Content uploaded by Marcel Aebi
Author content
All content in this area was uploaded by Marcel Aebi on Jul 15, 2015
Content may be subject to copyright.
ORIGINAL PAPER
Testing the‘‘Sexually Abused-Abuser Hypothesis’’in Adolescents:
A Population-Based Study
Marcel Aebi •Markus A. Landolt •Christoph Mueller-Pfeiffer •Ulrich Schnyder •
Thomas Maier •Meichun Mohler-Kuo
Received: 19 May 2014 /Revised: 18 October 2014 / Accepted: 10 November 2014
ÓSpringer Science+Business Media New York 2015
Abstract A long-standing belief in the literature on sex offend-
ers is that sexually victimized youths are at increased risk of becom-
ing sex offenders themselves. The present study tested the link
between past sexual abuse, either with or without contact, and
sexually offending behavior in a representative sample of male
and female adolescents while controlling for other types of abuse,
mental health problems, substance use, and non-sexual violent
behaviors. Self-reported data were collected from a nationally
representative sample of 6,628 students attending 9th grade
public school in Switzerland (3,434 males, 3,194 females, mean
age =15.50 years, SD =0.66 years). Exposure to contact and
non-contact types of sexual abuse was assessed using the Child
Sexual Abuse Questionnaire and sexually offending behavior
by the presence of any of three behaviors indicating sexual coer-
cion. Two-hundred-forty-five males (7.1 %) and 40 females
(1.2 %) reported having sexually coerced another person. After
controlling for non-sexual abuse, low parent education, urban ver-
sus rural living, mental health problems, substance use, and non-
sexual violent behavior, male adolescents who were victims of
contact sexual abuse and non-contact sexual abuse were signif-
icantly more likely to report coercive sexual behaviors. Females
who experienced contact or non-contact sexual abuse were also
found at increased risk of committing sexual coercion after con-
trolling for covariates. The present findings demonstrate a strong
relationship between past sexual abuse, with and without phys-
ical contact, and sexual-offending behavior in male and female
adolescents. Reducing exposure to non-contact sexual abuse (like
Internet-based sexual exploitation) should become a new area of
sexual violence prevention in youths.
Keywords Sexual victimization Sexual abuse
Sexual coercion Juvenile sex offender
Introduction
According to a recent meta-analysis worldwide prevalence rates
for child sexual abuse (CSA) are 7.9 and 19.2 % among male and
female adolescents in community and student samples (Pereda,
Guilera, Forns, & Gomez-Benito, 2009). There is no consensus
among researchers as to what defines CSA. Some only consider
abuse that involves actual physical contact, whereas others use a
broad range of sexual abusive behaviors with and without phys-
ical contact (Mohler-Kuo et al., 2014). Nowadays, sexual abuse
can also be committed online via computers and other media
devices. When considering non-contact sexual abuse (e.g., sexual
M. Aebi (&)
University Clinics of Child and Adolescent Psychiatry, University of
Zurich, Neptunstrasse 60, 8032 Zurich, Switzerland
e-mail: marcel.aebi@uzh.ch
M. A. Landolt
Department of Psychosomatics and Psychiatry, University Children’s
Hospital Zurich, Zurich, Switzerland
M. A. Landolt
Department of Child and Adolescent Health Psychology, Institute of
Psychology, University of Zurich, Zurich, Switzerland
C. Mueller-Pfeiffer U. Schnyder
Department of Psychiatry and Psychotherapy, University Hospital
Zurich, Zurich, Switzerland
C. Mueller-Pfeiffer T. Maier
Center of Education and Research (COEUR), Psychiatric Services of
the County of St. Gallen-North, Wil, Switzerland
C. Mueller-Pfeiffer
Department of Psychiatry, Massachusetts General Hospital and
Harvard Medical School, Boston, MA, USA
M. Mohler-Kuo
Institute of Social and Preventive Medicine, University of Zurich,
Zurich, Switzerland
123
Arch Sex Behav
DOI 10.1007/s10508-014-0440-x
harassment via the internet) prevalence rates of CSA increase to
17.2 and 40.2 % for male and female youth, respectively (Mohler-
Kuo et al., 2014). There is evidence from clinical and popula-
tion-based studies that, relative to non-abused youths, those who
have been sexually abused (most studies combine contact and non-
contact sexual abuse) are at higher risk for both physical and men-
tal health problems (e.g., Fergusson, Horwood, & Lynskey, 1996;
Irish, Kobayashi, & Delahanty, 2010; Maniglio, 2009) but are
also at increased risk for re-victimization (Bramsen, Lasgaard,
Koss, Elklit, & Banner, 2012). One large population study exam-
ined outcomes of non-sexual abuse, non-contact sexual abuse,
contact sexual abuse without penetration and contact sexual abuse
with penetration in childhood. They found that more severe forms
of sexual abuse were associated with more severe psychopatho-
logical and poorer socioeconomic outcomes in adulthood (Fer-
gusson, McLeod, & Horwood, 2013).
A long-standing belief in the literature on sex offenders is that
victims of child sexual abuse are also more likely to sexually
coerce another person. The link between the experience of sex-
ual victimization and subsequently committed sexual perpetra-
tion is also known as‘‘the sexually abused abuser hypothesis’’
(Seto et al., 2010) or the ‘‘victim-to-victimizer cycle’’ (Glasser
et al., 2001).
Evidence for the sexually abused abuser hypothesis comes
mainly from forensic studies based on adult and adolescent sex
and non-sex offender samples. Two large meta-analyses of adult
and adolescent offenders, respectively, reported a 3.4 and 2.8
times higher prevalence rate of sexual abuse in sex offenders com-
pared to non-sex offenders (Jespersen, Lalumiere, & Seto, 2009;
Seto & Lalumiere, 2010). Additionally, sexual victimization in
childhood was found to be associated with repeated sexual offend-
ing and sexual recidivism in male adolescents (Carpentier &
Proulx, 2011; Dennison & Leclerc, 2011) but not in adults
(Hanson & Morton-Bourgon, 2005).
As female children are more likely to have experienced sex-
ual abuse and only a very small minority of sex offences are com-
mitted by females, sexual abuse seems less specifically related to
sexual coercive behaviors in adolescent girls or females. Never-
theless, some studies also found high prevalence rates for sexual
abuse histories and a longer duration of childhood sexual abuse in
female sex offenders compared to non-sex offenders and male
sex offenders, respectively (Christopher, Lutz-Zois, & Reinhardt,
2007; Van der Put, Van Vugt, Stams, & Hendriks, 2013). In sum,
there is limited evidence for the sexually abused-abuser hypoth-
eses in females.
Further studies analyzed sexual offending prospectively in
child victims of sexual abuse (Ogloff et al., 2012;Widom&
Ames, 1994). For example, Ogloff et al. found that sexually
abused children were almost 8 times more likely to be charged
for a sexual offense in later life compared to non-abused con-
trols. Salter et al. (2003) found that sexual offending often occurs
in the years following the occurrence of abuse.
