“Am I at risk of cyberbullying”? A narrative review and conceptual framework
for research on risk of cyberbullying and cybervictimization: The risk and
needs assessment approach

Anna Costanza Baldry, David Farrington, Anna Sorrentino

PII: S1359-1789(15)00076-2
DOI: doi: 10.1016/j.avb.2015.05.014
Reference: AVB 919

To appear in: Aggression and Violent Behavior

Received date: 17 April 2015

Accepted date: 13 May 2015

Please cite this article as: Baldry, A.C., Farrington, D. & Sorrentino, A., “Am I at
risk of cyberbullying”? A narrative review and conceptual framework for research on
risk of cyberbullying and cybervictimization: The risk and needs assessment approach,
Aggression and Violent Behavior (2015), doi: 10.1016/j.avb.2015.05.014

This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.
 ACCEPTED MANUSCRIPT
1

“Am I at risk of cyberbullying”? A narrative review and conceptual framework for research on
risk of cyberbullying and cybervictimization: the risk and needs assessment approach
Anna Costanza Baldry(a)1, David Farrington(b) and Anna Sorrentino(a,c)
(a)
Department of Psychology, Second University of Naples
(b)
Institute of Criminology, University of Cambridge
(c)
Department of Humanities, University of Naples, Federico II

T
R IP
(Special Issue Editors for Aggressive and Violent Behavior “Bullying, Cyberbullying, and Youth Violence:

SC
Facts, Prevention, and Intervention”)

NU
ABSTRACT

Bullying and its electronic evolution, cyberbullying, are widespread problems among children and adolescents.
MA
Numerous studies have been conducted that address its prevalence, nature, possible impact and strategies to prevent
cyberbullying (Patchin, 2013). Some recent papers have reviewed existing instruments designed to measure
cyberbullying (Berne et al., 2012; Vivolo-Kantor, Martell, Holland &Westby, 2014), while others have reviewed risk
D

factors (Kowalski et al. 2014).

TE

The aim of the present study is to present what is known about of risk factors associate with cyberbullying and
cybervictimization by using an ecological framework, addressing the importance of adopting a risk and needs
P

assessment approach to identify early who is at risk and tailor interventions.

We conducted a review of cyberbullying risk factors, as well as assessment and measurement instruments and
CE

risk and needs assessment to identify which papers fulfilled the purpose of this study. Multiple online databases (i.e.,
PsychInfo, SocIndex and PubMed) were searched to identify relevant studies. The keyword search criteria were: (bull*
AC

or cyberbull* or school viol* or juvenile delinquency) AND (risk* or threat* or assess*) AND (measure* or *method*)
between 2000 and 2015. Article titles and abstracts were reviewed, and all articles that appeared relevant were
retrieved in full-text format and evaluated for inclusion in the review. In addition, articles accessed electronically were
hand-searched for other relevant studies. A total of 7199 potential articles were located. Of these, only 53 were
considered to be directly relevant and used for the purpose of the present work.
Because of the large variability of methods, construct definitions, measures and item wording used in the
different studies, a meta-analysis was not possible, therefore a narrative review approach was adopted to identify risk
factors according to the ecological theoretical framework.
Results regarding risk factors showed that individual as well as socio-family related factors were associated
with cyberbullying and cybervictimization with some slight differences for boys and girls. The strongest risk factor that
was associated with cyberbullying is school bullying. This review is of relevance because it is innovative in proposing a
conceptual framework for developing a risk and needs assessment tool for cyberbullying and cybervictimization.

1
Correspondent author: Anna Costanza Baldry, Department of Psychology, Second University of Naples, viale Ellittico, 31, 81100
Caserta – Italy. E.mail annacostanza.baldry@unina2.it
 ACCEPTED MANUSCRIPT
2

Key words:
Cyberbullying
Cybervictimization
Risk factors

T
Risk and needs assessment

R IP
Contents

SC
1. Introduction
1.1 Risk factors for cyberbullying and cybervictimization: static and dynamic
1.2 Aims of the current study

2. Review of risk factors

NU
2.1 Design and method for the literature search
MA
2.2 Selection procedure for relevant published articles
2.3 Theory driven selection method for risk factors: the ecological framework
2.3 Narrative Scheme Review System (NSRS)

3. Findings
D

3.1 Cyberbullying risk factors

3.1.1 Individual level
TE

3.1.2 Interpersonal level

3.1.3 Community/social level
P

3.2 Cybervictimization risk factors

CE

3.2.1 Individual level

3.2.2 Interpersonal level
3.2.3 Community/social level
AC

4. Discussion
4.1 Design and research limitations of the narrative review
4.2 Cyberbullying and cybervictimization profile
4.3 A new approach for predicting and managing risk and needs of cyberbullying and
cybervictimization

Acknowledgment
References

1. Introduction
 ACCEPTED MANUSCRIPT
3

In the last 10 years, a growing interest has developed in the media and among researchers and policy
makers on the massive development of online technology and its use and impact, and in particular in
youngsters. Online communication is not only any more a daily way to work, but it is mainly used to
communicate and interact with known and unknown people (peers and adults). And this is particularly of
relevance for youngsters. If in 2010 the number of children and teenagers having access to the internet at

T
home was 66% (reported by Tokunga, 2010), this proportion is growing yearly. National UK statistics report

IP
an increase in 2012 where 21 million households (80 per cent) had Internet access, compared with 19

R
million (77 per cent) in 2011 (Office for National Statistics, 2012). Competition between internet providers
and reduction of prices of IT (Internet Technology) devices have decreased prices and increased

SC
dissemination especially in young people; access to a computer and/or smartphones is overwhelming. This
far is all good news, when thinking about the free world, the advantages of online, technological

NU
communication and search. However, research and news reports do show another picture of web 2.0
communication. The dark side of children’s use of internet and associated technology is the risk of being
MA
bullied online or bullying others, so called cyberbullying, leading to short and long term negative
consequences (Patchin and Hinduja, 2006; Topcu, Erdur-Baker and Capa-Aydin, 2008; Ybarra et al, 2006)
and ultimately even suicide or attempted suicide (Van Geel, Vedder, Tanilon, 2014); (for a review on
D

consequences, Tokunaga, 2010).

TE

Several conceptual definitions have been provided on cyberbullying, summarized by Tokunga in

2010; this author provides an integrative definition which tries to capture all relevant aspects:
P

“Cyberbullying is any behavior performed through electronic or digital media by individuals or groups that
CE

repeatedly communicates hostile or aggressive messages intended to inflict harm or discomfort to others”
(Tokunga, 2010: 278). This definition as well as all the similar ones provided by other authors (Besley, 2009;
Finkelhor ate al., 2000; Juvonen and Gross, 2008; Li (2008); Patchin and Hinduja, 2006; Slonje and Smith,
AC

2007; Smith et al., 2008); Willar (2007); Ybarra and Mitchell, 2004) have similar features to the definition of
traditional bullying (Olweus, 1993) with regard to intention of harming and in regard to imbalance of
power. With regard to ‘repeated actions’, cyberbullying can differ. Some authors claim that it is enough to
experience one or two actions to be defined as cyberbullying (Gámez-Guadix et al., 2013; Juvonen & Gross,
2008; Kowalski & Limber, 2007; Li, 2007; Vieno, Gini, & Santinello, 2011; Ybarra & Mitchell, 2004). A single
cyber-attack (a video, a comment, a picture) can remain online or in a mobile phone for quite some time,
therefore prolonging the harm to the victims (Dooley, Pyzalski, & Cross, 2009; Raskauskas & Stoltz, 2007;
Slonje et al., 2013) and increasing access to a potential cyberbystanders who can in turn share the attack,
prolonging the victim’s distress (Dooley et al., 2009; Tokunga, 2010). The problem is not online
communication per se, but its use; as clearly stated by Kowalaski and colleagues in their extensive review,
“certain features of online communication including reproducibility, lack of emotional reactivity, perceived
 ACCEPTED MANUSCRIPT
4

uncontrollability, relative permanence, and 24/7 accessibility, make it more likely for online misbehavior to
occur” (Kowalaski et al., 2014: 2, emphasis added).
Once studies have extensively researched the nature and proportion of the problems, in the last
decade they addressed characteristics linked to such problems, trying to answer to the following questions:
which are the features of those children who bully or are bullied online?

T
As pointed out by Tokunga (2010) and Slonje, Smith and Frisen (2012), what is often missing in

IP
studies on cyberbullying is a clear sound theoretical foundation guiding studies. If based on sound valuable

R
theories, methods, variables and procedures can be adopted to test hypotheses, and more sound, reliable
and coherent information can be available, helping to develop useful cost-effective intervention strategies.

SC
Kowalaski et al. (2014), in their extensive review have overcome this limitation by adopting the General
Aggression Model (GAM) to review existing (supporting) studies addressing the different aspects related to

NU
this theory, adopted from studies on aggression (inter alia, Bandura, 1986; Crick and Dodge, 1994) in
relation to victimization and perpetration of cyberbullying.
MA
Another attempt to address and understand cyberbullying has been adopted in another recent
review conducted by Mehari, Farrell, and Le (2014). These authors suggest that the mean through which
aggression takes place may be best conceptualized as a new dimension on which aggression can be
D

classified, rather than addressing cyberbullying as a “distinct counterpart to existing forms of aggression”
TE

(Mehari et al., 2014: 2). Therefore, research on cyberbullying should be considered within the context of
theoretical and empirical knowledge on aggression among youngsters.
P

Another useful theoretical approach is the ecological system theory, based on Bronfenbrenner’s
CE

ecological framework (Bronfenbrenner, 1979, 1986, 1994) extensively used in the context of school bullying
in the review by Hong and Espelage (2012). The authors adopted this framework to present the relationship
between different individual, interpersonal and more broadly society dimension to show onset,
AC

development and recurrence of bullying and victimization among youngsters. The review showed empirical
findings on the risk factors associated with bullying and peer victimization at school within the context of
different levels of the ecological framework, addressing risks and needs of those involved in bullying (either
as victims or perpetrators), concluding that there is no one single risk factors, or cause to explain bullying,
but that all levels’ risk factors at all levels can have a role and influence and these vary from individual to
individual, and from context to context.
The current narrative review aims to increase the understanding of risk factors for cyberbullying
and cybervictimisztion’s and their relationship and interaction to understand who is most at risk and
provide a possible framework for a risk and needs assessment approach to cyberbullying and
cybervictimization (see Figure 1).
 ACCEPTED MANUSCRIPT
5

--------------------------------------------------- Insert Figure 1 about here -----------------------------------------

1.1 Risk factors for cyberbullying: static and dynamic

Risk factors for the purpose of the risk and needs assessment approach can be distinguished in two

T
main categories: static risk factors, that cannot change and will always remain the same for youngsters [e.g.

IP
gender, prior involvement in bullying or other antisocial behavior, prior victimization, to some extent,

R
impulsivity; past exposure to violence], and dynamic risk factors, that can change with time or as a
consequence of some form of intervention [e.g. school policies to prevent and reduce cyberbullying,

SC
parental supervision, academic achievement, empathy, time spent in online activities]. Some dynamic risk
factors, in this framework, can be considered as ‘vulnerable’ aspects of the child that even if they are

NU
associated with the cyberbullying, they can be considered as needs that can be address, to modify a risk,
that is changeable. Needs are considered as dimensions related to the individual, to his or her relationships
MA
and involvement in cyberbullying either as a cyberbully or cybervictim or as a bystander, that require
attention and intervention. These needs, according to the risk and needs assessment approach, if not
addressed, might not change spontaneously and will continue to influence a juvenile’s attitudes and
D

behavior such as involvement in cyberbullying and its justification or minimization.

TE

Any level of risk factors does not determine the occurrence or re-occurrence of a given behaviour;
risk factors are not necessarily causally related to a given behavior. A risk factor can explain a relationship:
P

it’s presence increases the likelihood of such behaviour (i.e. cyberbullying) taking place nor or in the future;
CE

it increases the likelihood but its absence does not eliminate the risk (Hart, 2013; Hoge, Vincent and Guy,
2012). The assessment of the influence of risk factors on a given behavior and the likelihood of its
occurrence or reoccurrence is called risk and needs assessment.
AC

We therefore will explain cyberbullying and cybervictimization and their associated risk factors
according to the risk and needs assessment approach, well known and adopted in the field of criminology,
especially with regard to juvenile delinquency but that has never been used or adopted in the field of
cyberbullying, within the ecological system framework, as we will discuss in the final part of this review.

