Youths Who Sexually Harm, A Multivariate Model of Behaviour

This article was downloaded by:[HEAL- Link Consortium]
On: 11 December 2007

Access Details: [subscription number 772811123]
Publisher: Routledge
Informa Ltd Registered in England and Wales Registered Number: 1072954
Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
Journal of Sexual Aggression

An international, interdisciplinary forum for
research, theory and practice
Publication details, including instructions for authors and subscription information:
http://www.informaworld.com/smpp/title~content=t713723837
Youths who sexually harm: A multivariate model of
behaviour
Louise Almond a; David Canter a
a
Centre for Investigative Psychology, School of Psychology, University of Liverpool,
Liverpool, UK
Online Publication Date: 01 November 2007

To cite this Article: Almond, Louise and Canter, David (2007) 'Youths who sexually
harm: A multivariate model of behaviour', Journal of Sexual Aggression, 13:3, 217 -
233
To link to this article: DOI: 10.1080/13552600701788608
URL: http://dx.doi.org/10.1080/13552600701788608
PLEASE SCROLL DOWN FOR ARTICLE
Full terms and conditions of use: http://www.informaworld.com/terms-and-conditions-of-access.pdf

This article maybe used for research, teaching and private study purposes. Any substantial or systematic reproduction,
re-distribution, re-selling, loan or sub-licensing, systematic supply or distribution in any form to anyone is expressly
forbidden.
The publisher does not give any warranty express or implied or make any representation that the contents will be
complete or accurate or up to date. The accuracy of any instructions, formulae and drug doses should be
independently verified with primary sources. The publisher shall not be liable for any loss, actions, claims, proceedings,
demand or costs or damages whatsoever or howsoever caused arising directly or indirectly in connection with or
arising out of the use of this material.
Journal of Sexual Aggression
(November 2007), Vol. 13, No. 3, pp. 217233
Downloaded By: [HEAL- Link Consortium] At: 08:09 11 December 2007
Youths who sexually harm: A multivariate

model of behaviour
Louise Almond* & David Canter

Centre for Investigative Psychology, School of Psychology, University of Liverpool, Liverpool, UK
Abstract This study investigates the variations in behaviour displayed by young people who
sexually harm, as previous research has shown that they are not a homogeneous sample. Three
conceptually distinct sets of behaviour were hypothesized, relating to various modes of interaction
between the young people with harmful sexual behaviour (HSB) and their victim, victim as object,
victim as person and victim as vehicle. Thirty-three behaviours were drawn from an extensive review
of the files of 300 youths who had sexually harmed. The pattern of co-occurrence of the presence or
absence of these behaviours was revealed by a multidimensional scaling procedure, Smallest Space
Analysis (SSA). The results provide empirical support for three distinct behavioural themes; 258
(86%) of the youths could be assigned to one dominant mode of interaction (victim as object, victim as
person or victim as vehicle). The findings have implications for the ways in which practitioners
differentiate between young people with HSB.
Keywords Adolescent; sexually harmful behaviour; Smallest Space Analysis
Introduction
Classifying all young people who sexually harm as ‘‘juvenile sex offenders’’ suggests that they
are a homogeneous and coherent group, yet research shows that these young people are highly
heterogeneous (Beckett, 1999). However, compared with adult abusers there is a paucity of
research regarding the behaviours of these young people. Epps (1999) states that information
about offence behaviour is frequently neglected during the process of managing and treating
sexually abusive children. He goes on to argue that collecting information about the types of
behaviour perpetrated against the victim may contribute to the estimation of risk and the
development of strategies to reduce risk, by identifying its consistent features.
Three major British descriptive studies address the issue of harmful sexual behaviour
(HSB) behaviour in young people. Manocha and Mezey (1998) describe 51 adolescents
referred to a specialist facility; Dolan, Holloway, Bailey and Kroll (1996) focus on 121 males
aged less than 18 years referred to adolescent forensic units; and, Richardson, Graham, Bhate
and Kelly (1995) describe 100 males aged 1118 from a variety of sources. Table I
summarizes the key characteristics from these three studies as they relate to behaviours
*Corresponding author: Dr Louise Almond, Centre for Investigative Psychology, School of Psychology, University
of Liverpool, Eleanor Rathbone Building, Bedford Street South, Liverpool L69 7ZA, UK. Tel: 0151 7941138. E-mail:
lalmond@liverpool.ac.uk
ISSN 1355-2600 print/1742-6545 online # 2007 National Organisation for the Treatment of Abusers
DOI: 10.1080/13552600701788608
218 L. Almond & D. Canter
Table I. Behaviours displayed during sexually abusive incidents committed by young people, in British descriptive
studies.
Manocha & Mezey Richardson

Behaviour (1998)% Dolan et al. (1996)% et al. (1995)%
Nature of abusive behaviour