One limitation of using clinical or criminal justice samples to
study the association between sexual victimization and sexual
perpetration is the lack of generalizability of the findings to the
general population. In fact, only victims who reported sexual
abuse to the authorities and only sex offenders who were charged
or convicted of their sexual offences can be included in clinical
or forensic studies. However, only a small proportion of child
sexual abuse is presumed to be reported to the authorities and only
a part of offenders is presumed to be arrested for their sexual abu-
sive behaviors (Gilbert et al., 2009; Maier, Mohler-Kuo, Landolt,
Schnyder, & Jud, 2013;Weinrott&Saylor,1991). Furthermore,
adolescent sexual offenders are more likely to be reported to the
police if they also had a history of non-sexual violence (Knight
&Prentky,1993). Criminal characteristics are possibly overrep-
resented among participants from studies in clinical or criminal
justice settings and may have influenced the findings. Therefore,
further research in population samples is needed (Seto et al.,
2010).
Few studies have tested the‘‘sexually abused abuser hypoth-
esis’’ in population-based samples. Three U.S. surveys of high-
school students support the link between contact CSA and sex-
ual aggression in adolescents (Borowsky, Hogan, & Ireland,
1997; Casey, Bednell, & Lindhorst, 2009; Lodico, Gruber, &
DiClemente, 1996). Adolescents who reported having been sex-
ually abused alsowere more likely to report sexual aggression
against a friend or dating partner. After controlling for drug use,
emotional problems and family/school variables, a history of CSA
remained a significant predictor of sexual aggr ession, with higher
odds ratios in boys (OR =2.36–2.71) than in girls (OR =1.51–
1.53). More recently, Seto et al. (2010) found similar results in
two male population samples of adolescents in Norway and Swe-
den: males with a history of CSA were more likely to report coer-
cive sexual behaviors, even when controlling for non-coercive
sexual behaviors (e.g., pornography use), substance use, and non-
sexual violent behavior.
Although the results of some population studies do support
the ‘‘sexually abused abuser hypothesis,’’ particularly for male
youths, further studies are needed that take into account risk fac-
tors for sexual coercion, such as exposure to non-sexual abuse
(Van der Put et al., 2013;Widom&Ames,1994), mental health
problems (Galli et al., 1999), substance use (Seto et al., 2010)
and non-sexual violent offending behavior (Aebi, Vogt, Plattner,
Steinhausen, & Bessler, 2012). Considering that the experience
of sexual abuse is associated with a wide range of psychosocial
outcomes (e.g., poor mental health, aggression, drug use) that by
themselves also increase the risk for sexual coercion, the inclu-
sion of additional risk factors becomes crucial for explaining
a more specific relation of sexual abuse and sexual coercive
behaviors.
Previous studies have focused mainly on contact sexual abuse
in childhood or adolescence, but have not considered non-
physical types of sexual abuse as possible predictors of sexual
Arch Sex Behav
123
coercion. Nowadays, adolescents are increasingly engaged in
new media and Internet-based communication platforms; and
with the rise of these new technologies, exposure to online forms
of non-contact sexual abuse has increased (Boonman, Grudzins-
kas, & Aebi, 2014; Mohler-Kuo et al., 2014). To our best knowl-
edge, no previous study tested the link between exposure to non-
contact sexual abuse and sexual offending behaviors. Therefore,
the present study aimed to test the‘‘sexually abused abuser hypoth-
esis’’ in a general population-based sample of adolescents.
We hypothesized that past exposure to either contact or non-
contact sexual abuse would be positively associated with sexual
coercion in both male and female adolescents, even after control-
ling for other risk factors.
Among victims of CSA, a close relationship with the offender
and the experience of severe forms of sexual abuse were found
related to later committed sexual coercion (Burton, Miller, & Tai
Shill, 2002; Christopher et al., 2007;Setoetal.,2010). We addi-
tionally examined whether multiple sexual abuse episodes, oral,
vaginal or anal penetration, abuse before age 12, abuse by a
stranger, or abuse by a family member predicted sexual coercion
in those youths who report contact sexual abuse.
Method
Participants and Procedure
The present study was based on a nationally representative sample
of 9th grade students attending public schools in Switzerland. In
the spring of 2009, a sample of 10,000 pupils from 560 classes at
228 public schools was randomly drawn from the most updated
list of the Swiss Federal Bureau of Statistics. The probability pro-
portional to size (PPS) cluster sampling method was used to
select classes and schools, accounting for school size. Sampling
was stratified according to the seven regions and 26 cantons of
Switzerland. Approval from 26 ethics committees and 26 depart-
ments of education had to be obtained. While 25 out of 26 ethics
committees approved the study, four education departments did
not grant permission, mainly because of other school surveys that
were being done during the same period of time. This resulted in
the loss of 28 schools encompassing 63 classes. In those cantons
that approved the study, the principals of selected schools were
invited by letter to participate in the survey. After being contacted,
23 schools (48 classes) refused to participate, leaving a final sam-
ple of 177 schools with 449 classes. Due to the absence of some
students because of illness (n=537) and students’ individual
refusals (n=63), 6,841 students participated in the survey. Due
to computer-related problems, 15 questionnaires subsequently
were lost and 39 questionnaires excluded because of invalid data.
A further 159 participants were excluded because of missing
data on mental health problem scales (n=122, see below) and/
or sexual coercion (n=37). Therefore, the final sample consisted
of 6,628 participants (Total response rate 89.1 %; 3,434 male
youths, 3,194 female youths) with a mean age of 15.50 years
(SD =0.66 years). Of these youths, 4,931 adolescents (74.4 %)
were of Swiss nationality.
From September 2009 to May 2010, data were anonymously
collected using computer-assisted self-report questionnaires. Two
trained assistants were present during data collection to ensure
an atmosphere of privacy and provide technical support, if nec-
essary. The assistants also were instructed to keep an eye on stu-
dents, in order to detect signs of stress or discomfort. Following
completion of the questionnaire, the students received an infor-
mation sheet with a list of readily accessible institutions that pro-
vided help for victims of trauma. A more detailed description of
study procedures is available in previous publications (Landolt,
Schnyder, Maier, Scho
¨nbucher, & Mohler-Kuo, 2013;Mohler-
Kuo et al., 2014).
All students who were included in the present study provided
informed consent. Because all students wereolder than 14 years,
according to Swiss law, parental consent was not required for par-
ticipation in this anonymous school survey.
Measures
Sexual Coercion
Sexual coercion was assessed using three dichotomous ques-
tions thatcould be answered either‘‘yes’’or‘‘no.’’The questions
were: (1) Have you ever force d someone else to undress himself/
herself or to show himself/herself naked against his/her will? (2)
Have you ever touched someone else’s private parts against their
will? (3) Have you ever urged someone else to do a sexual act,
even though that person did not want to? If any of these three
items was answered‘‘yes,’’ sexual coercion was coded as present.