1.2 Relevance of the current study

This narrative review identifies various risk factors associated with cyberbullying and cybervictimization for
boys and girls based on original studies and significant reviews addressing risk factors associated with
cyberbullying. This study also suggests to adopt a risk and needs assessment approach (Baldry and
Sorrentino, in press; Borum and Douglas, 2003; Hoge, Vincent and Guy, 2012), for best developing tailored,
efficient and effective intervention strategies to manage risk and reduce the risk of (re)occurrence of
misbehavior. In particular, we examined risk factors that have been found to be positively linked and
 ACCEPTED MANUSCRIPT
6

associated with perpetration or victimization of cyberbullying. By adopting an ecological theoretical

framework (Bronfenbrenner, 1979, 1986), we looked at the different levels of each (risk) factor and
classified it accordingly: individual (onto-), interpersonal (micro-), community (meso-) and societal (macro-)
and searched for consistency of findings and provide a possible explanatory model. We then used a new
approach to address cyberbullying, the risk and needs assessment approach, widely used in criminal

T
behavior and highly developed and used in the field of juvenile delinquency (Baldry and Kapardis, 2013;

IP
Hoge, Vincent, Guy, 2012). By classifying identified risk factors as static and dynamic factors (also identified

R
as ‘needs factors’), practitioners and policy makers, but also researchers, can build and promote
interventions based on the actual needs of children at risk or potentially at risk, and develop tailored

SC
interventions which can reduce the risk of cyberbullying in a cost-effective manner (Welsh and Farrington,
2006).

NU
MA
2. Design and method

2.1 Literature search

D

A systematic search was conducted for all works on the assessment and measurement of
TE

cyberbullying and cybervictimization published between 2000 and 2015.

To this aim, PsychInfo, SocIndex with Full Text and PubMed electronic databases were used. We
P

used as keyword search criteria: (bull*, vict*, cyberbull*, cybervict*, school viol* or juvenile delinquency)
CE

AND (risk* or threat* or assess*) AND (measure* or *method*). All search terms were used in combination
with each other in order to retrieve relevant publications. The abstracts of all relevant articles were
screened for eligibility for inclusion. Full publications, whose abstracts were considered appropriate for the
AC

present study, were retrieved and reviewed in order to identify the individual and contextual risk factors
associated with involvement in cyberbullying and cybervictimization. In addition, articles accessed
electronically were hand-searched for other relevant studies.
Because of the large variability in methods, constructs and item wording across the studies included
in our review, a meta-analysis was not possible. Also, specifically with regard to risk assessment or threat
assessment methods, no specific instruments were found, so a theoretical review framework was adopted
to specify a conceptual framework for the development of a risk and needs assessment approach to
cyberbullying and cybervictimization.

2.2 Selection procedure for relevant published articles

 ACCEPTED MANUSCRIPT
7

The selection procedure is summarized in Figure 2. A total of 7199 potential publications matching
the initial search criteria were identified; of those, 147 studies were considered relevant for abstract
screening. However only 87 studies of those that were retrieved and reviewed were eligible; 60 were not
actually dealing with risk factors cyberbullying or cybervictimization. Thirty-four additional publications
were excluded because they were duplicates did not meet our inclusion criteria, resulting in a final total of

T
53 articles complying fully with our inclusion criteria (see Figure 2 for more detailed information).

IP
In summary, publications included in this study met the following criteria:

R
(a) evaluated or investigated risk factors for youngsters’ involvement in cyber bullying as perpetrator
and/or victim; (b) assessed either static or dynamic risk factors for involvement in cyberbullying; (c)

SC
assessed cyberbullying through survey/questionnaire administration; (d) included participants aged 10-20;
(e) was developed and published between 2000 and 2015; and (f) included measurement information on

NU
risk factors for cyber bullying or cybervictimization and (g) were published in English.
Studies that did not meet the above mentioned criteria were excluded. In particular, we excluded
MA
publications that (i) did not include any form of measurement information; (ii) did not investigate risk
factors for cyberbullying and/or cybervictimization; (iii) analysed only the consequences following
youngsters’ involvement in cyberbullying; (iv) investigated risk factors and incidence of only traditional
D

bullying.
TE

2.3 Theory driven selection method of classification of risk factors: the ecological framework
P

Once studies were selected for inclusion in the sample, to conduct the narrative review for
CE

classification of risk factors we adopted the ecological conceptual framework (Bronfenbrenner, 1979,
1986). By classifying risk factors according to this theoretical framework, it is possible to look at the
relationship between risk factors and involvement in cyberbullying and cybervictimization and establish the
AC

likelihood of children being more at risk than others, at one or more of the four possible levels underlying
the ecological theory —the individual, the interpersonal, the community, and the societal (Bronfenbrenner,
1979, 1986; Hong and Espelage, 2012).
At the individual level, originally identified as the ontogenetic level, personal history and biological
factors are found to influence how children behave, and these factor show they can increase the likelihood
of becoming a victim or a perpetrator of cyberbullying. Among these factors we searched for gender, age,
use of new technologies, numbers of hours online, involvement in school bullying, self-esteem, empathy,
moral disengagement, impulsivity, academic achievement, and internet addiction.
At the interpersonal (microsystem) level, personal relationships such as family, friends and peers
may influence the risks of becoming a perpetrator or a victim of cyberbullying. For example, at the family
level, we looked at whether studies investigated the role of parental monitoring of online activities, sharing
information and procedures to increase ITC security, the child’s supervision in off/online activities, and
 ACCEPTED MANUSCRIPT
8

parental styles that can play a role and cyberbullying. With regard to peers and friends, positive or negative
perceived peer interactions, close friendship, isolation, and peer attitudes towards cyberbullying were also
included.
At the community and school level (mesosystem), the context in which social relationships occur,
such as schools or neighborhoods, could also influence cyberbullying. Risk factors searched for include the

T
school climate, the presence of rules on the use of cyber-communication, school curricula including (or not)

IP
intervention programs, and management strategies in cases of bullying or cyberbullying. Also, at the

R
community level, it can be of influence more general policies on cyber bullying, prevention programs, the
presence of services and strategies for prevention of violence can be important.

SC
Finally, societal/cultural risk factors (macrosystem) influence whether cyberbullying is encouraged
or inhibited. These include the growth of the economy, widespread cultural norms and beliefs about

NU
violence, societal responses on safe youth use of internet and companies’ online protection policies.
Unfortunately, as far as we could find, perhaps also due to the difficulty of assessing and evaluating
MA
society-community level risk factors for youngsters’ involvement in cyber bullying both as perpetrator and
victim, no studies have been published measuring the macrosystem role in cyberbullying. Because of the
lack of publications of society-community level risk factors, we included studies at least at one of the other
D

three ecological levels above mentioned and focused our review on risk factors within the context of the
TE

ontogenetic (individual), micro- (interpersonal), or meso-system (community) levels.

P

2.4 Narrative Scheme Reviewing System (NSRS)

CE

The current study not only reviewed existing knowledge about risk factors for cyberbullying and
cybervictimization, but also investigated these factors according to the ecological framework for possible
use in the risk and needs assessment approach. We created the Narrative Scheme Reviewing System (NSRS)
AC

procedure to classify variables (i.e. risk factors) according to our given theoretical framework. All relevant
studies selected, including the most relevant existing reviews addressing risk factors and cyberbullying,
were read and factors were categorized according to their level (individual, interpersonal or community).
The narrative approach is appropriate because it is theory driven and makes it possible to compare
different studies that used different methods and definitions for risk factors and cyberbullying. This
procedure also makes it possible to further classify factors according to whether they are static or dynamic
risk factors in their nature. This aspect that is hardly taken into consideration in the studies.

3. Findings: risk factors

 ACCEPTED MANUSCRIPT
9

Table 1 & Table 2 present the results of the narrative review. We organized studies and findings according
to which level the risk factor belongs to, according to the ecological framework, excluding the cultural
societal level since no studies were found directly measuring cyberbullying or cybervictimization in relation
to this. For each study we provide details about the methodology, the sample and countries and the type of
relationship with either cyberbullying or cybervictimization, separately. Direct numeric comparisons are not

T
possible because that different studies in many cases use different measures for same or similar constructs.

IP
Most studies reviewed are cross –sectional but some are longitudinal.

R
Insert Table 1 here

SC
3.1 Risk factors for cyberbullying

NU
3.1.1 Individual risk factors for cyberbullying (‘Ontogenetic’)
MA
3.1.1.1 Gender
Overall gender differences indicated that boys are more involved as cyberbullies than girls. This was
reported in several studies. Lapidot-Lefler & Dolev-Cohen’s (2015) online study in Israel conducted with 465
D

7-12 grade students reported higher rates for boys. The same was found by Erdur-Backer (2010) in his
TE

Turkish study with 276 students from 7th to 9th grade. The same was reported by Sourander et al. (2010),
Huang & Chou (2010), Li (2006), and in the review of 131 studies conducted by Barlett & Coyne (2014).
P

Gender is considered a static risk factor that does not change with time.
CE

3.1.1.2 SES
Wang, Iannotti & Nansel (2009) found a positive relationship between SES (Socio-Economic Status) and
AC

cyberbullying. When analyzing data from the 2005 Health Behavior Questionnaire in the USA with a sample
of 7182, 6-10 grades, it emerged that higher levels of socio-demographic status are associated with higher
prevalence rates of cyberbullying. SES is considered a dynamic factor since, to some extent, it can change in
time.

3.1.1.3 School commitment

Kowalski and Limber (2013) in their study with 931 students from 6 to 12th grade in the US and Ybarra &
Mitchell (2004a) in their US phone interview study with 1501, 10 to 17 year old students, both found that
cyberbullying is associated with low school commitment; this risk factor can be considered to be dynamic.

3.1.1.4 Technology use

 ACCEPTED MANUSCRIPT
10

At the individual level, different studies found that technology use is highly associated with cyberbullying.
All studies including such risk factors as ICT expertise found significant correlations. Higher internet
expertise is associated with a higher level of cyberbullying others as identified by Sticca, Ruggieri, Alasker &
Perren (2013) in their longitudinal study conducted with a Swiss sample of 835 students, and the same
positive correlation was reported by Warlave & Heirman (2011) in Belgium with 1.378 students aged 12 to

T
18 years old and by Hinduja & Patchin (2008) in the US. The same positive correlation applies for Internet

IP
use with Kowalski et al. (2014), Casas, Del Rey, & Ortega-Ruiz (2013), Sticca, Ruggieri, Alasker & Perren

R
(2013), Mishna, Khoury-Kassabri, Gadalla & Daciuk (2012), Warlave & Heirman (2011), Erdur-Backer (2010),
Hinduja & Patchin (2008), Ybarra & Mitchell (2004a) all reporting positive correlations with cyberbullying:

SC
the more time you spend online, the higher the correlation. Risky online behavior: The review by Kowalski
et al. (2014) concluded that most studies reviewed showed significant correlation between being so called

NU
‘risky behavior online’ and being a cyberbully. Risky behavior refers, among others to surfing websites not
suitable for children, communicating on social networks with unknown people. Specifically, this correlation
MA
is found in a Spanish study by Casas, Del Rey, & Ortega-Ruiz (2013) with 893, 11 to 19 year old students,), in
Canadian’s self-reported survey conducted with 2186 students (Mishna et al., 2012) and in Turkey with 276,
10 to 14 years old students (Erdur-Backer, 2010). All ICT related risk factors are clearly dynamic, given that
D

the type of behavior online, can be changed.

TE

3.1.1.5 Personality individual risk factors

P
CE

With regard to individual risk factors related to the personality of children, the most reported risk factor
associated with cyberbullying is low empathy. Kowalski et al. (2014) in their review systematically found an
association between low empathy and cyberbullying (as did Casas, Del Rey, & Ortega-Ruiz, 2013; Topcu and
AC

Erdur-Baker, 2012; Steffengen, König, Pfetsch & Melzer, 2011). More specifically Ang and Goh (2010) found
that girls with lower levels of cognitive empathy and boys with low levels of both cognitive and affective
empathy are more likely to cyberbully others. Steffengen and König (2009) in Luxembourg also found this
correlation. Also, self-esteem was studied by Modecki, Barber and Vernon (2013), who found in their
Australian longitudinal study that cyberbullies report lower level of self-esteem (Patchin & Hinduja, 2010).
Cyberbullies also reported lower levels of self-control compared to non-cyberbullies when asked about
several aspects of their lives in Czech Republic (Bayraktar et al. 2014). According to Modecki, Barber and
Vernon (2013) cyberbullies in Australian students also present a so called ‘early depressed mood’. Though
personality risk factors are difficult to address and modify, with special dedicated intervention and
treatment they could be changed, so we consider them to be dynamic.

