Content uploaded by Antonio Camacho
Author content
All content in this area was uploaded by Antonio Camacho on Aug 24, 2023
Content may be subject to copyright.
Computers in Human Behavior 148 (2023) 107924
Available online 19 August 2023
0747-5632/© 2023 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
Joint trajectories of cyberbullying perpetration and victimization:
Associations with psychosocial adjustment
Antonio Camacho
a
,
*
, Peter K. Smith
b
, Rosario Ortega-Ruiz
a
, Eva M. Romera
a
a
Department of Psychology, Universidad de C´
ordoba, Spain
b
Department of Psychology, Goldsmiths, University of London, London, United Kingdom
ARTICLE INFO
Handling editor: Bjorn de Koning
Keywords:
Social adjustment
Need for popularity
Perceived peer popularity
Perceived internet popularity
Adolescents
Longitudinal proles
ABSTRACT
Cyberbullying is one of the most disturbing characteristics regarding the relationship between adolescents on the
Internet. Although a longitudinal overview of the trajectories that adolescents may develop has been established,
there is a lack of understanding of these when both perpetration and victimization are considered together. The
present study aimed to analyze the joint trajectories between cyberbullying perpetration and victimization
among highly involved adolescents and to examine whether these proles are associated with social adjustment,
need for popularity and perceived popularity (off and online). A total of 3012 adolescents (M
AgeT1
=13.15, SD =
1.09; 50% girls) aged 11–16 participated in the study at four time points (each six months apart). The results of
growth mixture modelling yielded a four-class solution for cyberbullying victimization and perpetration sepa-
rately for those adolescents highly involved. When unied in a parallel process, this resulted in three distinct
proles: decrease both, increase perpetration, and increase both. Finally, multi-group growth mixture models
indicated that these proles showed differences in baseline and evolution of social adjustment, need for popu-
larity and popularity. The ndings support the relevance of considering the evolution of both perpetration and
victimization when preventing cyberbullying, as well as addressing the psychosocial adjustment and motivations
for behavior of those involved.
1. Introduction
The dynamic evolution of technology provides the opportunity for
alternative ways for adolescents to learn and develop social interactions.
However, problematic use of technology has resulted in a pattern of
bullying among adolescents. Cyberbullying is dened as an intentional
and repetitive act to harm someone who cannot easily defend her/
himself via electronic devices (Smith et al., 2008). Compared to
face-to-face bullying, cyberbullying has specic characteristics; ano-
nymity, exibility in time (the victim is available 24/7), unlimited
reproduction and a potentially larger audience (Slonje et al., 2013).
Given the psychosocial outcomes that may lead the involvement from
late adolescence to adulthood (Katsaras et al., 2018), understanding the
factors involved in the phenomenon has become a challenge for the
policy and education communities to address or reduce it. For the design
of effective prevention and intervention programs, emphasis has been
placed on the importance of considering involvement in both victimi-
zation and perpetration as well as their characteristics associated with a
heterogeneous nature (Sumter et al., 2012).
This phenomenon is particularly relevant during middle adolescence
because it peaks at this age range (Kowalski et al., 2014). Thus, it is
particularly interesting to examine the trajectories of cyberbullying and
its associated motivational and social factors during middle adolescence.
However, there may be important individual differences in involvement
concerning stability and developmental patterns (Yoo, 2021). Whereas
longitudinal studies on cyberbullying have been conducted, few have
undertaken growth trajectory analyses to analyze their association on
motivational and psychosocial factors. Considering cyberbullying perpe-
tration and cyberbullying victimization, knowledge about the proles of
involvement between both could provide useful insights into the specic
risk groups and psychosocial characteristics that underlie such patterns
of development. The present study considered the proles of the joint
trajectories of perpetration and victimization to analyze their stability
and changes over time, and the evolution of social adjustment, popularity
and need for popularity.
* Corresponding author. Av. San Alberto Magno, s/n, 14071, C´
ordoba, Spain.
E-mail address: antonio.camacho@uco.es (A. Camacho).
Contents lists available at ScienceDirect
Computers in Human Behavior
journal homepage: www.elsevier.com/locate/comphumbeh
https://doi.org/10.1016/j.chb.2023.107924
Received 17 April 2023; Received in revised form 10 August 2023; Accepted 18 August 2023
Computers in Human Behavior 148 (2023) 107924
2
1.1. Trajectories of cyberbullying
Cyberbullying is particularly prevalent through adolescence. During
early adolescence, general prevalence increases notably, until reaching a
peak middle adolescence and thereafter slightly decreasing (Zych &
Farrington, 2021). For example, a study using linear growth curve
analysis has reported an increase in cyberbullying involvement during
early and middle adolescence (Charalampous et al., 2021), while
research with the same methodology on middle and late adolescence
reported a decrease in cyberbullying behaviours (Cho & Rustu, 2020).
Such studies assumed that the trajectory represents the population as a
whole (homogeneity). However, there are differences in the trajectories
that individual pupils may develop. Therefore, longitudinal studies need
to be considered to account for the fact that within the sample there are
subpopulations with different trajectories.
Several studies have analyzed the heterogeneity of the cyberbullying
perpetration and victimization trajectories. Considering victimization,
Yoo (2021) identied three groups, uninvolved, increase and decrease.
Other studies have also identied three groups based on cyberbullying
perpetration, although varying in patterns: uninvolved, decrease, and
moderate to high (Kim et al., 2017; Song et al., 2020); uninvolved,
moderate stable, and decrease (Yoo, 2021); and uninvolved, decrease,
and increase (Cho & Glassner, 2021). These studies provide support for
heterogeneity in cybervictimization and cyberperpetration trajectories
independently. Nevertheless, no studies have explored both trajectories
as a joint process, leaving an important gap in our knowledge.
Several meta-analyses of cross-sectional studies have shown a mod-
erate to strong association between cyberbullying perpetration and
victimization (r =.43 to r =.51) (Kowalski et al., 2014; Lozano-Blasco
et al., 2020; Walters, 2021). In a representative cross-sectional study
among adolescents in southern Spain, 11% of participants were found to
be clustered in a group which had high involvement in both victimization
and perpetration in cyberbullying (Zych et al., 2018). This co-occurrence
and the transition between cyberbullying perpetration and victimization
requires that common developmental processes should be accounted for
when analyzing cyberbullying trajectories and their associated factors.
Although the heterogeneity of both involvements has not been
addressed in cyberbullying, the joint trajectory between victimization
and perpetration has been reported in previous studies for the phenom-
enon of bullying generally (Cho & Lee, 2020; de Vries et al., 2021; Zhou
et al., 2022). Such studies have found common trajectories between
bullying perpetration and victimization that suggest co-occurrence be-
tween both (e.g., increase both, decrease both), and trajectories that
differ (e.g., high perpetration, decrease victimization, decrease perpe-
tration). While bullying and cyberbullying are phenomena with
different characteristics, the overlap between victimization and perpe-
tration within each phenomenon is similar (r
Bullying
=.39, r
Cyberbullying
=
.44, see Walters, 2021, for a meta-analysis), so it is expected to nd joint
developmental trajectories between cyberbullying perpetration and
victimization. The present study focused on parallel process growth
mixture modelling to address the changes along time in cyberbullying
proles. This type of analysis is particularly relevant due to the specic
characteristics of cyberbullying, where there is a higher uctuation from
one prole to another, with victims and aggressors not xed and where
even a one-time involvement may be considered as relevant (Vande-
bosch & Van Cleemput, 2009).
Involvement in cyberbullying remains sporadic and adolescents tend
to be involved in specic cyberbullying situations (Camacho et al., 2023;
Modecki et al., 2014). Few adolescents are involved in a wide range of
cyberbullying situations; although their number is lower, their higher
level of involvement corresponds to higher levels of psychosocial risks
and requires a differentiated research analysis (Mishna et al., 2012).
