, child disclosure
explained around twice as much variance as solicitation,
control, or parent–child relationships, ␤ ϭ Ϫ.36,
p Ͻ .001, ␤ ϭ .17, p Ͻ .001, ␤ ϭ Ϫ.19, p Ͻ .001, and ␤ ϭ
Ϫ.13, p Ͻ .001, for disclosure, solicitation, control, and
parent–child relationships, respectively. We also tested
for gender effects by including, simultaneously,
parent–child relationships, the three information
sources, sex, the interaction of parent–child relationships
with sex, and interactions of the information
sources with sex. Neither sex nor any of the interactions
was significant in this model.
Similar results were found using the second
measure—parents’ judgments of family closeness,
which were taken for a subset of the sample (n ϭ 154).
Family closeness was a significant predictor of normbreaking
when entered alone, ␤ ϭ Ϫ.23, p Ͻ .01; however,
child disclosure and parental control significantly
predicted normbreaking, over and above the
family closeness measure, ␤ ϭ Ϫ.45, p Ͻ .001, ␤ ϭ
Ϫ.34, p Ͻ .001, for child disclosure and parental control,
respectively. In this limited sample, parental solicitation
was marginally significant, ␤ ϭ Ϫ.15, p Ͻ
.08. Again, when all of the variables were entered simultaneously,
child disclosure explained more than
twice as much variance as solicitation, control, or
family closeness, ␤ ϭ Ϫ.47, p Ͻ .001, ␤ ϭ .21, p Ͻ .02,
␤ ϭ Ϫ.24, p Ͻ .01, and ␤ ϭ Ϫ.15, p Ͻ .05, for disclosure,
solicitation, control, and family closeness, respectively.
We find no evidence, then, that child disclosure and
normbreaking are related only because adolescents
who are emotionally close to their parents will share
their experiences and also refrain from normbreaking.
DISCUSSION
In this study we argued that parental “monitoring” is
really knowledge that parents have about their children’s
activities outside the home and that this knowledge
comes partly from the parents’ own efforts to find
out what their children are doing (solicitation and
control) and partly from the child’s spontaneous and
willing divulgence of information (child disclosure).
These three knowledge sources accounted for close to
1082 Child Development
half of the variance in our “monitoring” measure for
both children’s and parents’ reports.
In contrast to the parenting skills or direct supervision
interpretation of monitoring that dominates the
literature (Dishion & McMahon, 1998; Snyder &
Patterson, 1987), however, this study suggests that
“monitoring” is not what the term implies: a parental
activity. It is more a child’s activity. Both children’s
and parents’ reports of “monitoring” were best explained
as children’s voluntary descriptions of their
free-time activities.
Parents want to prevent antisocial behavior in their
children or to stop it if it has begun. In the criminological
literature, parents of delinquents have been characterized
as poor and inconsistent monitors of their
children’s activities (Patterson & Stouthamer-Loeber,
1984; Snyder & Patterson, 1987). In this study, as in
previous studies, poor parental “monitoring” was
linked to more normbreaking behavior. However,
when we asked which of the three potential sources
of information best explained the child’s normbreaking
behavior, we found that child disclosure was the
strongest predictor. This was true both for children’s
and parents’ reports and for the broad normbreaking
measure as well as for police contact.
Parental monitoring, defined as direct control of the
child’s behavior, was rejected early on by criminologists
as an effective proactive strategy for parents (see
Wells & Rankin, 1988, for a review). The argument was
that adolescents spend so much of their free time at
places where parents are not present that parents cannot
control their children’s behavior directly (Nye,
1958). Thereafter, a sizable body of research on parental
“monitoring” that challenged the early view emerged.
Many studies showed that “monitoring” did play an
important role in constraining antisocial behavior, and
several review studies came to that conclusion (Loeber
& Stouthamer-Loeber, 1986; Snyder & Patterson, 1987).