3.1.1.6 Values
 ACCEPTED MANUSCRIPT
11

Another set of individual variables that have been frequently found to be associated with cyberbullying is
moral disengagement, reported in different study designs (correlational and longitudinal alike), consistently
showing the same positive correlation (Kowalski et al. 2014; Cappadocia, Craig, & Pepler, 2013; Menesini,
Nocentini & Camodeca, 2013; Pozzoli, Gini, & Vieno, 2012; Bauman , 2010) and Pornari and Wood, 2010).
Differently measured and defined moral disengagement refers to a denial of wrong doing, of moral

T
detachment about antisocial action, and blaming the victim, which is higher in cyberbullies. Cyberbullies

IP
tend also to breake rules more often (Sticca, Ruggieri, Alasker and Perren, 2013). In the same line of

R
(distorted) values, cyberbullies also reported normative beliefs about aggression, indicating minimization
and justification (Kowalski et al., 2014). Other maladaptive behaviors, such as alcohol drinking is reported

SC
by Cappadocia, Craig, & Pepler (2013) in Canada. These values can also be changed so are dynamic.

NU
3.1.1.7 Bullying in school

Most studies looking at possible risk factors for cyberbullying also measured school bullying and
MA
consistently found a positive association. Hemphill & Heerde (2014), Modecki et al., 2014), and the
extensive review by Kowalski et al. (2014), Cappadocia, Craig, & Pepler (2013), Kowalski & Limber (2013),
Sticca, Ruggieri, Alasker & Perren (2013), Del Rey, Elipe & Ortega-Ruiz(2012), Mishna, Khoury-Kassabri,
D

Gadalla & Daciuk (2012), Gradinger, Strohmeier, & Spiel (2009), Vandebosch & Van Cleemput (2009),
TE

Hinduja & Patchin (2008), Smith et al (2008) and Raskauskas & Stoltz (2007), all reported that cyberbullies
are also school bullies. Very few of these studies are longitudinal (with the exception of Cappadocia, et al.,
P

2013; Sticca et al., 2013) so it is not possible to say whether one starts as a school bully and then goes
CE

online or whether the two behaviors happen at the same time. The studies also do not report data about
respondents’ reporting whether they bullied and cyberbullied the same or different children.
Gender differences are often not reported; Gradinger, Strohmeier, and Spiel, 2009) looked at
AC

gender differences and reported that boys are more likely than girls to be school bullies and cyberbullies.
Being a victim of school bullying is also considered a risk factor for cybervictimization for girls as the two
behaviors have been found to be associated by Kowalski, Morgan and Limber (2012) in their US study with
over 4.000 students, and the same positive association was found in the telephone interview study done by
Ybarra & Mitchell (2004a). Being a cybervictim is also associated with cyberbullying, as found by Kowalski et
al. (2014), Wright & Li (2013) and Vandebosch & Van Cleemput (2009). This relationship reflects the “role
inversion hypothesis” which emphasizes the overlap between bullies and victims (Ybarra & Mitchell, 2004a).
Clearly any bullying behavior is dynamic in its nature.

3.1.2 Interpersonal level risk factors for cyberbullying (Microsystem)

3.1.2.1 Peer group risk factors.

 ACCEPTED MANUSCRIPT
12

The role and influence of the interpersonal social context is found in several studies associated with
cyberbullying. Who the children hang around with and the lack of prosocial peer influences were found
predictors in the Canadian longitudinal study of Cappadocia, Craig, & Pepler (2013) of future cyberbullying
involvement. This is true also for peer rejection, with cyberbullies more likely to be rejected than non-
cyberbullies (Bayraktar et al. 2014; Wright & Li, 2013) and experiencing greater social isolation as well as

T
lack of perceived peer support, found in the systematic review (Kowalski et al., 2014), and specifically also in

IP
the Spanish (Calvete et al., 2010) and US (Williams & Guerra, 2007) studies. Clearly peer factors can be

R
considered dynamic risk factors that can change in time or with tailored intervention.

SC
3.1.2.2 Family risk factors
At the interpersonal level, lack of parental support is also considered as a risk factor. Hemphill and Heerde

NU
(2014), in Australia, and Wang, Iannotti, & Nansel (2009) in the US, reported that parents of cyberbullies
provide less support and lack also emotional bonding (Ybarra & Mitchell, 2004a). These parents are
MA
reported as having poor parental attachments (Bayraktar et al., 2014), but also the family management in
general is reported as lacking some kind of rules and any forms of monitoring of the children’s life and
online activities specifically. This is reported not only in the overall review by Kowalski et al., (2014), but
D

also in the Australian longitudinal study by Hemphill & Heerde (2014) and in Ybarra & Mitchell’s (2004a)
TE

survey. When parents are poorly involved with the child’s internet use, Vandebosch and Van Cleemput
(2009) have found children more at risk of cyberbullying. Parental involvement in the reduction of
P

cyberbullying is essential and possible due to the fact that these risk factors are dynamic.
CE

3.1.3 Community level risk factors for cyberbullying (mesosystem)

AC

3.1.3.1 School risk factors:

The role of schools, their policies and school climate all have been reported having a relationship with the
risk of cyberbullying. A negative school climate (Kowalski et al., 2014; Williams & Guerra, 2007), specifically
defined as lack of teacher support and lack of clear rules and school safety, was reported in the Spanish self-
report study conducted with 893 adolescents (Casas, Del Rey, & Ortega-Ruiz, 2013) to be associated with
involvement in cyberbullying. Also, perceiving not being connected or bonded to school (Williams & Guerra,
2007) and general lack of school safety (Kowalski et al., 2014), are all reported as associated with
cyberbullying. Any school-related aspect related to rules and regulations is dynamic. However, where the
school is located and which area, for example a disruptive area with high levels of delinquency, are more to
be considered as static risk factors.
 ACCEPTED MANUSCRIPT
13

3.2 Risk factors for cybervictimization

With regard to risk factors associated with cybervictimization, we will here present results following
again the three levels: individual, interpersonal and community. We report from the literature review cross
sectional or longitudinal associations with involvement in cybervictimization, as presented in Table 2.

T
Insert Table 2 about here

R IP
3.2.1 Individual level risk factors (‘Ontogenetic level’)

SC
3.2.1.1 Gender

NU
Overall gender differences indicate that girls are more involved as cybervictims than boys. This was
reported in several studies. Bayraktar, Machackova, Dedkova, Cerna,& Ševčíková (2014)’s online survey
conducted with 2.092 students aged 12-18 years reported higher rates of cybervictimization among girls.
MA
The same was found by Holt, Fitzgerald, Bossler, Chee, Ng (2014) in their study with 4.315 students from
primary and secondary schools in Singapore. The same gender difference was reported by Sourander et al.
(2010), and Kowalski & Limber (2007), in which the same conclusions were drawn.
D
TE

3.2.1.2 Low school achievement

Hindujia & Patchin (2008) in their online survey involving 1.378 respondents aged less than 18 years
P

found that cybervictimization is associated with low school achievement. Similarly, Wang et al. (2014) in
CE

their study with 1.023 5th grade students and in the review of 25 studies conducted by Tokunaga(2010) it
was found that cybervictimization is associated with students’ low academic achievement. This risk factor
AC

can be considered dynamic, since can change over time.

3.2.1.3 Technology use

At the individual level, several studies found that technology use is highly associated with cyber
victimization. Internet use: can be considered a risk factor for cybervictimization as summarized by Kowalski
et al. (2014) in their review and in the studies by Zhou, Tang, Tian, Wei, Zhang, & Morrison (2013), Mishna,
Khoury-Kassabri, Gadalla & Daciuk (2012), Erdur-Backer (2010), Vandebosch & Van Cleemput (2009) and
Hinduja & Patchin (2008). They all report the existence of a positive correlation with cybervictimization,
that is, the more time youngsters spend online the more they are at risk of being victimized online. Risky
online behavior: The review concluded by Kowalski et al. (2014) showed a significant correlation between
being involved in so called ‘risky online behavior’ and being a cybervictim. These correlations were found in
Mishna et al. (2012), Walrave & Heirman (2011), Erdur-Backer (2010), Katzer, Fetchenhauer, & Belschak
 ACCEPTED MANUSCRIPT
14

(2009), Mesch (2009) and Hinduja & Patchin (2008), all reporting that cyber victimization is related to their
involvement in online risky behaviors.

3.2.1.4 Personality
With regard to individual risk factors related to a youngster’s personality, the most reported risk

T
factor associated with cybervictimization is low self-esteem. Bayraktar, Machackova, Dedkova, Cerna,&

IP
Ševčíková (2014), in their online survey involving 2.092 students aged 12-18 years from Czech Republic, and

R
Modecki, Barber & Vernon (2013) in their longitudinal study conducted with 1.364 Australian students aged
12-14 years, found that cyber victimization is related to low levels of self-esteem. The same positive

SC
correlation was found by Kowalski & Limber (2013), Patchin & Hinduja (2010), and Katzer, Fetchenhauer, &
Belschak (2009). Perceived social intelligence can be considered a risk factor for cybervictimization, as

NU
reported by Kowalski et al. (2014) in their review of 131 studies, and in particular in the study of Hunt,
Peters and Rapee (2012) conducted with 218 Australian students aged 8-15 years and by Schultze-
MA
Krumbholz & Scheithauer (2009) involving 71 German students. They found that cyber victimization is
related to low levels of perceived social intelligence namely children thinking of themselves of not clever or
smart. Emotional control has also been found as an individual risk factor for cyber victimization. The
D

longitudinal studies by Hemphill an Heerde (2014) and Hemphill et al. (2014) conducted with respectively
TE

927 Australian students aged 10-11 years and 673 Australian students aged 12-13 years, found that cyber
victimization is related to youngsters poor emotional control. Also empathy is found to be a risk factor for
P

being victimized online, as the study conducted by Schultze-Krumbholz & Scheithauer (2009) with 71
CE

German 7, 8 and 10 grade students, highlighted.

3.2.1.5 Morality
AC

At the individual level a risk factor related to youngsters’ values is moral disengagement. In their
review of 131 studies, Kowalski at al. (2014) found that cyber victimization is correlated with high levels of
moral disengagement. Moral values, and in particular moral disengagement, is a dynamic risk factor for
cyber victimization, that must be taken into account when planning and implementing prevention and
intervention programs focused on cybervictims.

3.2.1.6 Psychological status

With regard to youngsters’ psychological status, depression is associated with cybervictimization.
This relationship is highlighted by several studies included longitudinal ones, such as that of Cappadocia,
Craig, & Pepler (2013) conducted with 1.972 9 to 12 grade Canadian students, the Spanish study by Gámez-
Guadix, Orue, Smith and Calvete (2013) with 845 students aged 13-17 years, and the Australian conducted
by Modecki, Barber, & Vernon (2013) with 1.364 students aged 12-14 years. All these studies found that
 ACCEPTED MANUSCRIPT
15

cyber victimization was related to a high level of depression later in time. Kowalski et al. (2014) in their
meta-analysis of 131 studies identified as risk factors for cyber victimization also social anxiety and anger,
that is cyber victimization is related to high levels of social anxiety and feelings of anger. Also the presence
of psychosocial problems is a risk factor for cyber victimization as in the review by Tokunaga (2010). Such
risk factors can be considered dynamic ones, because they could be treated and are potentially changeable.

T
IP
3.2.1.7 Maladaptive behaviors

R
In their longitudinal study with 845 Spanish students aged 13-17 Gámez- Guadix, Orue, Smith, &
Calvete (2013) found that students’ substance use was linked to future cybervictimization. The same

SC
relationship was found by Hinduja & Patchin (2008).