Analyzing those highly involved in a wide range of cyberbullying situ-
ations may led to identify the associated psychosocial characteristics. In
contrast, getting involved in one specic situation may be motivated by
a momentary impulsive reaction or even by unawareness (Runions &
Bak, 2015), which does not have to characterize these adolescents as at
risk regarding their psychosocial characteristics. Therefore, the present
study aims to analyze the joint trajectories of perpetration and victimi-
zation only with those adolescents with a high involvement in either.
This would give insight into the psychosocial characteristics of those
adolescents who have been highly involved in cyberbullying and
whether they differ from those with low engagement or uninvolved
adolescents.
1.2. Cyberbullying and psychosocial adjustment
Cyberbullying as a serious issue involves not just individuals, but
rather the whole social and environmental framework. Previous
research on abusive patterns has demonstrated how social structure can
inuence individual adolescent behavior. This impact is particularly
noteworthy after the transition from early to middle adolescence, as it
changes the landscape for adolescents, requiring them to readjust their
social position within the new peer group. Within the social structure of
the peer group, the status of adolescents (i.e., popularity), motivation
towards social domination (i.e., need for popularity) or how they are
accepted within the peer group (i.e., social adjustment) has been asso-
ciated to being an online victim or perpetrator.
Popularity is a signicant social factor in the understanding of the
power imbalance, as it characterizes the social structure of peer groups
in adolescence. Popularity involves a high level of social dominance,
reputation and power within the peer group (Cillessen & Marks, 2011).
While popularity may be prosocial in nature (e.g., leadership), it may also
appear to be of a selsh character (e.g., aggressive, arrogant, or rude)
(Malamut et al., 2021). Adolescence is a period where behavior is highly
inuenced by the peer group. According to social identity theory (Hogg,
2016), popular adolescents are considered more valuable within the
peer group, through the recognition, consideration, support and admi-
ration of others, or the ability to inuence and attract others. Previous
studies have found that bullies often are more popular and able to
establish and maintain their dominance over others (Romera et al.,
2019). Popular adolescents may perceive themselves as safe to behave
offensively against others without worrying about adverse social effects
from the peer group (Vanden Abeele et al., 2017). Meanwhile, victims
tend to have lower levels in comparison with uninvolved adolescents
(Romera et al., 2019). Although adolescents generally consider bullying
perpetration as immoral, the association between perpetration and
popularity has been identied by them as a normative risk behavior (i.e.,
cool or extrovert image) during adolescence (Strindberg et al., 2020). In
that sense, bullying has been characterized as a deliberate strategy to
keep a position of dominance (Pouwels et al., 2018). In addition to its
intentional nature, the harm caused by the perpetrator may also involve
a conscious act accepted in the peer group, as socially motivated
behavior.
With Internet connectivity, cyberbullying has also been discussed as
socially oriented behavior playing a role in providing higher social
status (Vanden Abeele et al., 2017). Although cyberbullying behaviors
take place on the Internet, during adolescence it is intrinsically
contextualized in a social network as the classroom or school. Adoles-
cents often use social networks or chats on the Internet to support or
reinforce ofine relationships. Self-perceived peer popularity has been also
associated with a higher later involvement in cyberbullying perpetration
(Ranney & Troop-Gordon, 2020; Vanden Abeele et al., 2017; Wang &
Ngai, 2022; Wright, 2014). Furthermore, online perpetration has also
been reported as associated with popularity (Wegge et al., 2016). This
implies that popularity and online perpetration are associated factors
that reinforce each other in a bidirectional way. Online perpetration may
be targeted at a lower socially visible individual, as evidenced by the
lower popularity of those adolescents who experience cyberbullying
victimization (Festl & Quandt, 2013).
A complementary research approach, in addition to the achievement
of popularity, addresses the importance of the motivation to reach
A. Camacho et al.
Computers in Human Behavior 148 (2023) 107924
3
popularity (see meta-analysis by Samson et al., 2022). The achievement
of a key social position is a goal of adolescents (de Vries et al., 2021).
Adolescents commonly pursue increased status among their peers,
perhaps by providing power-related access to group resources. Adoles-
cents may engage in a certain pattern of behavior aimed at being
identied as popular within the group, which is known as the need for
popularity (Santor et al., 2000). According to social information processing
theory (Dodge, 2014), adolescents may take decisions based on their
expectations of the outcome of their behavior. Motivation encourages
individuals to target information about the goal to be achieved, to
appraise the suitability of alternatives, and to activate behavioral re-
sponses to achieve the goal. The need for popularity has been broadly
addressed in terms of its association with other risk factors involved on
the Internet (i.e., sexting, grooming, disclosure of feelings, body
dissatisfaction or porn use) (Del Rey et al., 2019; Kim, 2020; Swirsky
et al., 2022; Vanden Abeele et al., 2014).
Given that perpetration may be a tool that provides adolescents with
a valued social status of privilege, they may nd enough reason to
engage in perpetration to achieve the desired recognition within the
peer group (Romera et al., 2021). Previous research reported that high
levels of need for popularity are associated with higher involvement in
cyberbullying perpetration (Romera et al., 2017; Vanden Abeele et al.,
2017). Differences have been found in the literature on the association
between need for popularity and victimization. In a descriptive study
about the roles of cyberbullying, higher levels of need for popularity
were found for those adolescents involved in both cyberbullying perpe-
tration and victimization (Romera et al., 2016). The experience of only
victimization on the Internet has not been found to be related with higher
status motivations (Romera et al., 2017). This may be explained by the
fact that purely victimized adolescents may not prefer to be the focus of
interest, rather they may wish to be more unnoticed, avoiding the
chance of further victimization (Breslend et al., 2018). However, need for
popularity can be a risk factor for cybervictimization (Goagoses et al.,
2022; Wright et al., 2022) when the overlap between victimization and
perpetration is not controlled for. For this reason, need for popularity
may lead to further cyberbullying victimization since high levels may
also represent high involvement in perpetration.
Social adjustment provides a useful overview of adolescents’ t with
their immediate social context. It involves spending time with others,
positive relationships, and supportive and caring friends (Romera et al.,
2016). Social adjustment serves a central function in the socio-emotional
development of adolescents. Peer group membership protects in-
dividuals from contextual distress. Adolescents with low social adjust-
ment may perceive their interpersonal connections as impaired and
decrease interaction with peers (Ding et al., 2020). As a result, low social
adjustment has been associated with higher cyberbullying victimization
(Romera et al., 2016), even as an outcome (Espino et al., 2023). Ado-
lescents who have been involved in both perpetration and victimization
online also tend to have lower social adjustment compared to uninvolved
adolescents (Romera et al., 2016). Conversely, online perpetrators have
been identied with normative levels of social adjustment (i.e., like un-
involved adolescents) (Romera et al., 2016). Online perpetration may be
perceived as a way of attaining acceptance among peers. In groups
established on the assumption of immoral norms, group integration may
be strengthened through perpetration.
While there are cross-sectional studies on the co-occurrence of
involvement in cyberbullying perpetration and/or victimization with
popularity (face-to-face and online), need for popularity and social
adjustment, there are no studies that have analyzed whether and how
such correlates are associated over time.
1.3. The current study
The present study focused on two objectives. The rst objective was
to analyze the joint trajectories between perpetration and victimization in
those adolescents highly involved in cyberbullying. As evidenced by
previous meta-analyses (Kowalski et al., 2014; Lozano-Blasco et al.,
2020; Walters, 2021) and longitudinal research on bullying generally
(Cho & Lee, 2020; de Vries et al., 2021; Zhou et al., 2022), common
trajectories were expected to reveal the co-occurrence between cyber-
bullying perpetration and victimization (i.e., increase and/or decrease
perpetration and victimization) (Hypothesis 1.1). It is expected that
these common patterns would not be the case for all adolescents, so that,
as has been evidenced in bullying perpetration (Cho & Lee, 2020; de
Vries et al., 2021; Zhou et al., 2022), there would be some adolescents
whose trajectories over time would be different in perpetration and
victimization (i.e., increase perpetration and decrease victimization and
vice versa) or those who are only highly involved in one specic area
over time (Hypothesis 1.2). The second objective was to explore the
motivational and psychosocial characteristics associated with cyber-
bullying trajectories. It was expected that those adolescents with com-
mon perpetration and victimization trajectories, or only victimization
trajectories, showed inverse associations with the development of
popularity and social adjustment, but that there was no link with the need
for popularity (Hypothesis 2.1). Conversely, trajectories associated only
with perpetration would be positively associated with the development
of popularity, need for popularity, and social adjustment (Hypothesis 2.2).
