Content uploaded by Maria José D. Martins
Author content
All content in this area was uploaded by Maria José D. Martins on Aug 09, 2016
Content may be subject to copyright.
DOI: 10.4018/IJCBPL.2016070105
Copyright © 2016, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
Volume 6 • Issue 3 • July-September 2016
Maria José D. Martins, Polytechnic Institute of Portalegre, Portalegre, Portugal
Ana Margarida Veiga Simão, Faculty of Psychology, University of Lisbon, Lisbon, Portugal
Isabel Freire, Institute of Education, University of Lisbon, Lisbon, Portugal
Ana Paula Caetano, Institute of Education, University of Lisbon, Lisbon, Portugal
Armanda Matos, Faculty of Psychology and Science Education, University of Coimbra, Coimbra, Portugal
This study aims to clarify how the quality of the family environment is related to the involvement
in cyberbullying behaviors, either as a cyber-victim or as a cyber-aggressor, via a cross-sectional
research design. With this purpose a diagnostic questionnaire with questions about both the quality of
family environment and cyberbullying was conceived and administered to 3525 adolescents attending
6th, 8th and 11th grades at several schools in Portugal. The results suggested that two family aspects
seem to be equally important in protection against cyberbullying: perception of family support and
perception of rules within the family. A hierarchical regression analysis reveals that lack of family
support is more predictive of cyber-victimization and that a lack of family rules is more predictive
of cyber-aggression. The authors discuss the implications for the well-being of adolescents, as well
as the challenges that parents face in the supervision of adolescents’ use of digital technologies.
Cyberbullying, Family Environment, Family Supervision
The use of digital technologies (DT) has been recently increasing and proliferating, not only in
academic and work environments, but also in youth culture and leisure. These technologies bring
multiple benefits and opportunities, but may also entail risks such as cyberbullying, which has been
generally defined as repeated aggressive and intentional actions with the use of electronic devices (e.g.,
cell phones and computers) and associated programs (e.g., e-mail, the Internet, and social networks),
by means of sending messages and/or creating websites that insult, denigrate, threaten, or harass others
in some way (Amado, Freire, Matos, Vieira & Pessoa, 2012; Amado, Matos & Pessoa, 2009; Li, 2007;
Kowalski, Limber & Agaston, 2008; Smith, 2009; Willard, 2005). Many studies have suggested that
cyberbullying consists of an indirect form of bullying, and frequently represents continuations of face-
to-face bullying situations (Kowalski, Giummetti, Schroeder & Lattanner, 2014; Ortega, Calmaestra
& Mora-Merchan, 2008; Ortega, Elipe, Mora-Merchan, Calmaestra & Vega, 2009). Cyberbullying
can be predicted by previous attitudes similarly to what happens with face-to-face bullying (Boulton,
65
Volume 6 • Issue 3 • July-September 2016
66
Lloyd, Down & Marx, 2012). In contrast with other types of bullying, cyberbullying does not tend to
decrease with age or grade level; it may actually increase over time (Kowalski et al., 2014; Walker,
Sockman & Koehn, 2011) and can also be found in college and university students (Francisco, Veiga
Simão, Ferreira & Martins, 2015).
The EUKIDS Online project, a European study conducted by Livingstone, Haddon, Gorzig, and
Olafsson (2011) involving 25.000 children and teenagers, aged from 9 to 16, revealed that Internet
usage is part of children’s daily life in many European countries, as 93% of the respondents claimed
they use the Internet on a weekly basis and one-third of the respondents aged 9 to 10 affirmed that
they use the Internet on a daily basis. This study also revealed that children are increasingly accessing
the Internet at an earlier age, since older children (15-16 years old) reported they started using the
Internet from the age of 11, while younger children (9-11 years old) reported they began to use it
from the age of 9 (in a cross-sectional study). Moreover, this research demonstrated how children
access the Internet mostly at home (around 87%) and school (67%). Specifically, children usually
access the Internet on their desktop computer in their bedrooms (49%) or on a mobile device (33%).
Lastly, Livingstone and colleagues (2011) identified that about one-third of the respondents claimed
to know more about the Internet than their own parents; this finding is very relevant to this study.
Other works have reported similar tendencies regarding the use of digital technologies (Hertlein,
2012) and suggested that parents with less expertise and knowledge on DT have more difficulties
in monitoring teenagers’ activities (e.g., Fletcher & Blair, 2014; Sorbring, 2014). Sorbring (2014)
concluded that the parents who gave more importance to the Internet and who simultaneously had
less relevant knowledge, were those who worried mostly about their teenagers’ use of the Internet.
Kowalski and collaborators (2014) conducted a meta-analysis on cyberbullying research in
youths and concluded that the theoretical approaches that explain aggression could be used to explain
bullying and cyberbullying. Therefore, these authors sustain that both personal and situational factors
influence the occurrence of cyberbullying and/or cyber-victimization. As a result, they concluded
that the strongest associations with cyberbullying perpetration were normative beliefs about
aggression and moral disengagement, and that the strongest associations with cyber-victimization
were stress and suicidal ideation. In reference to situational factors, results suggested that parental
involvement, parental monitoring, and school characteristics influence both cyber-aggression and
cyber-victimization. School variables that were inversely linked to engagement in cyberbullying
included school climate and school safety. Regarding family variables, it seems that less frequent
cyber-victimization is associated with parental discussion about online behavior and that cyberbullies
had weaker emotional bonds with their parents and less frequent monitoring of online activities
(Kowalski et al., 2014).
In this study, the authors aim to understand how the family environment relates to the involvement
of pre-adolescents and adolescents in incidents of cyberbullying, either as victims or bullies, because
this topic has been less investigated and less deepened than socio-demographic and school variables
associated with cyberbullying (Kowalski et al., 2014).