Child Sexual Abuse
Past exposure to sexual abuse with or without physical contact was
assessed using the Child Sexual Abuse Questionnaire (Mohler-
Kuo et al., 2014). This self-report instrument consists of 15 ques-
tions to assess various forms of CSA. Life-time CSA and CSA over
the preceding 12 months can be assessed separately. The CASQ
was newly developed for the present study by taking into consid-
eration previous literature and instruments. The details of the
CSAQ development has been published elsewhere (Mohler-Kuo
et al., 2014). For the present study, life-time CSA was taken into
account. In addition, we considered CSA before age 12 in the
supplemental analyses. The first eight items of the CSAQ are re-
lated to sexual abuse without physical contact and include items like
‘‘have you been forced to show your private parts,’’‘‘have you been
sexually harassed by another person (either verbally, or by internet
chatting, telephone or text massages), or‘‘has anyone taken pictures
of your naked body against your will.’’ Another seven items are
elated to sexual abuse with physical contact and include items like
‘‘have you ever been touched in a sexual way against your will,’’
Arch Sex Behav
123
‘‘has anyone tried to penetrate you vaginally or anally with his
finger or using any object,’’ and‘‘has anyone tried to penetrate you
vaginally or anally with his penis.’’ Exposure to sexual abuse with
contact was coded as present if ‘‘yes’’ was reported for any of the
seven related items. Sexual abuse without physical contact was
coded in a similar way, but only for those participants who were not
exposed to contact sexual abuse. Internal consistencies (Cronbach
alphas) were 0.7 for contact sexual abuse and 0.6 for non-contact
sexual abuse. As the events of contact and non-contact sexual abuse
may not be closely related to each other, the internal consistencies
of the corresponding CASQ scales were only moderate. Therefore,
we did not use sum scores of the contact and non-contact sexual
abuse in the present study. The agreement on the presence of sexual
abuse between the CASQ (for both, contact and non-contact sexual
abuse) and the sexual victimization score from the Juvenile Vic-
timization Questionnaire (JVQ; see below) was found to be in a
moderate range (Cohen’s kappa 0.41, p\0.001). The CASQ in-
cludes more items of sexual abuse than the JVQ.
Physical Violence and Child Maltreatment
The Juvenile Victimization Questionnaire (JVQ, Hamby, Finkel-
hor, Ormrod, & Turner, 2005) was used to measure past exposure
to physical violence and/or maltreatment. The JVQ asks about
different forms of offense against youths covering five areas: (1)
conventional crime, (2) child maltreatment, (3) peer victimiza-
tion, (4) sexual victimization, and (5) indirect victimization. The
instrument had already demonstrated adequate test–retest re-
liability and construct validity in a U.S. national sample of 10 to
18-year old youths (Finkelhor, Hamby, Ormrod, & Turner, 2005).
For the present study, the physical assault composite score (ten
items, e.g., assaults with or without a weapon, prejudicial attacks,
date violence) and the child maltreatment score (five items, e.g.,
physical abuse, emotional abuse, neglect) were considered. Both
were coded as present if ‘‘yes’’ was responded to any of the re-
lated questions.
Demographic Variables
Parental education was assessed using two separate items mea-
suring paternal and maternal education on 5-point Likert scales
ranging from 1 (no vocational training) to 5 (a university degree).
Paternal and maternal scores were added together to create a sin-
gle summation score. If one parent’s education level was
unknown, the other parent’s score was doubled. If education
level was reported for neither parent, the score was coded as
‘‘ m i s s i n g . ’’ L o w p a r e n t a l e d u c a tion was categorized as ‘‘low’’if
the parental education summation score had a value of two or
three, and average to above-average otherwise. As the 627 (9.5 %)
cases with missing values for parental education did not differ
from those cases with average to above-average parental educa-
tion, with respect to sexual coercion (4.6 % vs. 4.0 %, v
2
=0.57,
p=0.45), missing cases were included in further analyses.
Living area was coded as‘‘urban’’ if the participant was living in
a community with more than 50,000 residents.
Mental Health
Mental health was assessed by means of the self-report version
of the Strengths and Difficulties Questionnaire (SDQ). This mea-
sure of mental health problems in children ages 11–17 is com-
prised of 20 items relating to emotional symptoms, conduct prob-
lems, hyperactivity, and peer problems (Goodman, 1997). Reli-
ability and validity of the instrument was found sufficient (Good-
man, 2001). In the present analyses, scores falling within the
clinical range defined in the manual were considered clinically
relevant.
Violent Behavior/Substance Use
Participants were asked if they had ever committed a robbery or
physically assaulted/attacked another person. If any of these
questions was answered ‘‘yes,’’ violent behavior was coded as
present. Frequency of alcohol consumption was reported on a 6-
point Likert-type scale from ‘‘every day’’to ‘‘less than once per
month’’; two or more times a week was chosen to indicate
‘‘regular alcohol use’’(Seto et al., 2010). Participants also were
asked if they had ever used cannabis or any ‘‘hard’’ drugs (e.g.,
cocaine, amphetamines, ecstasy, heroine, hallucinogens).
Statistical Analysis
Pearson v
2
-tests were used to compare predictor and outcome
variables between the two genders. Odds ratios (OR) were cal-
culated to assess the association between past exposure to con-
tact and non-contact sexual abuse, and engaging in any sexual
coercion behavior, while controlling for other covariates. These
analyses were stratified by gender. Hierarchical logistic regres-
sion analysis was performed, the first model entering exposure
to non-sexual abuse/demographic variables as independent fac-
tors (controlling for other types of adversities associated with
sexual abuse), the second entering mental health problems and
violent behaviors/substance (controlling for other possible out-
comes of sexually abusive experiences)use as binary independent
variables, and the third model entering all the covariates, with
reported coercive behaviors as dependent variable. Multi-
collinearity diagnostics were applied by analyzing variance
inflation factors (VIF; Myers, 1990). None of the predictors/covari-
ates showed multicollinearity. All analyses were performed using
SPSS Version 21 for Windows, were two-tailed, and utilized a
threshold for statistical significance of p=0.05.
Arch Sex Behav
123
Results
Descriptive Findings
Frequencies for all outcomes, predictors and covariates, by gen-
der, are shown in Table 1. In total, 245 (7.1 %) males and 40
(1.3 %) females reported to have sexually coerced another per-
son, either forcing them to undress, touching the other person’s
private parts against their will, or urging them to perform some
sexual act or acts. Notably, all coercive sexual behaviors were
more frequently reported by males. Among these‘‘sexual coerc-
ers,’’ 104 (42.4 %) males and 34 (85.0 %) females reported a
personal history of sexual victimization. In the full sample, rela-
tive to females, males more often reported conduct problems,
violent behaviors, regular alcohol consumption, and either can-
nabis or hard drug use. Conversely, females more often reported
exposure to all types of abuse. Additionally, females also reported
emotional problems more often than males. Out of 173 (5.0 %)
male and 484 (15.2 %) female victims of contact sexual abuse,
41 (23.7 %) and 22 (4.5 %) reported having enacted coercive
sexual behaviors, respectively. Out of 416 (12.1 %) male and
780 (24.4 %) female victims of non-contact sexual abuse, 63
(15.1 %) and 12 (1.5 %) reported coercive sexual behaviors (data
not shown in Table 1).