2. Method
2.1. Participants and procedure
The study used data from a six-month four wave longitudinal project
designed to investigate developmental risk and protective factors on
adolescents’ involvement in bullying and cyberbullying. A convenience
sample was recruited from 13 middle schools from the south of Spain.
Data were obtained from 3012 adolescents (52% rural; 48% urban). The
study was approved by the Bioethics and Biosafety Committee of the
Universidad de C´
ordoba. Each school was contacted and enquired about
their participation in the study. After approval by the management
team, governmental permissions were obtained. Each parent authorized
their child to take part in the study. This study is part of a broader
research project aimed at studying the risk and protective factors of
bullying and cyberbullying. The participants completed the question-
naire collection during school hours for approximately 40 min under the
supervision of experienced researchers in psychology and with the
attendance of a teacher. Participants received standardized instructions
and were informed of their voluntary, anonymous, and condential
participation. Two annual measurements were taken at the beginning
and at the end of both 2017/18 (grades 7, 8 and 9) and 2018/19 (8, 9
and 10) academic years, over 18 months. The mean age of the partici-
pants at the beginning of the research was 13.15 (SD =1.09). Girls
accounted for 50% of the sample. The participation at each measure-
ment point was 2788 (Time 1; T1), 2551 (T2), 2473 (T3) and 2360 (T4).
The main reason for attrition was that participants did not attend school
on the day of collection or because they moved to another school.
Missing data were addressed via Full Information Maximum Likelihood
(FIML) estimation, given the data was missing at random (MAR;
χ
2
/df =
1.40) (Bollen, 1989).
2.2. Measures
2.2.1. Cyberbullying
Cyberbullying perpetration and victimization were measured with
the Spanish version of the European Cyberbullying Intervention Project
Questionnaire (Del Rey et al., 2015), comprising 22 items. Each form of
cyberbullying was addressed by 11 items covering physical, verbal, and
relational cyberbullying: for example, “I created a fake account, pre-
tending to be someone else” (perpetration); “Someone posted embar-
rassing videos or pictures of me online” (victimization). Participants
were asked to report the frequency with which they had experienced
each situation over the last three months, with ve response options (0
A. Camacho et al.
Computers in Human Behavior 148 (2023) 107924
4
=never, 1 =once or twice, 2 =once a month, 3 =once a week, 4 =more
times a week). The measure reported a good internal reliability for each
time point (McDonald’s Omega,
ω
perpetration
; T1 =0.87, T2 =0.89, T3 =
0.89, T4 =0.89;
ω
victimization
; T1 =0.88, T2 =0.85, T3 =0.88, T4 =
0.89). In T1, the Conrmatory Factor Analysis (CFA) reported that the
two-factor structure of the scale has good psychometric properties
χ
2
(208) =2104.488, p <.001; CFI =0.918, TLI =0.909, and RMSEA =
0.058, 90% CI [0.056, 0.060].
2.2.2. Social adjustment
This was assessed using a subscale from the Adolescent Multidi-
mensional Social Competence Questionnaire (G´
omez-Ortiz et al., 2017).
The social adjustment subscale consists of eight items (e.g., “My class-
mates and friends know they can count on me when they have to
organize some kind of activity”). Adolescents responded on a 5-point
Likert-type scale ranging from 1 (completely false) to 7 (completely
true). This subscale has shown to have good reliability previously with
Spanish adolescents (Romera et al., 2022). In the present study, the
measure reported a good internal reliability for each time (McDonald’s
Omega,
ω
; T1 =0.86, T2 =0.88, T3 =0.88, T4 =0.90). In T1, good
psychometric properties of the scale were reported in the CFA:
χ
2
(20) =
499.678, p <.001; CFI =0.959, TLI =0.942, and RMSEA =0.093, 90%
CI [0.086, 0.100].
2.2.3. Need for popularity
This was measured with the Spanish version of the Need for Popu-
larity Scale (Del Rey et al., 2019; Santor et al., 2000). The instrument
comprises 12 items (e.g., “I’ve been friends with some people, just
because others liked them”) with 7 Likert-type response options from 1
=Strongly disagree to 7 =Strongly agree. The measure reported a good
internal reliability for each time (McDonald’s Omega,
ω
; T1 =0.87, T2
=0.89, T3 =0.89, T4 =0.90). In T1, good psychometric properties of
the scale were reported in the CFA:
χ
2
(44) =725.142, p <.001; CFI =
0.973, TLI =0.966, and RMSEA =0.075, 90% CI [0.070, 0.080].
2.2.4. Popularity
Adolescents rated popularity by being asked two items about how
popular they considered themselves to be among their peers and on the
Internet (Self-perceived peer popularity: “I am popular among the peers
in my class”; Self-perceived Internet popularity: “I am popular in the
virtual social networks”; r
T1
=.53, r
T2
=.57, r
T3
=.62, r
T4
=.66)
(Pozzoli & Gini, 2021). Participants answered each item on a scale
ranging from 1 (completely false) to 7 (completely true).
2.3. Statistical analyses
Three steps were taken to address the objectives of the study. In the
rst step, participants with a high level of involvement in cyberbullying
(using the average score of the items), at least at one time point in the
study, were selected. These criteria are stricter than other criteria that
selection those involved based on involvement in any form of cyber-
bullying. In our study, we used a stricter criterion, following studies that
recommend an averaged cut-off with multiple-item scales to prevent
targeting those adolescents who were involved only in one form of
cyberbullying (see Zych et al., 2016 for a systematic review). We rst
considered a strict cut-off of an average equal to or more than 2 (“once a
month”), while being prepared to relax this to ensure that at least 5% of
the sample were included so that heterogeneous trajectories could be
analyzed.
In the second step, a parallel process growth mixture modelling was
performed to capture the joint developmental trajectory of cyberbully-
ing perpetration and victimization simultaneously for those adolescents
involved in cyberbullying. From 2 to 6 classes were estimated. The most
optimal solution was adopted by comparing the different classes under
the criteria of: Akaike Information Criterion (AIC; lowest value),
Bayesian Information Criterion (BIC; lowest value), the adjusted
Bayesian Information Criterion (aBIC; lowest value), entropy (values
close to 1 support a better classication accuracy), and theoretical
meaningfulness (Nylund et al., 2007).
1
The loglinear parameterization
was applied to determine combinations between categorical latent
variables.
In the third step, multi-group analysis was used to examine differ-
ences in the trajectory of social adjustment, need for popularity, self-
perceived peer popularity and self-perceived Internet popularity
among the joint cyberbullying trajectory classes. The intercept and slope
were identied as indicators of trajectory. The intercept established if
social domain differed between the groups at the beginning of the study,
while slope reported the change of trajectory over time. Signicant
differences between intercept and slope among classes in comparison
with the uninvolved adolescents in cyberbullying were analyzed with
the Wald Test (Muth´
en & Muth´
en, 1998). Analyses were performed with
Mplus 8.7. (Muth´
en & Muth´
en, 1998), with robust maximum likelihood
(MLR) as an estimator. FIML was used to account for all available in-
formation without removing any missing information or replacing
missing data.
3. Results
First, those adolescents involved in cyberbullying were selected.
