Cyberbullying victimization among adolescencts: results of the International self-report delinquency study 3

ABSTRACT The study is based on a wide international research study, the International Self-Report Delinquency Study 3 (ISRD-3), with the aim of understanding if there is a relationship between having lived traumatic intrafamily experiences and potentially being a victim of cyberbullying. The data have been collected by a questionnaire ISRD-3 administered to a 20 European country sample of 57.463 students from 7th to 9th grade. Results show a statistically significant relationship between traumatic intrafamilial experiences such as a serious illness of one of the parents, episodes of violence suffered and assisted, separation or divorce of the parents, and being victimized in adolescence of cyberbullying.


Introduction
Social media is an important venue for interaction and conversation among youth all over the world. Fully 76% of all teens use social media. Facebook is the dominant platform, with 71% of all teens using it. Instagram and Snapchat have also become increasingly important, with 52% of teens using Instagram and 41% using Snapchat. Nearly two-thirds of teenagers around the world use Twitter and another third use Google Plus. Fewer teens use Vine or Tumblr.
Social media plays a critical role in connecting teens to new friends, allowing teens to learn more about new friends and get to know them better. Nearly two-thirds (64%) of teens who have made a new friend online say they have met new friends on a social media platform. Two-thirds (62%) of teens say they have shared their social media username with a brand-new friend as a way to stay in touch.
Beyond making new friends, social media is a major way that teens interact with their existing friends. More than nine-in-ten teens (94%) say they spend time with friends on social media. Fully 30% say they spend time with friends on social media every day, and another third (37%) say they do so every few days. When asked to rank the ways they communicate with friends, social media sites like Facebook or Twitter are one of the top ways of communicating with friends for two-thirds (66%) of teens (Bulck, 2007;Raskauskas & Stoltz, 2007;Adams et al., 2013;Pull & Swabey, 2014;UNICEF, 2017).
The familiarity of adolescents with the new communication channels, however, does not make them computer experts and connoisseurs of the network, especially of the damaging consequences that can derive from an unaware use of it. These new relational modalities often lead young people to hide behind the technological means to implement new forms of bullying that have developed alongside the traditional forms of bullying, taking the name of cyberbullying, a term defined in 2006 by Peter K. Smith and colleagues like 'Aggressive and intentional act, conducted by an individual or group of individuals, using various forms of electronic contact, repeated over time against a victim who has difficulty defending himself'.
Today we can specify cyberbullying as bullying that occurs through digital devices such as cell phones, computers, and tablets. For the US government group stopbullying.gov 'Cyberbullying can occur via SMS, SMS and apps or online in social media, forums or games where people can view, participate or share content. Cyberbullying includes posting, posting, or sharing negative, harmful, false, or petty content on someone else. It can include sharing personal or private information about someone else which causes embarrassment or humiliation. Some cyberbullying oversteps the line in illegal or criminal behavior.' The difference between traditional bullying and cyberbullying lies in the fact that, in the latter, aggressive acts hide behind forms of communication attributable to the web and do not happen in person, implying that everything that is published online with the intent to hurt and denigrate becomes visible to millions of users, spreads rapidly and in most cases cannot be removed, rendering the victim powerless. The subjects who, by sharing this material increase its diffusion and consequently feed the danger of the phenomenon, are the so-called bystanders.
The lack of physical contact between the cyberbully and the victim, the condition of anonymity in which the cyberbully finds himself at the time of the implementation of aggressive behaviours, and the ease with which the material can be disseminated and the unlimited number of users that is possible reach make this phenomenon particularly dangerous from the point of view of the consequences for the victim (Genta et al., 2013;Nixon, 2014).
A recent UNICEF poll of 170.000 young people has collected the frequency of online bullying in 30 countries: one in three has been the victim of cyberbullying in a lifetime 1 . This poll also tells us that one in five of these young people skipped the school due to online bullying.
In Italy, the vast majority of young Italian people, 61%, say that they have been the victims of some form of bullying, whether it be cyberbullying or the traditional kind, according to a survey released on Thursday by the Terre des Hommes association and the Scuolazoo website for students. It said that 68% have witnessed episodes of bullying. The survey said drugs and sexual violence were among the threats most feared by young Italians. It said six out of 10 adolescents did not feel safe online. It said 52.16% of girls and young women feared being the victim of revenge porn. The survey revealed that 93% of the adolescents felt alone in 2020, the year the COVID-19 pandemic exploded, up 10 points with respect to 2019 (ANSA).
The literature, which it mainly refers to traditional bullying, clearly shows us how bullied victims are 'double victims': at home as well as at school, because they often experience domestic violence, conflicting parental relationships and the absence of one or both parents (Baldry, 2003;Bauer et al., 2006;Boel-Studt & Renner, 2013). The family, in fact, can be both a protective factor and a risk factor in the development of child victims or aggressors since both are linked to parents with high scores in Parental Stress Scale (PSS; adaptation by Oronoz et al., 2007).
The family is the first social context within which the primary socialization process of its members develops, which could favour the legitimization of some forms of violence (De Vera, 2016).
The literature that has dealt with analysing the phenomenon of victimization by cyberbullying in adolescence has highlighted the close relationship between being victims and having experienced traumatic family experiences characterized by a potentially damaging parental relationship on an emotional level or by the absence of one or both caregivers (Baldry, 2003;Baldry & Farrington, 2000;Bauer et al., 2006;Bertol et al., 2014;Boel-Studt & Renner, 2013;Casas et al., 2013;Cassidy et al., 2013;Gatti et al., 2010).
Regarding , specifically, online bullying, Cortés et al. find that family conflicts, an authoritarian parenting style, and communication problems with parents are all risk factors for teens regarding cyberbullying. In this direction, we also find a recent longitudinal study conducted on 861 adolescents attending Cypriot schools, which highlighted how 'parenting' fulfils the function of predictor for all forms of bullying/victimization in adolescents (Charalampous et al., 2018).
From a second study conducted on a cohort of 18.341 students between the ages of 15 and 17, from six Chinese cities, enrolled between 2009 and 2010, it emerged that having witnessed episodes of domestic violence or having experienced a feeling abandonment by caregivers are associated with a high probability for children to become 'Internet victims'; higher probabilities are also found for those who have witnessed parental divorce and/or who live in environments with low family income and poor schooling (Chen et al., 2018;Smith et al., 2006).
Regarding the relationship between early traumatic experiences and victimization by cyberbullying, also a recent Italian study, based on the ISRD3 2 national database, found a significant association between being a victim of cyberbullying (in its sample of youngers between the ages of 12 and 16) and being a male who lived in a family with high aggressiveness as well as being female and having experienced three types of traumatic intrafamily events: illness of a parent, violence or serious conflict between parents and separation or divorce of the same (Gualco et al., 2021). Along the same lines, the authors want to understand if at the international level, could similar results be found . Experienced early traumatic life events could increase the vulnerability to be a victim of cyberbullying. There could also be a gender difference between groups. The study shown is based on data obtained from the International Self-Report Delinquency Study-3.

