Exposure to Internet Pornography among Children and Adolescents: A National Survey

Abstract

Estimates suggest that up to 90% or more youth between 12 and 18 years have access to the Internet. Concern has been raised that this increased accessibility may lead to a rise in pornography seeking among children and adolescents, with potentially serious ramifications for child and adolescent sexual development. Using data from the Youth Internet Safety Survey, a nationally representative, cross-sectional telephone survey of 1501 children and adolescents (ages 10-17 years), characteristics associated with self-reported pornography seeking behavior, both on the Internet and using traditional methods (e.g., magazines), are identified. Seekers of pornography, both online and offline, are significantly more likely to be male, with only 5% of self-identified seekers being female. The vast majority (87%) of youth who report looking for sexual images online are 14 years of age or older, when it is developmentally appropriate to be sexually curious. Children under the age of 14 who have intentionally looked at pornography are more likely to report traditional exposures, such as magazines or movies. Concerns about a large group of young children exposing themselves to pornography on the Internet may be overstated. Those who report intentional exposure to pornography, irrespective of source, are significantly more likely to cross-sectionally report delinquent behavior and substance use in the previous year. Further, online seekers versus offline seekers are more likely to report clinical features associated with depression and lower levels of emotional bonding with their caregiver. Results of the current investigation raise important questions for further inquiry. Findings from these cross-sectional data provide justification for longitudinal studies aimed at parsing out temporal sequencing of psychosocial experiences.

Full text

Exposure to Internet Pornography among Children
and Adolescents: A National Survey
MICHELE L. YBARRA, M.P.H., Ph.D.1, and KIMBERLY J. MITCHELL, Ph.D.2
ABSTRACT
Estimates suggest that up to 90% or more youth between 12 and 18 years have access to the In-
ternet. Concern has been raised that this increased accessibility may lead to a rise in pornog-
raphy seeking among children and adolescents, with potentially serious ramifications for
child and adolescent sexual development. Using data from the Youth Internet Safety Survey,
a nationally representative, cross-sectional telephone survey of 1501 children and adolescents
(ages 10–17 years), characteristics associated with self-reported pornography seeking behav-
ior, both on the Internet and using traditional methods (e.g., magazines), are identified. Seek-
ers of pornography, both online and offline, are significantly more likely to be male, with
only 5% of self-identified seekers being female. The vast majority (87%) of youth who report
looking for sexual images online are 14 years of age or older, when it is developmentally ap-
propriate to be sexually curious. Children under the age of 14 who have intentionally looked
at pornography are more likely to report traditional exposures, such as magazines or movies.
Concerns about a large group of young children exposing themselves to pornography on the
Internet may be overstated. Those who report intentional exposure to pornography, irrespec-
tive of source, are significantly more likely to cross-sectionally report delinquent behavior
and substance use in the previous year. Further, online seekers versus offline seekers are
more likely to report clinical features associated with depression and lower levels of emo-
tional bonding with their caregiver. Results of the current investigation raise important ques-
tions for further inquiry. Findings from these cross-sectional data provide justification for
longitudinal studies aimed at parsing out temporal sequencing of psychosocial experiences.
473
CYBERPSYCHOLOGY & BEHAVIOR
Volume 8, Number 5, 2005
© Mary Ann Liebert, Inc.
INTRODUCTION
AS THE INTERNET gains increasing prominence in
the lives of young people,1researchers have
begun investigating the influence that the Internet
environment may be having on child and adolescent
development.2Of particular interest is exposure to
sexual material. Child and adolescent exposure to
pornography is a controversial issue. Questions
about the contribution of pornography to deviant
sexual behavior, including sexual assault, negative
attitudes towards women, and the acceptance of de-
viant or aggressive sexual behavior among peers,
have been studied for decades. Results are mixed,
with some investigators arguing for clear and consis-
tent effects of exposure to pornography and subse-
quent sexually aggressive attitudes and behaviors,3
while others describe null or inconclusive findings.4,5
With specific relevance to young people, there is the
additional concern of negative effects on facets of
sexual development, such as sexual callousness, for
those who are exposed to pornography.6
1Internet Solutions for Kids, Inc., Irvine, California.
2Crimes Against Children Research Center, University of New Hampshire, Durham, New Hampshire.
14105c08.pgs 9/28/05 12:22 PM Page 473

Sexual development
Adolescent sexual development is complex and
dynamic. As children get older, they gain a greater
sense of their sexual self,7enhanced by an interplay
of biological and social changes as the individual
matures through childhood into adolescence. Al-
though puberty begins at different ages, virtually all
boys and girls have started the process by 14 years
of age.7Sexual interest increases with age and bio-
logical changes, with the average age of first sex ex-
perience in the United States being 15.8 years.8
Expression of sexual curiosity spans a continuum of
behaviors, from talking about sex, looking at sexual
materials, to actually engaging in sexual activity. Al-
though sexual activity can represent risks in and of
itself (e.g., sexually transmitted diseases), re-
searchers and other adolescent health professionals
have posited that exposure to pornography may be
harm-promoting in other domains as well.
Pornography research
The majority of pornography studies have been
conducted with adults.3–5 This is largely due to the
ethical and legal considerations of exposing chil-
dren and adolescents to potentially harmful mater-
ial. A handful of adolescent studies suggest no
relationship between pornography and behavior.
For example, a retrospective study of adolescent sex
offenders found no relationship between prior ex-
posure to pornography and the number of victims.9
Additionally, in a more intensive interview with a
sub-sample of youths, the majority of offenders de-
nied that their use of pornography in any way led to
the subsequent sex crime. Clearly, more research is
needed about child and adolescent consumption of
pornography. Importantly, as the Internet is used by
more and more young people,1the effects of access
to and exposure of online pornography on child
and adolescent development will be a child and
adolescent health issue of increasing importance.
Internet use among children and adolescents
More than 90% of young people between the
ages of 12 and 18 years use the Internet in the
United States.1A vast amount of information is
now widely and easily accessible to anyone who
has an Internet connection. Although positive as-
pects of the Internet are frequently cited, including
the availability of important and sometimes sensi-
tive health information10,11,13 the often unfettered
access to web sites may lead to an overall increase
in the numbers of young people seeking out porno-
graphic material. To safeguard against this type of
exposure, filtering and blocking software has been
developed to prevent access to specific sites, and
several child-oriented organizations recommend
the usage of such software on home computers.
The use of blocking software on public computers
such as those in public libraries remains controver-
sial because of free speech issues.
Prevalence of intentional Internet exposures
Several studies have asked youth about purpose-
ful exposure to sexual material online. A survey of
young people attending a private, urban school in
the Midwest reports that 21% have ever visited a
pornographic site for at least 3 min.12 A national tele-
phone survey of young people between the ages of
12 and 17 found that 15% have lied about their age to
gain access to a web site.10 Interestingly, this percent-
age is similar to the corresponding value observed
for adults in the same survey.10 Males, older youths,
those with greater months/ years of Internet experi-
ence, and those who use the Internet more inten-
sively (i.e., 5+ h/day) are more likely to report
purposefully seeking out these sites.10,12
Adolescent beliefs about Internet exposures to sexual
material
Although there are no studies about exposure to
sexual material on the Internet and resulting be-
havioral changes, a national study of older teen-
agers illuminates concerns about changes in the
attitudes and beliefs of young people.13 Over half
(59%) of respondents believe that viewing Internet
pornography may encourage young people to have
sex earlier. Almost one in two respondents (49%)
indicate that Internet pornography promotes nega-
tive attitudes towards women, with a similar per-
centage (49%) indicating that the images may
promote the perception that unprotected sexual ac-
tivity is “okay.” Beyond perception and belief, no
information is available about actual outcomes or
linkages between purposeful exposure to Internet
pornography and psychosocial or developmental
challenge.
Gaps in current literature
Despite the vast amount of adult literature about
intentional exposure to pornography,3–5 and the
emerging reports of adolescent pornography seek-
ing online,10,12–13 important questions remain. First,
beyond demographic characteristics, what is the
474 YBARRA AND MITCHELL
14105c08.pgs 9/28/05 12:22 PM Page 474

