Content uploaded by Asher Flynn
Author content
All content in this area was uploaded by Asher Flynn on Feb 24, 2019
Content may be subject to copyright.
Image-based sexual abuse: The extent, nature, and predictors of perpetration in a
community sample of Australian residents
Anastasia Powell
Justice and Legal Studies, RMIT University
Nicola Henry
Global, Urban and Social Studies, RMIT University
Asher Flynn
School of Social Sciences, Monash University
Adrian J. Scott
Department of Psychology, Goldsmiths, University of London; School of Arts and
Humanities, Edith Cowan University
Acknowledgement
The research upon which this article draws has been supported by funding from an Australian
Criminology Research Council Grant (CRG 08/15-16) and an Australian Research Council
Discovery Project Grant (DP170101433).
2
Abstract
Image-based sexual abuse (IBSA) involves three key behaviors: the non-consensual taking or
creation of nude or sexual images; the non-consensual sharing or distribution of nude or
sexual images; and threats made to distribute nude or sexual images. IBSA is becoming
increasingly criminalized internationally, representing an important and rapidly developing
cybercrime issue. This paper presents findings of the first national online survey of self-
reported lifetime IBSA perpetration in Australia (n = 4,053), with a focus on the extent,
nature, and predictors of perpetration. Overall, 11.1% (n = 411) of participants self-reported
having engaged in some form of IBSA perpetration during their lifetime, with men
significantly more likely to report IBSA perpetration than women. With regard to the nature
of perpetration, participants reported targeting men and women at similar rates, and were
more likely to report perpetrating against intimate partners or ex-partners, family members
and friends than strangers or acquaintances. Logistic regression analyses identified that
males, lesbian, gay or bisexual participants, participants with a self-report disability,
participants who accepted sexual image-based abuse myths, participants who engaged in or
experienced sexual self-image behaviors, and participants who had a nude or sexual image of
themselves taken, distributed, or threatened to be distributed without their consent were more
likely to have engaged in some form of IBSA perpetration during their lifetime.
Keywords: Image-Based Sexual Abuse, Cybercrime, Perpetration, Revenge Pornography,
Victimization, Non-consensual Pornography
3
Introduction
Image-Based Sexual Abuse (IBSA) refers to the taking, distributing, and/or making of
threats to distribute, a nude or sexual image without a person’s consent (see e.g. DeKeseredy
& Schwartz 2016; McGlynn & Rackley 2017; McGlynn, Rackley & Houghton 2017; Powell
& Henry 2016; Powell & Henry 2017; Powell, Henry & Flynn, 2018).
1
While some research
has drawn on alternative terms such as ‘revenge pornography’ (e.g. CCRI, 2014; Hall &
Hearn 2017; Salter & Crofts 2015) or ‘non-consensual pornography’ (e.g. Citron & Franks
2014; Poole 2015; Suzor, Seignior & Singleton 2017; Walker & Sleath 2017), such studies
have tended to focus primarily on the non-consensual distribution of nude or sexual images,
without consideration of the related non-consensual taking or creation of such images and/or
threats to distribute them. Furthermore, as Powell, Henry and Flynn (2018) have described,
far beyond the popular understanding of an image shared by a jilted ex-lover, IBSA occurs in
a range of contexts including: (1) relationship retribution, where a perpetrator misuses the
nude or sexual images of a current or former intimate partner in order to seek revenge or
cause distress following a relationship breakdown; (2) sextortion, where the perpetrator
threatens to create or distribute an intimate image of another person in order to obtain further
images, money, or unwanted sexual acts, regardless of whether or not the image exists; (3)
voyeurism, where perpetrators seek to create or distribute images as a form of sexual
gratification or social status building, including (but not limited to) ‘upskirting’ and ‘down-
blousing’
2
; (4) sexploitation, where the primary goal is to obtain monetary benefits through
the trade of non-consensual nude or sexual imagery; and (5) sexual assault, where
perpetrators and/or bystanders record sexual assaults and rapes on mobile phones or other
devices and/or distribute those images via mobile phone or online (see also Powell & Henry
2017).
1
We use the term image to capture both nude or sexual photos and videos.
2
Upskirting refers to the act of someone taking an image up a victim’s skirt or dress. Down-blousing refers to the
act of someone taking a photo down the victim’s shirt and/or of the victim’s cleavage.
4
An emerging body of research has sought to examine the extent and nature of IBSA
victimization. Some studies have found that, similar to other forms of intimate aggression,
women are more commonly the targets of IBSA as compared to men (e.g. Wood et al. 2015),
although other studies have found either similar victimization rates among both men and
women (e.g. Lenhart, Ybarra & Feeney-Price 2016; Powell & Henry 2016; Reed, Tolman &
Ward 2016), or somewhat higher victimization rates among men (e.g. Borrajo, Gámez-
Guadix & Calvete 2015; Priebe & Svedin 2012). In addition to examining the gendered
nature of IBSA, several studies have reported differing rates of IBSA victimization according
to sexuality, with minority participants more likely to report a person having shared a sexual
image of them without permission as compared to heterosexual participants (e.g. Lenhart,
Ybarra & Feeney-Price 2016; Priebe & Svedin 2012). Though research into IBSA
victimization is growing, by comparison there remains a dearth of research that has examined
the perpetration of IBSA with regards to its extent, nature, and potential predictors.
The study described in this article is the first to undertake a comprehensive
investigation that includes all three subsets of IBSA perpetration; that is, the non-consensual
taking or creation of a nude or sexual image; the non-consensual sharing or distribution of a
nude or sexual image; and threats made to distribute a nude or sexual image. Reporting on an
online panel survey of 4,053 Australian residents (aged 16 to 49 years), we examine IBSA as
it is increasingly conceptualized in the international literature, presenting original analyses of
self-reported perpetration behaviors and potential predictors. The article begins by briefly
summarizing the conceptualization of IBSA, before providing a more detailed review of the
(limited) international literature on IBSA perpetration, and subsequent aims of the study
reported here.
5
Literature Review
Prevalence of IBSA Perpetration
The majority of quantitative research in the broader field on technology and intimate
relationships has been on ‘sexting’ (the sending and/or receiving of nude or sexual images or
texts) among adolescents (see e.g. Crofts, Lee, McGovern & Miliovojevic 2015; Patrick,
Heywood, Pitts & Mitchell 2015; Stanley et al. 2018; Villacampa 2017). While most of this
research has focused on consensual forms of sexting, some studies have also sought to
investigate the prevalence of ‘non-consensual sexting’, where images are either taken or
shared without consent. For example, Patrick, Heywood, Pitts and Mitchell (2015) found that
10% of school students had sent ‘a sexually explicit nude or nearly nude photo or video of
someone else.’ Similarly, in a 2014 survey with undergraduate psychology students,
Strohmaier, Murphy and DeMatteo (2014) found that 11% of participants reported that a sext
had been sent on without their consent while they were deemed to be a minor (i.e. under the
age of 18). Crofts, Lee, McGovern and Miliovojevic‘s study (2015) found a slightly lower
rate with 6% of participants aged under 18 years reporting sending an image to another
person without consent. In Crofts et al.’s (2015) study, 20% of adolescents surveyed reported
that they had shown another person an image without the depicted person‘s consent, such as
by displaying the image on their mobile phone screen.