Prediction of Sexual Coercion in Male Youths
In univariate analyses, males who reported a history of personal
non-contact or sexual contact abuse were more likely to report
having engaged in coercive sexual behavior(s). In addition to
past exposure to physical violence and childhood maltreatment,
the presence of emotional problems, conduct problems, hyperac-
tivity, and personal violent behavior and substance use were all
positively associated with sexual coercion (Table 2). Furthermore,
exposure to non-contact and contact sexual abuse was significantly
associated with sexual coercion after controlling for all covari-
ates. Childhood maltreatment, conduct problems, violent behav-
ior,andharddruguseweresignificant predictors of sexual coer-
cion in the final full model (Table 2, fifth column).
Table 1 Frequencies of outcome, predictor, and covariate variables in male and female youths
Variables Male youths (n=3,434) Female youths (n=3,194) v
2
-test Total sample (N=6628)
Frequencies (%) Frequencies (%) Frequencies (%)
Sexual coercion
Forced someone else to undress 71 (2.1 %) 17 (0.5 %) 29.78*** 88 (1.3 %)
Touched private parts of someone else
against his/her will
202 (5.9 %) 25 (0.8 %) 130.11*** 227 (3.4 %)
Urged someone else to do sexual things 46 (1.3 %) 15 (0.5 %) 13.73*** 61 (0.9 %)
Any committed sexual coercion 245 (7.1%) 40 (1.3 %) 139.14*** 285 (4.3 %)
Exposure to sexual abuse
Exposure to sexual abuse with contact 173 (5.0 %) 484 (15.2 %) 189.62*** 657 (9.9 %)
Exposure to sexual abuse without contact 416 (12.1 %) 780 (24.4 %) 169.48*** 1,196 (18.0 %)
Exposure to nonsexual abuse
Exposure to physical violence 2,259 (65.8 %) 1,859 (58.2 %) 40.42*** 4,118 (62.1 %)
Exposure to child maltreatment 658 (19.2 %) 1,204 (37.7 %) 281.41*** 1,862 (28.1 %)
Demographics
Low parental education 853 (24.8 %) 806 (26.9 %) 3.76 n.s. 1,713 (25.8 %)
Urban versus rural living area 2,511 (73.1 %) 2,411 (75.5 %) 4.84* 4,922 (74.3 %)
Mental health problems
Clinical range of SDQ Emotional problems 82 (2.4 %) 350 (11.0 %) 199.48*** 432 (6.5 %)
Clinical range of SDQ Conduct problems 432 (12.6 %) 228 (7.1 %) 54.65*** 660 (10.0 %)
Clinical range of SDQ Hyperactivity 274 (8.0 %) 269 (8.4 %) 0.43 n.s. 543 (8.2 %)
Clinical range of SDQ Peer problems 142 (4.1 %) 112 (3.5 %) 1.77 n.s. 254 (3.8 %)
Violent behavior/substance use
Violent behavior (nonsexual) 1,853 (54.0 %) 806 (25.2 %) 568.40*** 2,659 (40.1 %)
Alcohol ([2?times p. week) 542 (15.8 %) 249 (7.8 %) 100.46*** 791 (11.9 %)
Ever used cannabis 944 (27.5 %) 552 (17.3 %) 98.66*** 1,496 (22.6 %)
Ever used hard drugs 547 (15.9 %) 384 (12.0 %) 20.92*** 931 (14.0 %)
*p\.05, ** p\.01, *** p\.001
Arch Sex Behav
123
Prediction of Sexual Coercion in Female Youths
Univariate analysis revealed that females who reported personal
victimization with either non-contact or sexual contact abuse
were more likely to report engaging in coercive sexual behav-
ior(s). The presence of child maltreatment also was associated
with sexual coercion, as were all of the SDQ problem scales and
violent behavior/substance use variables (Table 3, second col-
umn). After controlling for covariates, exposure to sexual con-
tact abuse remained significantly associated with sexual coer-
cion. Exposure to non-contact sexual abuse narrowly missed
meeting the a priori level of statistical significance, while the pre-
sence of emotional problems, hyperactivity, violent behavior,
regular alcohol use, and hard drug use remained positive predic-
tors of sexual coercion (Table 3, fifth column). This could be due
to a lack of statistical power because only 40 females (1.3 %)
reported having committed sexual coercion. Therefore, we per-
formed an additional logistic regression with a reduced number
of predictor variables to increase the statistical power of the pre-
dictors by combining covariates (any maltreatment or physical
abuse, any SDQ-problem scale in the clinical range, violent behav-
ior, and any alcohol, cannabis or drug use). In this model (not
shown in Table 3), past exposure to sexual abuse with contact
and past non-contact sexual abuse became statistically signif-
icant predictors of sexual coercion, with OR =8.12 (95 % CI
3.08–21.40) and OR 3.16 (95 % CI 1.13–8.83), respectively.
Additional Analyses of Risk Factors in Youths Who
Reported Sexual Abuse with Contact
Additional analyses in the subsample of victims of sexual abuse
with contact revealed that the presence of sexual abuse perpe-
trated by a stranger was a significant risk factor for sexual coer-
cion in male youths. Being a victim of multiple sexual abuse events
and being a victim of oral, vaginal, or anal penetration were sig-
nificant predictors of sexual coercion in female youths (Table 4).
Discussion
Consistent with previously published findings and our hypothesis,
victims of CSA were more likely than non-victims to report per-
sonally engaging in coercive sexual behaviors against another
person (Borowsky et al., 1997; Casey et al., 2009;Lodicoetal.,
1996;Setoetal.,2010). After controlling for other risk factors, the
OR was almost four (3.8) for adolescent boys exposed to sexual
Table 2 Predictors (covariates) of committed sexual coercion in male youths (N=3434)
Variables Univariate analyses Multivariate model
a
Multivariate model
b
Full model
OR (95 % CI) OR (95 % CI) OR (95 % CI) OR (95 % CI)
Exposure to sexual abuse
Exposure to sexual abuse with contact 5.96 (4.04–8.79)4.95 (3.31–7.39)4.04 (2.67–6.12)3.81 (2.49–5.82)
Exposure to sexual abuse without contact 3.42 (2.49–4.70)2.98 (2.15–4.12)2.48 (1.78–3.46)2.43 (1.73–3.41)
Exposure to non-sexual abuse
Exposure to physical violence 1.66 (1.23–2.34) 1.21 (0.88–1.66) – 0.78 (0.55–1.09)
Exposure to child maltreatment 2.43 (1.84–3.20)1.81 (1.34–2.43)– 1.49 (1.09–2.03)
Demographics
Low parental education 1.26 (0.94–1.67) 1.08 (0.80–1.45) – 1.05 (0.77–1.43)
Urban versus rural community 1.23 (0.91–1.67) 1.21 (0.90–1.68) – 1.20 (0.87–1.66)
Mental health problems
Clinical range of SDQ Emotional problems 2.78 (1.54–5.02) – 1.54 (0.78–3.05) 1.45 (0.73–2.87)
Clinical range of SDQ Conduct problems 3.91 (2.93–5.22)– 1.97 (1.41–2.76)1.91 (1.36–2.68)
Clinical range of SDQ Hyperactivity 2.30 (1.59–3.33) – 1.36 (0.90–2.06) 1.34 (0.89–2.03)
Clinical range of SDQ Peer problems 0.88 (0.44–1.74) – 0.70 (0.33–1.49) 0.68 (0.32–1.44)
Violent behavior/substance use
Violent behavior (nonsexual) 4.25 (3.05–5.93)– 2.83 (1.99–4.01)2.90 (2.02–4.18)
Alcohol ([2?times per week) 2.25 (1.68–3.02) – 1.02 (0.72–1.43) 1.02 (0.72–1.44)
Ever used cannabis 2.43 (1.87–3.16) – 1.35 (0.99–1.85) 1.36 (0.99–1.86)
Ever used hard drugs 3.14 (2.37–4.15)– 1.78 (1.29–2.45)1.79 (1.29–2.47)
CI 95 % confidence interval, OR odds ratios, CI that do not include 1.00 indicate a significant OR at p\0.05;s ignificant OR are in bold,
a
entering variables for
non-sexual abuse and demographics,
b
entering variables for mental health problems and substance use/violent behavior
Arch Sex Behav
123
contact abuse, while an OR of 2.4 was identified for boys exposed
to non-contact sexual abuse. Corresponding numbers for adoles-
cent girls were 5.9 and 2.7. These findings mirror those of a
previous study of Norwegian and Swedish male adolescents that
found sexually abused boys to be approximately three times more
likely to exhibit coercive sexual behaviors (Seto et al., 2010).