profile of young person who seek out pornography?
With the advent of the Internet, it is also important
to begin understanding specific characteristics of
children and adolescents who seek pornography on-
line, and investigate whether they differ signifi-
cantly from other young people. The Youth Internet
Safety Survey, a cross-sectional, nationally represen-
tative telephone survey of young people between
the ages of 10 and 17 years, provides a unique op-
portunity to compare online and offline pornogra-
phy seekers with other youth in an effort to inform
adolescent health and mental health professionals.
This investigation will report important cross-sec-
tional associations necessary to justify and inform
future, more complex, longitudinal surveys.
MATERIALS AND METHODS
The Youth Internet Safety Survey (YISS) was a na-
tionally representative telephone survey conducted
to quantify the online experiences of Internet-using
youth (n = 1,501).2It was conducted by researchers
at the Crimes against Children Research Center at
the University of New Hampshire. Data were col-
lected between the fall of 1999 and spring of 2000.
Study methods were approved and supervised by
the University of New Hampshire’s Human Subjects
Committee and adhered to the research guidelines
set forth by the Department of Justice.
Sampling method
The sample was identified using a national prob-
ability design. Phone numbers were generated by
another nationally representative study of youth
that was being conducted concurrently, the Second
National Incidence Study of Missing, Abducted,
Runaway, and Thrownaway Children (NISMART-
2).14 The phone numbers of households that were
identified by NISMART-2 as having a youth be-
tween the ages of 9 and 17 were forwarded to re-
searchers of the YISS (n = 6,594). A sample size of
1,500 was needed to achieve a sampling error of
+/2.5% at the 95% confidence interval. All phone
numbers received were dialed and contact was
made with 3,446 households. At the time the target
sample size of 1,500 was reached, 82% of contacted
and eligible households had agreed to participate.
The young person was encouraged to identify a
time for the interview during which he or she could
talk freely. Each participant received $10 and infor-
mation brochures about Internet safety as incentive
for their participation.
Study population
One youth and one caregiver in each household
were interviewed by telephone. Eligibility re-
quirements for the young person included: being
between the ages of 10 and 17 years, having used
the Internet at least once a month for the previous
6 months (at any location), being English speak-
ing, and having lived in the household for at least
2 weeks in the previous year. Caregivers were
self-identified as the one most familiar with the
young person’s Internet use. Adult consent was
required for the adult interview and both adult
consent and child assent were required for the
child interview.
The demographic characteristics of the sample
population were generally higher than the average
household in the United States, but were reflective
of households with Internet access at the time of
data collection.15,16 For example, more than three-
quarters of adult respondents reported at least
some college as the highest household education,
and half of the households surveyed had an annual
income of $50,000 or higher.2
Measures
Pornography seeking. Youths were asked four
questions about their pornography seeking behav-
ior. Those who indicated they had visited an
x-rated website on purpose were coded as report-
ing online-seeking behavior. Offline-only seeking
behavior was reflected in a positive response to at
least one of the three following actions: (1) seeking
x-rated books or materials; (2) watching x-rated
movies; or (3) calling a 900 telephone number.
Youth who reported both online and offline seeking
behavior were categorized as online-seeking youth.
Three categories were created: (1) online-seeking,
(2) offline-only seeking, and (3) non-seeking (refer-
ence group) youth.
Demographic characteristics. Parents reported the
youth’s sex, age, and household income. As pu-
berty typically starts between ages 8 and 14 for
girls, and ages 9 and 14 for boys,7we dichotomized
age to compare youth 13 years of age and younger
with those 14 years of age and older. Income was
collected and entered as one of four categories. To
ensure cell stability, this indicator was di-
chotomized to compare the highest group, $75,000
per year or greater, with all other households.
Youth indicated the race and ethnicity with which
they identified themselves. Race was categorized
as White, Black, or other race. Ethnicity was a
EXPOSURE TO INTERNET PORNOGRAPHY AMONG CHILDREN 475
14105c08.pgs 9/28/05 12:22 PM Page 475