NU
3.2.1.8 Bullying in school
Studies found that involvement in school bullying is a risk factor for cybervictimization. The majority
MA
of studies highlighted the existence of an overlap between the role of traditional victim and cyber victim.
Kowalski et al. (2014) in their review systematically found an association between traditional victimization
and cyber victimization. Hemphill, Tollit, Kotevski, & Heerde (2014) and Cappadocia, Craig, & Pepler (2013)
D

in their longitudinal studies conducted respectively with 673 Australian students aged 12-13 years and
TE

1.972 Canadian 9-12 grade students, found that being involved in traditional bullying as a victim was
related to cybervictimization. The same relationship was found by Holt, Fitzgerald Bossler, Chee, Ng (2014),
P

Hemphill & Heerde (2014), Del Rey, Elipe & Ortega-Ruiz (2012), Mishna et al. (2012), Gradinger,
CE

Strohmeier, & Spiel (2009) Vandebosch and Van Cleemput (2009) Hinduja and Patchin (2008) Smith et al.
(2008) and Raskauskas & Stoltz (2007). However, some studies highlighted, as risk factor for
cybervictimization, students’ involvement as traditional bullies, supporting the “role inversion hypothesis”
AC

(Ybarra & Mitchell, 2004a) mentioned earlier. Kowalski, Morgan & Limber’ s (2012) study, conducted with
4.531 6-12 grade American students found a positive association between cyber victimization and being a
school bully. Hindujia & Patchin (2008) and Ybarra & Mitchell (2004a) supported these results. Also,
youngsters’ cyberbullying is a risk factor for cybervictimization, meaning that a cyberbully is often also
victimized online. Cappadocia, Craig, & Pepler (2013), in their longitudinal study with 1.972 Canadian
students, found a that cyberbullying and cybervictimization of related to each other. This overlap was also
found by Jose, Kljakovic, Scheib, & Notter (2011) and Walrave & Heirman (2011) studies.

3.2.2 Interpersonal level risk factors (Microsystem)

3.2.2.1 Peer group risk factors.
At an interpersonal level, several studies have found that peer influence can constitute a risk factor
for cyber victimization. Peer antisocial influence is positively associated with cyber victimization; in fact
 ACCEPTED MANUSCRIPT
16

Hemphill & Heerde (2014) in their longitudinal study conducted with 927 Australian students aged 10-11
years, found that youngsters involvement as cyber victims is correlated with involvement with antisocial
friends. Also peer rejection is associated with cyber victimization. Bayraktar et al. (2014) and Katzer,
Fetchenhauer, & Belschak (2009) in studies conducted respectively with 2.092 students aged 12-18 years
and 1.700 5-11 grade students, report a positive relationship between cyber victimization and high levels of

T
peer rejection. Perceived peer support: Kowalski et al. (2014) in their review of 131 studies concluded that

IP
cyber victimization and low levels of perceived peer support are significantly correlated. The same

R
association was found by Wang, Iannotti and Nansel (2009) in their study conducted with 7.182 6-10 grade
American students.

SC
3.2.2.2 Family risk factors.

NU
Several studies investigated parents’ role and involvement in cyber victimization, in particular
parental control of technology, meaning parents’ habits and aptitude to check and supervise children’s
MA
understanding of technology and use of online activities. Kowalski et al. (2014)’s summary of findings show
that cybervictimization is associated reported connected with low levels of parental control of technology
(see also Aoyama, Utsumi & Hasegawa, 2012; Mesch, 2009). Parents’ rules on allowed online activities is
D

associated with higher risk of cybervictimization: a lack of clear parental rules on what to do online is
TE

positively associated with cyber victimization, as found by Mesch (2009) by surveying 935 American
students aged 12-17 years. Low communication with parents is a risk factor for cyber victimization.
P

Özdemir (2014) found a positive association between cyber victimization and less communication with
CE

parents by surveying 337 Turkish students aged 15-18 years. Also parental attachment was found to be a
risk factor for cyber victimization, as reported by Bayraktar et al. (2014). The study found that cyber
victimization is associated with poor parental attachment. Emotional parent-child relationship, and in
AC

particular parental anxious concerns is a risk factor for cyber victimization, Katzer, Fetchenhauer, &
Belschak (2009) found a positive relationship between cyber victimization and parental anxious concerns.
Family trouble: Tokunaga’s (2010) review of 25 studies systematically showed the existence of a
relationship between cyber victimization and the presence of family troubles.

3.2.1 Community level risk factors (Mesosystem)

Community level risk factors have rarely being studied. What has been done so far, is mainly
concerned with school risk factors.

3.2.1.1 School risk factors

Studies identified risk factors for cyber victimization by focusing on the schools’ role in preventing or
otherwise facilitating youngsters involvement in cyberbullying as victims. As the Kowalski et al. (2014)
 ACCEPTED MANUSCRIPT
17

review highlights, youngsters’ perception of school climate is a risk factor for cyber victimization, that is, a
perceived negative school climate is associated with cyber victimization. This result is supported by Wang
et al. (2014) in a study of 1023, 5 grade American students, who found a relationship between cyber
victimization and a negative school climate. School safety: Kowalski et al. (2014) found that cyber
victimization is related to a low perception.

T
IP
4. Discussion

R
4.1 Design and research limitations of the narrative review

SC
Narrative reviews do not summarize effect sizes as in a meta-analysis. In addition, using a
predetermined theoretical framework adopted for the criteria of selection of articles, might have left out

NU
some studies which indirectly contribute to the understanding of cyberbullying. Also, we only searched and
analysed English languages papers, and this might have left out other studies published only in other
MA
languages, or unpublished work. Regardless of these limitations, we think that our study is a significant
further theoretical contribution for better understanding cyberbullying and cybervictimization. Our
narrative review is innovative and unique because it enables to advance knowledge about cyberbullying
D

and cybervictimization by adopting a risk and needs assessment approach, in the ecological theory
TE

framework. Hong and Espelage (2012) conducted an ecological system analysis but looked at studies on
bullying and victimisation in school and did not classify studies by systematically looking at what role they
P

can play in predicting occurrence or reoccurrence of the miss behaviour. Addressing each level of risk factor
CE

and assess whether these are static or dynamic, instead, is a step forward for accurate interventions.
Compared to other reviews, Kowalski et al. (2014) adopts the General Aggression Model and is not
specifically addressing risks and needs as the current one, or the work by Tokunga (2010) focused on
AC

comparing and providing a meta synthesis between 25 reviewed studies up to 2009 on definitions and
characteristics of cyberbullying but this was done without a clear theoretical framework.
Cyberbullying is a complex phenomenon that has unique features in comparisons with school bullying,
though it is highly connected. As with many antisocial, disruptive behaviors or aggression in general, it is
not possible to identify a single causal association between one characteristic and the (antisocial) behavior.
To do this, longitudinal studies are needed, and not very many were found. A multiple variety of
characteristics, referred to as risk factors, can play a role and interact. In the present review, we looked at
what is known in the literature with regard to risk factors associated with cyberbullying and
cybervictimization, and we did this with two main issues in mind. On the one hand, we used the ecological
theoretical framework to identify and look at risk factors (Bronfammener, 1976, 1986). As outlined, this
theoretical framework is of high relevance and usefulness for studying cyberbullying since it takes into
consideration the complexity associated with this disruptive youngsters’ behaviour.
 ACCEPTED MANUSCRIPT
18

On the other hand, what we did was to look at the different levels on a risk factor approach. What is
making a child more at risk? Which are the vulnerable aspects of his life that can promote or facilitate the
occurrence or re-occurrence of cyberbullying either as a victim or as a perpetrator? This approach was used
to identify and then classify risk factors, according to whether they are static or dynamic. Dynamic risk
factors can be viewed as ‘needs’. They can change over time and susceptible to intervention.

T
IP
4.2 Cyberbullying and cybervictimization profiles

R
With these two framework in mind, results from the narrative approach showed that there is no one

SC
single risk factor or level of risk that is more relevant than another one; several seem to be associated. Also,
what emerged of interest is that risk factors associated with cyberbullies and with cybervictims are rather
similar, as if being or becoming a cyberbully or cybervictim are not separate processes (Kowalski et al.,

NU
2014; Wright & Li; 2013; Vandebosch & Van Cleemput; 2009). This confirms once more that traditional
victims as well as cybervictims could reverse roles; they share same characteristics because to some extent,
MA
they are the same children and will invert roles (Ybarra & Mitchell, 2004a).
Looking back at Figure 1 presented at the beginning of this paper, we can list what has been found
which risk factor is in fact of relevance and associated with the risk of being a cyberbully or a cybervictim.
D

Cyberbullies are more likely to be boys, have low school achievement and commitment, use internet
TE

devices a lot, lack emotional and cognitive empathy, do not have a great idea of themselves, and are more
impulsive, tend to break rules, conform to social supportive norms about aggression, be also involved in
P

bullying in school, and tend to be morally disengaged. They are more isolated and have no or scarce
CE

parental support or supervision that is specifically related to internet use. They attend schools that do not
have clear rules or policies against cyberbullying or use of the web, and cyberbullies do not in any case feel
AC

bonded to their school.

With regard to cybervictims, they can be boys or girls, tend to have low school achievement, are
online more than non cybervictims, so they are more exposed to risk, they lack social skills, have a low self-
esteem, report depression symptoms, as well as social anxiety, anger and also maladaptive behavior, they
are likely to be victimized at school but also to be bullies and cyberbullies. They lack social support at school
are rejected by other children and also at home have parents with whom they do not communicate a lot,
have poor support and with whom do not have strong bonds. Parents of cybervictims tend to have little
knowledge about internet use and safety regulations and have no or very little control over the children’s
activities online. The schools cybervictims attend lack a supportive climate and these victims feel detached
from their schools.
Overall, based on our findings it turned out that with a few exceptions, cyberbullies share similar risk
factors with those found in past studies on school bullies and the same applies for cybervictims and victims
(Hong and Espelange, 2012; Olweus, 1993).
 ACCEPTED MANUSCRIPT
19

These profiles show us that most of the risk factors associated with cyberbullying and
cybervictimization are dynamic in their nature. They could and can change, ‘naturally’ with time but more
likely and most effectively with tailored interventions. For this reason we conclude that to better
understand and address cyberbullying we need to develop a risk and need assessment tool, for
understanding who is most at risk, and what are the needs that have to be addressed and changed. Some

T
differences occurred. With regard to cyberbullying and cybervictimization a unique risk factors has to do

IP
with online activities and time spent online. Also, what was found contrary to the literature of school

R
bullies is that cybers report higher level of SES. This might be due to the fact that they have more electronic
devices, the most updated that serve different communication feature, increasing in this regard the risk of

SC
cyberbullying and cybervictimization.

NU
4.3 A new approach for predicting and managing risk and needs of cyberbullying and cybervictimization:
risk and needs assessment tool and future research
MA
From our narrative review we conclude that one of the most useful ways to view and understand
cyberbullying and cybervictimsation is by adopting a ‘risk and needs assessment approach’. Risk assessment
of any antisocial behavior is a procedure used to establish whether cyber menaces are likely to take place,
D

or take place again, and even more specifically, whether there is a risk of recidivism of something that is
TE

already taking place. Using a dynamic assessment approach means not simply to identify and sum up risk
factors present in individuals at school, at home, in their use of the internet, but to understand that certain
P

behaviors can be a dynamic combination of individual, environmental and social factors that interact one
CE

with each other in a specific moment in time for that single individual.
Assessing risk factors, and their presence and intensity is not enough to reduce risk, since their impact
AC

is relative to that child, that period of his or her life, and in combination with other factors; risk assessment
is rather complex and cannot be solved just by using an instrument/tool. However, linking identification
and understanding of risk factors at any of the individual, interpersonal, community or societal levels, as we
did, with an assessment approach of such risks to address and manage them, as in the field of juvenile
delinquency or of violence, is a promising and innovative way to address cyberbullying and prevent its
(re)occurrence. We propose to use this approach to address different levels of needs directly related to the
individual, his or her personal relationships or at the community/school level, to develop a dedicated risk
and needs assessment tool. This will facilitate risk management strategies and develop interventions to
reduce the impact of risk factors by addressing the different level of needs (dynamic risk factors) and
increase peers, parents, school, and community commitment and involvement.
This innovative approach would constitute a step forward to future research, and for the development
of a risk and needs assessment tool. Many studies have investigated prevalence and nature of
cyberbullying, and risk factors are becoming well established (Kowalski et al., 2014). What we found
 ACCEPTED MANUSCRIPT
20

innovative which had not been studied and addressed yet, and that we did try to achieve in the current
review, is to use of a risk and needs assessment approach, as in the field of criminology to address and
intervene in an efficient way with regard to (cyber)bullying to try to reduce it within an ecological
framework.
To develop such instruments, longitudinal studies should be set up within a specific theoretical

T
framework, assessing the presence of the different (four) levels of possible risks and needs, measuring also

IP
involvement in cyberbullying and cybervictimization. Data on these risk and needs factors should be

R
collected at least in a 6-9 month period of time, with reliable instruments measuring the different
dimensions. Those risk and needs factors that explain significant levels of variance (one standard deviation

SC
above the mean) will be included in a new instrument that could comprise 10-15 risk and needs factors.
The developed and validated instruments, as in the case of the risk and needs assessment tools developed

NU
for of juvenile delinquency (Borum, 2003) and other violent behaviors (Hart, 2010), would give the
opportunity to assess the risk of cyberbullying and cybervictimization. We think that this type of risk and
MA
needs assessment tool would be of use also for self-assessment, to give the opportunity to students to
understand if they are at risk of being or becoming victims or perpetrators of cyberbullying and whether
what is going on is of law, medium or high risk. This assessment tool would constitute a first step towards
D

intervention, addressing the identified using effective programs.