When selecting on the basis of cut-off (greater or equal to 2), less than
5% of participants were involved (3%), so this subsample was not suf-
cient for analyzing growth trajectories. Therefore, the slightly relaxed
criterion of 1.5 was set, which resulted in a sufcient sample of 224 (7%)
adolescents highly involved in cyberbullying perpetration and/or
victimization. In this sample, boys (66%) were more involved in
cyberbullying than girls (34%) (
χ
2 (1)
=23.43, p <.001). No differences
were found in involvement by age (
χ
2 (5)
=8.18, p >.05). Of those
adolescents, 24% were involved at some point in perpetration (33%
girls), 40% in victimization (43% girls), and 36% in both (25% girls).
These 224 adolescents were selected for the analysis on simultaneous
trajectories of cyberbullying perpetration and victimization.
Parallel process growth mixture modelling was developed to analyze
joint trajectories in cyberbullying perpetration and victimization (see
Table 1 for model t indices). While the ve-prole solution had better
entropy, the BIC value was considerably lower in the four-class solution,
which was nally retained. Fig. 1 overviews the following four classes
for cyberbullying perpetration: (a) decrease class (55%), (b) low and in-
crease class (23%), (c) middle and increase class (12%), and (d) high stable
class (10%). Regarding the categories of cyberbullying victimization, the
following classes were found: (a) high decrease class (48%), (b) low
decrease class (24%), (c) low increase class (15%), and (d) high increase
class (12%) (see Fig. 1).
One main interest of the study was to analyze the role of adolescents
in cyberbullying perpetration and victimization as parallel processes. A
total of 16 classes were identied according to the simultaneous tra-
jectory in cyberbullying perpetration and victimization (see Table 2 for
the joint probabilities membership). This led to a low probability of
simultaneity for several classes. Consequently, and because of the
theoretical rationale and providing enough probability for further
comparative analysis, the classes were clustered according to the com-
mon direction of the same involvement (i.e., “high decrease” and “low
decrease” classes in cyberbullying victimization were treated together as
a decrease) and the shared or different direction in the trajectories both
perpetration and victimization (i.e., shared direction: increase both
perpetration and victimization; different direction: increase perpetra-
tion and decrease victimization).
1
The Lo–Mendell–Rubin Test, the Vuong–Lo–Mendell–Rubin likelihood ratio
test, and the Bootstrapped Likelihood Ratio Test have not been reported
because it is not available for growth mixture modelling with more than one
categorical latent variable (Muth´
en & Muth´
en, 1998–2017).
A. Camacho et al.
Computers in Human Behavior 148 (2023) 107924
5
The rst class was labelled decrease both class (52%; n =125), for
adolescents whose levels of cyberbullying perpetration and victimiza-
tion tended to decrease simultaneously. In the second class, labelled as
increase perpetration class (14%; n =34), involvement in cyberbullying
perpetration tended to increase while victimization decreased. Finally, a
third class was labelled increase both (22%; n =46), because cyberbul-
lying perpetration and victimization increased simultaneously. Three
more classes could have been identied as sharing common trends but
were not considered due to the low number of participants in each
(increase victimization, n =6; high perpetration and decrease
victimization, n =13; and high perpetration and increase victimization
n =9).
Multi-group analyses were performed to address the second main
objective of the study, on the motivational and psychosocial character-
istics associated with cyberbullying trajectories. Those adolescents
largely uninvolved in cyberbullying (n =2788) comprised the reference
class. Table 3 reports the intercept and slope for each class based on their
initial levels and trajectories in social adjustment, need for popularity,
perceived peer popularity, and perceived Internet popularity while these
trends are shown in Fig. 2.
Wald Test results indicated that adolescents clustered in decrease both
class (d =0.38) and increase perpetration class (d =0.78), reported lower
initial levels of social adjustment compared to those adolescents not
involved in cyberbullying. Both the uninvolved class and the increase
perpetration class reported an ascending trajectory across the four time
points. The increase both class had a decreasing trend.
Based on need for popularity, the decrease both class (d =1.23), in-
crease perpetration class (d =0.73), and increase both class (d =0.82),
displayed higher initial levels compared to those adolescents not
involved in cyberbullying. The need for popularity decreased over time
for the decrease both class. According with perceived peer popularity, those
adolescents in increase both class reported higher initial levels compared
to uninvolved class (d =0.80). The perceived peer popularity increased over
time for the increase perpetration class, while the trajectory of those un-
involved was decreasing.
Regarding perceived Internet popularity, those adolescents clustered in
decrease both class (d =1.22) and increase both class (d =1.39) reported
higher initial levels compared to those adolescents not involved. The
Table 1
Model t indices of parallel process growth mixture models.
Number of proles AIC BIC aBIC Entropy Cyberbullying perpetration Cyberbullying victimization
1 2 3 4 5 6 1 2 3 4 5 6
2 3126.85 3253.08 3135.82 0.782 .81 .19 .82 .18
3 3034.36 3245.89 3049.40 0.802 .72 .17 .10 .65 .28 .08
4 2942.54 3273.47 2966.06 0.811 .49 .24 .15 .11 .43 .27 .17 .13
5 2921.65 3406.11 2956.08 0.832 .36 .26 .17 .12 .09 .33 .32 .14 .10 .10
6 2994.09 3666.18 3041.86 0.777 .25 .23 .20 .17 .11 .03 .31 .20 .16 .15 .09 .08
Note. AIC =Akaike Information Criterion; BIC =Bayesian Information Criterion; aBIC =adjusted Bayesian Information Criterion.
Fig. 1. Estimated cyberbullying perpetration and victimization trajectories.
Table 2
Joint probability of trajectory group membership.
Cyberbullying
perpetration
Cyberbullying victimization
High
decrease
Low
decrease
Low
increase
High
increase
Decrease 34% (n =
76)
a
18% (n =
49)
a
2% (n =5) 0% (n =1)
Low and increase 9% (n =
21)
b
4%
2
(n =
9)
b
7% (n =
15)
c
4% (n =8)
c
Middle and increase 1% (n =3)
b
0%
2
(n =
1)
b
6% (n =
13)
c
5% (n =
10)
c
High stable 4% (n =9) 2% (n =4) 0% (n =1) 4% (n =8)
Note.
a
Decrease both class.
b
Increase perpetration class.
c
Increase both class.
A. Camacho et al.
Computers in Human Behavior 148 (2023) 107924
6
Table 3
Multivariate growth mixture models for the joint trajectories of cyberbullying perpetration and victimization on motivation and psychosocial adjustment.
Variable Uninvolved (n =2788) Decrease both (n =116) Increase perpetration (n =34) Increase both (n =46) Wald Test
Intercept
1
χ
2
(1)
=7.67**
Social adjustment 5.61 5.35
1
5.07
2
5.49
2
χ
2
(1)
=6.96**
Need for popularity 1.87 2.77
3
2.39
4
2.45
5 3
χ
2
(1)
=50.38***
Perceived peer popularity 4.48 4.76 4.03 5.04
6 4
χ
2
(1)
=5.75*
Perceived Internet popularity 3.79 4.67
7
3.74 4.77
8 5
χ
2
(1)
=9.56*
Slope
6
χ
2
(1)
=4.23*
Social adjustment 0.02 0.04 0.19 −0.14
9 7
χ
2
(1)
=21.75***
Need for popularity 0.00 −0.24
10
0.05 0.17
8
χ
2
(1)
=13.33***
Perceived peer popularity ¡0.04 0.02 0.23
11
−0.16
9
χ
2
(1)
=6.55*
Perceived Internet popularity 0.01 0.03 0.30
12
−0.01
10
χ
2
(1)
=24.48***
11
χ
2
(1)
=6.83**
12
χ
2
(1)
=4.99*
Note. Signicant slopes are reported in bold (p <.05). Intercept and slope signicantly different from the uninvolved group (reference) in italic.
Fig. 2. Estimated cyberbullying classes on the multivariate growth mixture models.
A. Camacho et al.
Computers in Human Behavior 148 (2023) 107924
7
increase perpetration class increased over time in perceived Internet
popularity.