Different authors (e.g., Diaz-Aguado, 2004, 2005; Granot & Mayseless, 2001) suggested various
family aspects, which are usually associated with healthy psychosocial development and adjustment,
namely: the presence of warm and affective relationships, without overprotection, or, in other words,
strong and safe bonds between parents and their children; the presence of age-appropriate supervision
and care, that means adequate balance between parent support and autonomy stimulation; and
disciplinary practices in which rules are explicit and authority coexists with negotiation, or, in other
words, a consistent parenting education or an assertive discipline parenting that is neither authoritarian
nor permissive or inconsistent.
On the other hand, several studies (e.g., Blanc & Janoz, 2002; Fonseca, 2002; Ijzendoorn,
2002; Kõiv, 2012; Smith, 2005) revealed that negative parenting styles and/or parent-child insecure
attachment have consequences on children’s behavior, in the sense that they seem to predict aggression
and other social adjustment problems. Hence, several family characteristics have been generally
Volume 6 • Issue 3 • July-September 2016
67
associated with aggressive behaviors or victim conditions in children and adolescents. For instance,
coercive parenting discipline resorting to physical punishment, inconsistent parenting discipline that
oscillates between extreme coercion and allowance (mainly due to adult exhaustion), or an absence
of discipline may constitute a risk factor for aggressor or victim-aggressor conducts at school (Diaz-
Aguado, 2004, 2005; Smith, 2005). Furthermore, the nature of attachment between parents and
children, especially insecure attachment, disorganized attachment, or avoidant attachment, may also
be considered risk factors for behavioral problems, namely for aggressor or victim-aggressor conducts
at school (Granot & Mayseless, 2001; Ijzendoorn, 2002). In particular, Kõiv (2012), with a sample
of 1.921 Estonian pupils from 10 to 18 years, found that bullies were more likely to have an avoidant
attachment, whereas victims exhibited more insecure attachment than bullies. An overprotective family
may also constitute a risk factor for victimization at school (Olweus, 1995). Lastly, abuse among
family members can also contribute as a risk factor, since child abuse may generate victim-aggressor
behaviors at school, whereas exposure to conjugal violence may generate aggressive-bullying behaviors
(Corvo & de Lara, 2010; Schwartz, Dodge, Petti & Bates, 1997; Smith, 2005).
Specifically concerning relationships between cyberbullying and family relations, recent studies
revealed that involvement in cyberbullying is influenced by both peers’ and adults’ behavior (e.g.,
Fanti, Demetriou & Hawa, 2012; Hinduja & Patchin, 2013). Fanti, Demetriou and Hawa (2012)
posited that in Cyprus peers aged 11 to 14 influenced cyberbullying involvement, but parent support
significantly decreased the probability of involvement in such conducts. The authors concluded that
their results highlight the role of parents in protection against cyberbullying, because adolescents that
reported more family support also had fewer incidents of cyber-aggression and cyber-victimization
a year later (Fanti et al., 2012).
Hinduja and Patchin (2013) also studied the influence of peers and family in cyberbullying
involvement in North American teenagers. The authors revealed that peers influenced cyberbullying,
but the expectation of punishment for such conducts from adults reduced the likelihood of involvement.
According to Hinduja and Patchin (2013), cyberbullying offences are associated with the perceptions
teenagers have of their peers’ similar behavior, as well as the probability of adults’ reprimands. What
is more, respondents who were less likely to participate in cyberbullying believed that adults in their
life would punish them.
Mesch (2009) investigated the effect of exposure of American teenagers to online risks of
cyberbullying and parental mediation of this phenomenon. The author defined parental mediation
as parents’ actions to protect their children from exposure to Internet risks and he conceptualized
two types of parental mediation, namely, restrictive mediation and evaluative mediation. The first
does not include child participation and involves parents’ decisions about the amount of time, type
of programs, and location of Internet access. Evaluative mediation, on the contrary, includes child
participation through open discussion of issues related to Internet use. To be exact, parents decide
together with their children on what the rules are regarding the amount of time the Internet is used
and what web sites are appropriate or not. Mesch (2009) concluded that most parental mediation
measures, particularly the restrictive or even some evaluative measures, such as keeping computers on
shared divisions, are not effective measures regarding online activities. The only mediation measures
that seem to protect teenagers against cyberbullying are the evaluative ones that establish rules for
online navigation, and the permanent dialogue between parents and children regarding the nature
and contents of websites. The nature of online activities seems to impose difficulties in parental
mediation strategies that are usually effective when applied to other media (e.g., television), especially
concerning teenagers (Matos, 2006).
As such, the objectives of this work are twofold. Firstly, this research aims to study the relationship
between the quality of the family environment and the involvement in cyberbullying. Specifically,
this study search to understand what type of relationship exists between the quality of the family
environment, as perceived by teenagers, and the experience of victimization and/or aggression practices
through the use of digital technologies. Secondly, it proposes to identify which family aspects are more
Volume 6 • Issue 3 • July-September 2016
68
related to involvement in acts of cyberbullying. More specifically, this research essays to understand
whether the determination of rules for the use of digital technologies and a consistent parenting
discipline are more or less important than parental support as protection factors for cyberbullying
involvement (as cyber-aggressor or cyber-victims), as perceived by teenagers.
The sample consisted of 3.525 pupils from twenty-three middle and high schools located in nine of
the eighteen districts in Portugal, selected using convenience sampling. Of the 3.525 pupils in the
sample, 1.683 (47.7%) were male and 1.837 (52.1%) were female, distributed over three school grades:
6th, n = 1.125 (Mage = 11.2 years, SD = .97; 52.8% male and 47.2% female), 8th, n = 1.217 (Mage
= 13.2 years, SD = .82; 48.4% male and 51.6% female) and 11th, n = 1.172 (Mage = 16.4 years, SD
= 1; 42.3% male and 57.7% female). The whole sample ranged in age from 10 to 23 years (Mage =
13.6 years, SD = 2.3) and the median and mode was 13 years.