Penetrative acts 53.6 66
Non-penetrative acts 30.4 10 23
Non-contact acts 7.1 11 1
Vaginal penetration with penis 17.9 4
Vaginal penetration with finger 10.7 7
Attempted vaginal penetration 8.9 10
Anal penetration with penis 12.5 5
Anal penetration with finger 1.8
Touching victim’s genitals 18 44
Oralgenital contact 7.1 40
Exhibitionism 5.4 5 28
Co-perpetrator 9.8 7
Use of coercion
Physical force used 66.7 16.5 68
Verbal threats used 54.9 16.5 39
Emotional threats 33.3 21
Use or threaten to use weapon 15.7 9
Inducement 23.5 2.5 11
exhibited by young people during the sexually abusive incident. As Table I shows, young
people with HSB carry out the full range of sexually abusive behaviour from exhibitionism to
penile penetration. However, the exact frequencies from each study differ. Such differences
may result from a range of factors, including the agency contexts in which each sample was
drawn with, for example, data from community-based agencies reflecting less serious abusive
incidents than data from specialist secure units.
As Table I shows, each of the behaviours report only medium to low frequencies. This
indicates that young people with HSB are a heterogeneous group who display different
behaviours during the abusive incident. This suggests that there may be particular subgroups
of individuals who differ both in terms of the abusive behaviour and the methods of coercion
they display. Therefore, by examining the behaviours of these young people it may be possible
to distinguish meaningfully between individuals.
Current conceptualizations of the differences between young people with HSB often
utilize victim age as a method of differentiation. This is also common in adult sex offender
research. Research has shown that youths who sexually assault peers and adults differ from
those that target children. Hunter, Hazelwood and Slesinger (2000) conducted a criminal
records review of 126 juveniles arrested for a sexual offence. They found that youths who
target older victims offend principally against females who are generally strangers or
acquaintances. They more often abused in a group context and were also more likely to
commit the crime in conjunction with other criminal activity. In general, these peer/adult
offenders also displayed higher levels of aggression and violence and were more likely to use a
weapon. Hunter et al., however, did not investigate how these two subgroups of youths differ
in terms of the sexual behaviour they displayed.
Richardson, Kelly, Graham and Bhate (1997) also used victim age as a method of
differentiation, adding into this comparison on the basis of the victim’s relationship with the
young person. Richardson et al. (1997) therefore divided their sample of 100 youths into four
Youths who sexually harm 219
discrete subgroups: (1) the incest group, (2) the child group, (3) the peer/adult group and (4)
the mixed group. They found that all the groups committed a wide range of sexually abusive
behaviour; however, the nature of the sexually abusive behaviour was not found to be
discriminate between the four groups. The mixed and incest groups, however, had
perpetrated more abusive acts and had longer abusive careers. Young people in the incest
subgroup were also more likely to have offered inducements in the commission of their abuse,
whereas youths in the peer/adult subgroup tended to use or threaten to use a weapon.
Richardson et al.’s method of classification is, however, questionable, as they utilize victim age
in some subgroups and the victim relationship in others.
Although grouping on the basis of victim selection is common practice in adult sex
offender research (Knight & Prentky, 1990; Richardson et al., 1997), it has been criticized as a
method of differentiating young people with HSB by the National Organization for the
Treatment of Abusers (NOTA) who, in a position paper, stated:
Frequently adolescents who have sexually offended have not targeted a specific group of
victims, but, perhaps, one individual because of that person’s availability (NOTA,
2001, p. 3).
Researchers also differ in their definition of what constitutes a child victim. Some argue that it
is two years difference between the victim and the youth, others argue that it is five years. This
results in an age discrepancy between studies, which may be misleading. A problem also arises
with this method when youths offend against more than one type of victim, therefore requiring
the creation of a somewhat ambiguous mixed group. More importantly, however, few studies
utilize using victim characteristics as a method of differentiating between youths find any
significant differences in the sexually abusive behaviour being displayed.
Researchers have also classified young people with HSB in terms of their personality
characteristics. Smith, Monastersky and Deisher (1987) investigated a sample of 262 youths
and using cluster analysis they derived a typology based on the Minnesota Multiphasic
Personality Inventory (MMPI). They identified four distinct groups which they labelled (1)
immature, (2) personality disordered, (3) socialized delinquents and (4) conduct-disordered.
From a sample of 112 youths, Worling (2001) also used cluster analysis to derive a
personality-based typology very similar to Smith et al.’s (1987) groups. Recently, Richardson,
Kelly, Graham and Bhate (2004) produced a personality-based taxonomy of a sample of 112
youths, using cluster analysis and the Millon Adolescent Clinical Inventory (MACI). They
argue that there are five subgroups of young person with HSB: (1) normal, (2) antisocial, (3)
submissive, (4) dysthymic/inhibited and (5) dysthymic/negativistic.
Further statistical analysis carried out by Smith et al. (1987), Worling (2001) and
Richardson et al. (2004) found that there were no significant differences in the distribution of
personality clusters within the different offence types in terms of the victim age, gender or
relationship with the youth. Worling also found that there were no differences between the
personality clusters with regard to sexual recidivism. Therefore, these studies have failed to
find a link between offence and personality type.
Current methods often take a single one-dimensional approach to differentiating young
people with HSB. They utilize one factor, such as victim age, as a method of identifying
particular offence types; they then investigate the differences between these identified
subgroups. It can be argued that this single dimension approach has only limited value
when considering an issue as complex as human behaviour. Each youth obviously has a
different combination of variables and, depending on the factor being utilized, will be placed
in different subgroups. Howarth (1999) argues that what is needed is a broad macro-level,
multifaceted approach capable of responding to the diversity, as well as the complexity, of this
phenomenon.
In the adult literature, researchers have argued that variations between different types of
sexual offender are reflected in differences in offence behaviour. In particular, these models
argue that sexual assault involves interpersonal behaviour, with the variations in offending
behaviour indicating differences in the roles that the victims have for their offenders. It is
argued that these styles of offendervictim interaction vary from each other in terms of
interpersonal distance, and they represent underlying themes in offence behaviour (Canter,
Hughes & Kirby, 1998).
Canter (1994) analysed 105 cases involving rape in order to establish whether there was a
coherent pattern to the behaviour displayed by perpetrators. Canter’s 1994 model identified
three distinct behavioural themes that related to the role the victim plays for the offender. The
‘‘victim as object’’ theme refers to perpetrators’ lack of empathy for their victims. These
perpetrators view their victims as objects to be exploited for their own personal gain, not only
sexual but also instrumental, by using the opportunity to steal items from the victim. The
‘‘victim as person’’ theme refers to perpetrators who view their victims as significant
individuals. These perpetrators attempt to develop a pseudo-intimate relationship with their
victim by treating them as a reactive individual rather than a sexual object. For some of these
perpetrators, the desire for social contact is the primary purpose for the assault. The third
behavioural theme, according to Canter, is that of ‘‘victim as vehicle’’, which refers to
perpetrators viewing victims as a representation or surrogate for some generalized other,
rather than being targeted as a specific person or treated as an object. These perpetrators use
their victims as a vehicle for venting their anger and frustration. These incidents are
characterized by verbal and physical violence and often involve humiliation and demeaning of
the victim.
Canter (1994) argued that rape is essentially a crime of violation; the victim experiences a
range of violations from the highly intrusive penetrative sexual violation to personal
humiliation and physical assault. He went on to show that a pattern of frequencies exists
among the behaviours that occur during sexual assaults, which correspond to this scale of
violation. In general, Canter found that sexually violating behaviours were high-frequency
variables, and personally violating behaviours were low-frequency variables with physically
violating variables represented in between.
Canter’s (1994) model has since been refined by Canter, Bennell, Alison and Reddy
(2003). The refined model has four behavioural themes: ‘‘hostility’’, which relates to Canter’s
original ‘‘victim as vehicle’’ theme and ‘‘involvement’’, which relates to the ‘‘victim as person’’
theme. The ‘‘victim as object’’ theme was differentiated, in this more recent model, into two
themes: ‘‘control’’, which refers to the perpetrator exerting power over the victim by
demobilizing, them and ‘‘theft’’, which refers to the perpetrator using the opportunity for
instrumental gain. As in Canter’s (1994) model, Canter et al. (2003) suggested that a pattern
of frequencies exists among the variables, which corresponded to the various levels of violation
(sexual, physical and personal).
Analysing 97 cases involving a child victim, Canter et al. (1998) developed a multivariate
model of offence behaviour in child sexual abuse. The findings indicated that sexual offences
against children could be conceptualized into three distinct behavioural themes, which are
characterized by aggressive, intimate or criminalopportunist modes of offendervictim
interaction. The ‘‘aggressive’’ style of interaction involves the perpetrator perceiving the child
as a focus or vehicle for their aggression. The incidents often involve threats of harm and
physical aggression greater than that necessary to control the victim. The sexual activity is
characterized by domination and anger and necessarily involves the degradation of the victim.
This ‘‘aggressive’’ theme is therefore similar to Canter’s (1994) ‘‘victim as vehicle’’ theme.
A second behavioural theme found in the behaviour of paedophiles is the ‘‘intimate’’ style
of interaction, which involves the perpetrator perceiving the child as an alternative to an age-
appropriate partner. They view their relationship with the victim as comparable with a
conventional sexual relationship between adults. These perpetrators use methods such as
enticement, affection and the giving of gifts in order to develop a consenting sexual
relationship with the child. This is often referred to as ‘‘grooming’’. The sexual behaviour
develops over time from non-sexual physical contact into sexual touching in order to
desensitize the victim. The perpetrator performing sexual acts on the victim, such as oral sex
and mutual masturbation, characterizes this sexual activity. This ‘‘intimate’’ theme is
therefore similar to Canter’s (1994) ‘‘victim as person’’ theme.
The third theme found by Canter et al. (1998) is a ‘‘criminal/opportunist’’ style of
interaction, which involves the perpetrator using the victim solely as an outlet for their sexual
gratification. It is not the intention of these perpetrators to hurt their victim and the force used
will not exceed that necessary to control the victim. These incidents are opportunistic in
nature with little emotional attachment involved; these individuals have a general disregard for
the rights of others and the sexual abuse is just one of the criminal activities in which they are