4. Discussion
The present explores the joint trajectories of cyberbullying perpetration
and victimization and their association with motivational and social
factors. Based on longitudinal data from early and middle adolescence,
the study focused on trajectories in common or different between
perpetration and victimization in adolescents highly involved in
cyberbullying. The selection on the basis of an averaged cut-off with
multiple-items (Zych et al., 2016) allowed us to focus longitudinally on
only those adolescents who have had a high involvement in cyberbul-
lying, and to analyze the characteristics of psychosocial adjustment in
these adolescents who have a higher level of exposure.
While the previous literature has overcome some of the limitations of
the early research on developmental processes in cyberbullying through
the identication of heterogeneous trajectories, it remains to explore
such variety by controlling the possible overlap between perpetration
and victimization. The rst objective of this study was to address
whether the trajectories of cyberbullying perpetration and victimization
have a common or different development in involved adolescents. We
found that boys were more involved than girls, as also found in an earlier
meta-analysis, (3:1 ratio; Smith et al., 2019). The results of the parallel
process growth mixture modelling found a solution of four proles for
both dimensions of cyberbullying. In relation to the trajectories of
cyberbullying perpetration, the results of the development of the proles
are in line with previous studies on increasing (Cho & Glassner, 2021),
decreasing (Cho & Glassner, 2021; Kim et al., 2017), and stable high
proles in perpetration (Kim et al., 2017), and increasing and decreasing
proles in victimization (Song et al., 2020; Yoo, 2021). The nding that
a stable high prole was not found in victimization is in line with a
previous latent transition analysis, where cybervictims tended to be
more sporadic, as 70–81% were not subsequently involved (Tian et al.,
2023). Greater stability in perpetration could be explained by the
overlap with victimization that can be mutually reinforcing (Walters,
2021). This argues for the need to analyze both trajectories combined as
a way of identifying common patterns.
The growth mixture modelling derived 16 proles on parallel tra-
jectory between victimization and perpetration that were clustered into
three proles due to the overlap in the direction of development. The
proles found were increase both (perpetration and victimization), in-
crease perpetration (and decrease victimization), and decrease both. These
results highlight the heterogeneity of adolescents’ cyberbullying
involvement in terms of its temporal evolution (stability, increase or
decrease) and involvement in the phenomenon (victimization and/or
perpetration).
As reported in different meta-analyses on cyberbullying (Kowalski
et al., 2014; Lozano-Blasco et al., 2020; Walters, 2021) and on joint
trajectories in bullying (Cho & Lee, 2020; Zhou et al., 2022), two of the
proles found had common trajectories between perpetration and
victimization (Hypothesis 1.1). However, the overlap between perpe-
trator and victimization is different in cyberbullying due to its specic
characteristics. In this case the power imbalance may be related to the
anonymity of the perpetrator so that victims, afraid of retaliation, may
use the Internet to equally harm another (Runions & Bak, 2015). In
addition, known and visible perpetrators in cyberbullying are exposed to
others which increases their likelihood of being attacked by others and
becoming cybervictims at the same time (Kowalski et al., 2014). As in
previous studies, a tendency was found, where trajectories did not
converge but had opposite directions so that there was a group of ado-
lescents initially involved in many forms of cybervictimization (not
involved in perpetration) which was decreasing over time as perpetra-
tion increased (Hypothesis 1.2). A further interesting trend is to discuss
why these adolescents involved in victimization to a large extent were
not involved in perpetration at the same time, but only subsequently.
This can be explained by social learning (Bandura, 1986), so that the
experience of these events makes adolescents learn maladaptive coping
strategies; rather than acts of revenge that develop immediately as a
response to their own victimization (Runions & Bak, 2015), they may
develop patterns to interact aggressively with others in their own future
relationships. In addition, having suffered from victimization may lead
them to develop the necessary means to attack online without being
recognized.
Of particular interest in the longitudinal trend is to consider what
differentiating characteristics may distinguish the increase both and
increase perpetration groups, since both share the increase in perpe-
tration but differ in their involvement in victimization. It is also of in-
terest to identify such characteristics between the decrease both and
increase both proles because they follow opposite trajectories.
Regarding the development of popularity, signicant differences in
the initial levels and the trajectories were found between adolescents
involved in cyberbullying and those uninvolved. In line with previous
results (Wegge et al., 2016), the present research reinforces the account
of perceived popularity (independently of the context) as a relevant factor
to understand involvement and change in cyberbullying. For the increase
both class, although at baseline they did not show involvement in
cyberbullying, their initial levels of perceived peer and Internet popularity
were higher compared to uninvolved participants. While decrease both
class (high initial cyberbullying) had initially high levels of perceived
popularity, these levels remained stable over time, despite their decrease
in cyberbullying involvement. Furthermore, ndings provide evidence
that perceived popularity may be linked differently for perpetration and
victimization (Hypotheses 2.1 and 2.2). It appears that perpetration can
be associated and developed over time in the same way with perceived
peer and Internet popularity (Hypothesis 2.2). This is evidenced by a
signicant increase in perceived popularity over time in both contexts for
the increase perpetration class.
However, popularity was inversely correlated with victimization
(Hypothesis 2.1). Thus, increase perpetration class (high initial cyberbul-
lying victimization) exhibited lower levels of perceived peer popularity (not
in the Internet context) baseline compared to the uninvolved class. This
experience may have an immediate effect on their social relationships
with their peers and their perception in real life, due to increased
withdrawal or inhibition and thus affect their attraction, interest with
peers or social exclusion. However, the lack of differences at baseline
with perceived Internet popularity may be supported by the specic
characteristics of online context. On the Internet, adolescents can
interact with a much larger number of people than in their immediate
environment, which means that they can have a larger group of fol-
lowers and friends than in their everyday life (Breslend et al., 2018). In
addition, the self-identity that adolescents may develop on the Internet
may mean that their cyberbullying victimization may not affect their
perceived Internet popularity and attention because of the online
disinhibition.
While perceived popularity may be an indicator of social adjustment,
the motivations adolescents may have to achieve visibility are also
relevant. In the present study, need for popularity was found to be an
associated factor over time for adolescents’ later involvement in
cyberbullying. Being highly involved in cyberbullying at any of the time
points was associated with higher levels of need for popularity in any of
the groups compared to those uninvolved at the beginning of the study.
These results are in line with previous studies that have shown that
adolescent behaviors aimed at achieving status within the peer group
are linked to involvement in cyberbullying (Romera et al., 2016; Vanden
Abeele et al., 2017). However, the development of need for popularity
was not equal for all groups. It remained high and stable in the increase
perpetration and increase both groups. These results indicate that no
matter whether initially being victimized online or not (increase perpe-
tration and increase both groups), initial levels of need for popularity (as a
risk factor) were subsequently associated with involvement in cyber-
bullying perpetration.
A. Camacho et al.
Computers in Human Behavior 148 (2023) 107924
8
Moreover, such levels of need for popularity did not decrease over
time. As an associated factor, need for popularity only decreased when the
involvement in cyberbullying declined over time (Hypothesis 2.2).
During adolescence, cyberbullying perpetration may be a coercive strategy
motivated by obtaining the desired power that comes with popularity
among peers. Consistent with recent studies (Goagoses et al., 2022;
Wright et al., 2022), need for popularity was also found to be a risk factor
for cyberbullying victimization, as higher initial levels of need for popu-
larity were associated with an increase in victimization (at the same time
as perpetration increased). This may be explained because the imple-
mentation of different strategies (e.g., cyberbullying perpetration) to
gain status may make a student vulnerable to be the focus for cyber-
bullying victimization (Walters et al., 2021).
A further characteristic related to cyberbullying is the adolescents’
involvement in positive and supportive relationships with peers. The
results of the present study indicate that the different patterns of
engagement in cyberbullying found were closely correlated with social
adjustment. The baseline differences in social adjustment found are
consistent with previous cross-sectional research about the roles of
cyberbullying (Romera et a., 2016). Lower levels of initial social
adjustment were found for those adolescents with a high baseline
involvement in cyberbullying victimization, either together with low (in-
crease perpetration class) or high (decrease perpetration class) involvement
in cyberbullying perpetration (compared to the uninvolved group). Even
lower levels of social adjustment were found only when there was initially
low perpetration and high victimization (increase perpetration class). No
differences were found with the increase both class, due to their initial
low involvement in cyberbullying.