dichotomous (yes/no) measure reflecting the re-
port of Hispanic ethnicity.
Internet usage characteristics. Youths were asked
to estimate the average amount of time per day and
per week spent online. Due to indications of non-
linearity, each indicator was dichotomized at one
standard deviation above the sample mean (5+
days/week versus fewer; more than 2 h/day and
greater versus fewer). Each respondent also was
asked to rate the importance of the Internet in his
or her life on a five-point Likert scale ranging from
5 (Extremely important) to 1 (Not at all). This was di-
chotomized at very or extremely important versus
less. Self-rated Internet expertise was similarly di-
chotomized at “almost” or “very expert” versus
less.
Respondents were asked to indicate the activity
for which they used the Internet most. Twelve dif-
ferent activities were mentioned, including school
assignments, checking prices, and creating/main-
taining a web page. For the purposes of our analy-
ses, activities were categorized into 4 different
groups based upon their likelihood of interper-
sonal interaction: (1) e-mail, (2) instant messaging,
(3) chat room, and (4) all other (reference group).
Young people were also asked to identify the loca-
tion at which they used the Internet most fre-
quently. Five options were available, including
home, school, library, someone else’s home, and
another place. To ensure cell stability this was cate-
gorized into three groups based upon the likely
amount of monitoring: (1) library or school, (2)
someone else’s home or another place, and (3)
home (reference group).
Unwanted exposure to sexual material. In contrast
to intentional exposure, some young people re-
ported unwanted exposure to sexual content while
online in the previous year. The definition of un-
wanted exposure was established by Mitchell and
colleagues.17 First, young people were asked three
screening questions about whether they had re-
ceived an email, clicked on a link, or gone to a web-
site and seen pictures of naked people or people
having sex when the child had not wanted to see
that type of material. Next, those who indicated
any Internet victimization were asked more de-
tailed follow-up questions. Due to time constraints,
an algorithm was created to determine which of the
three main study outcomes (i.e., harassment, sexual
solicitation, exposure to sexual material) were ex-
plored further with the child: harassment incidents
were chosen first, sexual solicitation incidents were
chosen second, and unwanted exposures to sexual
material incident were chosen third. If the child re-
ported all three types of victimization, only the first
two were queried. As a result, some unwanted ex-
posure incidents were not the subject of more de-
tailed inquiry and therefore not included in
incidence rates. Post-hoc analyses indicated that
78 young people who reported an unwanted expo-
sure were not counted in the estimate for unwanted
exposure.
Parental Internet controls. Caregivers were asked
about potential steps that they might have taken to
safeguard their child from unwanted exposures on-
line. These same safeguards may have affected the
likelihood of intentional access to Internet pornog-
raphy. Caregivers were asked whether they had a
rule restricting their child from visiting porno-
graphic web sites (yes/no). Additionally, they were
asked whether in the previous year, they had at any
time checked the history function to view the web
sites their child had visited recently (yes/no). Care-
givers were asked to indicate whether blocking or
filtering software was used on the home computer.
Given indications in previous studies that reports
by young people (in comparison to parents) are
more strongly associated with exposures online,2,17
we also included youth-report of filtering or block-
ing software on the home machine.
Caregiver–child relationships. Caregiver–child re-
lationships were queried using nine questions.
Based upon exploratory factor analysis (all eigen-
values>=1), three different aspects of the relation-
ship emerged: emotional bonding (i.e., how well
caregiver and child get along, caregiver trust of
child, discussing problems with caregiver when
feeling sad or in trouble, and frequency of having
fun together), parental monitoring (i.e., frequency
with which caregiver knows where child is, and
with whom child is spending time), and coercive
discipline (i.e., frequency of ‘nagging’ child, taking
away privileges, and yelling). A sum score was
created for each of the three aspects of the care-
giver-child relationship. Due to indications of non-
linearity, all were then dichotomized at one
standard deviation above the sample mean versus
lower (high scores reflect a poorer rating).
Psychosocial characteristics. Clinical features of
depression were measured using nine questions
(yes/no) that were based upon the nine criteria for
major depression in the DSM IV.18 All questions re-
ferred to the previous month except for anhedonia,
which referred to all day nearly ever day for the
last two weeks. Youth who reported five or more
476 YBARRA AND MITCHELL
14105c08.pgs 9/28/05 12:22 PM Page 476

clinical features, one of which was anhedonia or
dysphoria, and functional impairment in at least
one area of life (i.e., school, personal hygiene, self-
efficacy) were categorized as reporting a high level
of depression.
Youth were asked five questions about their sub-
stance use behavior, namely whether and how fre-
quently they had used: cigarettes, marijuana,
inhalants, alcohol, and all other controlled sub-
stances. Youth who reported any drug use four or
more times a week in the previous year were coded
as seriously involved substance users.
Delinquent behavior was indicated if a youth en-
dorsed at least one of the four following behav-
iors/experiences: (1) police contact (i.e., interaction
with police that did not necessarily result in an ar-
rest), (2) physically assaulting another person,
(3) purposefully damaging property, or (4) stealing
property.
Youth who reported at least one negative life ex-
perience in the previous year were compared to
those indicating fewer: (1) caregiver divorce,
(2) caregiver loosing a job, (3) death in the family, or
(4) move of residence.
Physical or sexual victimization was included in
the analyses to reflect youth who indicated they
had been hit, beat, kicked, physically abused in any
way by a grown-up taking care of them, and/or
forced or made to do sexual things by someone, in-
cluding someone they didn’t know or even some-
one they knew well.
Statistical analysis
Standard data collection procedures, including
Computer Assisted Telephone Interviewing
(CATI), were used during the survey process. For
the purposes of the current analyses, all cases were
required to have at least 90% valid data across all
measures (i.e., 22 out of 24 variables). Seventeen
cases were dropped as a result, leading to a final
sample size of 1,484. Missing data and non-re-
sponses (i.e., don’t know and refused) were im-
puted using best-set regression.19 For most
variables, less than 1% of data were imputed;
household income (6.7% imputed) and race (1.6%
inputed) were exceptions. No significant differ-
ences between retained and not-retained cases
were observed in reported intentional exposure to
online pornography (X2(2) = .62, p > 05). A dummy
variable was created to adjust for potential and
unanticipated differences in the results between
cases with imputed data and all other cases.
First, chi-square tests were used to test signifi-
cant difference in distribution across the three
groups of pornography exposure (i.e., no exposure,
any online exposure, only offline exposure) based
upon youth-reported characteristics.
Second, in order to identify a cross-sectional pro-
file of pornography seeking youth, a parsimonious
multinomial logistic regression model of influential
characteristics was identified using likelihood ratio
tests (LRT). A saturated model was created by in-
cluding all characteristics of interest; then, each vari-
able was tested for its contribution to the model by
dropping it and then testing the statistical difference
between the saturated model and the new model.
Variables that did not significantly contribute to the
model (i.e., LRT p > .05) were dropped. Those that
indicated a large magnitude of association in the sat-
urated model (i.e., of 3.0 or greater Conditional
Odds Ratio [COR]) were retested at the final stage if
they were dropped during the variable testing pro-
cess. Those that continued to demonstrate a strong
association (i.e., COR of 3.0 or greater), regardless of
significance, were included in the model and left to
the reader for interpretation and valuation. Robust
standard errors were estimated given the clustering
of answers within household.
The third step in the statistical methods was to
identify significant differences in cross-sectional
characteristics between youth who seek pornogra-
phy online versus those who reported only offline-
only seeking behavior. A parsimonious logistic
regression model was created using likelihood ratio
tests as described above. Again, variables that sug-
gested a strong association were retested at the end of
model building and if they continued to demonstrate
strong association (i.e., COR 3.0 or greater), were in-
cluded in the final model irrespective of statistical
significance. Robust standard errors were estimated
given the clustering of answers within household.
RESULTS
Children and adolescents between the ages of
10 and 17 who used the Internet were surveyed
about their behaviors and experiences online and
offline. Each respondent was asked whether he or
she had intentionally viewed sexual material on the
Internet, as well as using traditional media (e.g.,
magazines). Based upon self-report, young people
were categorized into one of three groups: 1) non-
seekers (neither intentional online nor offline expo-
sure to pornography); 2) online seekers (any
intentional online exposure to pornography); or
3) offline-only seekers (intentional exposure to
pornography only via traditional, offline means).
Fifteen percent of young, regular Internet users re-
EXPOSURE TO INTERNET PORNOGRAPHY AMONG CHILDREN 477
14105c08.pgs 9/28/05 12:22 PM Page 477