TE

Still scarce is knowledge about effective programs for cyberbullying and cybervictimization (see e.g.
Mishna, Cook, Saini, Wu, & MacFadden, 2009), but research on risk and needs assessment could provide
P

valuable information about the best target for intervention and how to intervene most effectively.
CE

Acknowledgment
AC

This review was made possible by the DAPHNE Programme of the European Union. European Daphne
III Program 2009-2010 and 2001-2012 (Projects: JLS/2009-2010/DAP/AG/1340; JUST/2011/DAP/AG/3259)
Tabby In Internet. Threat Assessment of Bullying Behaviour among youngsters in Internet and Tabby Trip.
Threat Assessment of Bullying Behaviour among youngsters. TRansferring Internet Preventive procedures
in Europe. The contents of this article are the sole responsibility of the authors of the article and can in no
way be taken to reflect the views of the European Commission. No conflict of interest are present.
The authors are grateful to the invited editors for the support provided.

References2

2
References marked with an asterisk indicate studies used in the narrative review.
 ACCEPTED MANUSCRIPT
21

Ang, R. P., & Goh, D. H. (2010). Cyberbullying among adolescents: The role of affective and cognitive
empathy, and gender. Child Psychiatry & Human Development, 41(4), 387-397.doi: 10.1007/s10578-
010-0176-3.
Aoyama, I., Utsumi, S., & Hasegawa, M. (2012). Cyberbullying in Japan: Cases, government reports,
adolescent relational aggression and parental monitoring roles. In Q. Li, D. Cross, & P.K. Smith (Eds.)
Cyberbullying in the global playground: Research from international perspectives, (pp.183-201).

T
Malden, MA: Blackwell.

IP
Baldry, A. C., & Kapardis, A. (Eds.). (2013).Risk assessment for juvenile violent offending. Routledge.
Baldry, A. C., & Sorrentino A. (in press). Risk and Need Assessment. Encyclopedia of Juvenile Delinquency

R
and Justice, Wiley-Blackwel.
Bandura, A. (1986). Social foundations of thought and action. Englewood Cliffs, NJ: Prentice- Hall.

SC
Barlett, C., & Coyne, S. M. (2014). A meta-.analysis of sex differences in cyber-bullying behavior: The
moderating role of age. Aggressive behavior, 40(5), 474-488. doi:10.1002/ab.21555.
Bauman, S. (2010). Cyberbullying in a rural intermediate school: An exploratory study. The Journal of Early

NU
Adolescence, 30(6), 803-833. doi:10.1177/0272431609350927.
Bayraktar, F., Machackova, H., Dedkova, L., & Cerna, A. (2014). Cyberbullying The Discriminant Factors
Among Cyberbullies, Cybervictims, and Cyberbully-Victims in a Czech Adolescent Sample. Journal of
MA
interpersonal violence. doi:10.1177/0886260514555006.
Besley (2009). Cyberbullying. Retrieved from http://www.cyberbullying.org/
Berne, S., Frisén, A., Schultze-Krumbholz, A., Scheithauer, H., Naruskov, K., Luik, P., Katzer, C., Erentaite, R.,
& Zukauskiene, R. (2013). Cyberbullying assessment instruments: A systematic review. Aggression
D

and violent behavior, 18(2), 320-334. doi:10.1016/j.avb.2012.11.022.

TE

Borum, R. (2003). Managing At-Risk Juvenile Offenders in the Community Putting Evidence-Based Principles
Into Practice. Journal of Contemporary Criminal Justice, 19(1), 114-137.
doi:10.1177/1043986202239745.
P

Borum, R. & Douglas K. S. (2003). New directions in violence risk assessment. Psychiatric Times, 20, 102-
CE

103.
Bronfenbrenner, U. (1994). Ecological models of human development. In T. Husen, & T. N. Postlethwaite
(Eds.), International encyclopedia of education (pp. 1643–1647). (2nd ed.). New York: Elsevier
AC

Sciences.
Bronfenbrenner, U. (1986). Ecology of the family as a context for human development: Research
perspectives. Developmental Psychology, 22(6), 723–742. doi:10.1037/0012-1649.22.6.723.
Bronfenbrenner, U. (1979). Ecology of human development: Experiments by nature and design. Cambridge,
MA: Harvard University Press.
Calvete, E., Orue, I., Estévez, A., Villardón, L., & Padilla, P. (2010). Cyberbullying in adolescents: Modalities
and aggressors’ profile. Computers in Human Behavior, 26(5), 1128-1135.
doi:10.1016/j.chb.2010.03.017.
Cappadocia, M. C., Craig, W. M., & Pepler, D. (2013). Cyberbullying prevalence, stability, and risk factors
during adolescence. Canadian Journal of School Psychology, 28(2), 171-192.
doi:10.1177/0829573513491212.
Casas, J. A., Del Rey, R., & Ortega-Ruiz, R. (2013). Bullying and cyberbullying: Convergent and divergent
predictor variables. Computers in Human Behavior, 29 (3), 580-587. doi:10.1016/j.chb.2012.11.015.
Crick, N. R., & Dodge, K.A. (1994). A review and a reformulation of social information-processing
mechanism in children’s social adjustment. Psychological Bulletin, 115, 74-101. doi:10.1037/0033-
2909.115.1.74.
 ACCEPTED MANUSCRIPT
22

Del Rey, R., Elipe, P., & Ortega-Ruiz, R. (2012). Bullying and cyberbullying: Overlapping and predictive value
of the co-occurrence. Psicothema, 24(4), 608-613.
Dooley, J. J., Pyżalski, J., & Cross, D. (2009). Cyberbullying versus face-to-face bullying. Zeitschrift für
Psychologie/Journal of Psychology, 217(4), 182-188. doi:10.1027/0044-3409.217.4.182.
Erdur-Baker, Ö. (2010). Cyberbullying and its correlation to traditional bullying, gender and frequent and
risky usage of internet-mediated communication tools. New Media & Society, 12(1), 109-125.

T
doi:10.1177/1461444809341260.

IP
Finkelhor, D., Mitchell, K. J., & Wolak, J. (2000). Online victimization: A report on the nation’s youth.
Alexandria, VA: National Center for Missing and Exploited Children.

R
Gámez-Guadix, M., Orue, I., Smith, P. K., & Calvete, E. (2013). Longitudinal and reciprocal relations of
cyberbullying with depression, substance use, and problematic internet use among adolescents.

SC
Journal of Adolescent Health,53(4), 446-452. doi:10.1016/j.jadohealth.2013.03.030.
Gradinger, P., Strohmeier, D., & Spiel, C. (2009). Traditional bullying and cyberbullying: Identification of risk
groups for adjustment problems. Zeitschrift für Psychologie/Journal of Psychology, 217(4), 205-213.

NU
doi:10.1027/0044-3409.217.4.205.
Hemphill, S. A., & Heerde, J. A. (2014). Adolescent predictors of young adult cyberbullying perpetration and
victimization among Australian youth. Journal of Adolescent Health, 55(4), 580-587.
MA
doi:10.1016/j.jadohealth.2014.04.014.
Hemphill, S. A., Tollit, M., Kotevski, A., & Heerde, J. A. (2014). Predictors of Traditional and Cyber-Bullying
Victimization A Longitudinal Study of Australian Secondary School Students. Journal of interpersonal
violence, 1-24. doi:10.1177/0886260514553636.
D

Hinduja, S., & Patchin, J. W. (2008). Cyberbullying: An exploratory analysis of factors related to offending
TE

and victimization. Deviant behavior, 29(2), 129-156. doi:10.1080/01639620701457816.

Hart, S. D. (2008). The Structured Professional Judgment approach to violence risk assessment: Core
principles of SPJ. Paper presented at the Annual Meeting of the International Association of Forensic
P

Mental Health Services, Vienna, Austria.

CE

Hoge, R. D., Vincent, G., & Guy, L. (2012). Predictions and risk/needs assessments. In R. Loeber & D. P.
Farrington (Eds.), From juvenile delinquency to adult crime: Criminal careers, justice policy, and
prevention (pp. 150-183). New York: Oxford.
AC

Holt, T. J., Fitzgerald, S., Bossler, A. M., Chee, G., & Ng, E. (2014). Assessing the risk factors of cyber and
mobile phone bullying victimization in a nationally representative sample of Singapore
youth. International journal of offender therapy and comparative criminology. doi:
10.1177/0306624X14554852.
Hong, J. S., & Espelage, D. L. (2012). A review of research on bullying and peer victimization in school: An
ecological system analysis. Aggression and Violent Behavior, 17(4), 311-322.
doi:10.1016/j.avb.2012.03.003.
Huang, Y. Y., & Chou, C. (2010). An analysis of multiple factors of cyberbullying among junior high school
students in Taiwan. Computers in Human Behavior, 26(6), 1581-1590. doi:10.1016/j.chb.2010.06.005.
Hunt, C., Peters, L., & Rapee, R.,M. (2012). Development of a measure of the experience of being bullied in
youth. Psychological Assessment, 24, 156-165. doi:10.1037/a0025178.
Jose, P. E., Kljakovic, M., Scheib, E., & Notter, O. (2012). The joint development of traditional bullying and
victimization with cyber bullying and victimization in adolescence. Journal of Research on
Adolescence, 22(2), 301-309. doi:10.1111/j.1532-7795.2011.00764.x
Juvonen, J., & Gross, E. F. (2008). Extending the school grounds?—Bullying experiences in
cyberspace. Journal of School Health, 78(9), 496-505. doi:10.1111/j.1746-1561.2008.00335.x
 ACCEPTED MANUSCRIPT
23

Katzer, C., Fetchenhauer, D., & Belschak, F. (2009). Cyberbullying: Who are the victims?: A comparison of
victimization in internet chatrooms and victimization in school. Journal of Media Psychology:
Theories, Methods, and Applications, 21(1), 25-36. doi:10.1027/1864-1105.21.1.25.
Kowalski, R. M., Giumetti, G. W., Schroeder, A. N., & Lattanner, M. R. (2014). Bullying in the digital age: A
critical review and meta-analysis of cyberbullying research among youth. Psychological Bulletin,
140(4), 1073-1137. doi:10.1037/a0035618.

T
Kowalski, R. M., & Limber, S. P. (2013). Psychological, physical, and academic correlates of cyberbullying

IP
and traditional bullying. Journal of Adolescent Health, 53(1), S13-S20.
doi:10.1016/j.jadohealth.2012.09.018.

R
Kowalski, R. M., Morgan, C. A., & Limber, S. P. (2012). Traditional bullying as a potential warning sign of
cyberbullying. School Psychology International, 33(5), 505-519. doi:10.1177/0143034312445244.

SC
Kowalski, R. M., & Limber, S. P. (2007). Electronic bullying among middle school students. Journal of
Adolescent Health, 41(6), S22-S30.doi:10.1016/j.jadohealth.2007.08.017.
Lapidot-Lefler, N., & Dolev-Cohen, M. (2015). Comparing cyberbullying and school bullying among school

NU
students: prevalence, gender, and grade level differences. Social Psychology of Education, 18(1) 1-16.
doi:10.1007/s11218-014-9280-8.
Li, Q. (2008). A cross-cultural comparison of adolescents' experience related to cyberbullying. Educational
MA
Research, 50(3), 223-234. doi:10.1080/00131880802309333.
Li, Q. (2007). New bottle but old wine: A research of cyberbullying in schools. Computers in Human
Behavior, 23(4), 1777-1791. doi:10.1016/j.chb.2005.10.005.
Li, Q. (2006). Cyberbullying in schools a research of gender differences. School Psychology International,
D

27(2), 157-170. doi:10.1177/0143034306064547.