involved. These perpetrators are often intoxicated, which results in the lowering of their
impulse control and enhancing their perception of opportunities to offend. This ‘‘criminal/
opportunist’’ theme is therefore similar to Canter’s (1994) ‘‘victim as object’’ theme.
Studies carried out into the behaviour of perpetrators during the acts of rape (Canter,
1994; Canter & Alison, 2003) and child abuse (Canter et al., 1998) have, therefore, found
empirical support for a composite model of sexual behaviour consisting of three modes of
interaction (‘‘victim as object’’, ‘‘victim as person’’ and ‘‘victim as vehicle’’) which occur at
various levels of violation (‘‘personal’’, ‘‘physical’’ and ‘‘sexual’’). However, these studies have
investigated the behaviour of adult offenders. The question arises as to whether the behaviour
of young people reveals a similar multivariate model.
Hypotheses
Broadly, current research has shown that the frequencies of behaviours displayed by young
people with HSB are medium to low, which indicates the heterogeneity of the sample and the
possible existence of subgroups. However, current methods of classifying these youths have
generally failed to find significant differences between subgroups in terms of the behaviour
exhibited during the abusive incident. Research carried out on adult sex offenders has,
however, found differences in behaviour by utilizing the narrative role of the victim as a
method of differentiating between perpetrators.
The first hypothesis proposed by this study is that the frequency patterns of behaviours
across a set of sexual offences committed by young people will conform to a scale of violation,
as is the case in adult sex offender models of behaviour. This pattern will have a common
order of increasing violation with increasing frequency of behaviour. High-frequency variables
will relate to sexually violating behaviours, low-frequency variables will relate to personally
violating behaviours and physically violating behaviours will be represented in between.
The second hypothesis proposed is that the behaviours displayed by young people with
HSB can be differentiated into three distinct themes, ‘‘victim as object’’, ‘‘victim as person’’
and ‘‘victim as vehicle’’ which relate to the differing modes of interaction between perpetrators
and their victim, as is the case in adult sex offender models of behaviour. If these three themes
are distinct, then each theme will contain a set of conceptually related behaviours that will
occur together consistently. This threefold method of differentiation is not meant to represent
distinct types of youths, but rather differentiated themes that will be present, to some degree,
in all youths.
A third hypothesis proposed for this study is that the behaviours an individual perpetrator
exhibits will reflect one dominant theme of interaction, as the three themes represent different
modes of interpersonal interaction between the perpetrator and the victim. It is proposed,
therefore, that the resulting multivariate model will be able to identify the underlying structure
of the behaviour of young people with HSB and will help to distinguish one set of assaults
from another. This has possible implications for the investigative process, both in the
apprehension of an unknown perpetrator and in the linking of a series of incidents. In
addition, as each theme of behaviour reflects a different mode of interaction with the victim, it
is proposed that each mode of interaction would benefit from distinctly different treatment
approaches.
Method
Procedure
As the nature of any one agency will inevitably determine the cases included and excluded
from the study, information was obtained from two specialist agencies that work with young
people with HSB in order to improve generalizability and representativeness. The sample here
consists of 170 cases from ‘‘G-MAP’’ and 130 from the ‘‘5A project’’. The youths were
referred to the two agencies between 1988 and 2002. G-MAP is an independent provider
established in 1988 and based in Greater Manchester, although the service is accessible on a
national basis. The Barnardo’s 5A project has, since 1994, provided a service for young
people with HSB in Liverpool and, from 1999, throughout the Merseyside area.
The data set is based on an extensive and detailed review of subject files. A data collection
form, designed specifically for this study, was used to collect information regarding the
behaviours of the youth during the abusive incident. In total, 33 behavioural characteristics
were coded as present or absent for each individual. Full details of these characteristics are
outlined in the content dictionary (Appendix I). Ten randomly selected files were coded
independently resulting in a Cohen’s kappa value of 0.770, which is significant at the p B
0.0001 level, indicating acceptable inter-rater reliability.
As this study relies solely on case files, which were not gathered for the purpose of
scientific research, the consistency with which information is recorded and the attention to
detail will not be of the standard that is often taken for granted in a research environment
(Canter et al., 2003). Information contained within the case files may be missing or
incomplete. This lack of comprehensive information increases the potential for distortion.
Therefore, a dichotomous approach (yes/no values based on the presence/absence) was used
to ensure maximum clarity and reliability when using records not initially collected for
research purposes (Canter & Fritzon, 1998; Salfati, 2000).
Sample
There was a total of 300 youths in the study. All were aged 18 years or under at the time of
their referral to the specialist agencies. The mean age of the cohort was 14.3 years (standard
deviation 1.79) with an age range of 918 years. The data consisted of behavioural
information concerning the abusive incident for which the youth was referred. If the youth
was referred for more than one victim, then the information pertaining to the first victim was
recorded. Canter et al. (2003), in their study of stranger rape, used this one victim per
offender method to ensure that the empirical structure revealed by the analysis was not biased
by weighting being given to youths who may be displaying a particular pattern of behaviour
with multiple victims.
Statistical analysis
This study aimed to examine the harmful sexual behaviour displayed by young people by
utilizing multidimensional statistical procedures, which have been used previously to examine
the behaviour of adult sexual offenders (Canter, 1994; Canter et al., 2003). In order to do
this, the data set was analysed using a multidimensional scaling procedure, known as Smallest
Space Analysis (SSA) (Lingoes, 1973). SSA is based on the assumption that any underlying
structure will be appreciated most readily by examining the relationship that each variable has
with every other variable (Canter, 1985). In order to test the relationship each behavioural
characteristic has to every other behavioural characteristic, a measure of association is
calculated that indicates the degree of co-occurrence between each behaviour with every other
behaviour. As the data were archival and not collected for research purposes, it can never be
certain that absent information was simply not recorded. Therefore, Jaccard’s coefficient was
used, as this is a measure of association that does not take account of joint non-occurrences.
The measures of association between all variables are then rank-ordered to enable the SSA to
represent the relationships between the characteristics as effectively as possible in a low
dimensionality. These rank-ordered correlations are then represented as ranked distances in
an abstract ‘‘space’’. This allows the underlying structure of the behaviour to be examined,
enabling any themes within the characteristics to be identified.
The visual representation is generated such that the higher the correlation between any
two characteristics, the closer together the points representing them will appear on the spatial
plot (Guttman, 1968). The resulting patterns of characteristics can hence be examined and
thematic differentiations can be delineated. The aims of the study were built upon the
assumption that characteristics with similar underlying themes would be more likely to co-
occur than those that imply divergent themes. As mentioned earlier, this regional hypothesis
has been viewed previously as an appropriate way of interpreting co-occurrences of variables,
and has been used successfully to interpret related areas such as paedophilia (Canter et al.,
1998) and rape (Canter et al., 2003).
Results
An SSA was carried out on 33 behaviours across the 300 cases. The three-dimensional SSA
solution had a Guttman Lingoes coefficient of alienation of 0.16, indicating a good fit
between the SSA plot and the original association matrix. In both figures, vectors 1 and 2 of a
three-dimensional space are shown. Each point represents a behaviour displayed by the youth
during the abusive incident. The closer any two points are, the more likely it is that the
behaviours will co-occur in comparison with other behaviours. For illustration, it is very likely
that if the incident involved a co-perpetrator then there will be multiple acts of violence. By
contrast, it is unlikely that if a youth coerces his/her victim physically they would also use some
form of inducement.
Levels of violation
The first hypothesis stated that the frequency patterns of behaviours across a set of sexually
abusive incidents committed by young people would conform to a scale of violation. Canter et
al. (2003), in their study of stranger rape, argued that as an SSA plot is derived from the
correlations between the behaviours there is no inevitable link to their frequencies. Therefore,
any frequency structure that emerges is an empirical finding. As indicated in Figure 1, it is
possible to draw contours on the SSA that represent a general frequency pattern. This pattern
ranges from medium frequency behaviours in the centre of the plot to lower frequency
behaviours that radiate out towards the edges of the plot. As found by Canter et al. (2003), the
frequencies serve as a summary of behaviours displayed during the abusive incident, showing
that those behaviours further out from the core are the most differentiated, giving any
particular incident its specific characteristics, while those at the centre are conceptually
central to youths’ sexually harmful behaviour.
An examination of the behaviours within these general frequency contours provides
evidence for the facet of increasing violation. The variables ‘‘youth touches victim’s genitals’’
(46%) and ‘‘more than one incident with the victim’’ (36%) were found to form the core
region in the SSA, indicating that these two variables are the defining variables of youth
sexually harmful behaviour. In addition, a medium-frequency band (1030%) could be
identified that consists generally of sexually violating behaviours and a lower-frequency band
( B10%) could be identified that consists of physically violating behaviours.
Unlike the stranger rape research carried out by Canter et al. (2003), this model does not
contain a set of personally violating behaviours. The only personally violating behaviours
recorded in the case files of young people with HSB was the behaviour ‘‘concurrently with
another crime’’, a low-frequency behaviour, as was the case in the adult model. The lack of
personally violating behaviours may be a result of the fact that young people are at a different
26.vremclo(2)
30.trick(7)
32.ancrime(2)
8.attvpe(11)
SEXUAL
31.intox(5) 10-30%
9.vpenp(15) 28.induc(12)
5.anpenp(11)
1.exhibit(19)
4.oraly(15)
16.weapon(6) 29.bribe(6)
17.threat(22)
33.more1inc(36)
24.tcloth(2) 3.oralv(14) PHYSICAL
19.abulan(5) 15.phycoe(28)
25.yremclo(15) 12.touvgen(46) 13.touygen(15)
<10%
21.vbeyc(3)
22.mviol(4) 14.kiss(15) 2.gencon(9)
27.othery(10)
10.vpenf(19)
11.toub(17) CORE
20.sviol(4)
7.attanp(9)
>30%
18.thrfam(2)
23.bound(2)
6.anpenf(3)
FIGURE 1. A 1 2 projection of a three-dimensional Smallest Space Analysis (SSA), 33 behaviours displayed by 300 young
people with harmful sexual behaviour (HSB). The SSA indicates the degrees of violation. Variables labels are brief summaries of
content analysis categories for full definitions see Appendix I. Values in brackets are percentage frequencies.