ported intentional exposure to pornographic mate-
rial in the previous year. Specifically, 8% (n = 122)
reported online seeking, while an additional 7%
(n = 106) indicated offline-only seeking behavior.
Sixty-eight youth reported intentional exposure
both online and offline and were thus categorized
as online seekers. Self-reported youth characteris-
tics were examined for significant differences
across all three groups using chi-square tests;
results are found in Table 1 and detailed below.
Demographic characteristics
The age distribution was significantly different
across the three groups of youth (p < .001). Eighty-
seven percent of offline-only seekers were 14 years
or older compared to 74% of non-seekers and 60%
of online seekers. The percentage of males across
the three groups of pornography-seeking behavior
also differed significantly (p < .001); 87% of online
seekers were male compared to 79% of offline-only
seekers and 47% of non-seekers. Youth were similar
in terms of annual household income, race, and
ethnicity.
Internet usage characteristics
Almost all characteristics of Internet use differed
based upon the report of intentional exposure to
pornography. For example, most frequent Internet
activity was significantly related to the report of
pornography seeking behavior (p < .05). Seventeen
percent of online seekers used the Internet most
frequently to visit chat rooms as compared to 11%
of offline seekers and 8% of non-seekers. On the
other hand, the percentage of youth who used the
Internet most frequently for instant messaging was
relatively similar, with 12% of online seekers re-
porting such behavior versus 9% of offline seekers
and 10% of non-seekers. Both frequency and inten-
sity of Internet use significantly differed across
youth with various pornography seeking behavior.
Fifty-six percent of online seekers used the Internet
for an average of 4 days or more, compared to 35%
of offline seekers and 41% of non-seekers (p < .01).
Almost 20% of offline and online-only seekers
reported using the Internet for an average of
2 hours or more as compared to 12% of non-seek-
ers (p < .05). Self-ratings of Internet expertise (p <
.001) and the importance of the Internet (p < .05)
significantly varied across groups of youth, with
49% of online seekers reporting themselves as al-
most expert or expert as compared to 29% of
offline seekers and 32% of non-seekers, and 30% of
online seekers reporting the Internet very impor-
tant to themselves versus 22% of offline seekers
and 19% of non-seekers.
Parental Internet controls
None of the four measures of parental Internet
controls significantly differentiated youth by their
self-report of pornography seeking behavior. Simi-
larly high percentages (85–93%) of caregivers re-
ported a household rule about disallowing Internet
pornography sites across the three groups of young
people (p > .05). When asked whether a filter or
blocking software was installed on the computer,
27% of caregivers and 16% of youth online seekers,
versus 22% of caregivers and 19% of youth offline-
only seekers, and 23% of both caregivers and youth
non-seekers responded positively (p > .05). A non-
significant but positive trend was observed for the
report of checking the history function; 53% of care-
givers of online seekers compared to 43% of non-
seekers and 39% of offline-only seekers responded
positively (p>.05).
Caregiver–child relationships
Ratings for all three aspects of the caregiver-child
relationship were significantly different across the
three groups of pornography seeking youth. Al-
most one-third of online seekers rated their emo-
tional bond with their caregiver as poor compared
to 15% of offline seekers, and 10% of non-seekers
(p<.001). One quarter of online as well as 23% off-
line seekers reported low caregiver monitoring as
compared to 9% of non-seekers (p < .001). Frequent
coercive discipline was most commonly reported
by offline seekers, with 31% of youth indicating
such caregiver behavior, as compared to 23% of on-
line seekers and 17% of non-seekers (p < .001).
Psychosocial challenge
All indications of psychosocial challenge signifi-
cantly differed based upon self-report of pornogra-
phy seeking behavior among young, regular
Internet users. Overall, 25% of youth in the survey
reported an unwanted exposure to sexual material
at least once in the previous year. When examined
by pornography-seeking behavior, 53% of online
seekers reported unwanted exposure versus 35% of
offline-only seekers and 22% of non-seekers (p <
.001). Fifty percent of online seekers indicated
physical or sexual victimization versus 37% of
offline seekers and 31% of non-seekers (p < .001).
Delinquent behavior was reported four times more
often by pornography seekers, with 48% of online
478 YBARRA AND MITCHELL
14105c08.pgs 9/28/05 12:22 PM Page 478