The Diagnostic Questionnaire of Cyberbullying was created for this study. The questionnaire
initiates with the following explanation about cyberbullying: “Children and young people use new
digital technologies on a daily basis and this offers them great advantages. However, there are also
situations in which certain pupils treat others badly via the Internet or mobile phone. For example,
some classmates mistreat others by sending offensive, insulting, harassing, or threatening messages
(SMS, MMS, photos, videos, etc.) or other kinds of unpleasant or private messages or images. We
are studying this phenomenon, which is called cyberbullying (…)”. This explanation was necessary
because there is no Portuguese expression for this phenomenon, so it is currently designated with
the English word and has recently entered in the Portuguese vocabulary.
In this research, we analyzed the questions related to involvement in cyberbullying, as well as
those related to students’ perceptions regarding the family environment (12 questions presented in a
subscale of the quality of the family environment as perceived by teenagers). These questions were
prepared for this study after a qualitative research (Amado, et. al., 2009; 2012) under a funded project
(PTDC/CPE-CED/108563/2008), by the Portuguese Foundation for Science and Technology (FCT),
and were subsequently related to fulfill the purposes of the study.
The two questions concerning cyberbullying which were analysed were: “Have you been a victim
of offence, harassment, threats, or defamation by anyone through cell phones or the Internet in the
past year?” and “Have you offended, harassed, threatened, or defamed anyone through cell phones
or the Internet in the past year?” They were answered on a scale from 1 to 6 (1= never, and 6= every
day), according to the frequency of occurrences. These two questions were considered as the most
adequate measures of cyberbullying, since in previous studies by different authors to whom Kowalski
and colleagues’ (2014) meta-analysis referred single items were used to access cyberbullying. Also,
we are not trying to measure specific cyberbullying behavior as a construct, but, in fact, understand
if students would admit to having been involved in cyberbullying in a general sense and with what
frequency; further literature has demonstrated that students are often reluctant to report this type
of experience (Francisco et al., 2015). Then we attempted to understand if family variables could
somehow influence whether or not students reported this involvement.
As to the family environment, 12 items were included, 8 of which were affirmative statements and
4 were negative. The items reflected not only the perceptions of students regarding family dialogue,
parental support, and information sharing for problem solving, but also their perceptions of rules
regarding the use of DT and family rules in general. Answers to each question required the selection
of a single alternative among four options (1= completely disagree, 2= disagree, 3= agree, and 4=
Volume 6 • Issue 3 • July-September 2016
69
completely agree). The scale was elaborated based on data from the previously cited studies (e.g.,
Amado et al., 2009; 2012), and afterwards related items that may reveal a good family environment
were suggested by various experts in psychology and educational sciences.
Reliability analysis for the family environment scale was checked by means of a Cronbach’s alpha
coefficient, which resulted in an alpha value of 0.81 for the entire scale, indicating good internal
consistency and acceptable reliability. One of the items (i.e., “When I have a problem, my parents/
guardians take care of it”) had a lower correlation with others and when it was deleted from the
whole scale the alpha did not change, so the item was theoretically ambiguous. Therefore, it can be
eliminated from the scale, in future use.
Following the approval of the study by the Portuguese Ministry of Education, 23 schools and school
groups in mainland Portugal were contacted and the respective school boards were asked to collaborate
in the project. Authorization was also sought from the students’ parents/guardians, to enable them to
take part in the study. Teachers administered the questionnaires in paper-pencil form during the first
months of the 2012/2013 academic year, with assurances of confidentiality and anonymity included
in the initial instructions, which were part of the instrument. Completion of the measures by each
pupil took approximately 30 to 40 minutes. The data was analyzed using SPSS (version 20.0, 2011).
Table 1 shows the frequency of cyberbullying conducts as experienced by the participants, either as
cyber-victims or cyberbullies. Results showed that 7.6% (267) of the participants admitted to be victims
of offense, denigration, threats, and harassment through cell phones or the Internet. Approximately
2.6% (90) of these participants admitted to be cyber-victims at least once a month in the previous
year. The percentage of aggressors (adolescents who admitted to offend, denigrate, threat, or harass
someone through cell phones or the Internet) was lower (3.9%, 138 participants), with 1.5% (51)
admitting to be involved in cyberbullying as aggressors at least once a month in the previous year.
Considering the importance of the family in the psychosocial adjustment and education of children and
the fact that parents are, in general, less experienced with DT than their children (Livingstone et al.,
2011; Ponte, Jorge, Simões & Cardoso, 2012), it is pertinent to understand which relationships exist
Table 1. Frequency of cyberbullying conducts, as cyber-victims or cyber-aggressors
Frequency of Cyberbullying Conducts Cyber-Victims Cyber-Aggressors
Never happened to me 3242 (92.1%) 3385 (96%)
1-4 times a year 177 (5%) 87 (2.5%)
Once a month 28 (0.8%) 12 (0.3%)
Once a week 17 (0.5%) 13 (0.4%)
Several times in a week 35 (1%) 16 (0.5%)
Every day 10 (0.3%) 10 (0.3%)
No answer 12 (0.3%) 2 (0.1%)
Total 3.525 (100%) 3.525 (100%)
Volume 6 • Issue 3 • July-September 2016
70
between cyberbullying conducts (victims or aggressors) and the quality of the family environment. Over
75% of the participants agreed with the positive family environment statements, with the exception
of one (“My parents/guardians establish rules for the use of technologies”), which presented a higher
percentage of disagreement (36.3%). Similarly, over 75% of the students disagreed with the negative
family environment statements (lack of support and dialogue and individualistic behaviors).