To date, few empirical studies have explored non-consensual behaviors among adult
populations. This is despite some research (e.g. Borrajo & Gámez-Guadix 2015) indicating
that technology-based abuse between partners occurs more often between young adults,
rather than adolescents or pre-adolescents. While it is difficult to synthesize the findings of
the few existing studies, given the different sample sizes, definitions, and instruments used,
collectively these studies indicate an approximate range of between 12% and 30% of
participants who report sharing nude or sexual images without the consent of the person
depicted in the video or photo. In their Australian ‘sexting’ study, Crofts et al. (2015) found
6
that among participants aged over 19 years (n = 422), 16% had shown a sexual image to
another person who was not meant to see it, 4% had shared the image online, and 4% had
forwarded the image via MMS or email. While the authors differentiated between different
forms of ‘sharing,‘ in other studies it is unclear what ‘sharing‘ means. For instance, in an
Italian study on sexting and dating violence among a sample of 13 to 30 year olds (n =
1,334), the authors found that 13% of participants had shared a sexual image without another
person’s consent at least once (Morelli, Bianchi, Baiocco, Pezzuiti & Chirumbolo 2016). In a
study on ‘technology-based coercion’ (n = 795), Thompson and Morrison (2013) found that
16% of men had shared a sexually suggestive message or picture of someone without their
consent (n = 795). Another sexting study of American adults aged between 21 and 75 years
(n = 5,805) by Garcia et al. (2016) found that more than one in five participants (23%)
reported sharing a ‘sexy’ photo with someone else without consent. It is important to note
that the Garcia et al. study focused on sharing photos, whereas the Thompson and Morrison
study did not differentiate between sexual images (videos and/or photos) or text, and in all of
these examples, participants were asked only about sharing or distributing sexual material
without consent, and not about other related image-taking or threat behaviors.
Characteristics of IBSA Perpetrators
Overall, existing studies have rarely reported on gender, sexuality, or other differences
in relation to self-reported perpetration items. An exception is the aforementioned study by
Garcia et al. (2016), which found that more men (25%) than women (20%) had ‘shared a
received sexy photo with someone else.’ This study also found that gay men were twice as
likely as lesbian women to share such images (Garcia et al. 2016). Meanwhile, qualitative
studies on IBSA perpetration have further sought to examine the potential gendered nature of
these behaviors. In one study, Hall and Hearn (2017) examined the online comments that
accompanied the postings of non-consensual nude or sexual images on a popular ‘revenge
7
porn’ website. They found not only that most of the images posted were of women, and were
shared by men, but that the text accompanying many of the images occurred in a context of
homosocial interaction in which men communicated normative masculine identities;
effectively reinstituting themselves as ‘real men’ within the mostly male community of
website users. In a second study, Uhl, Rhyner, Terrance and Lugo (2018) undertook content
analysis of 134 non-consensual photos contained on seven different websites. The researchers
found that 92% of victims depicted in the images were women. Moreover, for over a third of
the images (36%), the text accompanying the image revealed the perpetrator’s stated reason
for sharing the image with the most common being that the woman was an ‘ex’ (22%), the
woman was ‘hot’ or ‘sexy’ (22%), or the woman was ‘a slut’ (15%), or unfaithful (6%) (Uhl
et al. 2018). These findings indicate that while some IBSA perpetration may indeed be
motivated by ‘revenge,’ in other instances, it may be more related to other motivations such
as status-seeking among online male-dominated communities.
A small number of studies have begun to explore attitudes towards IBSA perpetration
such as through participant responses to hypothetical scenarios and the potential inclination
of participants to seek revenge through the non-consensual distribution of intimate images
(e.g. Bothamley & Tully 2018; Hudson, Fetro & Ogletree 2014; Pina, Holland & James 2017;
Scott & Gavin 2018). For instance, in a study by Scott and Gavin (2018), when confronted
with two hypothetical scenarios (one in which the perpetrator was a man and the victim a
woman, and the other in which the perpetrator was a woman and the victim a man),
participants perceived IBSA to be more serious when the perpetrator was a man. Meanwhile,
Hudson, Fetro and Ogletree’s (2014) survey (n = 697) of young adults aged between 18 to19
years, recorded a statistically significant difference in attitudes towards the sending and
receiving of explicit images among women and men. They found that male participants held
more favorable attitudes towards sexting, as well as higher intentions to send sexually explicit
images than female participants. Finally, Pina, Holland and James (2017) found an
8
association between higher levels of ambivalent sexism, Machiavellianism, narcissism, and
psychopathy with a greater self-reported proclivity to engage in non-consensual sexual image
sharing. Though low numbers of male participants precluded a gender analysis in the study,
the authors note that the findings reflect those of broader sexual violence research in which
psychological characteristics such as sexism and narcissism are frequently found to be
associated both with perpetration, and to be higher among male participants.
In summary, there is currently only limited research into the extent and nature of IBSA
perpetration, and indeed the characteristics of IBSA perpetrators. Moreover, in all of the
survey studies mentioned here, some of which only included one IBSA perpetration item, the
primary focus is on the non-consensual sharing or distribution of nude or sexual images (or in
some cases, other text-based sexual material), and not on related forms of sexual image-based
perpetration. The limited number of studies conducted to date indicate a potential role of
gender and/or sexuality as predictors of IBSA perpetration. However, it is unclear what role
other factors (such as attitudinal and/or experiential characteristics) might have on the
likelihood of an individual engaging in perpetration behavior, and indeed how such factors
might interact with the key characteristics of interest as identified in the literature.
The Current Study
The current study represents one component of a larger research project into IBSA
victimization and perpetration. The project, funded by an Australian Criminology Research
Council Grant, aimed to examine the extent, nature, and impacts of IBSA, as well as the
effects of existing and proposed legislative reform in Australia. The project draws on a mixed
methods approach, including empirical data on victimization and perpetration, as well as the
experiences of key stakeholders (e.g. police, legal services, women’s information services,
domestic violence services, disability services, and sexual assault services). Here, we report
on findings from the perpetration component of this research project. In particular, this
9
component sought to investigate IBSA perpetration among a community sample of
Australian residents aged 16 to 49 years. This age range was selected for the study because it
represents both those at highest risk for sexual- and family- related violence (ABS 2012), as
well as the majority of mobile and Internet users (ACMA 2011). The current study therefore
aimed to examine: (1) the extent of self-reported IBSA perpetration, (2) the nature of self-
reported IBSA perpetration, and (3) the predictors of self-reported IBSA perpetration.
Method
Recruitment and Participants
Research Now, an online panel provider, invited 113,294 Australian residents to
participate in the research and 4,303 responded, representing a 3.8% response rate.
3
Of these
participants, 221 were excluded from the data analysis because of incomplete responses
regarding self-reported IBSA perpetration items and 29 were excluded because they
identified as transgender or non-binary gender (unfortunately, the number in this category
was insufficient for data analysis). The final sample comprised 4,053 Australian residents,
2,298 females and 1,755 males, with an average age of 34.55 years (SD = 8.95, range 16 to
49 years). The demographic characteristics of the final sample are presented in Table 1.
---Table 1 about here---
Overall, our sample compared favorably with the Australian Census on markers such as
gender (57% vs. 52% female) and indigeneity (97% vs. 97% non-Aboriginal), and languages
spoken other than English (16% vs. 21% other languages) (ABS 2017). However, it was
overrepresented by Australian born participants (88% vs. 74% Australian born), participants
with a high level of education (76% vs. 66% at least one non-school qualification), and
3
The low response rate was the consequence of difficulties recruiting men aged between 16 and 24 years.
10
participants with a lesbian, gay or bisexual (LGB) sexuality (88% vs. 97% heterosexual). It
was also slightly underrepresented by participants with a self-reported disability (12% vs.
18% assistance required) (ABS 2017).
All participants were informed that the purpose of the research was to examine attitudes
and experiences of sex, technology, and relationships. The research was approved by an
institutional ethics committee and two police ethical committees, following guidelines as
prescribed by the Australian National Statement on Ethical Conduct in Human Research.
Measures
In light of the very limited quantitative research into IBSA, a survey instrument was
developed by the research team that comprised a range of items including those pertaining to:
(1) demographic characteristics; (2) sexual image-based abuse myth acceptance; (3) online
dating behaviors; (4) sexual self-image behaviors; (5) IBSA victimization; (6) IBSA
perpetration; and (7) the nature of IBSA perpetration. The measures used for the purpose of
the current study are described below.