TE

Mehari, K. R., Farrell, A. D., & Le, A.-T. H. (2014). Cyberbullying Among Adolescents: Measures in Search of a
Construct. Psychology of Violence, 4(4), 399-415. doi:10.1037/a0037521.
Menesini, E., Nocentini, A., & Camodeca, M. (2013). Morality, values, traditional bullying, and cyberbullying
P

in adolescence. British Journal of Developmental Psychology, 31(1), 1-14. doi:10.1111/j.2044-

CE

835X.2011.02066.x.
Mishna, F., Khoury-Kassabri, M., Gadalla, T., & Daciuk, J. (2012). Risk factors for involvement in cyber
bullying: Victims, bullies and bully–victims. Children and Youth Services Review, 34(1), 63-70.
AC

doi:10.1016/j.childyouth.2011.08.032.
Mesch, G. S. (2009). Parental mediation, online activities, and cyberbullying. CyberPsychology &
Behavior, 12(4), 387-393. doi:10.1089/cpb.2009.0068.
Modecki, K. L., Minchin, J., Harbaugh, A. G., Guerra, N. G., & Runions, K. C. (2014). Bullying prevalence
across contexts: a meta-analysis measuring cyber and traditional bullying. Journal of Adolescent
Health, 55(5), 602-611. doi:10.1016/j.jadohealth.2014.06.007.
Modecki, K. L., Barber, B. L., & Vernon, L. (2013). Mapping developmental precursors of cyber-aggression:
Trajectories of risk predict perpetration and victimization. Journal of Youth and Adolescence, 42(5),
651-661. doi:10.1007/s10964-013-9938-0.
Office for National Statistics (2012). Measuring National Well-being - Children's Well-being. Retrieved from
http://www.ons.gov.uk/ons/rel/wellbeing/measuring-national-well-being/measuring-children-s-well-
being/art-measuring-children-s-well-being.html.
Olweus, D. (1993). Bullying at school: What we know and what we can do. Malden, MA: Blackwell
Publishing.
Özdemir, Y. (2014). Cyber victimization and adolescent self-esteem: The role of communication with
parents. Asian Journal of Social Psychology, 17(4), 255-263. doi:10.1111/ajsp.12070.
 ACCEPTED MANUSCRIPT
24

Patchin, J. W., & Hinduja, S. (2010). Cyberbullying and self‐esteem. Journal of School Health, 80(12), 614-
621. doi:10.1111/j.1746-1561.2010.00548.x.
Patchin, J. W., & Hinduja, S. (2006). Bullies move beyond the schoolyard a preliminary look at
cyberbullying. Youth violence and juvenile justice, 4(2), 148-169. doi:10.1177/1541204006286288.
Pornari, C. D., & Wood, J. (2010). Peer and cyber aggression in secondary school students: The role of moral
disengagement, hostile attribution bias, and outcome expectancies. Aggressive Behavior, 36(2), 81-

T
94. doi:10.1002/ab.20336.

IP
Pozzoli, T., Gini, G., & Vieno, A. (2012). Individual and class moral disengagement in bullying among
elementary school children. Aggressive behavior, 38(5), 378-388. doi:10.1002/ab.21442.

R
Raskauskas, J., & Stoltz, A. D. (2007). Involvement in traditional and electronic bullying among adolescents.
Developmental psychology, 43(3), 564-575. doi:10.1037/0012-1649.43.3.564.

SC
Schultze-Krumbholz, A., & Scheithauer, H. (2009). Social-behavioral correlates of cyberbullying in a German
student sample. Zeitschrift für Psychologie/Journal of Psychology, 217(4), 224.-226.
doi:10.1027/0044-3409.217.4.224.

NU
Slonje, R. & Smith, P. K. (2008). Cyberbullying: Another main type of bullying? Scandinavian Journal of
Psychology, 49, 147–154. doi:10.1111/j.1467-9450.2007.00611.x .
Slonje, R., Smith, P. K., & Frisén, A. (2013). The nature of cyberbullying, and strategies for
MA
prevention. Computers in Human Behavior, 29(1), 26-32. doi:10.1016/j.chb.2012.05.024.
Smith, P. K., Mahdavi, J., Carvalho, M., Fisher, S., Russell, S., & Tippett, N. (2008). Cyberbullying: Its nature
and impact in secondary school pupils. Journal of Child Psychology and Psychiatry, 49(4), 376-385.
doi:10.1111/j.1469-7610.2007.01846.x.
D

Sourander, A., Klomek, A. B., Ikonen, M., Lindroos, J., Luntamo, T., Koskelainen, M., .Ristkari, T., & Helenius,
TE

H. (2010). Psychosocial risk factors associated with cyberbullying among adolescents: A population-
based study. Archives of General Psychiatry,67(7), 720-728. doi:10.1001/archgenpsychiatry.2010.79.
Steffgen, G., & König, A. (2009). Cyber bullying: The role of traditional bullying and empathy. In The good,
P

the bad and the challenging. Conference Proceedings (Vol. 2, pp. 1041-1047).
CE

Steffgen, G., König, A., Pfetsch, J., & Melzer, A. (2011). Are cyberbullies less empathic? Adolescents'
cyberbullying behavior and empathic responsiveness. Cyberpsychology, Behavior, and Social
Networking, 14(11), 643-648. doi:10.1089/cyber.2010.0445.
AC

Sticca, F., Ruggieri, S., Alsaker, F., & Perren, S. (2013). Longitudinal risk factors for cyberbullying in
adolescence. Journal of Community & Applied Social Psychology, 23(1), 52-67.
doi:10.1002/casp.2136.
Tokunaga, R. S. (2010). Following you home from school: A critical review and synthesis of research on
cyberbullying victimization. Computers in Human Behavior, 26(3), 277-287.
doi:10.1016/j.chb.2009.11.014.
Topçu, C., Erdur-Baker, Ö., & Capa-Aydin, Y. (2008). Examination of cyberbullying experiences among
Turkish students from different school types. CyberPsychology & Behavior, 11(6), 643-648.
doi:10.1089/cpb.2007.0161.
Vandebosch, H., & Van Cleemput, K. (2009). Cyberbullying among youngsters: Profiles of bullies and
victims. New Media & Society,11(8), 1349-1371. doi:10.1177/1461444809341263.
Van Geel, M., Vedder, P., & Tanilon, J. (2014). Relationship between peer victimization, cyberbullying, and
suicide in children and adolescents: a meta-analysis. JAMA Pediatrics, 168(5), 435-442.
doi:10.1001/jamapediatrics.2013.4143.
Vieno, A., Gini, G., & Santinello, M. (2011). Different forms of bullying and their association to smoking and
drinking behavior in Italian adolescents. Journal of School Health, 81(7), 393-399. doi:10.1111/j.1746-
1561.2011.00607.x.
 ACCEPTED MANUSCRIPT
25

Vivolo-Kantor, A. M., Martell, B. N., Holland, K. M., & Westby, R. (2014). A systematic review and content
analysis of bullying and cyber-bullying measurement strategies. Aggression and Violent
Behavior, 19(4), 423-434. doi:10.1016/j.avb.2014.06.008.
Walrave, M., & Heirman, W. (2011). Cyberbullying: Predicting victimisation and perpetration. Children &
Society, 25(1), 59-72. doi:10.1111/j.1099-0860.2009.00260.x.
Wang, J., Iannotti, R. J., & Nansel, T. R. (2009). School bullying among adolescents in the United States:

T
Physical, verbal, relational, and cyber. Journal of Adolescent Health, 45(4), 368-375.

IP
doi:10.1016/j.jadohealth.2009.03.021.
Wang, W., Vaillancourt, T., Brittain, H. L., McDougall, P., Krygsman, A., Smith, D., Cunningham, C. E.,

R
Haltigan, J. D.. & Hymel, S. (2014). School climate, peer victimization, and academic achievement:
Results from a multi-informant study. School Psychology Quarterly, 29(3), 360-377.

SC
doi:10.1037/spq0000084.
Welsh, B. C. & Farrington, D. P. (2007). Preventing crime: What works for children, offenders, victims and
places (pp. 227–237). The Netherlands: Springer.

NU
Willard, N. E. (2007). Cyberbullying and cyberthreats: Responding to the challenge of online social
aggression, threats, and distress. Research Press.
Williams, K. R., & Guerra, N. G. (2007). Prevalence and predictors of internet bullying. Journal of Adolescent
MA
Health, 41(6), S14-S21. doi:10.1016/j.jadohealth.2007.08.018.
Wright, M. F., & Li, Y. (2013). The association between cyber victimization and subsequent cyber
aggression: The moderating effect of peer rejection. Journal of Youth and Adolescence, 42(5), 662-
674. doi:10.1007/s10964-012-9903-3.
D

Ybarra, M. L., Mitchell, K. J., Wolak, J., & Finkelhor, D. (2006). Examining characteristics and associated
TE

distress related to Internet harassment: Findings from the Second Youth Internet Safety Survey.
Pediatrics, 118, e1169–e1177. doi:10.1542/peds.2006-0815.
Ybarra, M. L., & Mitchell, K. J. (2004a). Online aggressor/targets, aggressors, and targets: A comparison of
P

associated youth characteristics. Journal of Child Psychology and Psychiatry, 45(7), 1308-1316.
CE

doi:10.1111/j.1469-7610.2004.00328.x.
Zhou, Z., Tang, H., Tian, Y., Wei, H., Zhang, F., & Morrison, C. M. (2013). Cyberbullying and its risk factors
among Chinese high school students. School Psychology International, 34(6),630-647.
AC

doi:10.1177/0143034313479692.
 ACCEPTED MANUSCRIPT
26

Figure 2.
Flow Diagram of all stages of literature review

RECORDS IDENTIFIED THROUGH

T
DATABASE SEARCHING, RECORDS RECORDS EXCLUDED (N=7052)

IP
SCREENED (N=7199)

R
SC
NU
FULL-TEXT ARTICLES ASSESSED FOR FULL-TEXT ARTICLES EXCLUDED, WITH
ELIGIBILITY (N=147) REASONS (N=60)
MA

DUPLICATES OF SAME STUDIES

D

ARTICLES ACCEPTED AGAINST DIFFERENT PUBLICATIONS, OR NOT

SELECTION CRITERIA (N=87) FULLY COMPLYING WITH
TE

REQUIREMENTS (N=34)
P
CE

ELIGIBLE ARTICLES REPORTING RISK

FACTORS FOR INVOLVEMENT IN CYBER
AC

BULLYING AS VICTIM OR PERPETRATOR

(N=53)
 ACCEPTED MANUSCRIPT
27

Figure 1.
Risk factors for cyberbullying and cybervictimization according to the ecological framework

T
IP
Level of identification of Risk Factors Cyber bullying -victimization risk factors Ecological

CR
levels

US
Macrosystem

N
MA
Cultural norms and believes
about violence, societal
responses on safe youth use of
internet, companies online

D
protection issues

P TE
CE
AC
 ACCEPTED MANUSCRIPT
28

Culture/Society

T
Policies on cyber bullying,

IP
Community prevention programs at the
school and community levels

CR
Support on strategies for
prevention of violence

School

US
School climate, rules on the use
of cyber-communication,

N
intervention programs,
Peer and friends
management strategies in

MA
cases of bullying

Family

D
Peer interactions, friendship,
isolation, attitudes towards

TE
cyberbullying Mesosystem
Individual
(personality and

P
biological) Monitoring, information,
CE procedures to increase ITC
security, child’s supervision
off/online activities, parental
AC
styles

Use of new technologies,

numbers of hours online,
involvement in bullying, low
self-esteem, empathy, moral
disengagement, impulsivity Microsystem

Ontogenetic

School climate, school rules on the use

of cyber-communication, intervention
programs, management strategies in
cases of bullying
 ACCEPTED MANUSCRIPT
29

Table 1.