An exploratory factor analysis (Table 2) was performed, using an extraction method of principal
component analysis and a promax rotation method with Kaiser normalization, and a two-factor
solution was selected, given the importance of the family aspects as referred to in the previously
cited studies. The option of eliminating the last item of the positive scale has to do with the fact
that it is not adequate to evaluate the quality of the family environment in adolescence. Its positive/
negative nature reveals ambiguity, not emphasizing the development of autonomy and being subject
to inclusion in the support items. Furthermore, it showed no statistical usefulness for analyzing the
results, since the results of the reliability analysis were not affected by its elimination. What is more,
there was no coherent factor solution with the referred item included in the scale.
The two-factor solution for the eleven items of the scale explained 48.3% of the total variance.
According to Hair, Anderson, Tatham, and Black (1998), this result can be considered acceptable
given the nature of social variables. Table 2 presents the factor loadings of the items. An analysis of the
items based on factor loadings, as well as theoretical issues, allowed for grouping the items as follows:
Factor 1: Perception of parental support (includes eight items, with a Cronbach’s alpha of 0.81) that
therefore will be designated as family support.
Factor 2: Perception of consistent parenting discipline and the existence of family rules, either general
or related to the use of digital technologies (includes three items, with a Cronbach’s alpha of
0.60) that will be designated as family rules.
The whole sample was split in four groups: non-involved, that is those who have reported never
being involved in cyberbullying; those who have reported being only cyber-victims; or only cyber-
Table 2. Exploratory factor analysis (EFA) for the scale of the quality of family environment as perceived by middle and high
school students
Items Factor 1 Factor 2
We support each other .73 39
We care about each other. .72 .31
We express affection for one another. .71 .37
I can count on my family when I need help. .71 .36
There is open dialogue among us. .69 .42
We rarely talk. .66 .08
It is uncommon for us to share our problems. .60 .16
Each one takes care of their own problems. .52 .24
My parents/guardians establish rules for the use of Digital Technologies. .22 .85
My parents/guardians seem to care about what I do with technologies. .39 .81
I feel I can do whatever I want. .29 .46
Eigenvalues 4.10 1.30
% of variance 36.9 11.4
Note: Factor loadings over .40 appear in bold.
Volume 6 • Issue 3 • July-September 2016
71
aggressors; or both cyber-victims and cyber-aggressors (this last group were 17.46% of the total of
those involved in some type of cyberbullying). Table 3 shows frequencies of participants, means,
and standard deviations from those who belonged to one of the four groups mentioned. All mean
values presented in Table 3 are above the theoretical mean (30) of the whole scale. Due to the great
difference in number of individuals pertaining to each group, the sample effect size was calculated
using the Eta square coefficient, revealing a medium value (hp
2 = .48).
A one-way ANOVA was computed to compare these four groups in terms of their perceptions
of the quality of the family environment and verify whether the differences between the four groups
were statistically significant. The results of this analysis revealed that for the factor “family support”,
as well as for the factor “family rules”, and for the total scale, the differences between groups were
statistically significant, F(3, 3111) = 13.87, p < .001; F(3, 3240) = 25.37, p < .001; and F(3, 3355)
= 30.01, p < .001, respectively. The non-involved revealed better perceptions of the quality of the
family environment, followed by victims, then by aggressors and, finally, by victims-aggressors, who
reported the worst perceptions related to the quality of their family environment.
A Post Hoc Tuckey test was performed in order to verify any possible statistically significant
differences between groups when compared two by two. The results of this analysis revealed that in
the factor “family support” the differences between non-involved and the other three groups (only
victims, only aggressors, and victims-aggressors) were all statistically significant (with respectively
p = .005, p < .001 and p < .001). Nonetheless, the differences between these last three groups (when
compared two by two) were not statistically significant. This means that victims’, aggressors’, and
victims-aggressors’ perceptions of family support were significantly globally worse than those of
non-involved.
As to the factor “family rules”, the differences between non-involved and the other three groups
(only victims, only aggressors, and victims-aggressors) were all statistically significant (with
respectively p < .001, p < .001 and p < .001), and the differences between only victims and the other
two groups (only aggressor and victims-aggressors) were also statistically significant (p < .001 and
p < .001, respectively). Therefore, aggressors’ and victims-aggressors’ perceptions of family rules
were statistically worse than the perceptions of those who were not involved and of those who were
only victims. Also, those who were only victims revealed statistically significant worse perceptions
than those who were not involved in any way, although not as bad as those of aggressors and victims-
aggressors.
Table 4 presents correlations between the frequency of victimization, aggression, age, and the perceived
quality of the family environment. There is a positive and statistically significant correlation between
the condition of victim and aggressor, which reveals that some teenagers are victims and aggressors
Table 3. Mean values and standard deviations in the two subscales of family environment (post-EFA) as a function of the type
of involvement in cyberbullying
Note: Missing values were 169. Only those who answered the whole of both scales were included in the analysis.
*p < .01
Volume 6 • Issue 3 • July-September 2016
72
simultaneously. A negative and statistically significant correlation between the quality of family
environment and involvement in cyberbullying as victims or aggressors could be observed, indicating
that cyber-victimization and cyber-aggression are inversely related to the perception of a good family
environment. Moreover, this inverse relationship is slightly stronger for cyber-aggression conducts,
especially regarding the perception of family rules.