Demographic Characteristics
Participants were asked to complete the following items: gender (female, male),
sexuality (heterosexual, bisexual, gay, lesbian), age (in years), nativity (Australian born,
overseas born), languages spoken other than English (English only, other languages),
indigeneity (non-Aboriginal, Aboriginal), education (high school or less, trade certificate,
university/college, postgraduate/advanced degree), and disability (no assistance required,
assistance required). The survey instrument included three items regarding disability, one
relating to assistance with self-care activities, one relating to assistance with body movement
activities, and one relating to assistance with communication activities (no, yes sometimes,
yes always). A composite variable was first created by summing the number of ‘yes’
11
responses (yes sometimes and yes always) to the three original items (M = 0.27, SD = 0.80,
range 0 to 3). This composite variable was then used to create a dichotomous ‘disability’
variable for the purpose of data analysis.
Sexual Image-Based Abuse Myth Acceptance
The sexual image-based abuse myth acceptance (SIAMA) scale was developed by the
research team and modelled on rape myth acceptance (Payne, Lonsway & Fitzgerald 1999;
see also Powell & Webster 2018 for a review). It contains 18 items and asks participants
about their attitudes towards minimizing/excusing the harms and blaming the victims of
IBSA (see Table A1 of the Appendix). All items were rated on the same 7-point Likert scale
where 1 = ‘strongly disagree’ and 7 = ‘strongly agree’ (no labels were provided for points 2,
3, 4, 5, and 6 on the scale). The SIAMA scale has been found to have two components: the
‘minimize/excuse’ component contains 12 items (M = 2.30, SD = 1.33, range 1 to 7, α = .94)
and the ‘blame’ component contains six items (M = 3.70, SD = 1.65, range 1 to 7, α = .86).
The higher the score, the greater the minimizing/excusing of harms and blaming of victims of
IBSA.
Online Dating Behaviors
Participants self-reported whether they had ever engaged in or experienced nine
different online dating behaviors (see Table A2 of the Appendix). All items were rated on the
same 5-point Likert scale where 0 = ‘never,’ 1 = ‘rarely,’ 2 = ‘sometimes,’ 3 = ‘often,’ and 4
= ‘frequently.’ Principal component analysis with varimax rotation was performed to
examine the underlying structure of the nine items and one component was identified that
accounted for 67.55% of variance (α = .94). An average composite variable was first created
for the nine items (M = 0.82, SD = 0.91, range 0 to 4). This average composite variable was
12
then used to create a dichotomous ‘online dating behaviors’ variable for the purpose of data
analysis.
Sexual Self-Image Behaviors
Participants self-reported whether they had ever engaged in or experienced 11 different
sexual self-image behaviors (see Table A2 of the Appendix). Again, all items were rated on
the same 5-point Likert scale where 0 = ‘never,’ 1 = ‘rarely,’ 2 = ‘sometimes,’ 3 = ‘often,’
and 4 = ‘frequently.’ Principal component analysis with varimax rotation was performed and
one component was identified that accounted for 74.37% of variance (α = .97). An average
composite variable was first created for the 11 items (M = 0.52, SD = 0.83, range 0 to 4). This
average composite variable was then used to create a dichotomous ‘sexual self-image
behaviors’ variable for the purpose of data analysis.
IBSA Victimization
Participants self-reported whether they had ever (since 16 years of age) had a nude or
sexual image of themselves taken, distributed, and/or threatened to be distributed without
their consent. Participants responded to nine items relating to the content of the image for
each of the three IBSA victimization contexts (taken, distributed, and threatened), using a
dichotomous (yes, no) question format (see Table A3 of the Appendix). Three composite
variables were first created by summing the number of ‘yes’ responses to the nine content
items in each of the three contexts (IBSA victimization [taken]: M = 0.58, SD = 1.54, range 0
to 9; IBSA victimisation [distributed]: M = 0.34, SD = 1.25, range 0 to 9; IBSA victimization
[threatened]: M = 0.30, SD = 1.20, range 0 to 9). For the purpose of data analysis, these
composite variables were then used to create three dichotomous variables: ‘IBSA
13
victimization (taken)’; ‘IBSA victimization (distributed)’; and ‘IBSA victimization
(threatened).’
IBSA Perpetration
Participants self-reported whether they had ever (since 16 years of age) taken,
distributed, and/or threatened to distribute a nude or sexual image of another person without
their consent. Participants responded to eight items relating to the content of the image for
each of the three IBSA perpetration contexts (taken, distributed, and threatened), using a
dichotomous (yes, no) question format (see Table A3 of the Appendix). Three composite
variables were first created by summing the number of ‘yes’ responses to the eight content
items in each of the three contexts (IBSA perpetration [taken]: M = 0.32, SD = 1.26, range 0
to 8; IBSA victimization [distributed]: M = 0.27, SD = 1.19, range 0 to 8; IBSA victimization
[threatened]: M = 0.21, SD = 1.07, range 0 to 8). For the purpose of the descriptive and chi-
square analyses, these composite variables were then used to create three dichotomous IBSA
perpetration variables: ‘IBSA perpetration (taken)’; ‘IBSA perpetration (distributed)’; and
‘IBSA perpetration (threatened).’ An additional composite variable was created by summing
the number of ‘yes’ responses to all 24 content items (M = 0.81, SD = 3.28, range 0 to 24),
and this composite variable was used to create a dichotomous ‘IBSA perpetration’ variable
for the purpose of the logistic regression analyses.
Nature of IBSA Perpetration
Participants who self-reported taking, distributing, and/or threatening to distribute a
nude or sexual image of another person without their consent were asked to complete nature
items regarding their most recent IBSA perpetration experience. These items included victim
gender (i.e. the gender of the person/people in the nude or sexual image: female, male, female
and male, don’t know), and perpetrator-victim relationship (i.e. their relationship to the
14
person/people in the nude or sexual image: intimate partner or ex-partner, family member,
friend [known face to face], friend [known online only], work colleague or ex-work
colleague, acquaintance, stranger, or don’t know).
Data Analysis
Data analyses were conducted on the unweighted sample in three stages using IBM
SPSS Statistics Version 24. First, descriptive and chi-square analyses, with phi as a measure
of effect size, were performed to examine the extent of self-reported IBSA perpetration. Chi-
square analyses were performed to determine whether or not there were differences in IBSA
perpetration (taken, distributed, and threatened) according to participant gender and sexuality.
Additional chi-square analyses using 3-way crosstabulations were performed to determine
whether or not there were any significant interactions for gender and sexuality with regard to
the extent of IBSA perpetration. Second, descriptive and chi-square analyses, with phi and
Cramer’s V as measures of effect size, were performed to examine the nature of self-reported
IBSA perpetration. Chi-square analyses were performed to determine whether or not there
were differences in victim gender and perpetrator-victim relationship according to participant
gender and sexuality. Additional chi-square analyses using 3-way crosstabulations were then
performed to determine whether or not there were any significant interactions for gender and
sexuality with regard to the nature of IBSA perpetration.
Third, logistic regression analyses were performed to examine the predictors of self-
reported IBSA perpetration. Hosmer, Lemeshow, and Studivant’s (2013) seven step
‘purposeful selection’ model building process was used to examine the relationship between
15 participant characteristics and the dichotomous IBSA perpetration variable. The 15
participant characteristics comprised eight demographic characteristics (gender, sexuality,
age, nativity, languages spoken other than English, indigeneity, education, and disability),
two attitudinal characteristics (minimize/excuse and blame), and five experiential
15
characteristics (online dating behaviors, sexual self-image behaviors, IBSA victimization
[taken], IBSA victimization [distributed], and IBSA victimization [threatened]). The seven
steps comprised: 1) performing univariable analyses to identify participant characteristics
with p-values less than 0.25; 2) entering these participant characteristics into an initial model
and removing any characteristics with p-values less than 0.05; 3) determining whether any
removed participant characteristics should be re-entered into the model; 4) entering
participant characteristics originally excluded into the model to determine whether any have
p-values less than 0.05; 5) creating a main effects model; 6) identifying any significant
interaction terms and creating the final model; and 7) testing the adequacy and fit of the final
model (Hosmer et al 2013). A p-value of 0.25 was used during Step 1 because research
suggests use of a more traditional significance level (e.g. a p-value of 0.05) may fail to
identify important predictor variables (Bendel & Afifi 1977; Mickey & Greenland 1989).