Consistent with previous research (Espino et al., 2023), the trajec-
tories found in the present study have evidenced that social adjustment
was differently associated with cyberbullying perpetration and victimiza-
tion. The increase both and increase perpetration classes have in common
an increase in perpetration over time. However, their trajectories on
social adjustment were signicant, but in opposite directions. This is
explained by their different relationship with perpetration and victimi-
zation. As perpetration increases and victimization decreases, there is an
increase in social adjustment. However, there is a decrease in social
adjustment in both increase class. These results further support that the
experience of online victimization severely affects peer relationships
(Hypothesis 2.1). However, even with disrupted offending behaviors
(and decreased victimization) such adolescents can enhance and
develop support and positive interactions within their peer group. Lower
social adjustment was stable by decrease both class. While the initial
involvement of this group in cyberbullying accounts for lower social
adjustment, such adjustment remained stable throughout the remaining
times despite both victimization and perpetration decreasing over time
(Hypothesis 2.2).
Understanding joint trajectories and diverse proles helps to identify
how adolescent cyberbullying activity is growing. This evidence is
critical to design more effective preventive and mitigating effects, as it
helps to determine which groups of adolescents are more likely to
perpetuate and/or become victims of cyberbullying in a consistent way
and what their associated outcomes are. The ndings highlight that
involvement in cyberbullying can be uctuating; there are common and
distinctive aspects in the evolution of adolescents’ involvement in
perpetration and victimization and how this is connected to their
motivation and psychosocial adjustment. The prole of increase
perpetration reects the social characteristics linked to perpetration and
victimization, because their initial victimization was linked to a greater
deterioration of social adjustment and popularity, which were
increasing at the same time as victimization decreased and perpetration
increased. It implies that at these ages students may learn that cyber-
aggression is an effective strategy to get out of victimization and gain
social position. Specically, the high social status linked to these
harmful actions fuels their greater involvement as cyberperpetrators
(Kowalski et al., 2014).
Coping socialization strategies also play an important role in the
prevention of cyberbullying perpetration, since as shown in the longi-
tudinal proles, these adolescents improve their social adjustment as
they engage in perpetration (at the same time as need for popularity and
perceived popularity). Through coping socialization (Bradbury et al.,
2018), adolescents learn to cope with stressful online events through the
communication they establish with their immediate context by sharing
these experiences. Given the predominant focus on the social group
during adolescence, peers are the most inuential context for the
development of coping strategies, it is necessary to empower those ad-
olescents who suffer most forms of cyberbullying through social support
and integration into the peer group.
4.1. Limitations
The ndings of the present study must be considered in the context of
some limitations. All measures used were self-reports with the risk of
social desirability associated. Future studies could consider collecting
information from other actors. For example, both peer popularity and
social adjustment could be peer reported. Also, some qualitative research,
for example interviews or focus groups with pupils at the end of the
study, might help interpret certain ndings (Smith, 2019).
Although the present study applies a longitudinal design, its ndings
could not be considered in light of evolutionary developmental patterns
due to the heterogeneity in the age of the participants (from seventh to
ninth grade at T1). Future studies could explore whether the develop-
ment of cyberbullying involvement is developmentally associated with
these social cues at homogeneous ages (e.g., by taking a particular grade
and tracking its evolution over time across several grades). Even more
rigorous studies through the establishment of cohorts could examine the
difference in evolution between different age groups. It would also be of
particular relevance to control for certain developmental stages that
could affect their involvement in cyberbullying, as well as their psy-
chosocial adjustment (e.g., transition from primary to middle school).
The decision to use 1.5 as the average score cut-off resulted in a largely
sufcient (for analysis) but relatively small percentage of the sample
(7%); nevertheless, when doing the joint trajectories of cyberbullying
some of the subgroups were underestimated and it was not possible to
interpret the evolution of the social characteristics and motivations of
this minority. These issues could be addressed in future studies that
work on larger representative samples so that in the selection of those
highly involved larger numbers may be obtained and greater heteroge-
neity in the joint trajectories between cyberbullying perpetration and
victimization could be estimated.
5. Conclusions
The present study has highlighted the relevance of analyzing
cyberbullying trajectories considering both factors of victimization and
perpetration. It has shown that, in adolescents with a high involvement
in at least one time point, their involvement may be common or dif-
ferential in both behaviors. While for some adolescents there is an
overlap between victimization and perpetration, for others their
involvement over time may be the inverse. Moreover, students’ indi-
vidual differences in cyberbullying involvement have been shown to
vary in psychosocial adjustment as a function of their involvement and
trajectories in perpetration and victimization, with the increase in psy-
chosocial adjustment and in cyberbullying perpetration being the most
signicant association.
Funding
This work was supported by the Spanish National Research Agency
(PID 2020-113911RBI00).
A. Camacho et al.
Computers in Human Behavior 148 (2023) 107924
9
CRediT authorship contribution statement
Antonio Camacho: Conceptualization, Formal analysis, Methodol-
ogy, Software, Visualization, Writing – original draft. Peter K. Smith:
Conceptualization, Supervision, Writing – original draft. Rosario
Ortega-Ruiz: Supervision, Validation, Writing – review & editing. Eva
M. Romera: Data curation, Funding acquisition, Investigation, Project
administration, Resources, Supervision, Validation, Writing – review &
editing.
Declaration of competing interest
The authors declare that they have no competing nancial and per-
sonal relationships with other people or organizations that have inu-
enced the work reported in this paper.
Data availability
Data will be made available on request.
Acknowledgment
A. C. acknowledges the Margarita Salas Universidad de C´
ordoba
postdoctoral grants funded by the Spanish Ministry of Universities with
European Union funds-NextGenerationEU. A.C. also thanks the Telethon
Kids Institute, University of Western Australia for hosting as Honorary
Team Member.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.chb.2023.107924.
References
Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory.
Prentice-Hall.
Bollen, K. A. (1989). Structural equations with latent variables. Wiley. https://doi.org/
10.1002/9781118619179
Bradbury, S. L., Dubow, E. F., & Domoff, S. E. (2018). How do adolescents learn cyber-
victimization coping skills? An examination of parent and peer coping socialization.
Journal of Youth and Adolescence, 47, 1866–1879. https://doi.org/10.1007/s10964-
018-0812-y
Breslend, N. L., Shoulberg, E. K., McQuade, J. D., & Murray-Close, D. (2018). Social costs
for wannabes: Moderating effects of popularity and gender on the links between
popularity goals and negative peer experiences. Journal of Youth and Adolescence, 47
(9), 1894–1906. https://doi.org/10.1007/s10964-018-0810-0
Camacho, A., Runions, K., Ortega-Ruiz, R., & Romera, E. M. (2023). Bullying and
cyberbullying perpetration and victimization: Prospective within-person
associations. Journal of Youth and Adolescence, 52(2), 406–418. https://doi.org/
10.1007/s10964-022-01704-3
Charalampous, K., Georgiou, S., Demetriou, C., Tricha, L., Nikiforou, M., &
Stavrinides, P. (2021). Comparing short-term growth in traditional and cyber forms
of bullying in early and mid-adolescent students. European Journal of Developmental
Psychology, 18(3), 412–428. https://doi.org/10.1080/17405629.2020.1798754
Cho, S., & Glassner, S. (2021). Impacts of low self-control and opportunity structure on
cyberbullying developmental trajectories: Using a latent class growth analysis. Crime
& Delinquency, 67(4), 601–628. https://doi.org/10.1177/0011128720950018
Cho, S., & Lee, J. R. (2020). Joint growth trajectories of bullying perpetration and
victimization among Korean adolescents: Estimating a second-order growth mixture
model–factor-of-curves with low self-control and opportunity correlates. Crime &
Delinquency, 66(9), 1296–1337. https://doi.org/10.1177/0011128719890271
Cho, S., & Rustu, D. (2020). Examining the impacts of low self-control and online
lifestyles on cyberbullying perpetration among Korean adolescents: Using parallel
process latent growth curve modeling. Children and Youth Services Review, 117,
Article 105288. https://doi.org/10.1016/j.childyouth.2020.105288
Cillessen, A. H. N., & Marks, P. E. L. (2011). Conceptualizing and measuring popularity.