However, previous studies were limited to sexual violence in part-
nerships (Casey et al., 2009) and/or primarily focused on male
youths (Seto et al., 2010). By the inclusion ofa female sample
and by considering non-contact sexual abuse as an additional
predictor, the currentstudy adds to our understandingof sexual
coercion committed by juveniles.
Exposure to Sexual Abuse and Sexual Coercion
The current findings were in line with previous studies based on
male adult and adolescent offender samples (Burton, 2003;Jes-
persen et al., 2009; Seto & Lalumiere, 2010) and support the
exi stence of a unique pathway from sexual maltreatment to sexual
coercivebehaviors in youths.Other forms of maltreatment that
often co-occur with sexual abuse (Edwards, Holden, Felitti, &
Anda, 2003) and other frequently reported outcomes of sexual
abuse (Maniglio, 2009; McGrath, Nilsen, & Kerley, 2011)do
not explain the link between sexual victimization and reported
sexual coercive behaviors.
Table 3 Predictors (covariates) of committed sexual coercion in female youths (N=3194)
Variables Univariate analyses Multivariate model
a
Multivariate model
b
Full model
OR (95 % CI) OR (95 % C I) OR (95 % CI) OR (95 % CI)
Exposure to sexual abuse
Exposure to sexual abuse with contact 15.27 (6.16–37.87)11.96 (4.61–31.04)5.99 (2.31–15.53)5.91 (2.18–15.98)
Exposure to sexual abuse without contact 5.01 (1.87–13.40)4.28 (1.56–11.75) 2.45 (0.87–6.88) 2.66 (0.92–7.66)
Exposure to non-sexual abuse
Exposure to physical violence 1.91 (0.95–3.83) 0.96 (0.46–2.00) – 0.49 (0.22–1.09)
Exposure to child maltreatment 3.49 (1.79–6.79) 1.91 (0.94–3.87) – 1.35 (0.64–2.82)
Demographics
Low parental education 1.47 (0.76–2.83) 1.26 (0.65–2.46) – 1.30 (0.64–2.62)
Urban versus rural community 1.12 (0.53–2.36) 1.05 (0.49–2.23) – 0.94 (0.42–2.11)
Mental health problems
Clinical range of SDQ Emotional problems 3.15 (1.56–6.36)– 2.68 (1.22–5.87)2.77 (1.25–6.16)
Clinical range of SDQ Conduct problems 5.13 (2.53–10.42) – 1.12 (0.48–2.61) 1.08 (0.46–2.54)
Clinical range of SDQ Hyperactivity 5.45 (2.78–10.69)– 2.66 (1.25–5.66)2.72 (1.27–5.84)
Clinical range of SDQ Peer problems 4.07 (1.56–10.60) – 2.15 (0.70–6.62) 2.13 (0.68–6.69)
Violent behavior/substance use
Violent behavior (nonsexual) 6.33 (3.25–12.33)– 3.67 (1.78–7.56)4.06 (1.92–8.59)
Alcohol ([2?times per week) 6.69 (3.45–12.98)– 2.45 (1.16–5.20)2.71 (1.26–5.82)
Ever used cannabis 5.46 (2.92–10.23) – 1.61 (0.77–3.40) 1.55 (0.73–3.28)
Ever used hard drugs 8.50 (4.53–15.96)– 3.08 (1.51–6.29)3.33 (1.61–6.87)
CI 95 % confidence interval, OR odds ratios; CI that do not include 1.00 indicate a significant OR at p\0.05;significant OR are in bold,
a
entering variables for
non-sexual abuse and demographics,
b
entering variables for mental health problems and substance use/violent behavior
Table 4 Predictors (covariates) of sexual coercion in a subsample of male and female youths reporting exposure to sexual abuse with contact
Variables (Male youths n=173) (Female youths n=484)
OR (95 % CI) OR (95 % CI)
Multiple episodes of sexual abuse 1.27 (0.60–2.68) 3.62 (1.21–10.87)
Oral, vaginal or anal penetration 2.42 (0.92–6.42) 2.95 (1.20–7.28)
First sexual abuse before 12 years 1.38 (0.60–3.18) 1.25 (0.45–3.48)
Perpetrator was a stranger 2.59 (1.18–5.72) 1.00 (0.29–3.47)
Perpetrator was a family member 1.08 (0.11–10.62) n/c
CI 95 % confidence interval, OR odds ratios; CI that do not include 1.00 indicate a significant OR at p\0.05; significant OR are in bold; n/cnot calculable
Arch Sex Behav
123
In the present study the experience of sexual abuse was found
equally important as a risk factor for sexual coercive behaviors
in female as well as in male adolescents. This finding expands
previous research based on female offender samples which
found mixed results regarding the experience of sexual victim-
ization as a risk factor for sexually coercive behaviors (e.g.,
Christopher et al., 2007; Glasser et al., 2001). Given the high
percentage of sexual abuse among females who reported sexual
coercion in the present study, the understanding of how sexual
abuse affected sexual aggression should become a major issue in
future research of female sex offenders.
Cognitive and emotional processes play a role in victims
becoming sexual perpetrators: According to social learning
theory, victims may‘‘learn’’ inadequate sexualbehaviors from
their perpetrators and reproduce these behaviors against other
children or adolescents (Bandura, 1978).Previous studies sup-
port the existence of social learning processes in sexually vic-
timized male juvenile sex offenders (Burton, 2003). Offence char-
acteristics of experienced sexual abuse were found to mirror
characteristics of sexual offending behaviors regarding victim’s
gender, committed sexual acts, and used modus operandi.