Assumption testing was performed prior to assessment of the initial and final models. This
testing revealed that the sample size and multicolinearity assumptions were not violated.
With regard to outliers and influential cases, although there were outliers in the initial and
final models, assessment using Cook’s distance revealed that these outliers did not have an
undue influence on the models (Tabachnick & Fidell 2013).
Results
Extent of IBSA Perpetration
Overall, 11.1% of participants self-reported engaging in one or more of the 24 IBSA
perpetration behaviors during their lifetime. Behaviors involving the taking of a nude or
sexual image (8.7%) were the most common, followed by behaviors involving the
distribution of a nude or sexual image (6.4%), and behaviors involving threats to distribute a
16
nude or sexual image (4.9%). Table 2 presents the lifetime prevalence of self-reported IBSA
perpetration behaviors.
---Table 2 about here---
A series of chi-square analyses were performed to examine whether or not there were
significant differences in the lifetime prevalence of self-reported IBSA perpetration by
participant gender and sexuality. These analyses revealed that male participants were more
likely than female participants to self-report ever taking (12.0% vs. 6.2%), distributing (9.1%
vs. 4.4%), and/or threatening to distribute (7.0% vs. 3.3%) a nude or sexual image of another
person without their consent: χ2(1, n = 4,053) = 42.01, p < .001, φ = .10, χ2(1, n = 4,053) =
36.87, p < .001, φ = .10, and χ2(1, n = 4,053) = 28.44, p < .001, φ = .08 respectively. They
also revealed that LGB participants were more likely than heterosexual participants to self-
report ever taking (17.2% vs. 7.5%), distributing (13.7% vs. 5.4%) and/or threatening to
distribute (9.5% vs. 4.3%) a nude or sexual image of another person without their consent:
χ2(1, n = 4,053) = 49.62, p < .001, φ = .11, χ2(1, n = 4,053) = 47.59, p < .001, φ = .11, and
χ2(1, n = 4,053) = 24.23, p < .001, φ = .08 respectively. Additional chi-square analyses
confirmed that the significant findings for participant gender were consistent across the
categories of sexuality (heterosexual, LGB); that the significant findings for participant
sexuality were consistent across the categories of gender (female, male); and that there were
no significant interaction effects for participant gender and sexuality for the three IBSA
perpetration variables (taken, distributed, and threatened).
Nature of IBSA Perpetration
Of the 352 participants who self-reported ever taking a nude or sexual image, 43.5%
targeted females and 37.2% targeted males. Similarly, of the 259 participants who self-
17
reported ever distributing a nude or sexual image, 36.7% targeted females and 35.5% targeted
males. Finally, of the 198 participants who self-reported ever threatening to distribute a nude
or sexual image, 38.9% targeted females and 28.3% targeted males. The remaining
participants targeted both females and males, or did not know the gender of their victim.
Table 3 presents the lifetime prevalence of self-reported IBSA perpetration by victim gender
and perpetrator-victim relationship.
---Table 3 about here---
With regard to perpetrator-victim relationship, most participants who self-reported ever
taking a nude or sexual image targeted an intimate partner or ex-partner, a family member, or
a friend. For IBSA perpetration (taken), 40.1% of participants targeted an intimate partner or
ex-partner, 20.5% targeted a friend, and 20.2% targeted a family member. For IBSA
perpetration (distributed), 29.8% of participants targeted a friend, 22.8% targeted an intimate
partner or ex-partner, and 20.1% targeted a family member. Finally, for IBSA perpetration
(threatened), 34.8% of participants targeted a friend, 24.7% targeted a family member, and
22.2% targeted an intimate partner or ex-partner. The remaining participants targeted a work
colleague or ex-work colleague, an acquaintance, a stranger, or did not know the nature of
their relationship with the victim.
A series of chi-square analyses were performed to examine whether or not there were
significant differences with regard to the nature of self-reported perpetration (i.e., victim
gender and perpetrator-victim relationship) by participant gender and sexuality. These
analyses revealed that LGB participants were more likely than heterosexual participants to
take, distribute, and threaten to distribute a nude or sexual image of a male and less likely
than heterosexual participants to take, distribute, and threaten to distribute a nude or sexual
image of a female, χ2(3, n = 352) = 18.74, p < .001, φv = .23, χ2(3, n = 259) = 11.86, p = .008,
18
φv = .21, and χ2(3, n = 198) = 10.87, p = .012, φv = .23 respectively. There were no
significant differences for participant gender with regard to victim gender or perpetrator-
victim relationship, and no significant differences for participant sexuality with regard to
perpetrator-victim relationship. Additional chi-square analyses confirmed that the non-
significant findings for participant gender with regard to perpetrator-victim relationship were
consistent across the categories of sexuality (heterosexual, LGB), and that the non-significant
findings for participant sexuality with regard to perpetrator-victim relationship were
consistent across the categories of gender (female, male). However, these additional analyses
also revealed significant interaction effects for participant gender and sexuality with regard to
the three IBSA perpetration variables. This included that there were significant differences
for participant sexuality with regard to victim gender for male participants, but there were no
significant differences for participant sexuality with regard to victim gender for female
participants. LGB males were more likely to take a nude or sexual image of a male (73.7%
vs. 25.5%) and less likely to do the same of a female (14.0% vs. 50.3%) than heterosexual
males, χ2(3, n = 210) = 41.98, p < .001, φv = .45. LGB males were also more likely to
distribute a nude or sexual image of a male (63.8% vs. 28.6%) and less likely to do the same
of a female (14.9% vs. 42.9%) than heterosexual males, χ2(3, n = 159) = 19.48, p < .001, φv =
.35. Finally, LGB males were more likely to threaten to distribute a nude or sexual image of a
male (61.3% vs. 24.2%) and less likely than heterosexual males to do the same of a female
(16.1% vs. 41.8%) than heterosexual males, χ2(3, n = 122) = 14.86, p = .002, φv = .35.
Predictors of IBSA Perpetration
Logistic regression analyses were performed to examine the relationship between 15
participant characteristics and the lifetime prevalence of self-reported IBSA perpetration.
Eight participant characteristics were demographic and included: gender, sexuality, age,
nativity, languages spoken other than English, indigeneity, education, and disability. Two
19
participant characteristics were attitudinal: minimize/excuse and blame. The remaining five
were experiential: online dating behaviors, sexual self-image behaviors, IBSA victimization
(taken), IBSA victimization (distributed), and IBSA victimization (threatened).
Univariable Analyses
A series of chi-square and t-test analyses were performed to identify which of the 15
participant characteristics to include in the initial model, and 12 characteristics were
identified. Five participant characteristics were demographic: gender, χ2(1, n = 4,053) =
42.44, p < .001, φ = .10, sexuality, χ2(1, n = 4,053) = 54.63, p < .001, φ = .12; age, t(527) = -
4.82, p < .001, d = .24; indigeneity, χ2(1, n = 4,046) = 82.78, p < .001, φ = .14; and disability,
χ2(1, n = 4,049) = 502.78, p < .001, φ = .35. Two participant characteristics were attitudinal:
minimize/excuse, t(458) = 15.76, p < .001, d = .92; and blame, t(533) = 8.52, p < .001, d =
.43. Five were experiential: online dating behaviors, χ2(1, n = 4,053) = 107.44, p < .001, φ =
.16; sexual self-image behaviors, χ2(1, n = 4,053) = 230.34, p < .001, φ = .24; IBSA
victimization (taken), χ2(1, n = 4,053) = 640.31, p < .001, φ = .40; IBSA victimization
(distributed), χ2(1, n =4,053) = 941.22, p < .001, φ = .48; and IBSA victimization
(threatened), χ2(1, n = 4,053) = 1133.20, p < .001, φ = .53. Three participant characteristics,
all demographic, did not reach the 0.25 level of significance and were therefore excluded
from the initial model: nativity, χ2(1, n = 4,047) = 0.59, p = .444, φ = .01; languages spoken
other than English, χ2(1, n = 4,051) = 0.06, p = .815, φ = .00; and education, χ2(3, n = 4,051)
= 0.49, p = .920, φv = .01. Table 4 presents frequencies and descriptives for the 15 participant
characteristics by lifetime prevalence of self-reported IBSA perpetration.