developmental, emotional and behavioural stage of their life than adults; it may also be an
indication that young people are less sophisticated when compared to adult offenders. It is
also possible, of course, that differences in procedure when interviewing youths and their
victims about the abusive incident may have different emphasis from those carried out with
adults. Therefore, the personally violating behaviours identified in the adult behavioural
model, ‘‘implies knowing the victim’’ and ‘‘forces victim to make sexual comments’’, may
either not be occurring or the information regarding these behaviours is not being requested
during the referral process.
Sexual violation
As indicated by Figure 1 and Table II, behaviours that constitute sexual violation form a
central region in the SSA plot. Sexually violating behaviours found in this moderate frequency
band include ‘‘vaginal penetration with penis’’, ‘‘attempted vaginal penetration’’, ‘‘anal
penetration with penis’’, ‘‘attempted anal penetration’’, ‘‘exhibitionism’’, ‘‘victim makes oral
contact with youth’s genitals’’, ‘‘youth oral contact with victim’s genitals’’, ‘‘victim touches
youth’s genitals’’, ‘‘genital contact without penetration’’, ‘‘vaginal penetration with finger’’,
‘‘touches victim’s breasts’’ and ‘‘youth kisses victim’’. The region also included behaviours
such as ‘‘inducement’’, ‘‘bribe’’, ‘‘verbal threat’’, ‘‘physical coercion’’ and ‘‘youth removes
Table II. Behaviours contained within the two levels of violation. Letters in capitals refer to abbreviations used in
Figures 1 and 2. For full definitions of the behaviours see Appendix I.
Core variables 30% Sexual violation 1030% Physical violation B10%
12. TOUches Victim’s 15. PHYsical COErcion (28%) 25. Youth REMoves victim’s CLOthes
GENitals (46%) (15%)
33. MORE than 1 17. THREAT (22%) 27. OTHER Youth/s involved (10%)
INCident (36%)
1. EXHIBITionism (19%) 30. TRICKery (7%)
10. Vaginal PENetration with Finger 16. WEAPON use or threat of (6%)
(19%)
11. TOUched victim’s Breasts (17%) 19. ABUsive LANguage (5%)
4. ORAL contact with Victim’s genitals 31. INTOXicated (5%)
(15%)
9. Vaginal PENetration with Penis 20. Single act of VIOLence (4%)
(15%)
13. Victim TOUches Youth’s GENitals 22. Multiple acts of VIOLence (4%)
(15%)
14. KISS (15%) 21. Violence BEYond that necessary to
Control the victim (3%)
3. ORAL contact with Youth’s genitals 6. ANal PENetration with Finger (3%)
(14%)
28. INDUCement (12%) 18. THReat to FAMily (2%)
5. ANal PENetration with Penis (11%) 23. BOUND (2%)
8. ATTempted Vaginal Penetration 24. Tears CLOTHes (2%)
(11%)
2. GENital CONtact (9%) 26. Victim REMOves own CLOTHes
(2%)
7. ATTempted ANal Penetration (9%) 32. concurrently with ANother CRIME
(2%)
29. BRIBE (6%)
victim’s clothes’’; these variables relate to the methods of coercion the youth used to ensure
the victim’s compliance in order that the sexual behaviours be carried out successfully.
Physical violation
Within the lower-frequency contour, these lesser frequent behaviours reflect the physical
violation of the victim. The physically violating actions in the region include ‘‘weapon’’,
‘‘tearing of clothes’’, ‘‘abusive and/or sexual language’’, ‘‘single act of violence’’, ‘‘multiple acts
of violence’’, ‘‘violence beyond that necessary to control the victim’’, ‘‘ threats to harm victim’s
family’’, ‘‘binding of the victim’’, ‘‘trickery’’ and ‘‘anal penetration with finger’’. This region
also included variables such as ‘‘other youth involved’’, ‘‘incident occurred in conjunction
with another crime’’ and ‘‘youth intoxicated’’. These behaviours may relate to why the
incident occurred.
Theme of behaviour
The second hypothesis stated that the behaviours displayed by young people with HSB would
be differentiated into three distinct themes relating to the differing modes of interaction
between a perpetrator and their victim. Figure 2 illustrates how the co-occurrences of
behaviours of the sample can be partitioned into three thematically similar subgroups.
26.vremclo
VICTIM AS
OBJECT 30.trick VICTIM AS
PERSON
32.ancrime
8.attvpe
31.intox
9.vpenp 28.induc
5.anpenp
1.exhibit
4.oraly
16.weapon 29.bribe
17.threat
33.more1inc
24.tcloth 3.oralv
19.abulan 15.phycoe
25.yremclo 12.touvgen 13.touygen
21.vbeyc
22.mviol 14.kiss 2.gencon
27.othery
10.vpenf
11.toub
20.sviol
7.attanp
VICTIM AS 18.thrfam
VEHICLE
23.bound
6.anpenf
FIGURE 2. A 1 2 projection of a three-dimensional SSA of 33 behaviours displayed by 300 youth with harmful sexual
behaviour (HSB). The Smallest Space Analysis (SSA), indicates the different themes of behaviour. Variables labels are brief
summaries of content analysis categories, for full definitions see Appendix I.
Table III. Behaviours contained within each of the themes. Letters in capitals refer to abbreviations used in Figures 1
and 2. For full definitions of the behaviours see Appendix I.
Victim as object Victim as person Victim as vehicle
17. THREAT (22%) 33. MORE than 1 INCident (36%) 12. TOUches Victim’s GENitals (46%)
9. Vaginal PENetration 1. EXHIBITionism (19%) 15. PHYsical COErcion (28%)
with Penis (15%)
16. WEAPON use or 4. ORAL contact with Youth’s genitals 10. Vaginal PENetration with Finger
threat of (6%) (15%) (19%)
31. INTOXicated (5%) 13. victim TOUches Youth’s GENitals 11. TOUched victim’s Breasts (17%)
(15%)
24. Tears CLOTHes (2%) 3. ORAL contact with Victim’s genitals 14. KISS (15%)
(14%)
32. concurrently with 28. INDUCement (12%) 25. Youth REMoves victim’s CLOthes
ANother CRime (2%) (15%)
5. ANal PENetration with Penis (11%) 27. OTHER Youth/s involved (10%)
8. ATTempted Vaginal Penetration 7. ATTempted ANal Penetration (9%)
(11%)
2. GENital CONtact (9%) 19. ABUsive LANguage (5%)
30. TRICKery (7%) 20. Single act of VIOLence (4%)
29. BRIBE (6%) 22. Multiple acts of VIOLence (4%)
26. Victim REMoves own CLOThes 6. ANal PENetration with Finger (3%)
(2%)
21. Violence BEYond that necessary to
Control the victim (3%)
18. THReat to FAMily (2%)
23. BOUND (2%)
K-R 20 .511 K-R 20 .561 K.R 20 .622
Victim as object
As can be seen from Figure 2 and Table III, there is a collection of behaviours in the top left of
the plot that imply that the youth is interacting with their ‘‘victim as object’’. These youths are
using their victim to satisfy their own gratifications*sexual gratification as indicated by the
variable ‘‘vaginal penetration with the penis’’ and instrumental gratification, indicated by the
variable ‘‘another crime committed at time of the abusive incident’’. This accords with
Canter’s (1994) ‘‘victim as an object’’ mode of interaction. As in Canter et al.’s (1998) model
of paedophilic behaviour (‘‘criminal opportunist’’ region) the variable ‘‘youth intoxicated’’ is
found within this region. Canter et al. (1998) argued that alcohol and or drug use may
facilitate sexual offending by lowering impulse control and enhancing the perception of
opportunity. The lack of other sexual behaviours in this region indicates that the sole
motivation for these youths is the accomplishment of the sexual act. As is the case in Canter’s
(1994) model, these youths control their victims through ‘‘verbal threats’’ and ‘‘threat and/or
use of a weapon’’. These behaviours are used to force the victim into a non-participative role.
The behaviour ‘‘tearing of victim’s clothes’’ also indicates the youth’s attempt to exert control
over the victim.
Victim as person
Behaviours in the top right of the plot indicate that the youth is interacting with the ‘‘victim as
person’’. The presence of the behaviour ‘‘more than one sexually abusive incident occurred’’
within this region indicates [as is the case in Canter’s 1994 and 2003 models of rape and in
Canter et al.’s 1998 model of paedophilic behaviour (‘‘intimate region’’)], that these youths
view their relationship with the victim as a conventional sexual relationship and therefore
interact with them as a person. This is also implied by the reciprocal sexual behaviours
contained within this region, which include ‘‘youth makes oral contact with the victim’s
genitals’’, ‘‘victim makes oral contact with youth’s genitals’’, ‘‘victim touches youth’s genitals’’,
‘‘exhibitionism’’ and ‘‘victim removes own clothes’’. These behaviours imply that the youth
views their victim as a reactive individual. These youths also use psychological methods of
coercion as opposed to physical methods indicated by the behaviours ‘‘inducement’’,
‘‘trickery’’ and ‘‘bribes’’. However, due to the fact that these incidents may occur over a
period of time, as seen by the variable ‘‘more than one incident’’, this results in the escalation
of the seriousness of the behaviours being displayed as indicated by the behaviours ‘‘genital
contact without penetration’’, ‘‘attempted vaginal penetration with the penis’’ and ‘‘anal
penetration with the penis’’.
Victim as vehicle
Behaviours at the bottom of the plot indicate that the youth is interacting with the ‘‘victim as
vehicle’’. These youths are using their victim to vent their anger and frustration. As is the case
in Canter’s 1994 and 2003 models of rape and Canter et al.’s 1998 model of pedophilia
(‘‘aggressive’’ region), behaviours in this region indicate the aggressive manner in which these
youths interact with their victim. These behaviours included ‘‘physical coercion’’, ‘‘single act
of violence’’, ‘‘multiple acts of violence’’, ‘‘violence beyond that necessary to control the
victim’’ (which included, in one case, attempted murder), ‘‘youth removes victim’s clothing’’
and ‘‘victim is bound’’. These youths also utilized language as a method of aggression, as
indicated by the presence of ‘‘abusive and sexually explicit language’’ as well as ‘‘threats to
harm the victim’s family’’. These verbally aggressive behaviours were also found in the ‘‘victim
as vehicle’’ region in the adult sexual offender models (Canter, 1994; Canter et al., 1998;
Canter et al., 2003). These aggressive behaviours are used by the youths to intimidate their
victim physically and mentally. The sexual behaviours in this region in the main involve
penetration, ‘‘vaginal penetration with the finger’’, ‘‘anal penetration with the finger’’ and
‘‘attempted anal penetration’’; the behaviours ‘‘youth touches victim’s genitals’’ and ‘‘touches
victim’s breasts’’ were also contained within this region. These sexual behaviours are used by
these youths to humiliate and demean their victims. This region also contained the
characteristic ‘‘more than one youth involved’’. Amir (1971) found that gang rape was
associated significantly with violence, especially in its extreme forms. Interestingly, the
behaviour ‘‘youth kisses the victim’’ is in this ‘‘victim as vehicle’’ region. In both Canter et al.’s
2003 model of rape and Canter et al.’s 1998 model of paedophilia it is located in the regions
comparable with ‘‘victim as person’’. Therefore, young people in this sample are kissing their
victim as an aggressive action (the behaviour correlates highly with other aggressive
behaviours as indicated by its location in this region), and this is not comparable with the
behaviour of adult sexual offenders.
Table III shows the behaviours contained within each theme and their relative
frequencies. The highest-frequency behaviours within each theme help define that theme.
The highest-frequency behaviours in the ‘‘victim as object’’ theme are ‘‘verbal threat’’ and
‘‘vaginal penetration with the penis’’; in the ‘‘victim as person’’ theme they are ‘‘more than one
incident with the victim’’ and ‘‘exhibitionism’’, and in the ‘‘victim as vehicle’’ theme, the high-
frequency behaviours are ‘‘touches victim’s genitals’’ and ‘‘physical coercion’’.
Kuder-Richardson 20 (K-R 20) coefficients were calculated in order to provide an index
of internal reliability. The K-R 20 is equivalent to the more common Cronbach’’s alpha
coefficient but can be used with dichotomous data. Table III gives the K-R 20 coefficients for
each of the three themes; these values are reasonable considering that the data were not
collected for research purposes. They also equate with other published models in this area
which have used this coefficient, such as Canter et al. (2003) and Porter and Alison (2004).
These coefficients are not sufficiently high enough to demonstrate clear evidence of a scale,
therefore the characteristics within each theme do not represent a scale of increasing
behaviours within each theme. However, they are sufficiently high enough to indicate that the
themes are meaningful and coherent (Alison & Stein, 2001).
Classifying youths who sexually harm