In A. H. N. Cillessen, D. Schwartz, & L. Mayeux (Eds.), Popularity in the peer system
(pp. 25–56). Guilford. https://doi.org/10.4324/9781315861104-4.
Del Rey, R., Casas, J. A., Ortega-Ruiz, R., Schultze-Krumbholz, A., Scheithauer, H.,
Smith, P. K., Thompson, F., Barkoukis, V., Tsorbatzoudis, H., Brighi, A., Guarini, A.,
Pyzalski, J., & Plichta, P. (2015). Structural validation and cross-cultural robustness
of the European cyberbullying intervention project questionnaire. Computers in
Human Behavior, 50, 141–147. https://doi.org/10.1016/j.chb.2015.03.065
Del Rey, R., Ojeda, M., Casas, J. A., Mora-Merch´
an, J. A., & Elipe, P. (2019). Sexting
among adolescents: The emotional impact and inuence of the need for popularity.
Frontiers in Psychology, 10, 1–11. https://doi.org/10.3389/fpsyg.2019.01828
Ding, Y., Li, D., Li, X., Xiao, J., Zhang, H., & Wang, Y. (2020). Proles of adolescent
traditional and cyber bullying and victimization: The role of demographic,
individual, family, school, and peer factors. Computers in Human Behavior, 111,
Article 106439. https://doi.org/10.1016/j.chb.2020.106439
Dodge, K. A. (2014). A social information processing model of social competence in
children. In M. Perlmutter (Ed.), Cognitive perspectives on children’s social and
behavioral development (pp. 85–134). Psychology Press.
Espino, E., Guarini, A., & Del Rey, R. (2023). Effective coping with cyberbullying in boys
and girls: The mediating role of self-awareness, responsible decision-making, and
social support. Current Psychology. https://doi.org/10.1007/s12144-022-04213-5
Festl, R., & Quandt, T. (2013). Social relations and cyberbullying: The inuence of
individual and structural attributes on victimization and perpetration via the
Internet. Human Communication Research, 39(1), 101–126. https://doi.org/10.1111/
j.1468-2958.2012.01442.x
Goagoses, N., Schipper, N., & Koglin, U. (2022). Callous-unemotional traits, social goal
orientations, and bullying perpetration: Exploring concurrent associations during
adolescence. International Journal of Adolescence and Youth, 27(1), 135–148. https://
doi.org/10.1080/02673843.2022.2037441
G´
omez-Ortiz, O., Romera, E. M., & Ortega-Ruiz, R. (2017). Multidimensionality of social
competence: Measurement of the construct and its relationship with bullying roles.
Revista de Psicodid´
actica, 22(1), 37–44. https://doi.org/10.1387/
RevPsicodidact.15702
Hogg, M. A. (2016). Social identity theory. In S. McKeown, R. Haji, & N. Ferguson (Eds.),
Understanding peace and conict through social identity theory (pp. 3–17). Springer.
https://doi.org/10.1007/978-3-319-29869-6_1.
Katsaras, G. N., Vouloumanou, E. K., Kourlaba, G., Kyritsi, E., Evagelou, E., & Bakoula, C.
(2018). Bullying and suicidality in children and adolescents without predisposing
factors: A systematic review and meta-analysis. Adolescent Research Review, 3,
193–217. https://doi.org/10.1007/s40894-018-0081-8
Kim, M. (2020). Instagram sele-posting and young women’s body dissatisfaction:
Investigating the role of self-esteem and need for popularity. Cyberpsychology:
Journal of Psychosocial Research on Cyberspace, 14(4). https://doi.org/10.5817/
CP2020-4-4
Kim, J., Song, H., & Jennings, W. G. (2017). A distinct form of deviance or a variation of
bullying? Examining the developmental pathways and motives of cyberbullying
compared with traditional bullying in South Korea. Crime & Delinquency, 63(12),
1600–1625. https://doi.org/10.1177/0011128716675358
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. https://doi.org/10.1037/
a0035618
Lozano-Blasco, R., Cort´
es-Pascual, A., & Latorre-Martínez, M. P. (2020). Being a
cybervictim and a cyberbully – the duality of cyberbullying: A meta-analysis.
Computers in Human Behavior, 111, Article 106444. https://doi.org/10.1016/j.
chb.2020.106444
Malamut, S. T., van den Berg, Y. H. M., Lansu, T. A. M., & Cillessen, A. H. N. (2021).
Bidirectional associations between popularity, popularity goal, and aggression,
alcohol use and prosocial behaviors in adolescence: A 3-year prospective
longitudinal study. Journal of Youth and Adolescence, 50, 298–313. https://doi.org/
10.1007/s10964-020-01308-9
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. https://doi.org/10.1016/j.childyouth.2011.08.032
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. https://doi.org/
10.1016/j.jadohealth.2014.06.007
Muth´
en, L. K., & Muth´
en, B. O. (1998). Mplus user’s guide (8th ed.). Muth´
en & Muth´
en.
Nylund, K. L., Asparouhov, T., & Muth´
en, B. O. (2007). Deciding on the number of classes
in latent class analysis and growth mixture modeling: A Monte Carlo simulation
study. Structural Equation Modeling, 14(4), 535–569. https://doi.org/10.1080/
10705510701575396
Pouwels, J. L., Lansu, T. A. M., & Cillessen, A. H. N. (2018). A developmental perspective
on popularity and the group process of bullying. Aggression and Violent Behavior, 43,
64–70. https://doi.org/10.1016/j.avb.2018.10.003
Pozzoli, T., & Gini, G. (2021). Longitudinal relations between students’ social status and
their roles in bullying: The mediating role of self-perceived social status. Journal of
School Violence, 20(1), 76–88. https://doi.org/10.1080/15388220.2020.1850462
Ranney, J. D., & Troop-Gordon, W. (2020). The role of popularity and digital self-
monitoring in adolescents’ cyberbehaviors and cybervictimization. Computers in
Human Behavior, 102, 293–302. https://doi.org/10.1016/j.chb.2019.08.023
Romera, E. M., Bravo, A., Ortega-Ruiz, R., & Veenstra, R. (2019). Differences in perceived
popularity and social preference between bullying roles and class norms. PLoS One,
14(10), Article e0223499. https://doi.org/10.1371/journal.pone.0223499
Romera, E. M., Cano, J. J., García-Fern´
andez, M., & Ortega-Ruiz, R. (2016).
Cyberbullying: Social competence, motivation and peer relationships. Comunicar, 48,
71–79. https://doi.org/10.3916/C48-2016-07
Romera, E. M., Herrera-L´
opez, M., Casas, J. A., Ortega-Ruiz, R., & G´
omez-Ortiz, O.
(2017). Multidimensional social competence, motivation, and cyberbullying: A
cultural approach with Colombian and Spanish adolescents. Journal of Cross-Cultural
Psychology, 48(8), 1183–1197. https://doi.org/10.1177/0022022116687854
Romera, E. M., Luque, R., Ortega-Ruiz, R., G´
omez-Ortiz, O., & Camacho, A. (2022).