seekers and 42% of offline seekers reporting this be-
havior in the previous year, as compared to 11% of
non-seekers (p < .001). Higher percentages of
young people reported seriously involved sub-
stance use who also reported pornography seeking
versus non-seeking, with 37% of online-seekers
versus 26% of offline-only seekers and 10% of non-
seekers reporting such use (p < .001). Twice as
many online seekers (11%) reported clinical fea-
tures of major depression compared to offline (4%)
and non-seekers (5%) (p < .05). Two in five offline-
only seekers (42%) reported at least one negative
life experience in the previous year versus 31% of
online seekers and 27% of non-seekers (p < .01).
A cross-sectional profile of youth characteristics
associated with intentional exposure to pornography
In order to identify a cross-sectional profile
unique to young people who consume pornogra-
phy, a parsimonious multinomial logistic regres-
sion model of related youth characteristics was
estimated. Characteristics that significantly con-
tributed to the overall model were retained, pro-
ducing a regression model of variables that
together helped explain the conditional odds of re-
porting online seeking behavior and offline-only
seeking behavior versus non seeking behavior. All
variables in the model were adjusted for all others
listed in the model. Results are displayed in Table 2
and described below.
Males were significantly more likely to report
pornography seeking than females. Indeed, boys
were more than 7 times as likely to report online-
seeking (p < .001) and 4 times as likely to report
offline-only seeking (p < .001) as compared to
otherwise similar females. Older youth also were
significantly more likely to report intentional expo-
sure. Youth 14 years and older were almost three
times as likely to report online seeking behavior
compared to otherwise similar, younger youth (p <
.001). No significant differences in age were noted
between youth who reported offline-only seeking
and non-seeking behavior.
All Internet usage characteristics failed to signifi-
cantly differentiate reports of pornography seeking
behavior. On the other hand, the caregiver-child re-
lationship was an important influencer in estimat-
ing the likelihood of reporting pornography
exposure. Youth who reported a poor emotional
bond with their caregiver were twice as likely also
to report online-seeking behavior versus otherwise
similar youth who reported a strong emotional
bond (p < .01). Frequent coercive discipline was sig-
nificantly related to 67% higher adjusted condi-
tional odds of reporting offline-only seeking behav-
ior versus non-seeking behavior (p < .05).
Indications of psychosocial challenge were sig-
nificantly related to increased adjusted conditional
odds of both online and offline-only seeking behav-
ior. Delinquent behavior was associated with a
4-fold increase in adjusted conditional odds of re-
porting either online-seeking behavior (p < .001) or
offline-only seeking behavior (p < .001) compared
to non-seeking behavior after adjusting for all other
influential characteristics. The report of substance
use was related to more than a two-fold increase in
adjusted conditional odds in disclosing online (p <
.001) as well as offline-only (p < .01) seeking behav-
ior compared to otherwise similar youth who re-
ported negligible substance use. Young people who
reported unintentional exposure to sexual material
online were more than 2.5 times as likely to report
intentional exposure online compared to otherwise
similar young people who did not report uninten-
tional exposure (p < .001).
Differences in youth characteristics among youth who
report online and offline-only seeking behavior
In order to understand the cross-sectional differ-
ences in youth characteristics among young regular
Internet users who reported online seeking behav-
ior versus offline-only seeking behavior, a parsimo-
nious logistic regression model was identified (n =
228). Variables retained in the model together
helped explain the difference in the odds of report-
ing online versus offline-only behavior. Results are
displayed in Table 3 and discussed below.
Fifty-four percent of all self-identified pornogra-
phy seekers in the sample looked for sexual images
online. Demographic characteristics, Internet usage
characteristics, aspects of the caregiver-child rela-
tionship, parental Internet controls, and psychoso-
cial characteristics each significantly discriminated
online seekers and offline seekers. Hispanic youth
were almost three times as likely to report online
seeking versus offline seeking behavior versus oth-
erwise similar youth of non-Hispanic ethnicity (p =
.02). Older youth were significantly more likely to
report online versus offline seeking behavior after
adjusting for all other significant characteristics,
with those 14 and older almost 2.5 times as likely as
younger youth to report online-seeking behavior
(p= .03). Unsurprisingly, youth who rated them-
selves as almost expert or expert at using the Inter-
net were twice as likely to indicate online versus
offline pornography seeking behavior, after adjust-
ing for all other significant characteristics (p = .03).
Youth who reported frequent Internet use also were
EXPOSURE TO INTERNET PORNOGRAPHY AMONG CHILDREN 479
14105c08.pgs 9/28/05 12:22 PM Page 479

TABLE 1. COMPARISONS OF DEMOGRAPHIC AND INTERNET USE CHARACTERISTICS BY SELF-REPORTED PORNOGRAPHY-SEEKING BEHAVIOR (N= 1,484)
No purposeful Offline Online
exposure exposure only exposure Statistical
(85%, n= 1256) (7%, n= 106) (8%, n= 122) comparison
Demographic characteristics
Race
White (reference group) 77.3% (971) 73.6% (78) 76.2% (93) 2(4) = 1.9
Black 10.8% (135) 11.3% (12) 9.0% (11)
Other 11.9% (150) 15.1% (16) 14.8% (18)
Older age (14 years and older) 60.0% (753) 73.6% (78) 86.9% (106) 2(2) = 40.0***
Male 47.3% (594) 79.3% (84) 86.9% (106) 2(2) = 101.9***
Household income 23.0% (289) 21.7% (23) 27.9% (34) 2(2) = 1.6
Hispanic ethnicity 7.3% (91) 4.7% (5) 8.2% (10) 2(2) = 1.2
Internet usage characteristics
Most frequent log in location
Home (reference group) 65.6% (824) 51.9% (55) 67.2% (82) 2(4) = 9.0
School/library 22.9% (288) 30.2% (32) 20.5% (25)
Other house/place 11.5% (144) 17.9% (19) 12.3% (15)
Most frequent Internet activity
All other (reference group) 55.7% (699) 57.6% (61) 48.4% (59) 2(6) = 15.5*
E-mail 26.7% (335) 21.7% (23) 22.1% (27)
Instant messaging 9.9% (124) 9.4% (10) 12.3% (15)
Chat room 7.8% (98) 11.3% (12) 17.2% (21)
14105c08.pgs 9/28/05 12:22 PM Page 480