The authors recognize that every offence has the potential of drawing on all the regions,
therefore each offence can be said to have a profile of behaviours, not simply one type of
behaviour. However, the study’s third hypothesis proposed that the behaviours an individual
youth exhibits would reflect one dominant theme of interaction. Indeed, the utility of the
thematic structure proposed above is enhanced considerably if such dominant themes are found
for most youths. In order to test this hypothesis, as well as test whether the proposed framework
serves as a useful way of classifying young people with HSB, each of the 300 in the data set were
examined individually to ascertain if they could be assigned to a dominant theme of behaviour.
A youth may display behaviours from more than one theme; however, because the themes
represent psychologically distinct behavioural themes, it was hypothesized that the majority of
behaviours a youth displayed would be contained within one dominant theme. Every incident
was assigned a score for each of the three themes reflecting the percentage of ‘‘victim as object’’,
‘‘victim as person’’ and ‘‘victim as vehicle’’ behaviours that occurred in that incident. The
criterion used for assigning an individual to a particular theme was taken from Salfati and
Canter (1999), who used this method to classify stranger murders. To be assigned to a dominant
theme, the percentage score for that theme has to be greater than or equal to the sum of the other
two themes combined. Using this procedure, 86% (i.e. 258 of the 300 abusive incidents
perpetrated by a young person) could be assigned to a dominant theme. As Figure 3 shows, out
of the 258 (86%) incidents that could be assigned a dominant theme, 124 (41%) were classified
as ‘‘victim as person’’, 89 (30%) ‘‘victim as vehicle’’ and 45 (15%) ‘‘victim as object’’.
Summary of findings
This study hypothesized that the frequency patterns of behaviours across a series of sexual
offences committed by young people would conform to a scale of violation. In the SSA
Victim as Object
Victim as Vehicle
Victim as Person
0 20 40 60 80 100 120 140