---Table 4 about here---
20
Logistic Regression Analyses
The initial model contained 12 participant characteristics and was statistically
significant, χ2(12, n = 4,042) = 944.14, p < .001. It correctly classified 93.4% of cases (98.8%
with no self-reported IBSA perpetration, 45.9% with self-reported IBSA perpetration) and
explained between 20.8% (Cox & Snell R square) and 43.3% (Nagelkerke R square) of
variance in the lifetime prevalence of self-reported IBSA perpetration. Four non-contributing
participant characteristics were removed from the model during Step 2, and no additional
characteristics or interaction effects were found to contribute to the creation of a
parsimonious model during Steps 3 to 7. The final model therefore contained eight participant
characteristics and was statistically significant, χ2 (8, n = 4,049) = 935.29, p < .001. It
correctly classified 93.4% of cases (98.7% with no self-reported IBSA perpetration, 46.1%
with self-reported IBSA perpetration) and explained between 20.6% (Cox & Snell R square)
and 42.9% (Nagelkerke R square) of variance in the lifetime prevalence of self-reported
IBSA perpetration. A summary of the initial and final models is presented in Table 5.
---Table 5 about here---
Overall three demographic characteristics were significant predictors of self-reported
IBSA perpetration. Male participants had 78% greater odds than female participants, and
LGB participants had 54% greater odds than heterosexual participants, to self-report having
engaged in IBSA perpetration (OR = 1.78, 95% CI = 1.37 to 2.31 and OR = 1.54, 95% CI =
1.11 to 2.13 respectively), controlling for other participant characteristics in the model.
Furthermore, participants with a self-reported disability had 106% greater odds than
participants without a self-reported disability to self-report having engaged in IBSA
perpetration (OR = 2.06, 95% CI = 1.49 to 2.86), controlling for other participant
characteristics in the model. One attitudinal characteristic was a significant predictor of self-
21
reported IBSA perpetration. A one-point increase in participants’ blame scores was
associated with 15% greater odds of self-reporting having engaged in IBSA perpetration (OR
= 1.15, 95% CI = 1.06 to 1.25), controlling for other participant characteristics in the model.
Finally, four experiential characteristics were significant predictors of self-reported
IBSA perpetration. Participants who had engaged in or experienced sexual self-image
behaviors had 210% greater odds than participants who had not engaged in or experienced
sexual self-image behaviors to self-report having engaged in IBSA perpetration (OR = 310,
95% CI = 2.19 to 4.38), controlling for other participant characteristics in the model.
Furthermore, participants who had a nude or sexual image of themselves taken without their
consent had 178% greater odds than participants who had not experienced this subtype of
victimization to self-report having engaged in IBSA perpetration (OR = 2.78, 95% CI = 2.05
to 3.77), controlling for other participant characteristics in the model. Similarly, participants
who had a nude or sexual image of themselves distributed without their consent had 135%
greater odds than participants who had not experienced this subtype of victimization to self-
report having engaged in IBSA perpetration, and participants who had a nude or sexual image
of themselves threatened to be distributed without their consent had 367% greater odds than
participants who had not experienced this subtype of victimization to self-report having
engaged in IBSA perpetration (OR = 2.35, 95% CI = 1.60 to 3.44 and OR = 4.67, 95% CI =
3.17 to 6.87 respectively), controlling for other participant characteristics in the model.
Discussion and Implications
This article has presented data on the first international study that specifically
investigates the extent and nature of IBSA perpetration in a community sample of Australian
residents (aged 16 to 49 years). With regard to the extent and nature of IBSA perpetration, we
found that 1 in 10 participants self-reported having engaged in at least one of the 24 IBSA
behaviors surveyed. Males were significantly more likely than females to self-disclose
22
engaging in IBSA perpetration behaviors. Perpetrators were similarly likely to report that
their victim was female as male, which is broadly consistent with the limited available
research into IBSA victimization by gender (Gámez-Guadix et al. 2015; Lenhart, Ybarra &
Price-Feeney 2016; Reed, Tolman & Ward 2016; Henry, Powell & Flynn, 2017). Sexuality
was a significant finding in relation to perpetration, with LGB participants more likely than
heterosexual participants to self-report perpetration of any IBSA perpetration behavior. These
findings suggest that as in other forms of intimate aggression, gender and sexuality are
particularly relevant in understanding the extent and nature of IBSA experiences.
Male and female perpetrators were also more likely to report that the victim was an
intimate partner or ex-partner, family member or friend, than a stranger or acquaintance. In
practical terms, the vast majority of victims were known to the perpetrator in some way, and
a substantial proportion of these (approximately half) were intimate partners or ex-partners.
What this suggests is that IBSA perpetration represents both a method of harassment or abuse
in the context of intimate relationships, and of harassment or abuse in non-partner contexts of
family relationships and friendships. Importantly, such a finding may indicate that multiple
strategies for responding to and preventing IBSA in these different relational contexts may be
needed.
With regard to potential predictors of IBSA perpetration, 8 of the original 15 participant
characteristics were found to relate to the lifetime prevalence of self-reported IBSA
perpetration. Demographic characteristics included gender, sexuality and disability, whereby
males, LGB participants and participants with a self-reported disability were more likely to
have engaged in some form of IBSA perpetration during their lifetime. Attitudinally,
participants who accepted sexual image-based abuse myths and blamed victims of IBSA for
the harms they experience were more likely to have engaged in IBSA perpetration. This
finding is broadly consistent with the much more developed field of attitudinal research as
related to other forms of sexual aggression, including rape myths, as discussed at the outset of
23
this paper (see Payne, Lonsway & Fitzgerald 1999; Pina et al. 2017; Powell & Webster
2018).
The most noteworthy finding reported in this study is the strength of the relationship
between participants having experienced IBSA victimization and reporting engaging in IBSA
perpetration themselves. Experiential characteristics included sexual self-image behaviors
and three measures of IBSA victimization, whereby participants who engaged in or
experienced sexual self-image behaviors, as well as participants who had a nude or sexual
image of themselves taken, distributed, and/or threatened to be distributed without their
consent were more likely to have engaged in some form of IBSA perpetration during their
lifetime. Care should be taken, however, to avoid inferring a causal relationship between
these characteristics. Rather, this study adds substantially to related findings on IBSA
victimization, which suggest that both of these experiences occur within a broader context of
sexual image-taking and/or sharing. It is vital to note that self-reported rates of both IBSA
perpetration (1 in 10 reported here), and victimization (1 in 5, see Henry, Powell & Flynn,
2017) are much lower than participation in sexual selfie behaviors overall. In other words,
although engaging in sexual self-imagery behaviors, perhaps unsurprisingly, increases the
odds of either perpetration or victimization experiences (as might be anticipated through the
increased potential for misuse of a nude or sexual image), a majority of participants engage in
sexual self-imagery behaviors and do not engage in IBSA perpetration, nor do they
experience IBSA victimization. Such findings have important implications for policy
responses and particularly prevention of IBSA.