Today, the idea that monitoring measures reflect
parents’ efforts to control and manage their children is
widespread (see Dishion & McMahon, 1998; Snyder &
Patterson, 1987).Authors argue that more and stronger
surveillance by parents could reduce antisocial behavior
in children (Fletcher et al., 1995; Patterson &
Stouthamer-Loeber, 1984; Romer et al., 1994; Snyder &
Patterson, 1987; Weintraub & Gold, 1991). Patterson
and Stouthamer-Loeber (1984) suggested that juvenile
delinquents’ parents are “indifferent trackers of their
sons’ whereabouts” (p. 1305). Fletcher et al. (1995) concluded
that strong parental “monitoring” could both
deter adolescents from engaging in drug use in the first
place and reduce the risks of further use, and Romer et
al. (1994) suggested that parents of friendship groups
should join together to monitor their children’s freetime
activities to prevent risky sexual behavior. However,
these conclusions are not supported by the
present study. They were based on an inferential leap:
that parental knowledge must have come from surveillance
and direct control. The present study suggests
otherwise. Researchers, therefore, have likely drawn
premature conclusions from monitoring findings.
Researchers have also used “monitoring” measures
such as Patterson and Stouthamer-Loeber’s (1984)
five-item questionnaire to build models to explain how
parents influence their children’s antisocial behavior.
For example, Barber (1996) recently distinguished theoretically
between psychological and behavioral control.
He argued that they would be differently related
to internalizing and externalizing behavior; psychological
control should create internalizing problems
and behavioral control should prevent externalizing
problems, and, indeed, he found that behavioral control
was negatively related to delinquency. As a measure
of behavioral control, however, he used a 5-item
“monitoring” scale that measured parental knowledge,
the rationale being that “monitoring . . . appears
to be a particularly reliable and powerful index of
family management and regulation” (p. 3301). In reality,
although the “monitoring” scale taps parents’ awareness
of their children’s whereabouts, it cannot be assumed
to represent parents’ behavioral attempts to
control the child. The alternative interpretation—that
children with externalizing problems hide their
normbreaking behavior from their parents more than
other children, which results in their parents knowing
less—is more likely in light of our findings.
At least two plausible theoretical explanations could
account for our main finding: that high disclosers
were lower on normbreaking. One is a temperament
explanation. Perhaps some children are just temperamentally
prone to be agreeable, conventional people
who have nothing to hide from their parents and who
communicate willingly and openly. In this formulation,
temperament would be a third variable that causes
good behavior and also causes children to confide in
their parents, but there might be no causal link between
disclosure and good behavior at all. However, if this
was all that was going on in our data, child disclosure
should not have been important among children who
hung out on the streets in the evening, but it was.
Clearly, a richer explanation is needed.
Another explanation is that parents have done
something to build up the kinds of relationships that
facilitate communication, and those relationships are
what prevent bad behavior. This explanation is conceptually
linked to early formulations of control theory
that take the child’s attachment to the parents as
the crucial aspect of behavioral control (Hirschi, 1969).
Stattin and Kerr 1083
The basic idea is that the child’s emotional bond to the
parents prevents the child from doing anything that
might compromise the relationship—anything of
which parents might disapprove or that might embarrass
them. Consistent with this, attachment measures
have been shown to be important predictors of
delinquency (e.g., Benda & Whiteside, 1995; SokolKatz,
Dunham, & Zimmerman, 1997). Attachment
might be a third variable that causes children to want
to tell their parents all about their lives and also keeps
them from getting into trouble, thus explaining both
disclosure itself and the link between disclosure and
antisocial behavior. Our data are not supportive of
this idea, however. Child disclosure remains the
strongest predictor of normbreaking, even after controlling
for parent- and child-rated indicators of
parent–child relationships.
A limitation of both of these views is that they postulate
directional effects that are probably too simplistic.
Socialization is not a unidirectional process (Maccoby
& Martin, 1983; Magnusson & Stattin, 1998).
Developmental psychology has moved from a “topdown”
view of parenting to a more interactive view of
parent–child processes that recognizes the necessity of
reciprocity, cooperation, coordination, and coregulation.