Furthermore, the presence of shame has been linked to anger
arousal, irritability, and aggression (Tangney, Wagner, Fletcher,
&Gramzow,1992). Some highly ashamed victims of sexual
abuse may not be able to express these feelings adequately and
convert their shame to anger. Adolescents who converted shame
(i.e., little expressed sense of shame, high tendency to blaming
others) were found to demonstrate more violent delinquent be-
havior than peers who expressed their sense of shame (Gold,
Sullivan, & Lewis, 2011). Additionally, a sexually traumatized
offender may feel empowered when sexually coercing another
person. Such forms of appetitive aggression have been found to
decrease vulnerability to post-traumatic stress symptoms in
former victims of trauma (e.g., Weierstall et al., 2013). Despite
the strong link between exposure to sexual abuse and sexual
coercion, it is worth mentioning that the majority of self-
reported sexual coercers (particularly males) did not report any
personal experiences with sexual abuse. Hence, other etiologi-
cal factors, like innate hostilities or deviant sexual arousal, may
become relevant in explaining sexual aggression in non-abused
offenders (Hunter & Becker, 1994).
Sex Specific Risk Factors for Sexual Coercion
Having been maltreated, having a conduct disorder, having
shown violent non-sexual behaviors, and using either alcohol or
illicit drugs also were found to be related to sexual coercion.
Behavioral and emotional difficulties, as well as substance use,
have frequently been described in victims of maltreatment and
CSA (Maniglio, 2009). In some of these adolescents, the com-
mitted sex-coercing behavior most likely represents part of a
more comprehensive pattern of antisocial behaviors (Aebi et al.,
2012). Disinhibitory effects of alcohol and drug use also may
lower thresholds for criminal behaviors, including sexual vio-
lence (Martin, 2001).
Although we found common risk factors to be of relevance, the
threshold to commit sexual offenses appeared to be much higher
for girls than for boys. Remarkably in our study, hyperactivity and
emotional problems were predictive of sexual coercion in girls,
but not in boys. Delinquent girls had more and a broader spectrum
of psychiatric disorders and comorbidities than delinquent boys
(e.g., Plattner et al., 2007). The present findings support the as-
sumption that a very small group of sexually abused girls with
severe psychiatricproblems—such as reduced impulsecontrol
and depressed mood—are at increased risk of coercive sexual
behaviors.
Sexual Non-contact Abuse and Sexual Coercion
Previous studies on CSA have primarily focused on victimiza-
tion via sexual contact, with the perpetrator an adult and the
victim younger than 18 years; but they generally have not con-
sidered non-contact forms of sexual abuse (Freyd et al., 2005;
Seto et al., 2010). The CASQ (Mohler-Kuo et al., 2014) includes
types of sexual abuse related to the Internet and new media (e.g.,
online sexual harassment) but is not limited to these forms of
non-contact abuse. With the development of new media tech-
nology, the use of the internet and smart phones is currentlyvery
prevalent among adolescents. In Switzerland, where the current
study took place, 95 % of youths have a mobile phone and 97%
their own Internet access (Willemse, Waller, Su
¨ss, Genner, &
Huber, 2012). Chatting about sexual topics and exchanging porno-
graphic pictures have become readily feasible via social media
platforms. These technologies facilitate new forms of sexual
offense (Short, Black, Smith, Wetterneck, & Wells, 2012). There
is some evidence that juveniles’ sexual attitudes and behaviors
may be influenced to a higher degree by Internet content than by
other sources (Lo & Wei, 2005). In 10–15year old children,
besides exposure to Internet pornography, the potential for
sexual harassment via the Internet and text messaging was foun d
to be related to current and later aggressive sexual behavior
(Ybarra, Mitchell, Hamburger, Diener-West, & Leaf, 2011).
Our findings support the importance of non-contact sexual
victimization as an additional risk factor for coercive sexual
behaviors in adolescents.
Further Risk Factors in Victims of Contact Abuse
Of those adolescents who reported sexual abuse with contact,
only a minority reported engaging in sexual coercion them-
selves. Male adolescents whose abuse was perpetrated by a
stranger were at higher risk of engaging in coercive sexual
behaviors. In contrast, in a previous study among Norwegian
and Swedish adolescents, boys who were sexually abused by a
strangerwere less likely to reportsexual aggression (Setoet al.,
2010). In addition, a study of adolescent offenders found boys
Arch Sex Behav
123
who had been sexually abused over several years, boys who
were penetrated and boys who were related to the offender at
higherrisk to be in the sex offenderthan in the non-sexoffender
group (Burton et al., 2002). These disparate findings may be
explained by different ways of measuring experiences of sexual
abuse and different sample characteristics. For example,in the
current study, a broader spectrum of abuses was assessed than in
the study reported by Seto et al. (2010).
We found female adolescents who reported multiple victim-
izations and the experience of oral, anal, or vaginal penetration to
be at increased risk of themselves engaging in coercive sexual
behaviors. This finding supports an additive model, in which the
frequency and severity of experienced victimization influences
the probability of sexual coercion.
Study Strengths and Limitations
The present study is based on an anonymous survey of a large,
nationally representative sample of youths attending schools
within a well-developed, high-income country. The large num-
ber of participants allowed for extensive multivariate testing of
multiple risk factors with sufficient statistical power. However,
our survey was cross-sectional, so that causal relationships be-
tween life-time CSA and subsequent sexual coercion cannot be
established. In fact, we cannot rule out the possibility that some
adolescents may have engaged in coercive sexual behaviors
before being sexually victimized. Furthermore, the present
findings are based solely upon anonymous self-reports and are
subject to possible recall bias and potential confounding related
to the relative willingness of participants to report CSA and
coercive behaviors; for example, could ou r adolescent girls have
been more or less willing than their male counterparts to report
such activities? In addition, sexual behaviors—like the viewing
of pornography, age of first intercourse, and number of sexual
partners—were not assessed in the present survey and, therefore,
were not considered as covariates. Finally, the present findings
may not be generalizable to adolescents at high risk for sexual
abuse, as antisocial and sexually deviant youths were probably
under-reported in this school sample.
Conclusions
The present findings highlight the extreme importance of CSA
prevention in children and adolescents, given that victims them-
selves may become perpetrators, maintaining an endless cycle.
Consequently, a comprehensive assessment of several risk factors
in children and adolescents seems necessary to estimate the po-
tential for future or ongoing sexual coercion. It should be taken into
account that the use of modern media technologies is a dominant
and influential activity in youths. Clinicians may use instruments
that also target online sexual experiences and victimization in
youths (Boonman et al., 2014). Specific interventions targeting
children with victimization experiences and/or deviant sexual
behaviors might ultimately improve mental health and reduce
sexual aggression among adolescents, both short-term and well
into adulthood (Gibson & Leitenberg, 2000).
Acknowledgments The authors are gratefulto all the students who par-
ticipated in the study. We also thank Ursula Meidert from the Institute of
Social and Preventive Medicine, University of Zurich, who helped with data
collection. The data presented in this article were assessed in the context of
the Optimus Study. The Optimus Study was initiated and funded by the
UBS OPTIMUS FOUNDATION (http://www.optimusstudy.org).
Conflictofinterests All authors declare that they have no conflicts of
interest regarding this manuscript.
References
Aebi, M., Vogt, G., Plattner, B., Steinhausen, H. C., & Bessler, C. (2012).
Offender types and criminality dimensions in male juveniles con-
victed of sexual offenses. Sexual Abuse: A Journal of Research and
Treatment, 24(3), 265–288. doi:10.1177/1079063211420449.