Number of Perpetrators
FIGURE 3. Distribution of cases across dominant themes of behaviour.

configuration the variables with the highest frequencies are located in the centre of the plot
(see Figure 1). The behaviours that constitute sexual violation form the moderate frequency
contour and behaviours that constitute physical violation form the low-frequency contour.
The pattern of behaviour therefore has a common order of increasing violation with increasing
frequency.
The study’s second hypothesis was that the behaviour displayed by young people with
HSB could be differentiated into three distinct themes. The results of the SSA indicate that
the behaviour displayed by these young people could be conceptualized as various modes of
interpersonal interaction: ‘‘victim as object’’, ‘‘victim as person’’ and ‘‘victim as vehicle’’ (see
Figure 2). Each of these three modes of interaction is distinguishable, being made up of a
subset of conceptually related behaviours.
The study’s third hypothesis proposed that the behaviours an individual youth exhibits
would reflect one dominant theme of interaction. The fact that the majority of cases, 86%,
reflected one dominant theme of behaviour provides empirical support for the hypothesis that
this multivariate model represents the underlying structure of the behaviour displayed by
young people during the sexually abusive incident (see Figure 3). This multivariate model,
therefore, allows one set of assaults to be distinguished from another.
Implications of the present findings

The emergence of a multivariate model of behaviour in respect of young people with HSB has
a number of important implications. First, the current model may be able to aid practitioners
who working in the field with such young people, as it will allow the majority of individuals to
be classified as demonstrating one of three dominant themes of behaviour. Effective
classification schemes can help criminal justice agents in their decision-making, improve
service delivery and make more effective use of the limited resources made available to
specialist agencies.
This model could, potentially, enable the identification of youth with HSB who are at a
high risk of recidivism by examining the re-offending rates of the different themes. Using this
model it may also be possible to identify whether the risk is of further sexual, violent or non-
sexual offending. For example, young people who interact with their ‘‘victim as vehicle’’ may
be more likely to commit further violent offences, due to the high levels of aggression shown
during the abusive incident, while youths who interact with their ‘‘victim as object’’ may be
more likely to commit further sexual offences, as their primary motivation was the
accomplishment of the sexual act.
Masson and Hackett (2003) emphasized the need for services to attend to the diversity of
this heterogeneous population. Using this model there is considerable potential for the
development of various treatment programmes that are responsive to the variations that exist
between individuals, in terms of their individual behavioural profile. Youths who recognize
human, personal qualities in their victims need a different sort of treatment from those youths
who treat their victims as inanimate objects.
Through future research it may be possible to construct particular profiles of each of
these themes based on the examination of victim characteristics. These may then be used to
identify high-risk situations for a particular individual and, therefore, aid in the placement
decision-making process. For example, if youths who interact with their ‘‘victim as person’’
usually offend against younger siblings then they should not be placed into a care setting with
other young children, as this represents a high-risk scenario. Almond, Canter and Salfati
(2006) produced a multivariate model of perpetrator characteristics which resulted in three
themes of background: abuse, delinquency and impairment. Future research should now
investigate the links that exist between these two multivariate models as this may aid the police
in their investigations. If an unknown perpetrator displays a dominant behaviour theme
during an abusive incident, and if research has shown that links exist between the theme of
behaviour and particular perpetrator characteristics, this may be invaluable in suspect
elicitation and prioritization. Therefore, linking these two models together will help build a
richer picture of perpetrators and their actions, which has further implications for both
treatment and investigations.
Acknowledgements
The authors would like to thank all the staff at G-MAP and the 5A project for all their
assistance and for allowing them access to their files.
References
Alison, L. & Stein, K. (2001). Vicious circles: Accounts of stranger sexual assault reflect abusive variants of
conventional interaction. Journal of Forensic Psychiatry, 12, 515538.
Almond, L., Canter, D. & Salfati, G. (2006). Youths who sexually harm: A multivariate model of perpetrator
characteristics. Journal of Sexual Aggression, 12, 118.
Amir, M. (1971). Patterns in Forcible Rape. Chicago: University of Chicago Press.
Beckett, R. (1999). Evaluation of adolescent sexual abusers. In M. Erooga & H. Masson (Eds.), Children and Young
People Who Sexually Abuse Others. Challenges and Responses (pp. 204224). London: Routledge.
Canter, D., Bennell, C., Alison, L. & Reddy, S. (2003). Differentiating sex offences: A behaviourally based thematic
classification of stranger rapes. Behavioural Sciences and the Law, 21, 157174.
Canter, D. & Alison, L. (2003). Converting evidence into data: The use of law enforcement archives as unobtrusive
measurement. Qualitative Report, 8, 151176.
Canter, D. & Fritzon, K. (1998). Differentiating arsonists: A model of firesetting actions and characteristics. Legal and
Criminological Psychology, 3, 7396.
Canter, D., Hughes, D. & Kirby, S. (1998). Paedophilia: pathology, criminality, or both? The development of a
multivariate model of offence behaviour in child sexual abuse. Journal of Forensic Psychiatry, 9, 532555.
Canter, D. (1994). Criminal Shadows. London: Harper Collins Publishers.
Canter, D. (1985). Facet Theory Approaches to Social Research. New York: Springer Verlag.
Dolan, M., Holloway, J., Bailey, S. & Kroll, L. (1996). The psychosocial characteristics of juvenile sexual offenders
referred to an adolescent forensic service in the UK. Medicine. Science and Law, 36, 343352.
Epps, K. (1999). Looking after young sexual absuers: Child protection, risk management and risk reduction. In M.
Erooga & H. Masson (Eds.), Children and Young People Who Sexually Abuse Others. Challenges and Responses (pp. 67
85). London: Routledge.
Guttman, L. (1968). A general nonmetric technique for finding the smallest co-ordinate space for a configuration of
points. Psychometrika, 33, 469506.
Home Office (2003). Criminal Statistics for England and Wales 2002. London: Home Office.
Howarth, V. (1999). Foreword. In M. Erooga & H. Masson (Eds.), Children and Young People Who Sexually Abuse
Others. Challenges and Responses (pp. xvxvii). London: Routledge.
Hunter, J., Hazelwood, R. & Slesinger, D. (2000). Juvenile-perpetrated sex crimes: Patterns of offending and
predictors of violence. Journal of Family Violence, 15, 8193.
Knight, R. & Prentky, R. (1990). Classifying sexual offenders: The development and corroboration of taxonomic
models. In W. Marshall, D. Laws & H. Barbaree (Eds.), Handbook of Sexual Assault: Issues, Theories, and Treatment of
the Offender (pp. 2352). New York: Plenum.
Lingoes, J. (1973). The Guttman Lingoes Nonmetric Program Series. Ann Arbor, MI: Mathesis.
Manocha, K. & Mezey, G. (1998). British adolescents who sexually abuse: A descriptive study. Journal of Forensic
Psychiatry, 9, 588608.
Masson, H. & Hackett, S. (2003). A decade on from the NCH report (1992): Adolescent sexual aggression policy,
practice and service delivery across the UK and Republic of Ireland. Journal of Sexual Aggression, 9, 109124.
National Organization for the Treatment of Abusers (2001). Work With Young People who have Committed Sexual
Offences. NOTA position paper.
Porter, L. & Alison, L. (2004). An interpersonal model of sexually violent gang behaviour. Aggressive Behaviour, 30,
449468.
Richardson, G., Kelly, T., Graham, F. & Bhate, S. (2004). A personality-based taxonomy of sexually abusive
adolescents derived from the Millon Adolescent Clinical Inventory (MACI). British Journal of Clinical Psychology,
43, 285298.
Richardson, G., Kelly, T., Graham, F. & Bhate, S. (1997). Group differences in abuser and abuse characteristics in a
British sample of sexually abusive adolescents. Sexual Abuse: A Journal of Research and Treatment, 9, 239257.
Richardson, G., Graham, F., Bhate, S. & Kelly, T. (1995). A British sample of sexually abusive adolescents; Abuser
and abuse characteristics. Criminal Behaviour and Mental Health, 5, 187209.
Salfati, C. (2000). The nature of expressiveness and instrumentality in homicide. Homicide Studies, 4, 265293.
Salfati, G. & Canter, D. (1999). Differentiating stranger murders: Profiling offender characteristics from behavioural
styles. Behavioral Sciences and the Law, 17, 391406.
Smith, W., Monastersky, C. & Deisher, R. (1987). MMPI-based personality types among juvenile sexual offenders.
Journal of Clinical Psychology, 43, 422430.
Worling, J. (2001). Personality-based typology of adolescent male sexual offenders: Differences in recidivism rates,
victim-selection characteristics and personal victimization histories. Sexual Abuse: A Journal of Research and
Treatment, 13, 149166.
Appendix I.
Content dictionary
Behaviour Short name Definition
1. Exhibitionism exhibit The youth shows their genitals to the victim