Positive peer perception, social anxiety and classroom social adjustment as risk
A. Camacho et al.
Computers in Human Behavior 148 (2023) 107924
10
factors in peer victimization: A multilevel study. Psicothema, 34(1), 110–116.
https://doi.org/10.7334/psicothema2021.37
Romera, E. M., Ortega-Ruiz, R., Runions, K., & Camacho, A. (2021). Bullying
perpetration, moral disengagement and need for popularity: Examining reciprocal
associations in adolescence. Journal of Youth and Adolescence, 50, 2021–2035.
https://doi.org/10.1007/s10964-021-01482-4
Runions, K. C., & Bak, M. (2015). Online moral disengagement, cyberbullying, and cyber-
aggression. Cyberpsychology, Behavior, and Social Networking, 18(7), 400–405.
https://doi.org/10.1089/cyber.2014.0670
Samson, J. E., Delgado, M. A., Louis, D. F., & Ojanen, T. (2022). Bullying and social goal-
setting in youth : A meta-analysis. Social Development, 31(4), 945–961. https://doi.
org/10.1111/sode.12595
Santor, D. A., Messervey, D., & Kusumakar, V. (2000). Measuring peer pressure,
popularity, and conformity in adolescent boys and girls: Predicting school
performance, sexual attitudes, and substance abuse. Journal of Youth and
Adolescence, 29(2), 163–182. https://doi.org/10.1023/A:1005152515264
Slonje, R., Smith, P. K., & Fris´
en, A. (2013). The nature of cyberbullying, and strategies
for prevention. Computers in Human Behavior, 29(1), 26–32. https://doi.org/
10.1016/j.chb.2012.05.024
Smith, P. K. (2019). Research on cyberbullying: Strengths and limitations. In
H. Vandebosch, & L. Green (Eds.), Narrative approaches in research and interventions
addressing cyberbullying (pp. 9–27). Springer. https://doi.org/10.1007/978-3-030-
04960-7_2.
Smith, P. K., L´
opez-Castro, L., Robinson, S., & G¨
orzig, A. (2019). Consistency of gender
differences in bullying in different cross-cultural surveys. Aggression and Violent
Behavior, 45, 33–40. https://doi.org/10.1016/j.avb.2018.04.006
Smith, P. K., Mahdavi, J., Carvalho, M., Fisher, S., Russell, S., & Tippett, N. (2008).
Cyberbullying: Its nature and impact in secondary school pupils. The Journal of Child
Psychology and Psychiatry and Allied Disciplines, 49(4), 376–385. https://doi.org/
10.1111/j.1469-7610.2007.01846.x
Song, H., Lee, Y., & Kim, J. (2020). Gender differences in the link between cyberbullying
and parental supervision trajectories. Crime & Delinquency, 66, 1914–1936. https://
doi.org/10.1177/0011128720912371
Strindberg, J., Horton, P., & Thornberg, R. (2020). Coolness and social vulnerability:
Swedish pupils’ reections on participant roles in school bullying. Research Papers in
Education, 35(5), 603–622. https://doi.org/10.1080/02671522.2019.1615114
Sumter, S. R., Baumgartner, S. E., Valkenburg, P. M., & Peter, J. (2012). Developmental
trajectories of peer victimization : Off-line and online experiences during
adolescence. Journal of Adolescent Health, 50(6), 607–613. https://doi.org/10.1016/
j.jadohealth.2011.10.251
Swirsky, J. M., Rosie, M., & Xie, H. (2022). Correlates of early adolescents’ social media
engagement: The role of pubertal status and social goals. Journal of Youth and
Adolescence, 51(1), 74–85. https://doi.org/10.1007/s10964-021-01494-0
Tian, L., Huang, J., & Huebner, E. S. (2023). Proles and transitions of cyberbullying
perpetration and victimization from childhood to early adolescence: Multi-
contextual risk and protective factors. Journal of Youth and Adolescence, 52(2),
434–448. https://doi.org/10.1007/s10964-022-01633-1
Vandebosch, H., & Van Cleemput, K. (2009). Cyberbullying among youngsters: Proles
of bullies and victims. New Media & Society, 11(8), 1349–1371. https://doi.org/
10.1177/1461444809341263
Vanden Abeele, M., Campbell, S. W., Eggermont, S., & Roe, K. (2014). Sexting, mobile
porn use, and peer group dynamics: Boys’ and girls’ self-perceived popularity, need
for popularity, and perceived peer pressure. Media Psychology, 17(1), 6–33. https://
doi.org/10.1080/15213269.2013.801725
Vanden Abeele, M., Van Cleemput, K., & Vandebosch, H. (2017). Peer inuence as a
predictor of producing and distributing hurtful images of peers and teachers among
Flemish adolescents. Journal of Children and Media, 11(1), 69–87. https://doi.org/
10.1080/17482798.2016.1233123
de Vries, E., Kaufman, T. M. L., Veenstra, R., Laninga-Wijnen, L., & Huitsing, G. (2021).
Bullying and victimization trajectories in the rst years of secondary education:
Implications for status and affection. Journal of Youth and Adolescence, 50(10),
1995–2006. https://doi.org/10.1007/s10964-020-01385-w
Walters, G. D. (2021). School-age bullying victimization and perpetration: A meta-
analysis of prospective studies and research. Trauma, Violence, & Abuse, 22(5),
1129–1139. https://doi.org/10.1177/1524838020906513
Wang, L., & Ngai, S. S. (2022). Cyberbullying perpetration among Chinese adolescents:
The role of power imbalance, fun-seeking tendency, and attitude toward
cyberbullying. Journal of Interpersonal Violence, 37(23–24), NP21646–NP21671.
https://doi.org/10.1177/08862605211062988
Wegge, D., Vandebosch, H., Eggermont, S., & Pabian, S. (2016). Popularity through
online harm: The longitudinal associations between cyberbullying and sociometric
status in early adolescence. The Journal of Early Adolescence, 36(1), 86–107. https://
doi.org/10.1177/0272431614556351
Wright, M. F. (2014). Longitudinal investigation of the associations between adolescents’
popularity and cyber social behaviors. Journal of School Violence, 13(3), 291–314.
https://doi.org/10.1080/15388220.2013.849201
Wright, M. F., Wachs, S., Huang, Z., Kamble, S. V., Soudi, S., Bayraktar, F., Li, Z., Lei, L.,
& Shu, C. (2022). Longitudinal associations among machiavellianism, popularity
goals, and adolescents’ cyberbullying involvement: The role of gender. The Journal of
Genetic Psychology, 183(5), 482–493. https://doi.org/10.1080/
00221325.2022.2095251
Yoo, C. (2021). What are the characteristics of cyberbullying victims and perpetrators
among South Korean students and how do their experiences change? Child Abuse &
Neglect, 113, Article 104923. https://doi.org/10.1016/j.chiabu.2020.104923
Zhou, Y., Zheng, H., Liang, Y., Wang, J., Han, R., & Liu, Z. (2022). Joint developmental
trajectories of bullying and victimization from childhood to adolescence: A parallel-
process latent class growth analysis. Journal of Interpersonal Violence, 37(3–4),
NP1759–NP1783. https://doi.org/10.1177/0886260520933054
Zych, I., Beltr´
an-Catal´
an, M., Ortega-Ruiz, R., & Llorent, V. J. (2018). Social and
emotional competencies in adolescents involved in different bullying and
cyberbullying roles. Revista de Psicodid´
actica, 23(2), 86–93. https://doi.org/
10.1016/j.psicoe.2017.12.001
Zych, I., & Farrington, D. P. (2021). Stability and change in bullying and cyberbullying
throughout the school years. In P. K. Smith, & J. O’Higgins Norman (Eds.), The wiley
blackwell handbook of bullying: A comprehensive and international review of research and
intervention (Vol. 2, pp. 20–36). Wiley Blackwell. https://doi.org/10.1002/
9781118482650.ch36.
Zych, I., Ortega-Ruiz, R., & Marín-L´
opez, I. (2016). Cyberbullying: A systematic review of
research, its prevalence and assessment issues in Spanish studies. Psicologia
Educativa, 22(1), 5–18. https://doi.org/10.1016/j.pse.2016.03.002
A. Camacho et al.