Frequency of use (4+ days/week) 40.8% (512) 34.9% (37) 55.7% (68) 2(2) = 12.4**
Internet expertise (almost expert/expert) 31.7% (398) 29.3% (31) 49.2% (60) 2(2) = 16.1***
Importance of Internet to self (very/extremely important) 19.1% (240) 21.7% (23) 30.3% (37) 2(2) = 8.8*
Intensity of use (More than 2–3 h/day) 12.3% (154) 19.8% (21) 18.0% (22) 2(2) = 7.5*
Parental Internet controls
Rule about no pornography sites 88.7% (1114) 84.9% (90) 92.6% (113) 2(2) = 3.4
Check the history function 42.8% (538) 38.7% (41) 52.5% (64) 2(2) = 5.2
Filter/block (parent report) 22.5% (282) 21.7% (23) 27.1% (33) 2(2) = 1.4
Filter/block (youth report) 22.5% (282) 18.9% (20) 16.4% (20) 2(2) = 2.9
Caregiver-child relationship
Frequent coercive discipline 17.2% (216) 31.1% (33) 23.0% (28) 2(2) = 14.1***
Poor emotional bond 10.0% (125) 15.1% (16) 30.3% (37) 2(2) = 44.8***
Low monitoring 8.6% (108) 22.6% (24) 25.4% (31) 2(2) = 48.0***
Psychosocial characteristics
Physical or sexual victimization 31.1% (390) 36.8% (39) 50.0% (61) 2(2) = 18.8***
Negative life experience (2+) 27.2% (341) 41.5% (44) 31.2% (38) 2(2) = 10.3**
Delinquent behavior 11.1% (139) 41.5% (44) 47.5% (58) 2(2) = 162.3***
Seriously involved substance use 9.7% (122) 26.4% (28) 36.9% (45) 2(2) = 89.6***
Clinical features of depression 4.6% (58) 3.8% (4) 10.7% (13) 2(2) = 8.8*
Unwanted exposure to Internet pornography 21.8% (274) 34.9% (37) 52.5% (64) 2(2) = 60.9***
*p < .05; ** p < .01; *** p < .001
14105c08.pgs 9/28/05 12:22 PM Page 481

twice as likely to report online-seeking behavior
compared to otherwise similar youth who use the
Internet less frequently (p = .02). Associated with
Internet use, one aspect of parental Internet con-
trols significantly differed between online and
offline-only seekers. Young people whose care-
givers reported household rules about not visiting
Internet pornography websites were 3.5 times as
likely to cross-sectionally report online seeking be-
havior compared to otherwise similar young people
without such a household Internet rule (p = .01). All
other things being equal, those who rated their emo-
tional bond with their caregiver as poor were more
than twice as likely to be online seekers versus of-
fline seekers (p = .03). On the other hand, youth
who reported frequent coercive discipline were 65%
482 YBARRA AND MITCHELL
TABLE 2. PARSIMONIOUS MULTINOMIAL LOGISTIC REGRESSION OF PORNOGRAPHY-SEEKING BEHAVIOR
AMONG YOUNG REGULAR INTERNET USERS (N= 1,484)
Offline exposure only (n= 106) Online exposure (n= 122)
ACOR (95% CI) P-value ACOR (95% CI) P-value
Demographic characteristics
Male 4.06 (2.42, 6.80) <.001 7.41 (4.07, 13.49) <.001
Older age (14–17 years) 1.46 (0.90, 2.37) 0.13 2.79 (1.51, 5.14) <.001
Caregiver-child relationship
Poor emotional bond 0.77 (0.41, 1.44) 0.41 2.02 (1.19, 3.42) <.01
Frequent coercive discipline 1.67 (1.03, 2.71) 0.04 0.77 (0.45, 1.31) 0.33
Psychosocial characteristics
Delinquent behavior 3.94 (2.47, 6.27) <.001 4.17 (2.61, 6.65) <.001
Seriously involved substance use 2.12 (1.23, 3.66) <.01 2.60 (1.57, 4.29) <.001
Unintentional exposure to 1.43 (0.92, 2.22) 0.11 2.63 (1.71, 4.07) <.001
Internet pornography
ACOR (adjusted conditional odds ratio) refers to the conditional odds of reporting that specific type of
seeking behavior versus non-seeking behavior, taking into account the likelihood of falling into the other
two seeking behavior categories.
TABLE 3. PARSIMONIOUS LOGISTIC REGRESSION MODEL OF ONLINE SEEKING (54%)
VERSUS OFFLINE-ONLY SEEKING (46%) AMONG YOUNG REGULAR INTERNET USERS WHO
INTENTIONALLY SEEK EXPOSURE TO PORNOGRAPHY (N= 228)
AOR (95% CI) P-value
Demographic characteristics
Hispanic 3.32 (1.22, 9.06) 0.02
Older age (14 years and older) 2.35 (1.11, 4.95) 0.03
Internet characteristics
Internet expertise (almost or expert) 2.05 (1.09, 3.87) 0.03
Frequency of use (4+ days/week) 2.11 (1.13, 3.92) 0.02
Caregiver–child relationship
Poor emotional bond 2.41 (1.10, 5.29) 0.03
Frequent coercive discipline 0.34 (0.17, 0.68) <.01
Parental Internet controls
Rule about no pornography sites 3.53 (1.36, 9.18) 0.01
Psychosocial characteristics
Clinical features of major depression 3.51 (0.97, 12.66) 0.06
AOR (adjusted odds ratio) refers to the odds of reporting that specific type of seeking behavior versus
non-seeking behavior, taking into account the likelihood of falling into the other two seeking behavior
categories.
14105c08.pgs 9/28/05 12:22 PM Page 482