Bandura, A. (1978). Social learning theory of aggression. Journal of
Communication, 28(3), 12–29.
Boonman, C., Grudzinskas, A. J., & Aebi, M. (2014). Juveniles, the Internet,
and sexual offending. In F. M. Saleh, A. J. Grudzinskas, & A. Judge
(Eds.), Adolescent sexual behavior in the digital age. Considerations for
clinicians, legal professionals and educators (pp. 161–179). Oxford,
UK: Oxford University Press.
Borowsky, I. W., Hogan, M., & Ireland, M. (1997). Adolescent sexual
aggression: Risk and protective factors. Pediatrics, 100(6), E7. doi:10.
1542/peds.100.6.e7.
Bramsen, R. H., Lasgaard, M., Koss, M. P., Elklit, A., & Banner, J. (2012).
Adolescent sexual victimization: A prospective study on risk factors
for first time sexual assault. European Child and Adolescent Psy-
chiatry, 21(9), 521–526. doi:10.1007/s00787-012-0290-9.
Burton, D. L. (2003). Male adolescents: Sexual victimization and subse-
quent sexual abuse. Child and Adolescent Social Work Journal, 20(4),
277–295. doi:10.1023/A:1024556909087.
Burton, D. L., Miller, D.L., & Tai Shill, C.(2002). A sociallearning theory
comparisonof the sexualvictimizationofadolescentsexual offenders
and nonsexualoffending maledelinquents. ChildAbuse and Neglect,
26, 893–907. doi:10.1016/S0145-2134(02)00360-5.
Carpentier, J., & Proulx, J. (2011). Correlates of recidivism among ado-
lescents who have sexually offended. Sexual Abuse: A Journal of
Researchand Treatment, 23(4), 434–455.doi:10.1177/1079063211
409950.
Casey, E. A., Bednell, B., & Lindhorst, T. P. (2009). Predictors of sexually
coercive behavior in a nationally representative sample of adolescents.
Journal of Interpersonal Violence, 24(7), 1129–1147. doi:10.1177/
0886260508322198.
Christopher, K., Lutz-Zois, C. J., & Reinhardt, A. R. (2007). Female sexual-
offenders: Personality pathology as a mediator of the relationship
between childhood sexual abuse history and sexual abuse perpetration
against others. Child Abuse and Neglect, 31(8), 871–883. doi:10.1016/
j.chiabu.2007.02.006.
Dennison, S., & Leclerc, B. (2011). Developmental factors in adolescent
child sexualoffenders: A comparisonof nonrepeat and repeatsexual
offenders. Criminal Justice and Behavior, 38, 1089–1102. doi:10.
1177/0093854811417076.
Edwards, V. J., Holden, G. W., Felitti, V. J., & Anda, R. F. (2003). Relationship
between multiple forms of childhood maltreatment and adult mental
health in community respondents: Results from the adverse childhood
Arch Sex Behav
123
experiences study. American Journal of Psychiatry, 160(8), 1453–1460.
doi:10.1176/appi.ajp.160.8.1453.
Fergusson, D. M., Horwood, L. J., & Lynskey, M. T. (1996). Childhood
sexualabuse and psychiatricdisorder in youngadulthood: II. Psychi-
atric outcomes of childhood sexual abuse. Journal of the American
Academy of Child and Adolescent Psychiatry, 35(10), 1365–1374.
doi:10.1097/00004583-199610000-00024.
Fergusson, D. M., McLeod, G. F., & Horwood, L. J. (2013). Childhood
sexualabuse and adultdevelopmentaloutcomes: Findingsfrom a 30-
year longitudinal study in New Zealand. Child Abuse and Neglect,
37(9), 664–674. doi:10.1016/j.chiabu.2013.03.013.
Finkelhor, D., Hamby, S. L., Ormrod, R., & Turner, H. (2005). The Juvenile
Victimization Questionnaire: Reliability, validity, and national norms.
Child Abuse and Neglect, 29(4), 383–412. doi:10.1016/j.chiabu.2004.
11.001.
Freyd, J. J., Putnam, F. W., Lyon, T. D., Becker-Blease, K. A., Cheit, R. E.,
Siegel, N. B., & Pezdek, K. (2005). Psychology: The science of child
sexual abuse. Science, 308(5721), 501. doi:10.1126/science.1108066.
Galli,V.,McElroy,S.L.,Soutullo,C.A.,Kizer,D.,Raute,N.,Keck,P.E,Jr.,
& McConville, B. J. (1999). The psychiatric diagnoses of twenty-two
adolescents who have sexually molested other children. Comprehen-
sive Psychiatry, 40(2), 85–88. doi:10.1016/S0010-440X(99)90110-4.
Gibson, L. E., & Leitenberg, H. (2000). Child sexual abuse prevention
programs: Do they decrease the occurrence of child sexual abuse?
Child Abuse and Neglect, 24(9), 1115–1125. doi:10.1016/S0145-
2134(00)00179-4.
Gilbert, R., Kemp, A., Thoburn, J., Sidebotham, P., Radford, L., Glaser, D.,
& Macmillan, H. L. (2009). Recognising and responding to child mal-
treatment. Lancet, 373(9658), 167–180. doi:10.1016/S0140-6736(08)
61707-9.
Glasser, M., Kolvin, I., Campbell, D., Glasser, A., Leitch, I., & Farrelly, S.
(2001). Cycle of child sexual abuse: Links between being a victim and
becoming a perpetrator. British Journal of Psychiatry, 179, 482–494.
doi:10.1192/bjp.179.6.482.
Gold, J., Sullivan, M. W., & Lewis, M. (2011). The relation between abuse
and violent delinquency: The conversionof shame to blame in juvenile
offenders. Child Abuse and Neglect, 35(7), 459–467. doi:10.1016/j.
chiabu.2011.02.007.
Goodman, R. (1997). The Strengths and Difficulties Questionnaire: A
research no te. Journal of Child Psychology and Psychiatry, 38(5),
581–586. doi:10.1111/j.1469-7610.1997.tb01545.x.
Goodman, R. (2001). Psychometric properties of the strengths and
difficulties questionnaire. Journal of the American Academy of Child
and Adolescent Psychiatry, 40(11), 1337–1345. doi:10.1097/0000
4583-200111000-00015.
Hamby, S., Finkelhor, D., Ormrod, R., & Turner, H. (2005). The Juvenile
Victimization Questionnaire (JVQ): Administration and scoring manual.
Durham, NH: Crimes Against Children Research Center.
Hanson, R. K., & Morton-Bourgon, K. E. (2005). The characteristics of
persistent sexual offenders: A meta-analysis of recidivism studies.
Journal of Consulting and Clinical Psychology, 73(6), 1154–1163.
doi:10.1037/0022-006X.73.6.1154.
Hunter, J. A., & Becker, J. V. (1994). The role of deviant sexual arrousal in
juvenile sexual offending. Criminal Justice and Behavior, 21(1), 132–
149. doi:10.1177/0093854894021001009.
Irish, L., Kobayashi, I., & Delahanty,D. L. (2010). Long-term physical health
consequences of childhood sexualabuse: A meta-analytic review. Jour-
nal of Pediatric Psychology, 35(5), 450–461. doi:10.1093/jpepsy/jsp
118.