In order to understand which variables would be predictive of involvement in cyberbullying and,
particularly, to identify which family factors would be more predictive of cyber-victimization and
cyber-aggression, a hierarchical regression analysis was performed with an enter method used for
each block. Gender and age were entered as control variables because several authors found that these
variables are usually related with these types of behavior (Fanti et al., 2012; Walker et al., 2011).
Concerning cyber-victimization as an outcome, both control and study variables were statistically
significant, F(2,3337) = 11.002, p < .01 at step 1 and F(4,3337) = 13.273, p < .01 at step 2 of the
hierarchical regression analysis (Table 5).
Table 5. Hierarchical Regression (HR) for variables predicting cyber-victimization as an outcome
Note: Gender was dummy coded, 0 for girls and 1 for boys.
* p < .01. ** p < .001.
Table 4. Descriptive statistics and Pearson correlations among the study variables
Variables 1 2 3 4 5
1. Cyber-victimization .233* .050* -.119* -.060*
2. Cyber-aggression - .104* -.087* -.116*
3. Age - -.069* -.318*
4. Family support - .426*
5. Family rules -
M 1.13 1.07 13.7 27.08 8.83
SD .57 .45 2.29 4.05 1.97
* p < .01
Volume 6 • Issue 3 • July-September 2016
73
Results from the hierarchical regression on Table 6 show that the independent study variables that
were considered explained 1.4% (AdjR2) of the variation in cyber-victimization. In Step 1 of the HR,
the predictive power of gender (t = -2.974, p < .01) and age (t = 3.465, p = .001) was significant for
predicting cyber-victimization. Hence, older and female students were significantly more associated
with cyber-victimization.
Adding family support and family rules in the second step of the analysis revealed that the
predictors of cyber-victimization were, in general, family support (t = - 4.914; p < .001), gender (t =
-3.435; p = .001), and age (t = 2.764; p < .01). These results suggest that the lack of family support
was significantly associated with the experience of being a cyber-victim, and that older and female
students were at a higher risk of being cyber-victims.
In relation to cyber-aggression as an outcome variable, both control and study variables were also
statistically significant, F(2,3344) = 26.1, p < .01 at step 1 and F(4,3344) = 20.299, p < .01 at step 2.
The hierarchical regression for predicting cyber-aggression was similarly performed in two steps
and revealed that the independent variables (study variables) that were considered explained 2.3%
(AdjR2) of the variation in cyber-aggression (see Table 6). In Step 1, the predictive power of gender
(t = 3.685, p < .001) and age (t = 6.401, p < .001) was significant for predicting cyber-aggression.
Hence, older and male students were significantly associated with cyber-aggression.
By adding family support and family rules in the second step of the analysis, it can observed that
the best predictor of cyber-aggression was age (t = 4.706, p < .001), followed by family rules (t =
-3.154, p < .01), gender (t = 3.025, p < .01), and family support (t = -2.597, p < .01). These results
suggest that older and male students were more likely to become bullies and that those adolescents
with less family rules and less family support were at a higher risk of perpetrating cyber-aggression.
The results of this study indicated that around 8% of adolescents had been victims of cyberbullying, 4%
had been involved in cyberbullying conducts as aggressors, and around one sixth of these adolescents
Table 6. Hierarchical Regression (HR) for variables predicting cyber-aggression as an outcome
Note: Gender was dummy coded, 0 for girls and 1 for boys.
* p < .01. ** p < .001.
Volume 6 • Issue 3 • July-September 2016
74
were both cyber-victims and cyber-aggressors (1.62% of the total sample). These results are similar to
the outcomes obtained by Livingstone et al. (2011), who conducted a research on teenagers of several
European countries and found that around 7% of children were involved in those problems, but inferior
to the results obtained by other studies, that estimate the range of these behaviors between 10% and
40% (Kowalski et al., 2014). Though the frequencies obtained in this research seemed low, these
results should evoke some concern, as the anonymity, invisibility, and rapid proliferation of Internet
messages may impose serious consequences to the individuals who are involved in cyberbullying–
namely depression, fear, low self-esteem, social isolation, and suicidal thoughts for the victims, and
desensitization or indifferent feelings on the aggressors’ behalf (Ortega et al., 2009; Ortega, et. al.,
2012). This data seems to be consistent with the idea that teaching curriculums for digital competences
must include safety and ethical standards regarding the use of DT.
Our results indicate that adolescents’ perceptions of a good family environment are related to
a lower probability of involvement in cyberbullying, either as cyber-aggressors or cyber-victims,
similarly to other studies (e.g., Fanti et al., 2012; Hinduja & Patchin, 2013), but the research that
has been presented in this work has distinctively revealed that both family support and family rules
are family dimensions that protect against involvement in cyberbullying. Thus, perceptions of
parenting rules, namely concerning the use of DT, and perceptions of parental support and affective
relationships within the family seem to significantly contribute to decrease the likelihood of teenage
involvement in cyberbullying. Moreover, results contribute to clarify which family aspects seem to
protect adolescents regarding the different roles experienced in cyberbullying. More specifically, this
study has highlighted that the explicitness of family rules, particularly regarding the use of DT, helps
to prevent involvement in cyberbullying, especially for cyber-aggressors, and that parental support
helps to prevent cyber-victimization. Therefore, results suggest that family support, but not family
rules, help students to protect and defend themselves from cyber-victimization, and also suggest that
consistent parenting rules combined with support are necessary to protect against cyber-aggression.
This research has some limitations, namely its cross-sectional nature and the single focus on the
role of the family. A future study design could include other aspects besides the role of family, such
as the role of school ethos and the influence of peers in their involvement in cyberbullying because of
their probabilistic multifactor causes and influences. The evaluation of the role of these other factors
could contribute to an increase in the variance of the results obtained in the hierarchical regression and
better explain this phenomenon. The family environment scale also needs more items to increase the
psychometric properties of the scale, especially in what concerns the factor related with family rules.