Specifically, in seeking to prevent IBSA perpetration, the findings reported here
suggest that sexual self-image taking and exchanging is common among the 16 to 49 year-
olds surveyed. As such, prevention education which is foremost directed at abstaining from
sexual self-image taking and exchanging is at odds with the majority of participants’ lived
experiences. Rather, prevention may be better directed at, for example, education regarding
24
safer sexual self-image practices as well as the unethical (and indeed increasingly criminal)
nature of sharing nude or sexual images without another person’s consent. Importantly, such
education would appear to be relevant for the broader general community of young and
middle-aged adults, at least in the Australian context.
Despite the advances represented by the present study, there are some limitations that
should be mentioned to guide future research efforts. First, this study involved a non-
generalizable community sample recruited via an online panel. While online panel providers
make efforts at recruiting a diverse population, some research suggests that online panel
samples may under-represent some subgroups compared with others (AAPOR 2010). Indeed,
as acknowledged in this article, our sample was overrepresented by Australian born
participants, participants with a high level of education, and participants with a LGB
sexuality (according to Census data from the ABS 2017). Future research should thus seek to
validate these findings among a more representative sample of the general population and
further examine the experiences of different subgroups.
Second, in light of the approximately 1 in 10 IBSA perpetration rate reported here, even
our relatively large sample of over 4,000 participants precluded robust comparative analyses
of perpetrator subgroups (such as between genders within specific age groups, or by
sexuality), in which participant numbers became too low. As such, future research should
consider sampling strategies that provide sufficient participant numbers of self-reported IBSA
perpetrators so as to allow for such subgroup comparative analyses.
Third, while this survey has provided unique insights into the possible extent, as well as
the nature of IBSA perpetration, there are limitations to understanding the experiences,
perspectives, and motivations of IBSA perpetrators. In particular, if community education
and prevention initiatives are to be developed, an in-depth understanding of the contexts and
rationalizations of IBSA perpetration would be highly valuable to policy and program
development.
25
A fourth area for future research might thus comprise qualitative fieldwork with those
engaged in IBSA perpetration behaviors. Though, it should be noted, such research is
difficult to operationalize in practice and studies in other areas of sexual offending are often
limited to forensic samples due in part to the challenges in recruiting perpetrators. Given the
rate of disclosure in this anonymous online survey method, it is possible that an anonymous
online or digital interview method might be better suited than traditional face-to-face
interviews for addressing this important research gap. There is also a potentially important
role of those who may be sent a sexual or nude image of another person without their
consent, that has not been explored in detail in this study and represents a valuable area for
future research, particularly given the possible prevention opportunities of improving
‘bystander’ interventions in IBSA.
Finally, attitudes minimizing or excusing IBSA, and blaming the victim, were
significant in the overall model predicting IBSA perpetration, and as such, are worthy of
further investigation. In particular, it may be that attitudes towards IBSA share patterns in
common with other forms of sexual aggression (see e.g. Pina et al. 2017). Future research
might further examine the nature of potential inter-relationships between gender, IBSA
supportive attitudes, sexist ideology, proclivity, and self-reported perpetration behaviors.
Conclusion
The findings of this study suggest several important directions for policy, and indeed
prevention, in order to address IBSA perpetration. While much prevention education material
to date has been focused on school-age young people, this study suggests that prevention and
legal education may well benefit from being tailored for the specific contexts of different
perpetrator subgroups, with particular patterns of IBSA perpetration emerging according to
demographic characteristics, such as gender, sexuality, disability, and indigeneity. However,
given the overlap between IBSA perpetration and victimization reported here, it is important
26
to address referral and support information in ways that take care not to blame or minimize
the harms experienced by victims, while at the same time not excusing the behaviors of
perpetrators. Rather, both of these groups within the community may benefit from referral,
support, and legal information pathways. While no direction can be attributed to the
relationship between IBSA perpetration and victimization found here, it is possible that one
may be a response or reaction to the other. In the Australian legal context in which
recognition and/or redress options for IBSA are variable across jurisdictions, and either not
effectively utilized or their effectiveness remains somewhat unknown, it may well be the case
that neither victims nor perpetrators of IBSA are fully aware of the potential legal
consequences of the non-consensual taking, distributing, and/or making of threats to
distribute nude or sexual images.
There is a clear need to continue to examine and understand the varied contexts and
subgroups engaged in, and affected by, IBSA. As countries globally continue to grapple with
the extent, nature, impacts, and legal ramifications of IBSA, it is crucial that policy,
prevention, as well as legal and other supports, are targeted appropriately to those who need
them most.
27
References
Australian Bureau of Statistics (ABS) 2012. Personal Safety Survey. Canberra: Australian
Bureau of Statistics
Australian Bureau of Statistics (ABS) 2017. 2016 Census. Canberra: Australian Bureau of
Statistics. Available at: http://www.abs.gov.au/websitedbs/D3310114.nsf/Home/Census
Australian Communications and Media Authority (ACMA) 2011. Communications Report
2010–11 Series: The emerging mobile telecommunications service market in Australia.
Canberra: Commonwealth of Australia.
American Association for Public Opinion Research (AAPOR) 2010. Research synthesis
AAPOR report on online panels. Public Opinion Quarterly 74(4): 711-781.
Bendel, RB & Afifi AA 1977). Comparison of stopping rules in forward regression. Journal
of the American Statistical Association 72: 46-53.
Bothamley S & Tully R J 2018. Understanding revenge pornography: Public perceptions of
revenge pornography and victim blaming. Journal of Aggression, Conflict and Peace
Research, 10, 1-10.
Citron DK & Franks MA 2014. Criminalizing revenge porn. Wake Forest Law Review 49(2):
345–391.
Crofts T, Lee M, McGovern A & Milivojevic S 2015. Sexting and young people. London,
UK: Palgrave Macmillan.
Cyber Civil Rights Initiative (CCRI) 2014. End revenge porn: A campaign of the Cyber Civil
Rights Initiative. https://www.cybercivilrights.org/wp-
content/uploads/2014/12/RPStatistics.pdf
DeKeseredy WS, Schwartz MD 2016. Thinking sociologically about image-based sexual
abuse: The contribution of male peer support theory. Sexualization, Media, & Society
2(4): 2374623816684692.
28
Gámez-Guadix M, Almendros C, Borrajo E & Calvete E 2015. Prevalence and association of
sexting and online sexual victimization among Spanish adults. Sexuality Research and
Social Policy 12(2): 145-154.
Garcia, JR, Gesselman AN, Siliman SA, Perry B. L, Coe K & Fisher HE 2016. Sexting
among singles in the USA: Prevalence of sending, receiving, and sharing sexual
messages and images. Sexual Health 13(5): 428-435.
Hall M & Hearn J 2017. Revenge pornography and manhood acts: a discourse analysis of
perpetrators’ accounts. Journal of Gender Studies, OnlineFirst, 1-13.
Henry N, Powell A & Flynn A 2017. Not just ‘revenge pornography’: Australians’
experiences of image-based abuse. A summary Report. Melbourne: RMIT University.
Hosmer, DW, Jr Lemeshow S & Sturdivant RX 2013. Applied logistic regression (3rd ed).
Hoboken, NJ: John Wiley & Sons, Inc.
Hudson H, Fetro J & Ogletree R 2014. Behaviorial indicators and behaviors related to sexting
among undergraduate students. American Journal of Health Education 45(3): 183-195.
Lenhart A, Ybarra M & Price-Feeney M 2016. Online harassment, digital abuse and
cyberstalking in America.
https://www.datasociety.net/pubs/oh/Online_Harassment_2016.pdf
Payne DL, Lonsway KA & Fitzgerald LF 1999. Rape myth acceptance: Exploration of its
structure and its measurement using the Illinois Rape Myth Acceptance Scale. Journal
of Research in Personality 33(1): 27-68.
McGlynn C & Rackley E 2017. Image-based sexual abuse. Oxford Journal of Legal Studies
37(3): 534-561.
McGlynn C, Rackley E & Houghton R 2017. Beyond ‘revenge porn’: The continuum of
image-based sexual abuse. Feminist Legal Studies 25(1): 25-46.
Mickey J & Greenland S 1989. A study of the impact of confounder-selection criteria on
effect estimation. American Journal of Epidemiology 129: 125-137.