A good parent–child relationship, which should
operate preventively, should be a two-way process, including
both the parents’ solicitation of knowledge
and control of their children’s behavior and the children’s
willingness to make their parents part of their
lives. A parent–child relationship that protects children
from antisocial behavior is unlikely to be built on
the parents’ or the children’s actions alone. For instance,
parents’ attempts to solicit information from an
unwilling child might have a limited effect on the
child’s behavior. In fact, some youngsters might view
parental solicitation as an intrusion into their privacy
and a means for parents to control their behavior. Parents’
well-meaning attempts to find out about their
children’s activities have to be matched by the children’s
willingness to reveal what they are doing,
where they are going, and whom they are with. A few
times, this has been expressed in the monitoring literature
(Crouter, et al., 1990; Weintraub & Gold, 1991).
Crouter et al. (1990) defined monitoring as similar to
Pulkkinen’s (1982, p. 656) child-center guidance and
emphasized the dual nature of the concept: “Parents
who are good monitors have made the effort to establish
channels of communication with their child, and
as a result of their relationship with the child, they are
knowledgeable about the child’s daily experiences. In
order to be an effective monitor, however, parental interest
is not enough: A child must be willing to share
his or her experiences and activities with the parent.
Seen in this light, parental monitoring is a relationship
property.” This discussion of child disclosure and how
the parent–child relationship might facilitate it is rare
in the monitoring literature (for another exception, see
Crouter et al., 1999). Typically, the parent–child relationship
is discussed as a factor that determines what
parents will do to monitor their children (e.g., Dishion &
McMahon, 1998) and the child’s active role is ignored.
This study leaves certain questions to be answered by
future research. The cross-sectional design of the study
precludes any detailed insights into causal connections.
Only longitudinal studies will enable us to identify the
origins of child disclosure or to say anything definitive
about whether parents’ knowledge actually prevents
problems. Future studies should also include comparisons
of mothers and fathers within the same families because
recent evidence suggests that their knowledge
may differ under certain circumstances (Crouter et al.,
1999). This particular area of research would be especially
interesting for future Swedish studies because
Swedish law encourages and facilitates fathers’ involvement
more than in the United States, where differences
in parental knowledge have been found.
The present study has a number of strengths, however.
One is the large sample size and the high response
rates from both children and parents. Another
is that the study deals with both parents’ and children’s
reports, which do not overlap precisely but
which nonetheless reveal the same overall picture
about parents’ knowledge, where it comes from, and
how these factors are connected to normbreaking. This
study challenges the common assumption of what parental
monitoring is; it reveals that parents’ direct control
over children’s behavior is not as important as the
youngsters’ own voluntary disclosure of information
about their lives. Thus, this study suggests that a bidirectional
model of parent–child interactions is needed.
Our findings carry implications for future monitoring
research and for parenting. In the future, researchers
should not use the term monitoring to refer
to measures of parental knowledge. Monitoring
should be reserved for measures that actually tap parents’
active efforts. If parental knowledge is measured,
it should be labeled parental knowledge. We should
recognize that parental knowledge is extremely important,
for reasons which are as yet unknown. In future
criminological investigations, more effort should
go toward identifying aspects of parent–child relationships
that are associated with the child’s disclosure
of information. A focus on the adolescent’s point
of view is also needed. For example, one issue that
has received little attention in the monitoring literature
is adolescents’ trust in their parents—whether they
feel that their parents are willing to listen to them, are
1084 Child Development
responsive, and would not ridicule or punish if they
confided in them. Another issue is whether adolescents
and their friends develop common attitudes
and norms about how much they should confide in
their parents about their daily activities and experiences.
These and other possible influences on child
disclosure should be investigated.
Concerning implications for parenting, general
advice to parents should be given cautiously. Parents
normally use different socializing strategies simultaneously
and try to find a balance between them. Parents’
awareness of their children’s activities is certainly
important for preventing negative behavior.