As to the best way for parents to mediate children’s DT activities, several authors (e.g., Livingstone
et al., 2011; Nguyên & Mark, 2014; Ponte et al., 2012; Simões, 2012) have shown evidence that a
considerable number of children and teenagers, especially those of lower socioeconomic levels, have
declared not to experience any mediation from their parents in using DT and most of them actually
use it in their bedrooms or in other private divisions of the household. Livingston and collaborators
(2011) suggest this is mainly due to the generational differences concerning digital competences.
With this in mind, how can parents help their children deal with cyberbullying situations and mediate
their online activities?
The complex nature of online activities defies models of parental supervision. This and other
research studies have suggested that, particularly for teenagers, an affective relationship, dialogue,
parental interest in children’s activities, and taking opportunities to learn digital technologies with
children constitute the most effective ways of supervision and help parents to know what children
actually do while they are using DT (Morais, 2012; Simões, 2012; Sorbring, 2014). In contrast,
restrictive mediation strategies, such as filters, restrictions, and bans, might be less effective mediation
measures when it comes to supervising the use of Internet and technologies, because adolescents can
use them in locations or conditions out of parents’ supervision, like libraries or mobile phones from
colleagues (Kowalski, Limber & Agaston, 2008; Mesch, 2009). The fact that parents seem to know
less about DT than their children should not be an obstacle in monitoring children’s online activities
Volume 6 • Issue 3 • July-September 2016
75
and supporting them in their difficulties and worries. If family relationships are based on mutual
support, trust, dialogue, and explicitness of rules concerning the use, perks, and perils of DT, these
will contribute more effectively to preventing cyberbullying involvement as it seems to occur with
other domains of parental education with adolescents (Poulin & Nadeau, 2012).
Moreover, parents might benefit from programs and manuals that better help them deal with
and intervene in situations of cyberbullying or other problems associated with the use of Internet
and social networks (Farrington & Ttofi, 2009; Jäger, 2010; Jäger, Stelter, Amado, Matos & Pessoa,
2012; Kowalski, Limber & Agaston, 2008). These manuals can improve parents’ skills on digital
technologies, focus their attention, and make them aware of the risks their children are exposed to
when using the Internet. In addition, these guidebooks allow parents for skills to supervise children
and adolescents’ Internet activities and to intervene efficaciously when their children suffer from or
practice cyberbullying. In any instance, families that promote rules through dialogue with children
under a warm and affective relationship seem to provide protection against cyberbullying.
This research revealed that the three groups involved in cyberbullying (victims, aggressors, and
victims–aggressors) presented worse quality of family environment when compared to those who
were not involved in terms of perceptions of family support and family rules. Aggressors’ and victims-
aggressors’ perceptions revealed less family rules, when compared to victims and non-involved, and
victims presented less family rules, when compared to non-involved, but had more family rules than
aggressors and victims–aggressors. The three groups involved in cyberbullying reported less family
support than those non-involved in those problems.
In conclusion, a lack of family support seems to predict cyber-victimization; regarding cyber-
aggression, a lack of both family support and rules are predictors for cyber-aggression, however
the variance explained by these variables was low. This can be probably explained by the fact that
in adolescence the desire of autonomy from parents, together with the increase of group and peer
influence, could enhance peer influence as a more important factor in the prediction of the involvement
of adolescents in these kinds of conducts than family variables. In fact, several studies suggested that
youth social adjustment and development result from multiple sources of environmental influence
(e.g., Anderson, Sabatelli & Kosutic, 2007; Fanti et al., 2012; Sasson & Mesch, 2014). Future studies
should balance and access the interaction of several variables, like the influence of different family
aspects, peers, and school ethos.
Acknowledgements: This study was conducted in the frame of the project “Cyberbullying – a
diagnosis of the situation in Portugal,” financed by The Foundation for Science and Technology
(program PTDC/CPE-CED/108563/2008) in a partnership with the University of Coimbra and the
University of Lisbon, coordinated by professor João Amado.
Volume 6 • Issue 3 • July-September 2016
76
Amado, J., Freire, I., Matos, A., Vieira, C., & Pessoa, T. (2012). O cyberbullying e a escola: uma análise da
situação em Portugal. [The cyberbullying and the school: analysis of the situation in Portugal]. In Atas do III
Colóquio Luso-Brasileiro de Sociologia da Educação - Problemas contemporâneos da educação no Brasil e
em Portugal: desafios à pesquisa [Proceedings of the III Congress Luso-brasilien of Socioloy of Education –
Contemporary problems of education in Portugal and Brasil: challenges to research]. Rio de Janeiro, Brasil.
Amado, J., Matos, A., & Pessoa, T. (2009). Cyberbullying: Um novo campo de investigação e de formação.
[Cyberbullying: A new field of research and training]. In B. Silva, A. Lozano, L. Almeida, & M. Uzquiano
(Eds.), In Atas do Congresso Galeco-Português de Psicopedagogia [Proceedings of the X International Congress
Galician-Portuguese of Psychopedagogy] (pp.262-373). Retrieved from http://www.educacion.udc.es/grupos/
gipdae/documentos/congreso/Xcongreso/pdfs/t1/t1c11.pdf
Anderson, S., Sabatelli, R., & Kosutic, I. (2007). Families, urban neighborhood youth centers, and peers as
contexts for development. Family Relations, 56(4), 346–357. doi:10.1111/j.1741-3729.2007.00464.x
Blanc, M., & Janoz, M. (2002). Regulação familiar da conduta delinquente em adolescentes. [Family regulation
of delinquent conduct in adolescents] In A. C. Fonseca (Ed.), Comportamento anti-social e família. Uma
abordagem científica [Anti-social behavior and family. A scientific approach] (pp. 37–92). Coimbra: Almedina.