2. Genital contact without penetration gencon The youth simulates having intercourse
without penetrating the victim
3. Youth makes oral contact with the victim’s oralv The youth makes contact with the victim’s
genitals genitals using his/her mouth
4. Victim makes oral contact with the youth’s oralp The victim makes contact with the youth’s
genitals genitals using his/her mouth
5. Anal penetration with penis anpenp The youth penetrates the victim’s anus with
his penis
6. Anal penetration with finger anpenf The youth penetrates the victim’s anus with
his/her finger
7. Attempted anal penetration attanp The youth attempts to penetrate the victim’s
anus
8. Attempted vaginal penetration attvp The youth attempts to penetrate the victim’s
vagina
9. Vaginal penetration with penis vpenp The youth penetrates the victim’s vagina with
his penis
10. Vaginal penetration with finger vpenf The youth penetrates the victim’s vagina with
his/her finger
11. Touches victim’s breasts toub The youth touches the victim’s breasts
12. Youth touches the victim’s genitals touvgen The youth touches the victim’s genitals
13. Victim touches youth toupgen The victim touches the youth’s genitals
14. Youth kissed the victim kiss The youth kisses the victim
15. Physical coercion phycoe The youth uses physical force during the
incident (e.g. hitting, slapping, restraining)
16. Threat/use of weapon weapon The youth threatens and/or uses a weapon in
order to control the victim
17. Verbal threat threat The youth verbally threatens the victim
18. Threat to victim’s family thrfam The youth threatens to harm victim’s family
19. Youth uses abusive or sexually explicit abulan The youth uses abusive and/or sexually
language explicit language during the incident
20. Single act of violence sviol The youth carries out a single act of violence
against the victim (e.g. single slap)
<?th5pt?Appendix
Appendix I[/b] ([it]Continued[/it])>
I (Continued)
Behaviour Short name Definition
21. Violence used beyond the level necessary to vbeyc Violence is used beyond the level necessary to
control the victim control the victim (i.e. for gratuitous
purposes)
22. Youth carried out multiple acts of violence mviol The youth carries out multiple acts of
against victim during one incident violence against the victim (e.g. multiple
punches)
23. Victim is bound bound The victim is bound at any time during the
incident with any material (not including
restraint by youth’s hands)
24. Tears clothing tcloth The youth damages the victim’s clothing by
removing them in a violent manner
25. Youth removes victim’s clothing premclot The youth forcefully removes the victim’s
clothing
26. Victim removes own clothing vremclot The victim removes their own clothing
27. Another youth involved otherper Another youth/s were involved in the incident
28. Inducement induc The youth uses some form of inducement to
coerce the victim
29. Youth bribes the victim bribe The youth buys the victims gifts or gives
them money
30. Trickery trick The youth tricks the victim by playing
‘‘games’’
31. Youth intoxicated at time of the offence intox The youth was intoxicated with alcohol/and
or drugs when the incident occurred
32. Sexual assault occurred in conjunction with ancrime The incident occurred in conjunction with
another crime another crime (e.g. burglary)
33. More than one incident with the same victim more1inc The youth was involved in more than one
sexually abusive incident with the same
victim