Design
ISRD3 uses a standardized self-report survey conducted in school settings among students in grades 7,8 and 9 (or equivalent grades for children age 12-16), randomly selected from schools in two medium or large cities (in each city, 300 students per grade are being surveyed -a total of 900 children per city).
The data to create the database was entered into the computer using the EPIDATA software.

Sample
Students in grades 7, 8 and 9 (or equivalent grades for children age 12-16) randomly selected from schools in two medium or large cities. Schools were selected from a complete list of middle and high schools based on their region, urbanity, school type (public vs. private, vocational, technical and academic schools, racial composition and size). The students were extracted using a systematic random sampling method based on the list provided by each school. The first name was discovered by chance, and then the others were selected for a fixed interval. Students were admitted to participate in the research on a voluntary basis upon signature, by the student's parents or who in parent's place, of a letter concerning informed consent.

Participants
Students were admitted to participate in the research on a voluntary basis upon signature, by the student's parents or who in parent's place, of a letter concerning informed consent. 61.218 students have participated in the study.

Tool
The tool used was a self-report questionnaire compiled anonymously.
The questionnaire has closed-ended questions (some answers are dichotomous, others multiple, others on a Likert scale).
The questionnaire included questions about lifestyle, socio-demographic characteristics, victimization and deviant or law-offending behaviour.
The questionnaire was completed in class in the presence of the researcher and the teacher if possible.
The instruction was reported on the first page of the questionnaire: 'Hello, this survey is one of the first to give a voice to young people from different countries (about thirty currently), to allow them to express their opinions and compare situations of different countries.
In particular, we would like to learn more about some details about your life, the school you attend, how you spend your free time and the problems you may face in these situations. The questions in the questionnaire relate to your personal experience and opinions and you are free to answer or not answer. Consider that the questionnaire is completely anonymous; your name will not appear in any document and your parents, as well as your teachers, will not be able to know the answers you will give. Even our research group will not know who gave the answers. If there are any unclear questions, do not hesitate to contact the assistant present in the classroom. Don't think too much before answering the questions, just respond spontaneously. We sincerely thank you for agreeing to participate in our investigation. Before starting, please enter the number that will be shown in the fields below '.

Data
The ISRD-3 database relating to 20 countries was analysed: Armenia, Austria, Belgium, Bosnia and Herzegovina, Croatia, the Czech Republic, Denmark, Estonia, Finland, France and Germany, (Italy, not used for the analysis N = 3508), Indonesia, Lithuania, the Netherlands, Serbia, Slovakia, Switzerland, Ukraine, United Kingdom, Venezuela. N = 57,463.

Hypothesis
Based on the most recent international research on the subject (Cassidy et al., 2013;Charalampous et al., 2018;Chen et al., 2018), the authors expect that the victimized children have experienced one or more of the potentially traumatic events within the family context.

Goals
The study aimed at understanding if having traumatic childhood intra-family experiences is associated with a greater risk of being victims of cyberbullying in adolescence.
Specific objectives: 1. Understanding which traumatic childhood intra-family experiences may be a major risk factor for cyberbullying victimization in adolescence; 2. Investigate whether the possible association between traumatic childhood intra-family experiences and cyberbullying victimization differs with respect to the gender variable.
3. Observe if at an international level the same trend found in the Italian study mentioned in the introduction is observed.