Boulton, M., Lloyd, J., Down, J., & Marx, H. (2012). Predicting university students’ self-reported engagement
in cyber-bullying and traditional bullying from attitudes. Cyberpsychology, Behavior, and Social Networking,
15, 141–147. doi:10.1089/cyber.2011.0369
Corvo, K., & deLara, H. (2010). Towards an integrated theory of relational violence: Is bullying a risk factor for
domestic violence? Aggression and Violent Behavior, 15(3), 181–190. doi:10.1016/j.avb.2009.12.001
Diaz-Aguado, M.J. (Ed.), (2004). Prevención de la violencia y lucha contra la exclusión desde la adolescencia.
Intervención a través de la familia [[Prevention of violence and social exclusion in adolescence. Intervention
with the family]] (Vol. 3). Madrid: Instituto da Juventud.
Diaz-Aguado, M.J. (2005). Prevención de la violencia y de la exclusión a través de la intervención familiar.
[Prevention of school violence through familiar intervention]. In J. Sanmartín (Ed.), Violencia e escuela
[Violence and school] (pp. 121-135). IX Reunión Internacional sobre Biología e Sociología de la Violencia.
[International Conference about Biology and Sociology of Violence]. Valencia: Centro Reina Sofía para el
Estudio de la Violencia.
Fanti, K., Demetriou, A., & Hawa, V. (2012). A longitudinal study of cyberbullying: Examining risk and protective
factors. European Journal of Developmental Psychology, 9(2), 168–181. doi:10.1080/17405629.2011.643169
Farrington, D., & Ttofi, M. (2009). School-based programmes to reduce bullying and victimization. London:
The Campbell Collaboration. Retrieved from www.campbellcollaboration.org
Fletcher, A., & Blair, B. (2014). Maternal authority regarding early adolescents’ social technology use. Journal
of Family Issues, 35(1), 54–74. doi:10.1177/0192513X12467753
Fonseca, A. C. (2002). Comportamento anti-social e família: novas abordagens para um velho problema. [Anti-
social behavior and family: new approaches to an old problem] In A. C. Fonseca (Ed.), Comportamento anti-
social e família [Anti-social behavior and family]. (pp. 1–14). Coimbra: Almedina.
Francisco, S., Veiga Simão, A. M., Ferreira, P. C., & Martins, M. J. D. (2015). Cyberbullying: The hidden side
of college students. Computers in Human Behavior, 43, 167–182. doi:10.1016/j.chb.2014.10.045
Granot, D., & Mayseless, O. (2001). Attachment security and adjustment to school in middle childhood.
International Journal of Behavioral Development, 25(6), 530–541. doi:10.1080/01650250042000366
Hair, J. F., Anderson, R. E., Tatham, R. L., & Black, W. C. (1998). Multivariate data analysis (5th ed.). Upper
Saddle River: Prentice Hall.
Hertlein, K. (2012). Digital dwelling: Technology in couple and family relationships. Family Relations, 61(3),
374–387. doi:10.1111/j.1741-3729.2012.00702.x
Volume 6 • Issue 3 • July-September 2016
77
Hinduja, S., & Patchin, J. (2013). Social influences on cyberbullying behaviours among middle and high school
students. Journal of Youth and Adolescence, 42(5), 711–722. doi:10.1007/s10964-012-9902-4 PMID:23296318
Ijzendoorn, M. H. (2002). Vinculação, moralidade emergente e agressão: para um modelo socioemocional
e desenvolvimental do comportamento anti-social. [Attachment, emergent morality and aggression: for a
socioemotional and developmental model of anti-social behavior] In A. C. Fonseca (Ed.), Comportamento anti-
social e família [Anti-social behavior and family] (pp. 269–299). Coimbra: Almedina.
Jäger, T. (Ed.), (2010). Taking action against cyberbullying – Training Manual. Landau: Verlag Empirische
Padagogik. Retrieved from http://www.cybertraining-project.org/book/
Jäger, T., Stelter, C., Amado, J., Matos, A., & Pessoa, T. (Eds.). (2012). Cyberbullying - Um manual de formação
de pais. [Cyberbullying: a Parents’ training manual]. Retrieved from http://ct4p.zepf.eu/CT4P_Training_
manual_PT.pdf
Kõiv, K. (2012). Attachment styles among bullies, victims and uninvolved adolescents. Psychological Research,
2(3), 160–165.
Kowalski, R., Giummetti, G., Schroeder, A., & Lattanner, M. (2014). Bullying in the digital age: A critical review
and meta-analysis of cyberbullying research among youth. Psychological Bulletin, 2014, 1–61. doi:10.1037/
a0035618 PMID:24512111
Kowalski, R., Limber, S., & Agaston, P. (2008). Cyberbullying. Oxford: Blackwell Pub.
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
Livingstone, S., Haddon, L., Gorzig, A., & Olafsson, K. (2011). Risks and safety on the Internet: the perspective
of European children. Full findings. Retrieved from www.eukidsonline.net
Matos, A. P. M. (2006). Televisão e violência. (Para) novas formas de olhar [Television and violence. New
approaches]. Coimbra: Almedina.
Mesch, G. (2009). Parental mediation, online activities, and cyberbullying. Cyberpsychology & Behavior, 12(4),
387–393. doi:10.1089/cpb.2009.0068 PMID:19630583
Morais, T. (2012). Viciados no quarto? [Addicted to room?)] In C. Ponte, A. Jorge, J. Simões, & D. Cardoso
(Eds.), Crianças e Internet em Portugal [Children and Internet in Portugal] (pp. 75–86). Coimbra: Minerva.