However, our results suggest that in addition to controlling
the child’s whereabouts, parents should try to
optimize conditions for the child to disclose information
about his or her everyday experiences. Parental
characteristics that have been associated with juvenile
delinquency, such as aggression; hostility; cruel,
neglecting attitudes toward the child; negativism; and
permissiveness of aggression (see Snyder & Patterson,
1987, for a review) are all behaviors that would discourage
child disclosure. A more child-centered approach,
which would result in a two-way dialogue,
might encourage children to share their mental lives
with their parents. Parents must be aware of how adolescents
feel and think as they formulate their upbringing
strategies. Hirschi (1969) suggested that the
parents’ “psychological presence” in the child would
keep the child from behaving badly, but that may not
be achievable if the child’s point of view and possible
reactions are not psychologically present in the parents
as they interact with their child.
ACKNOWLEDGMENTS
This research was supported by grants from the
Swedish Council for Planning and Coordination of
Research and the Swedish Council for Research in the
Humanities and Social Sciences. During the preparation
of this manuscript, Margaret Kerr was supported
by a grant from the Wenner-Grenska Samfundet.
AUTHOR ADDRESSES AND AFFILIATIONS
Corresponding author: Håkan Stattin, Department
of Psychology, University of Örebro, 701 82 Örebro,
Sweden; e-mail: hs@psychmax.psychology.su.se.
Margaret Kerr is also at the University of Örebro.
REFERENCES
Aseltine, R. H. (1995). A reconsideration of parental and
peer influences on adolescent deviance. Journal of Health
and Social Behavior, 36, 103–121.
Barber, B. K. (1996). Parental psychological control: Revisiting
a neglected construct. Child Development, 67, 3296–3319.
Benda, B. B., & Whiteside, L. (1995). Testing an integrated
model of delinquency using LISREL. Journal of Social Science
Research, 21, 1–32.
Biglan, A., Duncan, T. E., Ary, D. B., & Smolkowski, K.
(1995). Peer and parental influences on adolescent tobacco
use. Journal of Behavioral Medicine, 18, 315–330.
Cernkovich, S. A., & Giordano, P. C. (1987). Family relationships
and delinquency. Criminology, 24, 295–321.
Chassin, L., Pillow, D. R., Curran, P. J., Molina, B. S. G., &
Barrera, M., Jr. (1993). Relation of parental alcoholism to
early adolescent substance use: A test of three mediating
mechanisms. Journal of Abnormal Psychology, 102, 3–19.
Cohen, D. A., & Rice, J. C. (1995). A parent-targeted intervention
for adolescent substance use prevention: Lessons
learned. Evaluation Review, 19, 159–180.
Crouter, A. C., Helms-Erikson, H., Updegraff, K., &
McHale, S. M. (1999). Conditions underlying parents’
knowledge about children’s daily lives in middle childhood:
Between- and within-family comparisons. Child
Development 70, 246–259.
Crouter, A. C., MacDermid, S. M., McHale, S. M., & PerryJenkins,
M. (1990). Parental monitoring and perceptions
of children’s school performance and conduct in dualand
single-earner families. Developmental Psychology, 26,
649–657.
Crouter, A. C., Manke, B. A., & McHale, S. M. (1995). The
family context of gender intensification in early adolescence.
Child Development, 66, 317–329.
Dishion, T. J., Capaldi, D., Spracklen, K. M., & Li, F. (1995).
Peer ecology of male adolescent drug use: Developmental
processes in peer relations and psychopathology [Special
issue]. Development and Psychopathology, 7, 803–824.
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.
Flannery, D. J., Vazsonyi, A. T., Torquati, J., & Fridrich, A.
(1994). Ethnic and gender differences in risk for early adolescent
substance use. Journal of Youth and Adolescence,
23, 195–213.
Fletcher, A. C., Darling, N., & Steinberg, L. (1995). Parental
monitoring and peer influences on adolescent substance
use. In J. McCord (Ed.), Coercion and Punishment in LongTerm
Perspectives (pp. 259–271). Cambridge, MA: Cambridge
University Press.
Fridrich, A. H., & Flannery, D. J. (1995). The effects of ethnicity
and acculturation on early adolescent delinquency.
Journal of Child and Family Studies, 4, 69–87.