29
Morelli M, Bianchi D, Baiocco R, Pezzuti L & Chirumbolo A 2016. Sexting, psychological
distress and dating violence among adolescents and young adults. Psicothema 28(2):
137-142.
Patrick K, Heywood W, Pitts M & Mitchell A 2015. Demographic and behavioural correlates
of six sexting behaviours among Australian secondary school students. Sexual Health
12(6): 480-487.
Pina A, Holland JE & James M 2017. The malevolent side of revenge porn proclivity: Dark
personality traits and sexist ideology. International Journal of Technoethics 8(1): 30-
43.
Poole E 2015. Fighting back against non-consensual pornography. University of San
Fransisco Law Review, 49, 181-214.
Powell A & Henry N 2017. Sexual violence in a digital age. Basingstoke, UK: Palgrave
Macmillan.
Powell A & Henry, N 2016. Technology-facilitated sexual violence victimization: Results
from an online survey of Australian adults. Journal of interpersonal violence
(OnlineFirst).
Powell A, Henry, N & Flynn A 2018. Image-based sexual abuse, in DeKeseredy WS,
Rennison CM & Hall-Sanchez AK (eds), The Routledge international handbook of
violence studies. New York: Routledge: 305-315.
Powell A & Webster K 2018. Cultures of gendered violence: An integrative review of
measures of attitudinal support for violence against women. Australian & New Zealand
Journal of Criminology 51(1): 40-57.
Priebe G & Svedin CG 2012. Online or off-line victimization and psychological well-being: a
comparison of sexual-minority and heterosexual youth. European Child & Adolescent
Psychiatry 21(10): 569-582.
30
Reed L, Tolman R & Ward M 2016. Snooping and sexting: Digital media as a context for
dating aggression and abuse among college students. Violence Against Women 22(13):
1556-1576.
Salter M & Crofts T 2015. Responding to revenge porn: Challenges to online immunity, in
Comella L & Tarrant S (eds), New views on pornography: Sexuality, politics and the
law. Santa Barbara, CA: Praeger Publishers: 233–253.
Scott AJ & Gavin J 2018. Revenge pornography: The influence of perpetrator-victim sex,
observer sex and observer sexting experience on perceptions of seriousness and
responsibility. Journal of Criminal Psychology 8(2): 162-172.
Stanley N, Barter C, Wood M, Aghtaie N, Larkins C, Lanau A & Överlien C 2018.
Pornography, sexual coercion and abuse and sexting in young people’s intimate
relationships: a European study. Journal of interpersonal violence, 33(19): 2919-2944.
Strohmaier H, Murphy M & DeMatteo D 2014. Youth sexting: Prevalence rates, driving
motivations, and the deterrent effect of legal consequences. Sexuality Research and
Social Policy 11: 245-255.
Suzor NP, Seignior B & Singleton J 2017. Non-consensual porn and the responsibilities of
online intermediaries. Melbourne University Law Review 40(3): 1057-1097.
Tabachnick BG & Fidell LS 2013. Using multivariate statistics (6th ed). Sydney: Pearson.
Thompson MP & Morrison DJ 2013. Prospective predictors of technology-based sexual
coercion by college males. Psychology of Violence 3(3): 233-246.
Uhl CA, Rhyner KJ, Terrance CA & Lugo NR 2018. An examination of nonconsensual
pornography websites. Feminism & Psychology 28(1): 50-68.
Villacampa C 2017. Teen sexting: Prevalence, characteristics and legal treatment.
International Journal of Law, Crime and Justice, 49, 10-21.
31
Walker K & Sleath E 2017. A systematic review of the current knowledge regarding revenge
pornography and non-consensual sharing of sexually explicit media. Aggression and
Violent Behavior 36: 9-24.
Wood M, Barter C, Stanley N, Aghtaie N & Larkins C 2015. Images across Europe: The
sending and receiving of sexual images and associations with interpersonal violence in
young people’s relationships. Children and Youth Services Review 59: 149-160.
32
Table 1
Demographic characteristics
%
n
Gender
Female
56.7
2298
Male
43.3
1755
Sexuality
Heterosexual
88.3
3577
LGB
11.7
476
Nativity
Australian born
76.7
3105
Overseas born
23.3
942
Languages spoken other than English
English only
83.5
3384
Other languages
16.5
667
Indigeneity
Non-Aboriginal
97.5
3943
Aboriginal
2.5
103
Education
High school or less
24.0
973
Trade certificate
25.9
1050
University/college
32.8
1330
Postgraduate/advanced degree
17.2
698
Disability
No assistance required
87.9
3559
Assistance required
12.1
490
Note. Six participants did not respond to the nativity item, two participants did not respond to the languages
other than English item, seven participants did not respond to the indigeneity item, two participants did not
respond to the education item, and four participants did not respond to the disability item.
33
Table 2
Lifetime Prevalence of Self-Reported IBSA Perpetration Behaviors
Taken
% (n)
Distributed
% (n)
Threatened
% (n)
IBSA perpetration
Where they were partially clothed or semi-nude
6.5 (263)
5.1 (208)
3.8 (152)
Where the person’s breasts, including their nipples, were visible
4.6 (186)
3.4 (139)
3.1 (125)
Where they were completely nude
4.4 (180)
3.6 (146)
2.6 (107)
Where the person’s genitals were visible
3.8 (155)
3.4 (138)
2.5 (101)
Where they were engaged in a sex act
3.6 (145)
3.2 (128)
2.5 (101)
Where they were showering, bathing or toileting
3.8 (152)
2.9 (116)
2.4 (99)
Which was up their skirt (e.g. ‘up-skirting’)
3.1 (125)
2.7 (108)
2.1 (85)
Which was of their cleavage (e.g. ‘down-blousing’)
2.7 (111)
2.6 (104)
2.2 (89)
Any nude/sexual images taken
8.7 (352)
6.4 (259)
4.9 (198)
34
Table 3
Lifetime Prevalence of Self-Reported IBSA Perpetration by Participant Gender and Perpetrator-Victim
Relationship
Taken
% (n)
Distributed
% (n)
Threatened
% (n)
Victim gender
Female
43.5 (153)
36.7 (95)
38.9 (77)
Male
37.2 (131)
35.5 (92)
28.3 (56)
Female and male
10.8 (38)
17.4 (45)
21.7 (43)
Don’t know
8.5 (30)
10.4 (27)
11.1 (22)
Perpetrator-victim relationship
Intimate partner or ex-partner
40.1 (141)
22.8 (59)
22.2 (44)
Family member
20.2 (71)
20.1 (52)
24.7 (49)
Friend (known face-to-face)
19.0 (67)
17.4 (45)
22.7 (45)
Friend (known online only)
6.0 (21)
12.4 (32)
12.1 (24)
Work colleague or ex-work colleague
4.0 (14)
5.4 (14)
5.1 (10)
Acquaintance
2.3 (8)
5.4 (14)
7.1 (14)
Stranger or don’t know
8.5 (30)
16.6 (43)
6.1 (12)
35
Table 4
Frequencies and Descriptives for the 15 Participant Characteristics by Lifetime Prevalence of Self-Reported
IBSA Perpetration
Yes
No
Total
%
n
%
n
%
n
Demographic characteristics
Gender
Female
7.4
171
92.6
2127
100.0
2298
Male
13.7
240
86.3
1515
100.0
1755
Sexuality
Heterosexual
8.9
317
91.1
3260
100.0
3577
LGB
19.7
94
80.3
382
100.0
476
Nativity
Australian born
10.3
320
89.7
2785
100.0
3105
Overseas born
9.4
89
90.6
853
100.0
942
Languages spoken other than English
English only
10.2
345
89.8
3039
100.0
3384
Other languages
9.9
66
90.1
601
100.0
667
Indigeneity
Non-Aboriginal
9.5
373
90.5
3570
100.0
3943
Aboriginal
36.9
38
63.1
65
100.0
103
Education
High school or less
9.8
95
90.2
878
100.0
973
Trade certificate
10.0
105
90.0
945
100.0
1050
University/college
10.6
141
89.4
1189
100.0
1330
Postgraduate/advanced degree
10.0
70
90.0
628
100.0
698
Disability
No assistance required