Jespersen, A. F., Lalumiere, M. L., & Seto, M. C. (2009). Sexual abuse
history among adult sex offenders and non-sex offenders: A meta-
analysis. Child Abuse and Neglect, 33(3), 179–192. doi:10.1016/j.chia
bu.2008.07.004.
Knight, R. A., & Prentky, R. A. (1993). Exploring characteristics for
classifyingjuvenilesex offenders. In H. E. Barbaree,W. L. Marshall,
& S. M. Hudson (Eds.), The juvenile sex offender (pp. 45–83). New
York, NY: Guilford.
Landolt,M. A., Schnyder,U., Maier, T.,Scho
¨nbucher,V., & Mohler-Kuo,
M. (2013). Trauma exposure and posttraumatic stress disorder in
adolescents: A national surveyin Switzerland. Journal of Traumatic
Stress, 26(2), 209–216. doi:10.1002/jts.21794.
Lo,V., & Wei, R. (2005). Exposureto internetpornographyand Taiwanese
adolescents’ sexual attitudes and behavior. Journal of Broadcasting
and Electronic Media, 49, 221–237. doi:10.1207/s15506878jobem
4902_5.
Lodico, M. A., Gruber, E., & DiClemente, R. J. (1996). Childhood sexual
abuse and coercive sex among school-based adolescentsin a midwest-
ern state. Journal of Adolescent Health, 18(3), 211–217. doi:10.1016/
1054-139X(95)00167-Q.
Maier, T., Mohler-Kuo, M., Landolt, M. A., Schnyder, U., & Jud, A. (2013).
The tip of the iceberg. Incidence of disclosed cases of child sexual
abuse in Switzerland: Results from a nationwide agency survey. Inter-
national Journal of Public Health, 58(6), 875–883. doi:10.1007/
s00038-013-0498-6.
Maniglio, R. (2009). The impact of child sexual ab use on health: A systematic
review of reviews. Clinical Psychology Review, 29(7), 647–657. doi:
10.1016/j.cpr.2009.08.003.
Martin, S. E. (2001). The links between alcohol, crime and the criminal
justice system: Explanations, evidence and interventions. American
Journalon Addictions,10(2), 136–158. doi:10.1080/105504901750
227796.
McGrath, S. A., Nilsen, A. A., & Kerley, K. R. (2011). Sexual victim-
ization in childhood and the propensity for juvenile delinquency and
adult criminal behavior: A systematic review. Aggression and Violent
Behavior, 16, 485–492. doi:10.1016/j.avb.2011.03.008.
Mohler-Kuo, M., Landolt, M. A., Maier, T., Meidert, U., Scho
¨nbucher, V., &
Schnyder, U. (2014). Child sexual abuse revisited: A population-based
cross-sectional study among Swiss adolescents. Journal of Adolescent
Health, 54(3), 304–311. doi:10.1016/j.jadohealth.2013.08.020.
Myers, R. (1990). Classical and modern regression with applications (2nd
ed.). Boston MA: Duxbury.
Ogloff, J. R. P., Cutajar, M. C., Mann, E., Mullen, P. E., Wei, F. T. Y.,
Hassan, H. A. B., & Yih, T. H. (2012). Child sexual abuse and
subsequent offending and victimisation: A 45year follow-up study.
Trends and Issues in Crime and Criminal Justice, 440,1–6.
Pereda,N.,Guilera,G.,Forns,M.,&Gomez-Benito,J.(2009).The
prevalence of child sexual abuse in community and student samples: A
meta-analysis. Clinical Psychology Review, 29(4), 328–338. doi:10.
1016/j.cpr.2009.02.007.
Plattner, B., Kraemer, H. C., Williams, R. P., Bauer, S. M., Kindler, J.,
Feucht, M., & Steiner, H. (2007). Suicidality, psychopathology,and
gender in incarcerated adolescents in Austria. Journal of Clinical
Psychiatry, 68(10), 1593–1600. doi:10.4088/JCP.v68n1019.
Salter, D., McMillan, D., Richards, M., Talbot, T., Hodges, J., Bentovim,
A., & Skuse, D. (2003). Development of sexually abusive behaviour in
sexually victimised males: A longitudinal study. Lancet, 361(9356),
471–476. doi:10.1016/S0140-6736(03)12466-X.
Seto,M.C.,Kjellgren,C.,Priebe,G.,Mossige,S.,Svedin,C.G.,&
Langstrom, N. (2010). Sexual coercion experience and sexually co-
ercive behavior: A population study of Swedish and Norwegian male
youth. Child Maltreatment, 15(3), 219–228. doi:10.1177/
1077559510367937.
Seto,M.C.,&Lalumiere,M.L.(2010). What is so special about male
adolescent sexual offending?A review and test of explanations through
meta-analysis. Psychological Bulletin, 136(4), 526–575. doi:10.1037/
a0019700.
Short, M. B., Black, L., Smith, A. H., Wetterneck, C. T., & Wells, D. E.
(2012).A review of internetpornography use research: Methodology
and content from the past 10 years. Cyberpsychology,Behavior, and
Social Networking, 15(1), 13–23. doi:10.1089/cyber.2010.0477.
Arch Sex Behav
123
Tangney, J. P., Wagner, P., Fletcher, C., & Gramzow, R. (1992). Shamed
into anger? The relation of shame and guilt to anger and self-reported
aggression. Journal of Personality and Social Psychology, 62(4), 669–
675. doi:10.1037/0022-3514.62.4.669.
Van der Put, C., Van Vugt, E., Stams, G. J., & Hendriks, J. (2013). Psy-
chosocial and developmental characterist ics of female adolescents who
have committed sexual offenses. Sexual Abuse: A Journal of Research
and Treatment, Epup ahead of print. doi:10.1177/1079063213492342
Weierstall, R., Hinsberger, M., Kaminer, D., Holzhausen, L., Madikane, S.,
& Elbert, T. (2013). Appetitive aggression and adaption to a violent
environment among youth offenders. Peace and Condlict: Journal of
Peace Psychology, 19(2), 138–149. doi:10.1037/a0032489.
Weinrott, M. R., & Saylor, M. (1991). Self-reportof crimes committed by sex
offenders. Journal of Interpersonal Violence, 6(3), 286–300. doi:10.
1177/088626091006003002.
Widom, C. P., & Ames, M. A. (1994). Criminal consequences of childhood
sexual victimization. Child Abuse and Neglect, 18(4), 303–318. doi:10.
1016/0145-2134(94)90033-7.
Willemse, I., Waller, G., Su
¨ss,D.,Genner,S.,&Huber,A.-L.(2012).
JAMES—Jugend, Aktivita¨ ten, Medien—Erhebung Schweiz. Retrieved
February 20, 2014, 2014, from www.psyxhologie.zhaw.ch/JAMES
Ybarra, M. L., Mitchell, K. J., Hamburger,M., Diener-West, M., & Leaf,
P. J. (2011).X-rated materialand perpetrationof sexually aggressive
behavioramong childrenand adolescents:Is there a link? Aggressive
Behavior, 37(1), 1–18. doi:10.1002/ab.20367.
Arch Sex Behav
123