Hanson, K. (1989). Family climate: An adjective checklist
for family diagnosis [Familjeklimat. En adjektivlista for
familjediagnostik] (Report number 1). Lund, Sweden:
Department of Applied Psychology, Lund University.
Hirschi, T. (1969). Causes of delinquency. Berkeley, CA: University
of California Press.
Loeber, R., & Stouthamer-Loeber, M. (1986). Family factors
as correlates and predictors of juvenile conduct problems
and delinquency. In M. Tonry & N. Morris (Eds.),
Stattin and Kerr 1085
Crime and justice: An annual review of research. (Vol. 7).
Chicago: University of Chicago Press.
Maccoby, E. E., & Martin, J. A. (1983). Socialization in the
context of the family: Parent-child interaction. In E. M.
Hetherington (Ed.), P. H. Mussen (Series Ed.), Handbook
of child psychology: Vol. 4. Socialization, personality, and
social development. (pp. 1–101). New York: Wiley.
Magnusson, D., & Stattin, H. (1998). Person-context interaction
theories. In R. M. Lerner (Ed.), W. Damon (Series
Ed.) Handbook of child psychology: Vol. 1. Theoretical
models of human development (pp. 685–759). New York:
Wiley.
McCord, J. (1986). Instigation and insulation: How families
affect antisocial aggression. In J. Block, D. Olweus, &
M. R. Yarrow (Eds.) Development of Antisocial and Prosocial
Behavior (pp. 343–357). New York: Academic Press.
Metzler, C. W., Noell, J., Biglan,A.,Ary, D., & Smolkowski, K.
(1994). The social context for risky sexual behavior among
adolescents. Journal of Behavioral Medicine, 17, 419–438.
Nye, F. I. (1958). Family relationships and delinquent behavior.
New York: Wiley.
Otto, L. B., & Atkinson, M. P. (1997). Parental involvement
and adolescent development. Journal of Adolescent Research,
12, 68–89.
Patterson, G. R., & Stouthamer-Loeber, M. (1984). The correlation
of family management practices and delinquency.
Child Development, 55, 1299–1307.
Pulkkinen, L. (1982). Self-control and continuity from childhood
to adolescence. In P. B. Baltes & O. G. Brim (Eds.),
Life span development and behavior (Vol. 4, pp. 63–105).
New York: Academic Press.
Read, A. W., et al. (Eds.) (1995). The New International Webster’s
Comprehensive Dictionary of the English Language (encyclopedic
edition). Chicago: Trident Press International.
Reid, J. B., & Patterson, G. R. (1989). The development of antisocial
behavior patterns in childhood and adolescence:
Personality and aggression [Special issue]. European
Journal of Personality, 3, 107–119.
Romer, D., Black, M., Ricardo, I., Feigelman, S., Kaljee, L.,
Galbraith, J., Nesbit, R., Hornik, R. C., & Stanton, B.
(1994). Social influences on the sexual behavior of youth
at risk for HIV exposure. American Journal of Public
Health, 84, 977–985.
Sampson, R. J., & Laub, J. H. (1994). Urban poverty and the
family context of delinquency: A new look at structure
and process in a classic study: Children and poverty
[Special issue]. Child Development, 65, 523–540.
Snyder, J., & Patterson, G. (1987). Family interaction and delinquent
behavior. In H. C. Quay (Ed.), Handbook of Juvenile
Delinquency. New York: Wiley.
Sokol-Katz, J., Dunham, R., & Zimmerman, R. (1997). Family
structure versus attachment in controlling adolescent
deviant behavior: Asocial control model. Adolescence, 32,
199–215.
Well, L. E., & Rankin, J. H. (1988). Direct parental controls
and delinquency. Criminology, 26, 263–285.
Weintraub, K. J., & Gold, M. (1991). Monitoring and delinquency.
Criminal Behavior and Mental Health, 1, 268–281.
White, M. J., & Kaufman, G. (1997). Language usage, social
capital, and school completion among immigrants and
native-born ethnic groups. Social Science Quarterly, 78,
385–398.