T
Summary of risk factors according to the ecological framework’s levels for cyberbullying

IP
Study Sample Method Positive association with Cyber Perpetration

CR
N Age/grade Nationality Individual-level risk factor

N US
S
Gender

MA
Lapidot-Lefler & Dolev- 465 7-12 grade Israel Online survey Being a boy
Cohen (2015)

D
Barlett & Coyne (2014)* Review of 109 studies Meta-analysis Being a boy

TE
+
Erdur-Backer (2010) 276 10-14 years Turkey Self-reported Being a boy

P
questionnaire

+
CE
Huang & Chou (2010) 545 7-9 grade Taiwan Anonymous Being a boy
AC
survey

+
Sourander et al (2010) 2.215 13-16 years Finland Questionnaire Being a boy

+
Kowalski & Limber (2007) 3.767 6-8 grade USA Self-reported Being a girl
questionnaire

+
Li (2006) 264 7-9 grade Canada Anonymous Being a boy
survey

S
SES
 ACCEPTED MANUSCRIPT
30

Wang, Iannotti & Nansel 7.182 6-10 grade USA Health Behavior Positive relationship with SES
+
(2009) in School-Aged
Children 2005

T
IP
Survey

CR
D
School commitment

+
Kowalski & Limber (2013) 931 6-12 grade USA Anonymous Low school commitment

US
survey

N
MA
+
Ybarra & Mitchell (2004a) 1.501 10-17 years USA Interview via Low school commitment
telephone

D
TE
D
Technology use

P
CE Self-reported ICT
expertise
AC

Sticca, Ruggieri, Alsaker & 835 Mean age Swiss Longitudinal Higher self-reported ICT expertise
+a
Perren (2013) =13.2 study

+
Walrave & Heirman (2011) 1.318 12-18 years Belgium Self-reported Higher self-reported ICT expertise
survey

+
Hinduja & Patchin (2008) 1.378 <18 years USA Online survey Higher self-reported ICT expertise

Internet use
 ACCEPTED MANUSCRIPT
31

Kowalski et al. (2014)* Review of 131 studies Meta-analysis

Casas, Del Rey, & Ortega- 893 11-19 years Spain Self-reported Internet addiction
+
Ruiz (2013) questionnaire

T
IP
Sticca, Ruggieri, Alsaker & 835 Mean age Swiss Longitudinal Reported more Internet use

CR
+a
Perren (2013) =13.2 study

Mishna, Khoury-Kassabri, 2.186 10-17 years Canada Self-reported Reported more Internet use

US
+
Gadalla & Daciuk (2012) questionnaire

N
+
Walrave & Heirman (2011) 1.318 12-18 years Belgium Self-reported Reported more Internet use

MA
survey

+
Erdur-Backer (2010) 276 10-14 years Turkey Self-reported Girls reported more frequent use

D
questionnaire

TE
+
Hinduja & Patchin (2008) 1.378 <18 years USA Online survey Reported more Internet use(no gender differences

P
explained)

Ybarra & Mitchell (2004a)

+
1.501 10-17 years
CE
USA Interview via Reported more Internet use(no gender differences
AC
telephone explained)

Risky online behaviors

Kowalski et al. (2014)* Review of 131 studies Meta-analysis Reported more risky internet use (no gender differences
explained)

Casas, Del Rey, & Ortega- 893 11-19 years Spain Self-reported Reported more risky internet use (no gender differences
+
Ruiz (2013) questionnaire explained)

Mishna, Khoury-Kassabri, 2.186 10-17 years Canada Self-reported Reported more risky internet use (no gender differences
 ACCEPTED MANUSCRIPT
32

+
Gadalla & Daciuk (2012) questionnaire explained)

+
Erdur-Backer (2010) 276 10-14 years Turkey Self-reported Boys reported more risky internet use

T
questionnaire

IP
D
Personality

CR
Empathy

US
Kowalski et al. (2014)* Review of 131 studies Meta-analysis Low empathy

N
Casas, Del Rey, & Ortega- 893 11-19 years Spain Self-reported Low empathy

MA
Ruiz (2013)+ questionnaire

Topcu &Erdur-Baker 795 13-18 years Turkey Self-reported Boys with low levels of empathy

D
+
(2012) questionnaire

TE
Steffgen, König, Pfetsch & 2.070 7-13 grade Luxemburg Self-.reported Low empathy

P
+
Melzer (2011) survey

Ang &Goh (2010)

+
396 12-18 years
CE
Singapore Self-reported Cognitive empathy Girls with low levels of cognitive empathy
survey
AC

Affective empathy Boys with low levels of both cognitive and affective
empathy

Steffgen &König (2009) 2.070 7-13 grade Luxemburg Self-reported Low empathy
survey
 ACCEPTED MANUSCRIPT
33

Self-esteem

Modecki, Barber, & Vernon 1.364 12-14 years Australia Longitudinal Low levels of self-esteem
+
(2013) study

T
IP
+
Patchin & Hinduja (2010) 1.963 10-16 years USA Self-reported Low levels of self-esteem

CR
survey

US
Self-control

Bayraktar, Machackova, 2.092 12-18 years Czech Online survey Low levels of self-control

N
Republic

MA
Dedkova, Cerna,&
Ševčíková (2014)

D
D
Values

TE
Moral disengagement

P
Kowalski et al. (2014)* Review of 131 studies Meta-analysis High levels of moral disengagement
CE
Cappadocia, Craig, & Pepler 1.972 9-12 grade Canada Longitudinal High levels of antisocial behaviors
AC
+a
(2013) study

Menesini, Nocentini & 390 14-18 years Italy Self reported Immoral and disengaged behaviors
+
Camodeca (2013) questionnaire

Pozzoli, Gini, & Vieno 663 Mean age =9 Italy Self reported High levels of moral disengagement
(2012) questionnaire

+
Bauman (2010) 221 5-8 grade USA Self-reported High levels of moral disengagement
survey
 ACCEPTED MANUSCRIPT
34

+
Pornari & Wood (2010) 339 7-9 grade UK Questionnaire High levels of moral disengagement

Rule braking

T
Sticca, Ruggieri, Alsaker & 835 Mean age Swiss Longitudinal Rule braking behaviors

IP
+a
Perren (2013) =13.2 study

CR
Normative beliefs about

US
aggression

N
Kowalski et al. (2014)* Review of 131 studies Meta-analysis Presence of normative beliefs about aggression

MA
D
Psychological states

D
Depression

TE
Modecki, Barber ,& Vernon 1.364 12-14 years Australia Longitudinal Early depressed mood
+a
(2013) study

P
CE Maladaptive behaviors
D
AC
Cappadocia, Craig, & Pepler 1.972 9-12 grade Canada Longitudinal Alcohol drinking High level of alcohol drinking
+a
(2013) study

D
Bullying

Bullies

a
Hemphill & Heerde (2014) 927 10-11 years Australia Longitudinal Being a traditional bully
study

Kowalski et al. (2014)* Review of 131 studies Meta-analysis Being a traditional bully
 ACCEPTED MANUSCRIPT
35

Modecki, Minchin, Review of 80 studies Meta-analysis Being a traditional bully

Harbaugh,

Guerra, & Runions (2014)*

T
IP
Cappadocia, Craig, & Pepler 1.972 9-12 grade Canada Longitudinal Being a traditional bully

CR
+a
(2013) study

+
Kowalski & Limber (2013) 931 6-12 grade USA Anonymous Being a traditional bully

US
survey

N
Sticca, Ruggieri, Alsaker & 835 Mean age Swiss Longitudinal Being a traditional bully

MA
+a
Perren (2013) =13.2 study

Del Rey, Elipe & Ortega- 274 12-18 years Spain Self-reported Being a traditional bully

D
Ruiz(2012) questionnaire

TE
Mishna, Khoury-Kassabri, 2.186 10-17 yeas Canada Self-reported Being a traditional bully

P
+
Gadalla & Daciuk (2012) questionnaire

Gradinger, Strohmeier, & 761 14-19 years

CE
Austria Self-report Boys involvement as traditional bullies
AC
+
Spiel, (2009) questionnaire

Vandebosch & Van 2.052 11-18 years Belgium Online Survey Being a traditional bully
+
Cleemput (2009)

+
Hinduja & Patchin (2008) 1.378 <18 years USA Online survey Being a traditional bully

+
Smith et al (2008) 533 11-16 years UK Self-reported Being a traditional bully
questionnaire

+
Raskauskas & Stoltz (2007) 84 13-18 years USA Self-reported Being a traditional bully
 ACCEPTED MANUSCRIPT
36

questionnaire

T
Victims

IP
CR
Kowalski, Morgan& Limber 4.531 6-12 grade USA Survey Girls victims of traditional bullying
+
(2012)

US
+
Ybarra & Mitchell (2004a) 1.501 10-17 years USA Interview via Being a traditional victim
telephone

N
MA
Cyber victims

Kowalski et al. (2014)* Review of 131 studies Meta-analysis Being a cyber victim

D
TE
+a
Wright & Li (2013) 261 6-8 grade USA Longitudinal Being a cyber victim

P
study
CE
AC
Vandebosch & Van 2.052 11-18 years Belgium Online Survey Being a cyber victim
+
Cleemput (2009)

Study Sample Method Microsystem Positive association with Cyber Perpetration

N Age/ Grade Interpersonal-level risk

factor

D
Peer group risk factors

Antisocial influences
 ACCEPTED MANUSCRIPT
37

Cappadocia, Craig, & Pepler 1.972 9-12 grade Canada Longitudinal Fewer pro-social peer influences
+a
(2013) study

Peer rejection

T
IP
Bayraktar, Machackova, 2.092 12-18 years Czech Online survey High level of peer rejection

CR
Republic
Dedkova, Cerna,&
Ševčíková (2014)

US
+a
Wright & Li (2013) 261 6-8 grade USA Longitudinal High level of peer rejection

N
study

MA
Perceived peer support

D
Calvete, Orue, Estévez, 1.431 12-17 years Spain Self-reported Lack of perceived peer support

TE
+
Villardón,& Padilla (2010) questionnaire

P
+
Williams & Guerra (2007) 3.339 5,8,11 grade USA Electronic Lack of perceived peer support

CE questionnaire

Study Sample Method Microsystem Positive association with Cyber Perpetration

AC

N Age/grade Nationality Interpersonal-level risk

factor

D
Family risk factors

Parental support

a
Hemphill & Heerde (2014) 927 10-11 years Australia Longitudinal Low parental support
 ACCEPTED MANUSCRIPT
38

study

Wang, Iannotti, & Nansel 7.182 6-10 grade USA Health Behavior Low parental support
+
(2009) in School-Aged

T
IP
Children (HBSC)
2005

CR
Survey

N US
Emotional bond

MA
+
Ybarra & Mitchell (2004a) 1.501 10-17 years USA Interview via Poor emotional bond with a caregiver
telephone

D
TE
Parental attachment

P
Bayraktar, Machackova, 2.092 12-18 years
CE
Czech Online survey Poor parental attachment
Republic
AC

Dedkova, Cerna,&
Ševčíková (2014)

Family management
(rules and children
monitoring)

a
Hemphill & Heerde (2014) 927 10-11 years Australia Longitudinal Poor family management
study
 ACCEPTED MANUSCRIPT
39

Kowalski et al. (2014)* Review of 131 studies Meta-analysis Poor family management

+
Ybarra & Mitchell (2004a) 1.501 10-17 years USA Interview via Poor family management
telephone

T
IP
Parents’

CR
involvement with the
child’s internet use

US
Vandebosch & Van 2.052 11-18 years Belgium Online Survey Poor parents involvement

N
+
Cleemput (2009)

MA
Study Sample Method Mesosystem Positive association with Cyber Perpetration

D
N Age/grade Nationality Community-level risk

TE
factor

P
D
School risk factors
CE Perception of school
climate
AC

Kowalski et al. (2014)* Review of 131 studies Meta-analysis Negative school climate

Casas, Del Rey, & Ortega- 893 11-19 years Spain Self-reported Lack of teachers support, clear rules and school safety
+
Ruiz (2013) questionnaire

+
Williams & Guerra (2007) 3.339 5,8,11 grade USA Electronic Negative school climate
questionnaire
 ACCEPTED MANUSCRIPT
40

T
IP
CR
Connection to school

+
Williams & Guerra (2007) 3.339 5,8,11 grade USA Electronic Low perception of being connected to school

US
questionnaire

N
School safety

MA
Kowalski et al. (2014)* Review of 131 studies Meta-analysis Low levels of school safety

D
TE
S D
Note. Denotes the Static Risk Factors associated with youngsters involvement in cyberbullying. Denotes the Dynamic Risk Factors associated with youngsters involvement in cyberbullying. *