Youths Who Sexually Harm, A Multivariate Model of Behaviour

Diunggah oleh

Informasi Dokumen

Deskripsi Asli:

Hak Cipta

Format Tersedia

Bagikan dokumen Ini

Bagikan atau Tanam Dokumen

Opsi Berbagi

Apakah menurut Anda dokumen ini bermanfaat?

Apakah konten ini tidak pantas?

Hak Cipta:

Format Tersedia

Youths Who Sexually Harm, A Multivariate Model of Behaviour

Diunggah oleh

Hak Cipta:

Format Tersedia

This article was downloaded by:[HEAL- Link Consortium]

On: 11 December 2007

Journal of Sexual Aggression

Online Publication Date: 01 November 2007

PLEASE SCROLL DOWN FOR ARTICLE

Full terms and conditions of use: http://www.informaworld.com/terms-and-conditions-of-access.pdf

Youths who sexually harm: A multivariate

Louise Almond* & David Canter

Manocha & Mezey Richardson

Nature of abusive behaviour

Core variables 30% Sexual violation 1030% Physical violation B10%

Victim as object Victim as person Victim as vehicle

Classifying youths who sexually harm

0 20 40 60 80 100 120 140

FIGURE 3. Distribution of cases across dominant themes of behaviour.

Implications of the present findings

Behaviour Short name Definition

1. Exhibitionism exhibit The youth shows their genitals to the victim

Behaviour Short name Definition

Anda mungkin juga menyukai