Nguyên, T. T., & Mark, L. (2014). Cyberbullying, sexting, and on-line sharing a comparison of parent and school
faculty perspectives. Journal of cyber behavior, psychology and learning, 4(1), 76-86.
Olweus, D. (1995). Bullying at school. What we know and what we can do. Oxford: Blackwell.
Ortega, R., Calmaestra, J., & Mora-Merchan, J. (2008). Cyberbullying. International Journal of Psychology
Psychological Therapy, 8(2), 183-192. doi:10.27/0044-3409.217.4.197
Ortega, R., Elipe, P., Mora-Merchan, J., Calmaestra, J., & Vega, E. (2009). The emotional impact on victims of
tradicional bullying and cyberbullying. The Journal of Psychology, 217(4), 197–204.
Ortega, R., Elipe, P., Mora-Merchán, J. A., Genta, M. L., Brighi, A., Guarini, A. K., & Tippett, N. etal. (2012).
The Emotional impact of bullying and cyberbullying on victims: A European cross-national study. Aggressive
Behavior, 38(5), 342–356. doi:10.1002/ab.21440 PMID:22782434
Ponte, C., Jorge, A., Simões, J., & Cardoso, D. (Eds.). (2012). Crianças e Internet em Portugal [Children and
Internet in Portugal]. Coimbra: Minerva.
Poulin, F., Nadeau, K., & Scaramella, L. V. (2012). The role of parents in young adolescents’ competence with
peers: An observational study of advice giving and intrusiveness. Merrill-Palmer Quarterly, 58(4), 437–462.
doi:10.1353/mpq.2012.0021
Sasson, H., & Mesch, G. S. (2014). Parental mediation, peer norms and risky online behaviors among adolescents.
Computers in Human Behavior, 33, 32–38. doi:10.1016/j.chb.2013.12.025
Schwartz, D., Dodge, K. A., Petti, G. S., & Bates, J. E. (1997). The early socialization of aggressive victims of
bullying. Child Development, 68(4), 665–675. doi:10.2307/1132117 PMID:9306645
Volume 6 • Issue 3 • July-September 2016
78
Maria José D. Martins is a professor at the College of Education of the Polytechnic Institute of Portalegre in
Portugal, where she teaches in the frame of social work, primary teachers and kindergartners degrees, and also
coordinates a master degree about children and adolescents at risk. She is a researcher member of the UIDEF.
She has several articles, book chapters and a book published in Portuguese and English.
Ana Margarida Veiga Simão is a professor at the Psychology Faculty, University of Lisbon in Portugal, where
she teaches in the frame of Educational Psychology and Teacher Education degrees and coordinates the Master
Degree and the Inter-University Doctoral Program on Educational Psychology. She is a researcher member of the
UIDEF. She has several articles, books and books chapters published in Portuguese and in English.
Isabel Freire, PhD in Education Sciences, is a professor and researcher at the Institute of Education of the University
of Lisbon, Portugal, where she coordinates doctoral courses in Education (specialization in Teacher Education) and
master courses in Education and Formation. She is a researcher member of the UIDEF. She has several scientific
articles, book chapters and books, published in Portuguese, English, French and Spanish.
Ana Paula Caetano is an associate professor at the Institute of Education of the University of Lisbon in Portugal,
where she teaches, being coordinator of a master degree on Education and Training – Cultural and Social
Development and of a master degree on Intercultural Education. She is a researcher member of the UIDEF. In the
last years she was involved in formal and non-formal education research projects. She published several papers,
books and books chapters in Portuguese, English and French.
Armanda Matos is Professor at the Faculty of Psychology and Educational Sciences of Coimbra University in
Portugal, and a researcher member of the CEIS 20 (Coimbra’s University Centre for 20th Century Interdisciplinary
Studies). She has teaching responsibilities in the field of media education and media literacy and her focus of
research and interest is in the relationship between media/education/citizenship, especially, violence in schools and
cyberbullying. She has several scientific articles, book chapters and books, published in Portuguese and English.
Simões, J. (2012). Mediações dos usos da Internet. Resultados nacionais do inquérito EU KIDS ONLINE.
[Mediations of Internet uses. National’s results of the enquiry EU KIDS ONLINE]. In C. Ponte, A. Jorge, J.
Simões, & D. Cardoso (Eds.), Crianças e Internet em Portugal [Children and Internet in Portugal] (pp. 121–143).
Coimbra: Minerva.
Smith, P. (2005). Violencia escolar e acoso: fatores de riesgo familiares. [School violence and bullying: family risk
factors]. In J. Sanmartín (Ed.), Violencia e escuela [Violence and school] (pp. 59-76). IX Reunión Internacional
sobre Biología e Sociología de la Violencia. [International Conference about Biology and Sociology of Violence].
Valencia: Centro Reina Sofía para el Estudio de la Violencia.
Smith, P. (2009). Cyberbullying – Abusive relationships in cyberspace. The Journal of Psychology, 217(4),
180–181. doi:10.1027/0044-3409.217.4.180
Sorbring, E. (2014). Parents’ concerns about their teenage children’s Internet use. Journal of Family Issues,
35(1), 75–95. doi:10.1177/0192513X12467754
SPSS, Inc. (2011). SPSS Version 20.0. Chicago, IL: Author.
Walker, C., Sockman, B., & Koehn, S. (2011). An exploratory study of cyberbullying with undergraduate
university students. TechTrends, 55(2), 31–38. doi:10.1007/s11528-011-0481-0
Willard, N. (2005). Educator’ guide to cyberbullying and cyberthreats. Retrieved from http://www.accem.org/
pdf/cbcteducator.pdf