6.2
220
93.8
3339
100.0
3559
Assistance required
38.8
190
61.2
300
100.0
490
36
Experiential characteristics
Online dating behaviors
No
2.0
21
98.0
1053
100.0
1074
Yes, one or more
13.1
390
86.9
2589
100.0
2979
Sexual self-image behaviors
No
2.4
45
97.6
1833
100.0
1878
Yes, one or more
16.8
366
83.2
1809
100.0
2175
IBSA victimization (taken)
No
4.3
140
95.7
3132
100.0
3272
Yes
34.7
271
65.3
510
100.0
781
IBSA victimization (distributed)
No
5.3
194
94.7
3457
100.0
3651
Yes
54.0
217
46.0
185
100.0
402
IBSA victimization (threatened)
No
5.4
202
94.6
3525
100.0
3727
Yes
64.1
209
35.9
117
100.0
326
Yes
No
Total
M
SD
M
SD
M
SD
Demographic characteristics
Age
32.68
8.23
34.76
9.01
34.55
8.95
Attitudinal characteristics
Minimize/excuse
3.51
1.69
2.16
1.20
2.30
1.33
Blame
4.30
1.48
3.63
1.66
3.70
1.65
37
Table 5
Summary of the Initial and Final Logistic Regression Models Predicting Lifetime Prevalence of Self-Reported IBSA Perpetration
B
SE
p
OR
95% CI
B
SE
p
OR
95% CI
Demographic characteristics
Gender
0.53
.14
<.001
1.71
[1.31, 2.23]
0.57
.13
<.001
1.78
[1.37, 2.31]
Sexuality
0.42
.17
.012
1.52
[1.10, 2.11]
0.43
.17
.010
1.54
[1.11, 2.13]
Age
0.01
.01
.462
1.01
[0.99, 1.02]
-
-
-
-
-
Indigeneity
0.49
.32
.120
1.64
[0.88, 3.06]
-
-
-
-
-
Disability
0.60
.18
.001
1.83
[1.29, 2.59]
0.72
0.17
<.001
2.06
[1.49, 2.86]
Attitudinal characteristics
Minimize/excuse
0.10
.06
.102
1.10
[0.98, 1.24]
-
-
-
-
-
Blame
0.09
.05
.071
1.10
[0.99, 1.21]
0.14
.04
.001
1.15
[1.06, 1.25]
Experiential characteristics
Online dating behaviors
0.49
.27
.072
1.63
[0.96, 2.78]
-
-
-
-
-
Sexual self-image behaviors
0.89
.20
<.001
2.44
[1.64, 3.62]
1.13
.18
<.001
3.10
[2.19, 4.38]
IBSA victimization (taken)
1.03
.16
<.001
2.79
[2.05, 3.78]
1.02
.16
<.001
2.78
[2.05, 3.77]
IBSA victimization (distributed)
1.50
.20
<.001
4.47
[3.02, 6.61]
0.83
.20
<.001
2.35
[1.60, 3.44]
IBSA victimization (threatened)
0.80
.20
<.001
2.23
[1.51, 3.29]
1.54
.20
<.001
4.67
[3.17, 6.87]
38
Note. Reference categories: gender = female, sexuality = heterosexual, indigeneity = non-Aboriginal, disability = no assistance required, online dating behaviors = no, sexual
self-image behaviors = no, IBSA victimization (taken) = no, IBSA victimization (distributed) = no, IBSA victimization (threatened) = no.
39
Appendix
Table A1
Sexual Image-Based Abuse Myth Acceptance (SIAMA) Scale Items
Minimize/excuse
1. Women should be flattered if a partner or ex-partner shows nude pics of her to some close friends
2. A woman should share a nude image of herself with her partner, even if she doesn’t really want to, for
the good of the relationship
3. If a guy shares a nude or sexual pic of his partner with his friends when he’s drunk, he can’t really be
held responsible
4. A man shouldn’t get upset if his partner sends nude pics of him to others
5. Although most women wouldn’t admit it, they generally find it a turn-on for a guy to share nude pics of
her with his mates
6. A woman shouldn’t get upset if her partner sends nude pics of her to others
7. If a woman shows her friends a nude or sexual image of her partner, it just shows how proud she is of
him
8. It’s only natural for a guy to brag to his mates by showing them a nude or sexual image of his partner
9. If a woman is willing to send a nude or sexual image to a man she just met, then it’s no big deal if he
goes a little further by showing it to his mates
10. Women tend to exaggerate how much it affects them if a nude or sexual image of them gets out online
11. A man’s reputation is boosted among his mates if he shares nude pics of a sexual partner
12. Men don’t usually mean to pressure a partner into sending nude pics, but sometimes they get too sexually
carried away
Blame
1. If a person sends a nude or sexual image to someone else, then they are at least partly responsible if the
image ends up online
2. A woman who sends a nude or sexual image to her partner, should not be surprised if the image ends up
online
3. If a man sends a nude or sexual image to someone he just met, he should not be surprised if the image
ends up online
40
4. Celebrities and well-known media personalities who take sexy images of themselves should not expect
that those images will remain private
5. People should know better than to take nude selfies in the first place, even if they never send them to
anyone
6. If a man sends a nude or sexual image to a partner, he can’t expect it will remain private
41
Table A2
Online Dating and Sexual Self-Image Behavior items
Online dating behaviors
1. Flirted with someone online
2. Asked someone out online for a first date
3. Asked someone out by sending them a text message or email
4. Used the internet or email to maintain a long-distance romantic relationship
5. Used an online dating website
6. Used a dating or hook-up app on your mobile phone
7. Asked someone you first met online to meet-up for sex
8. Went on a date with someone you met through an online dating website or app
9. Sent someone a flirty or sexy text or chat message
Sexual self-image behaviors
1. Sent a nude or sexual photo or video of yourself to a current sexual partner
2. Sent a nude or sexual photo or video of yourself to a person you only knew online
3. Sent someone you just met a nude or sexual photo or video to flirt with them
4. Let a sexual partner or date take a nude or sexual photo or video of you
5. Asked someone to send you a nude or sexual photo or video
6. Made a nude or sexy video with a sexual partner
7. Sent someone a nude or sexual photo or video when you didn’t really want to
8. Felt pressured to send a nude or sexual photo or video when you really didn’t want to
9. Received a nude or sexual photo or video of another person when you hadn’t requested it (not including
spam)
10. Received a photo or video of someone’s genitals when you hadn’t requested it (not including spam)
11. Discovered that an image was drawn, ‘photoshopped’ or manipulated to represent you in a sexual way
42
Table A3
The Nine IBSA Victimization and Eight Perpetration Items relating to the Content of the Nude or Sexual Image
Nine IBSA victimization items
1. Where you are partially clothed
2. Where your breasts, including your nipples, are visible
3. Where you are completely nude
4. Where your genitals are visible
5. Where you are engaged in a sex act
6. Where you are showering, bathing or toileting
7. Which is of a sex act that you did not agree to
8. Which is up your skirt (e.g., ‘up-skirting’)
9. Which is of your cleavage (e.g., ‘down-blousing’)
Eight IBSA perpetration items
1. Where they were partially clothed or semi-nude
2. Where the person’s breasts, including their nipples, were visible
3. Where they were completely nude
4. Where the person’s genitals were visible
5. Where they were engaged in a sex act
6. Where they were showing bathing or toileting
7. Which was up their skirt (e.g., ‘up-skirting’)
8. Which was of their cleavage (e.g., ‘down-blousing’)
Note. Only those participants who identified as female (or non-binary gender) were provided IBSA
victimization items 2, 8 and 9.