less likely to indicate online versus offline porno-
graphy seeking behavior (p < .01). A nonsignificant
trend was also observed for those who reported
clinical features of depression to be 3.5 times as
likely to also report online versus offline-only seek-
ing behavior (p = .06).
DISCUSSION
Using data from the Youth Internet Safety Sur-
vey, youth-reported pornography seeking behavior
was examined for cross-sectional differences in in-
dividual, Internet usage, and psychosocial charac-
teristics. In this study, we take a ’non-traditional’
look at characteristics related to intentional expo-
sure to pornography. Instead of focusing on beliefs
and behaviors related to sex or women,3an analysis
of personal characteristics is offered. This initial in-
vestigation reports cross-sectional linkages, the
first step in investigating longitudinal, develop-
mental trajectories of youth sexual development
and exposure to Internet pornography.
Demographic characteristics of pornography seekers
As reported by previous studies,10,12 seekers of
pornography, both online and offline, are signifi-
cantly more likely to be male. In fact, almost one-
quarter of all males report at least one intentional
exposure in the previous year as compared to 5% of
all females. There is a dearth of research informa-
tion about this small but perhaps significant group
of females; more research is needed.
Overall, older youth are more than twice as
likely as younger youth to report an intentional
pornographic exposure (i.e., 20% of youth between
14 and 17 years of age, versus 8% of youth between
the ages of 10 and 13). Among older youth, online
exposures are favored over traditional venues,
whereas younger youth report traditional expo-
sures more frequently than those found on the In-
ternet. Findings may be an indication that the
majority of youth who seek pornography are sim-
ply age-appropriately curious about sex. Perhaps
the age differential in terms of online exposure may
be an indication that controls Internet pornography
sites have enacted, specifically the necessity of a
credit cared number for access to the web site, have
helped keep young children out of x-rated web
sites. It is also true that older youth are more inde-
pendent with their Internet use, enabling more
freedom to explore. On the other hand, younger
people may be more likely to not disclose their In-
ternet seeking behavior, although why this bias
wouldn’t also affect the report of offline-only seek-
ing behavior is not clear.
Online versus offline seekers
Important differences in psychosocial function-
ing between young people who view pornography
online versus offline (e.g., magazines) are noted. Re-
sults suggest that those who use the Internet also
may be experiencing emotional challenge. These
young people are 3.5 times as likely also to report
clinical features of major depression and 2.4 times
as likely also to report a poor emotional bond with
their caregiver. It is unlikely that exposure to
pornography is a direct cause of these challenges.
Given reports that young people who are depressed
become socially isolated20 however, it is possible
that these young people are choosing the Internet
versus traditional methods because of the de-
creased social demands. On the other hand, some
online exposures may be more explicit or in some
other way disturbing, and this may contribute to
one’s depression. Further research is warranted.
Pornography exposure and aggression
Risk associated with the consumption of pornog-
raphy is a complex issue. A meta-analysis of natu-
ralistic studies of non-offending men reports that
for the majority of men, pornography exposure,
even at very frequent levels, are not linked to in-
creased levels of sexual aggression.3Among men
who have “predisposing risk levels” towards ag-
gressive sexual behavior, those who frequently
consume pornography have more than four times
greater levels of sexual aggression compared to
their peers who infrequently seek out pornogra-
phy.3 Indeed, what many agree upon is that where
an association between pornography and sexual
deviance does exist, it is likely with a person who
has more global challenge.4This is especially per-
tinent to the current findings as those youth who
report intentional exposure, both online as well as
offline-only, to pornography are significantly more
likely to cross-sectionally report delinquent be-
havior and substance use. Future research should
focus on longitudinal studies that parse out tem-
porality of these characteristics. It is possible that
young people who are delinquent seek out
pornography as a ‘symptom’ of their behavior.
Also, pursuing the role of alcohol is necessary
based upon current findings and the results of a
survey of sex-offending adolescents that suggests
increased alcohol use is significantly related to a
greater number of victims.9
EXPOSURE TO INTERNET PORNOGRAPHY AMONG CHILDREN 483
14105c08.pgs 9/28/05 12:22 PM Page 483

Replacing or increasing
The Internet represents a new mode of delivery
of pornographic material. Has this potentially
greater availability of pornography led to an over-
all increase in intentional exposure among adoles-
cents, or instead, has it simply replaced another
mode (e.g., magazines, movies) that would have
been used if the Internet not been available? In the
current sample, 4.5% of young regular Internet
users report both online and offline-only seeking
behavior, 3.6% report only online behavior, and
7.2% report offline only behavior. These results
suggest that more youth are accessing pornogra-
phy offline versus online. Being cross-sectional,
these findings cannot speak to trends in behavior
over time however, and it is possible that overall,
fewer young people sought out pornography be-
fore it was available online. Further research is
needed.
Parental involvement
Among adolescents who report intentionally
looking at pornography, the report of household
rules delineating no pornographic web sites is re-
lated to a three-fold increased odds of reporting on-
line seeking behavior compared to offline-only
seeking behavior. It is possible and likely that this
is a reflection of household rules following an inci-
dent (i.e., a caregiver discovers the young person
has gone to websites and then institutes a house-
hold rule).
Web site filters and other parental controls are
oft-cited remedies for caregivers hoping to shield
their youth from unwanted exposures. None of the
three safeguards examined in the current study
however (i.e., rules, blocking software, checking
the history function), differentiate between pornog-
raphy-seeking and non-seeking youth. Certainly
parental involvement is important, but additional
actions may be necessary. This may be especially
true for those adolescents who report poor care-
giver-child relationships along with pornography-
seeking behavior. Intervention and prevention
efforts aimed at adults are necessary but not suffi-
cient. Efforts directly aimed at children and adoles-
cents are also needed.
Limitations
This investigation is the first to scientifically
compare youth who intentionally seek sexual ma-
terial online and offline. It is not however, without
limitations. There were a unequal number of ques-
tions about offline versus online exposures. Three
questions were asked about traditional exposures
to pornography as compared to one question about
online exposures. This may have led to an over-
representation of offline versus online seeking
behavior. If this led to misclassification, the results
presented in the current investigation may be an at-
tenuated reflection of the true associations between
online seeking behavior and the personal charac-
teristics examined. Overall, 15% of young people
report some form of purposeful exposure to
pornography, 8% of whom report online seeking.
This one-year prevalence rate is lower than previ-
ous reports (i.e., 15% by Lenhart et al., 200110; 21%
by Stahl & Fritz, 200212), perhaps because of the or-
dering of the questions. As the survey was de-
signed to identify unwanted exposure to sexual
material, this outcome was queried before inten-
tional exposure. It may also be because the wording
of the questions about sexual material were clear
about their intent compared to other surveys that
have used less pointed questions and instead cho-
sen proxies by asking e.g., whether the youth has
lied about their age. Perhaps asking the question
clearly leads to response bias for youth who do not
want to admit specifically to seeking sexual mater-
ial. What is sacrificed in sensitivity however, is
gained in specificity. We can be reasonably confi-
dent that youth who respond in the affirmative
truly have intentionally sought out sexual material
online. Further, because questions about both on-
line and offline exposure were equally explicit, it is
unlikely that differential reporting resulted.
Additional limitations that need to be kept in
mind when interpreting the data include the fact
that the survey is not representative of all young
people; only Internet users were included in the
survey, possibly undercounting the number of
young people who intentionally seek offline expo-
sure to pornography. Second, this survey is cross-
sectional, thereby precluding temporal references.
It cannot be said therefore, that seeking pornogra-
phy lead to later delinquency for example, or that
delinquency led to seeking pornography. Third, the
measure of exposure is a dichotomous (yes/no) in-
dicator. It is possible that intensity of exposure re-
veals differences in psychosocial challenge and this
is an important area of future research. Fourth, the
data were collected between the fall of 1999 and
spring of 2000. It is possible that youth behavior
has changed online as the number of youth access-
ing the Internet has increased. Based upon reports
of general Internet use however (e.g., e-mailing, in-
stant messaging), it is less likely that behavior has
changed and more that access has increased.
484 YBARRA AND MITCHELL
14105c08.pgs 9/28/05 12:22 PM Page 484