P
+ a
Denotes reviews on Risk Factors for youngsters involvement in cyber bullying. Denotes publications cited in reviews on Risk Factors for youngsters involvement in cyberbullying. Denotes
CE
longitudinal studies. Unless differently specified, no gender differences were found or investigated.
AC
 ACCEPTED MANUSCRIPT
41

Table 2

T
Summary of risk factors according to the ecological framework’s levels for cybervictimization

IP
Study Sample Method Positive association with Cyber victimization

CR
N Age/grade Nationality Individual-level risk factor

US
S
Gender

N
Bayraktar, Machackova, 2.092 12-18 years Czech Online survey Being a girl

MA
Dedkova, Cerna,& Republic
Ševčíková (2014)

D
Holt, Fitzgerald, Bossler, 4.315 Primary/ Singapore Self-report Being a girl

TE
Chee, Ng (2014) secondary questionnaire
schools

P
Zhou, Tang, Tian, Wei,
Zhang, & Morrison (2013)
1.438 10-12 grade CE
China Anonymous
survey
Being a boy
AC

+
Erdur-Backer (2010) 276 14-18 years Turkey Self-reported Being a boy
questionnaire

+
Sourander et al. (2010) 2.215 13-16 years Finland Questionnaire Being a girl

+
Kowalski & Limber (2007) 3.767 6-8 grade USA Self-reported Being a girl
questionnaire

D
School problems
 ACCEPTED MANUSCRIPT
42

+
Hinduja & Patchin (2008) 1.378 <18years USA Online survey More school problems

T
D
Academic achievement

IP
CR
Wang et al. (2014) 1.023 5 grade USA Paper and pencil Lower academic achievement
survey

US
Tokunaga(2010)* Review of 25 studies/articles Meta-analysis Low academic commitment

N
MA
D
Technology use

D
Internet use

P TE
Kowalski et al. (2014)* Review of 131 studies
CE Meta-analysis Reported more Internet use

Zhou, Tang, Tian, Wei, 1.438 10-12 grade China Anonymous Reported more Internet use
AC

Zhang, & Morrison (2013) survey

Mishna, Khoury-Kassabri, 2.186 10-17 years Canada Self-reported Reported more Internet use
+
Gadalla & Daciuk (2012) questionnaire

+
Erdur-Backer (2010) 276 14-18 years Turkey Self-reported Girls reported more internet use
questionnaire

Vandebosch & Van 2.052 11-18 years Belgium Online Survey Reported more Internet use
+
Cleemput (2009)
 ACCEPTED MANUSCRIPT
43

+
Hinduja & Patchin (2008) 1.378 <18years USA Online survey Reported more Internet use

+
Ybarra & Mitchell (2004a) 1.501 10-17 years USA Interview via Reported more Internet use
telephone

T
IP
Risky internet use

CR
Kowalski et al. (2014)* Review of 131 studies Meta-analysis Reported more risky internet use

US
Mishna, Khoury-Kassabri, 2.186 10-17 years Canada Self-reported Reported more risky internet use
+
Gadalla & Daciuk (2012) questionnaire

N
MA
+
Walrave & Heirman (2011) 1.318 12-18 years Belgium Self-reported Reported more risky internet use
survey

D
+
Erdur-Backer (2010) 276 14-18 years Turkey Self-reported Boys reported more risky internet use

TE
questionnaire

P
Katzer, Fetchenhauer, & 1.700 5-11 grade Germany Survey Reported more risky internet use
Belschak (2009)
+
CE
+
Mesch (2009) 935 12-17 years USA Survey Reported more risky internet use
AC

+
Hinduja & Patchin (2008) 1.378 <18years USA Online survey Reported more risky internet use

D
Personality

Perceived social
intelligence

Kowalski et al. (2014)* Review of 131 studies Meta-analysis Low perceived social intelligence

Hunt, Peters, &Rapee 218 8-15 years Australia Personal Low perceived social intelligence
 ACCEPTED MANUSCRIPT
44

+
(2012) Experiences
Checklist (PECK)

Schultze-Krumbholz & 71 7-8-10 Germany Self-reported Low perceived social intelligence

T
+

IP
Scheithauer (2009) grade questionnaire

CR
Empathy

Schultze-Krumbholz & 71 7-8-10 Germany Self-reported Low levels of empathy

US
+
Scheithauer (2009) grade questionnaire

N
Self-esteem

MA
Bayraktar, Machackova, 2.092 12-18 years Czech Online survey Low levels of self-esteem
Dedkova, Cerna,& Republic

D
Ševčíková (2014)

TE
+
Kowalski & Limber (2013) 931 6-12 grade USA Anonymous Low levels of self-esteem

P
survey
CE
Modecki, Barber & Vernon 1.364 12-14 years Australia Longitudinal Developmental decrease in self-esteem
AC
(2013) study

+
Patchin & Hinduja (2010) 1.963 10-16 years USA Self-reported Low levels of self-esteem
survey

Katzer, Fetchenhauer, & 1.700 5-11 grade Germany Survey Low levels of self-concept
+
Belschak (2009)

Emotional control
 ACCEPTED MANUSCRIPT
45

a
Hemphill & Heerde (2014) 927 10-11 years Australia Longitudinal Poor emotional control
study

Hemphill, Tollit, Kotevski, & 673 12-13 years Australia Longitudinal Emotional dysregulation

T
a

IP
Heerde (2014) study

CR
D
Values

US
Moral disengagement

Kowalski at al. (2014)* Review of 131 studies Meta-analysis High levels of moral disengagement

N
MA
D
Psychological states

Depression

D
TE
Cappadocia, Craig, & 1.972 9-12 grade Canada Longitudinal High level of depression
a
Pepler (2013) study

P
Gámez- Guadix, Orue, 845 13-17 years Spain Longitudinal High level of depression
Smith, & Calvete (2013)
a
CE study
AC
+
Kowalski & Limber (2013) 931 6-12 grade USA Anonymous High level of depression
survey

Modecki, Barber, & 1.364 12-14 years Australia Longitudinal Early depressed mood
Vernon (2013) study

Social anxiety

Kowalski et al. (2014)* Review of 131 studies Meta-analysis High level of social anxiety

Anger
 ACCEPTED MANUSCRIPT
46

Kowalski et al. (2014)* Review of 131 studies Meta-analysis Feeling angry

Psychosocial problem

T
Tokunaga(2010)* Review of 25 studies/articles Meta-analysis Presence of psychosocial problem

IP
D

CR
Maladaptive behaviors

Substance use

US
Gámez- Guadix, Orue, 845 13-17 years Spain Longitudinal Substance use

N
a
Smith, & Calvete (2013) study

MA
+
Hinduja & Patchin (2008) 1.378 <18years USA Online survey Substance use

D
Bullying

TE
Victims

P
Hemphill, Tollit, Kotevski, & 673 12-13 years Australia Longitudinal Being a traditional victim
Heerde (2014)
a CE study
AC
a
Hemphill & Heerde (2014) 927 10-11 years Australia Longitudinal Being a traditional victim
study

Holt, Fitzgerald, 4.315 Primary/ Singapore Self-report Being a traditional victim

secondary questionnaire
Bossler, Chee, Ng (2014)
schools

Kowalski et al. (2014)* Review of 131 studies Meta-analysis Being a traditional victim
 ACCEPTED MANUSCRIPT
47

Cappadocia, Craig, & 1.972 9-12 Grade Canada Longitudinal Being a traditional victim
a
Pepler (2013) study

Del Rey, Elipe & Ortega- 274 12-18 years Spain Self-reported Being a traditional victim

T
IP
Ruiz(2012) questionnaire

CR
Mishna, Khoury-Kassabri, 2.186 10-17 years Canada Self-reported Being a traditional victim
+
Gadalla & Daciuk (2012) questionnaire

US
Gradinger, Strohmeier, & 761 14-19 years Austria Survey Being a traditional victim
+

N
Spiel, (2009)

MA
Vandebosch & Van 2.052 11-18 years Belgium Online Survey Being a traditional victim
+
Cleemput (2009)

D
+
Hinduja & Patchin (2008) 1.378 <18years USA Online survey Being a traditional victim

TE
+
Smith et al (2008) 533 11-16 years UK Anonymous self- Being a traditional victim

P
reported
CE questionnaire
AC
+
Raskauskas & Stoltz (2007) 84 13-18 years USA Self-reported Being a traditional victim
questionnaire

Bullies

Kowalski, Morgan & Limber 4.531 6-12 grade USA Survey Boys involvement as bullies
+
(2012)

+
Hinduja & Patchin (2008) 1.378 <18years USA Online survey Being a traditional bully
 ACCEPTED MANUSCRIPT
48

+
Ybarra & Mitchell (2004a) 1.501 10-17 USA Interview via Being a traditional bully
telephone

Cyber bullying

T
IP
Cappadocia, Craig, & 1.972 9-12 Grade Canada Longitudinal Being a cyber perpetrator

CR
a
Pepler (2013) study

Jose, Kljakovic, Scheib, & 1.700 11-16 years New Zeland Longitudinal Being a cyber perpetrator

US
+a
Notter (2011) study

N
+
Walrave & Heirman (2011) 1.318 12-18 years Belgium Self-reported Being a cyber perpetrator

MA
survey

Study Sample Method Microsystem

D
TE
N Age/Grade Nation Interpersonal-level risk
factor

P
CE Peer group risk factors
D

Antisocial influences
AC

a
Hemphill & Heerde (2014) 927 10-11 years Australia Longitudinal Association with antisocial friends
study

Peer rejection

Bayraktar, Machackova, 2.092 12-18 years Czech Online survey High level of peer rejection
Dedkova, Cerna,& Republic
Ševčíková (2014)
 ACCEPTED MANUSCRIPT
49

Katzer, Fetchenhauer, & 1.700 5-11 grade Germany Survey Low level of popularity
+
Belschak (2009)

Perceived peer support

T
IP
Kowalski et al. (2014)* Review of 131 studies Meta-analysis Low peer support

CR
Wang, Iannotti, & Nansel 7.182 6-10 grade USA Health Behavior Low peer support
+
(2009) in School-Aged

US
Children (HBSC)

N
2005 Survey

MA
Study Sample Method Microsystem Positive association with Cyber victimization

N Age/Grade Nationality Interpersonal-level risk

D
factor

TE
D
Family risk factors

P
CE Parental control of
technology
AC

Kowalski et al. (2014)* Review of 131 studies Meta-analysis Low levels of parental control

Aoyama, Utsumi & 133 9-12 grade Japan Self-reported Low levels of parental control
+
Hasegawa (2012) survey

+
Mesch (2009) 935 12-17 years USA Survey Low levels of parental web sites monitoring

Parents’ rules on allowed

online activities
 ACCEPTED MANUSCRIPT
50

+
Mesch (2009) 935 12-17 years USA Survey Lack of parental clear rules on allowed online activities

T
Communication with

IP
parents

CR
Özdemir (2014) 337 15-18 years Turkey Survey Less communication with parents

US
Parental attachment

N
Bayraktar, Machackova, 2.092 12-18 years Czech Online survey Poor parental attachment

MA
Dedkova, Cerna,& Republic
Ševčíková (2014)

D
TE
Emotional parent-child

P
relationship

Katzer, Fetchenhauer, & 1.700 5-11 grade

CE
Germany Survey Parents anxious concerned
+
Belschak (2009)
AC

Family trouble

Tokunaga(2010)* Review of 25 studies/articles Meta-analysis Presence of family trouble

Study Sample Method Mesosystem Positive association with Cyber victimization

N Age/grade Nationality Community-level risk

factor
 ACCEPTED MANUSCRIPT
51

D
School risk factors

Perception of school

T
climate

IP
Kowalski et al. (2014)* Review of 131 studies Meta-analysis Negative school climate

CR
US
Wang et al. (2014) 1.023 5 grade USA Paper and pencil Poor perception of school climate
survey

N
MA
School safety

Kowalski et al. (2014)* Review of 131 studies Meta-analysis Low perception of school safety

D
TE
S D
Note. Denotes the Static Risk Factors associated with youngsters involvement in cybervictimization. Denotes the Dynamic Risk Factors associated with youngsters involvement in cybervictimization *
+ a
Denotes reviews on Risk Factors for youngsters involvement in cybervictimization. Denotes publications cited in reviews on Risk Factors for youngsters involvement in cybervictimization. Denotes

P
longitudinal studies. Unless differently specified, no gender differences were found or investigated.
CE
AC