Lastly, there is a potential bias in the sampling
frame, as potential households were identified as
part of a larger study. All phone numbers were
forwarded to YISS researchers to minimize bias
however, including those households that declined
to participate in the NISMART-2 study.
Implications
Instead of an avalanche of young children turn-
ing to the Internet as a venue for easy access to
pornography, the current findings suggest that
the vast majority of minors who use the Internet
to look for sexual images are likely 14 years of age
and older. Younger children who have looked at
pornography are more likely to report traditional
exposures, such as magazines or television. Based
upon the current findings, concerns about a large
percentage of extremely young children using the
Internet to expose themselves to sexual images
before they are developmentally ready, may be
overstated.
Results from the current study suggest that
among young, regular Internet users, those who re-
port intentionally seeking pornography may be
facing multiple challenges, including delinquent
behavior and substance use. Further, there is an in-
creased trend for youth who report clinical features
of depression to be more likely to report online-
seeking versus offline-only seeking behavior. Thus,
intentional exposure to pornography may be one
behavior among many for young people struggling
in their adolescence. Child and adolescent health
professionals, and others interacting with youth
should be sensitive to the possibility that at least
for some young people, seeking out pornography
either online or offline, has implications beyond
their sexual development.
Future studies should focus on parsing out
whether online pornography seeking represents a
shift in behavior, or is leading to an overall increase
in the number of youth who are accessing pornog-
raphy. Additionally, the results of this study need
to be replicated; longitudinal studies are also called
for to parse out the temporality and contribution of
events. For example, does intentional exposure of
pornography actually lead to delinquent behavior
or is it a ’symptom’ of a young person acting out?
Adolescents who use the Internet to look at sex-
ual material may be manifesting age-appropriate
sexual curiosity. Based upon the cross-sectional
profile of young people who report pornography-
seeking behavior, however, it also appears that the
behavior may be a marker for greater challenge for
some young people. Results of the current investi-
gation have raised important questions for further
inquiry.
ACKNOWLEDGMENTS
We would like to thank colleagues at the UNH
Family Research Lab and Crimes against Children
Research Center, and James C. (Jim) Anthony, Pro-
fessor and Chair, Department of Epidemiology,
Michigan State University, for their review,
thoughtful comments, and instructive suggestions
on drafts of the manuscript.
REFERENCES
1. UCLA Center for Communication Policy. (2003).
UCLA internet report—year three. UCLA Center for
Communication Policy.
2. Finkelhor, D., Mitchell, K., and Wolak, J. (2000). Online
victimization: a report on the nation’s young people.
National Center for Missing & Exploited Children.
3. Malamuth, N.M., Addison, T., Koss, M. (2000).
Pornography and sexual aggression: are there reli-
able effects and can we understand them? Annual Re-
view of Sexual Research 11:26–91.
4. Fukui, A., & Westmore, B. (1994). To see or not to see:
The debate over pornography and its relationship to
sexual aggression. Australian and New Zealand Journal
of Psychiatry 28:600–606.
5. Kutchinsky, B. (1991). Pornography and rape: theory
and practice? Evidence from crime data in four coun-
tries where pornography is easily accessible. Interna-
tional Journal of Law and Psychiatry 14:47–64.
6. Zillmann, D. (2000). Influence of unrestrained access to
erotica on adolescents’ and young adults’ dispositions
towards sexuality. Journal of Adolescent Health 27:41–44.
7. Ponton, L.E., & Judice, S. (2004). Typical adolescent
sexual development. Child and Adolescent Psychiatric
Clinics of North America 13:497–511.
8. Berne, J., & Huberman, B. (1999). European ap-
proaches to adolescent sexual behavior and responsi-
bility. Washington, D.C., Advocates for Youth.
9. Becker, J., Stein, R.M. (1991). Is sexual erotica associ-
ated with sexual deviance in adolescent males? Inter-
national Journal of Law and Psychiatry 14:85–95.
10. Lenhart, A., Rainie, L., & Lewis, O. (2001). Teenage
life online: the rise of the instant-message generation
and the Internet’s impact on friendships and family
relations. Pew Internet & American Life Project.
11. Fox, S., Rainie, L., Horrigan, J., et al. (2000). The on-
line healthcare revolution: How the web helps
Americans take better care of themselves. Pew Inter-
net & American Life Project.
12. Stahl, C., & Fritz, N. (2002). Internet safety: adoles-
cents’ self-report. Journal of Adolescent Health 285:
3011–3014.
EXPOSURE TO INTERNET PORNOGRAPHY AMONG CHILDREN 485
14105c08.pgs 9/28/05 12:22 PM Page 485

13. Rideout, V. (2001). Generation Rx.com: how young
people use the Internet for health information. Kaiser
Family Foundation.
14. Hammer, H., Finkelhor, D., & Sedlak, A.J. (2002).
NISMART-2 Household Survey Methodology Tech-
nical Report. Washington, DC, U.S. Department of
Justice, Office of Justice Programs, Office of Juvenile
Justice and Delinquency Prevention.
15. National Public Radio, Kaiser Family Foundation,
and Kennedy School of Government. (2000). Na-
tional survey of American adults on technology and
National survey of American kids on technology.
16. UCLA Center for Communication Policy. (2001). The
UCLA Internet report 2001—surveying the digital future.
17. Mitchell, K., Finkelhor, D., & Wolak, J. (2003). The ex-
posure of youth to unwanted sexual material on the
Internet: a national survey of risk, impact, and pre-
vention. Youth & Society 34:330–358.
18. American Psychiatric Association. (1994). Diagnostic
and statistical manual of mental disorders, 4th ed. Wash-
ington, DC: American Psychiatric Association.
19. StataCorp. (2000). Stata Statistical Software. College
Station, TX: Stata.
20. Kazdin, A.E., & Marciano, P.L. (1998). Childhood and
adolescent depression. In: Mash, E.J., & Barkley,
R.A., (eds), Treatment of childhood disorders, 2nd ed.
New York: The Guilford Press, pp. 211–248.
Address reprint requests to:
Dr. Michele Ybarra
Internet Solutions for Kids, Inc.
74 Ashford
Irvine, CA 92618
E-mail: Michele@isolutions4kids.org
486 YBARRA AND MITCHELL
14105c08.pgs 9/28/05 12:22 PM Page 486