Measures
The data analysis was carried out by taking the item 'has anyone ever made fun of you or harassed or offended you by e-mail, instant messaging systems (Facebook, Messenger, etc.), in chat, on an Internet site or through SMS?', with a dichotomous reply 'yes . . . how many times in the last 12 months', ore 'no'. To create the 'cyberbullying victimization' variable, positive responses from students who reported having been victims of cyberbullying three or more in their lifetime were taken into consideration.
This variable was crossed between the one relating to exposure to traumatic events within the family unit identified by taking the item into consideration, always with a dichotomous answer 'yes', 'no', 'Have you ever experienced one or more of the following events? ': • Death of your father or your mother; • A serious illness of one of your parents or someone close to you; • One of your parents' alcohol/drug problems; • Episodes of physical violence between your parents; • Serious and repeated conflicts between your parents; • Your parents' separation or divorce.

Statistical analysis and study limits
The analysis of the responses to the questionnaire was conducted through a non-parametric chisquare statistical test, which made it possible to calculate the association between variables measured on categorical scales and to obtain a probability measure of the identified relationship.
In the data analysis, the p-value assumed as the level of significance was 0.05. The SPSS 21 programme was used to analyse the data.
It is important to emphasize that the Chi-square statistical test has limits determined by the fact that the mere verification of the significance of the relationship with the application of a hypothesis test does not allow to describe the strength of the relationship; another aspect to note is that this statistical analysis in the interpretation of the results does not allow to evaluate the effect of other intervening variables.
Furthermore, the Authors are aware of the fact that the study of the family and of one's own internal relational dynamics is extremely complex, so any attempt to describe and understand them cannot but be limited. Table 1 shows the association between last year cyberbullying victimization and intra-family traumatic childhood experiences.

A. Traumatic intrafamilial events and cyberbullying victimization
Looking at the percentage frequency, we note that almost half of the sample (49.3%) who suffered cyberbullying in the last year had experienced serious illnesses of parents.
High percentages are found for serious conflict between parents (33.1%) and separation/divorce of parents (32.2%). Looking at the p-value, we can observe a statistically significant association (p ≤ 0.05) between having experienced a serious illness of one of the parents, having parents with alcohol and drug problems, having witnessed violence or serious conflicts between them or their separation/divorce, and being a victim of cyberbullying in adolescence. These experiences, therefore, constitute an important risk factor on which to intervene with treatment and prevention programmes. Table 2 shows the association between last year cyberbullying victimization and intra-family traumatic childhood experiences according to the gender variable. N = 28,612 male and N = 28,851 female students participated in the study. The sample is therefore homogeneous according to the gender variable.

Traumatic intrafamilial events and cyberbullying victimization, according to gender variable
Having made this premise, from Table 2 we can observe that the female gender has a greater risk than the male gender of undergoing cyberbullying. N = 4935 of female samples versus N = 2962 of male samples (almost double) declared to have been a victim of cyberbullying in the last year. From Table 2 we note how the females who reported having suffered cyberbullying in the last year had even greater experiences of childhood intra-family trauma compared to the male sample. 45.1% of males who experienced severe parental illness have also been victims of cyberbullying in the past year, as have 51.9% of girls. 10.6% of boys and 12.5% of girls who had parents with alcohol and drug problems were cyberbullied. As well as 26.4% of males and 37.1% of females, who have witnessed severe conflicts between their parents, have been victims of online violence. Finally, 28.6% of boys and 32.7% of girls who witnessed the separation and/or divorce of their parents were digitally victimized.
Looking at the p-value, we can observe the same trend as in Table 1, also in regard to the gender variable.
Serious illness of parents, alcohol/drugs problems of parents, violence between parents, serious conflicts between parents and separation or divorce of parents are significantly associated (p ≤ 0.05) with cyberbullying in adolescence.

Discussion
The study showed tried to bring out those characteristics that better than others could prove useful in identifying a possible association between traumatic events experienced in the family context and victimization suffered by 'cyberbullies' in adolescence. The data in the literature on the subject have shown that having undergone traumatic experiences in the intra-family context constitutes a heavy risk factor for victimization from cyberbullying in adolescence (Cassidy et al., 2013;Charalampous et al., 2018;Cheli & Giacopuzzi, 2009;Chen et al., 2018;Domitrovich & Bierman, 2001;Espelage et al., 2012).
However, discordant data emerge regarding gender: while some studies show that the risk of victimization is higher for females who have been exposed to traumatic events in childhood (Baldry, 2003;Sharma et al., 2017), other research has not found significant differences in this area between boys and girls (Boel-Studt & Renner, 2013;Holt et al., 2009).
Therefore, it appeared useful to thoroughly investigate possible associations between these variables in order to identify any predictive factors of victimization.
The results of the research carried out confirmed the marked diffusion of the phenomenon at an international level. 7897 out of 57,463 participants, reported having suffered cyberbullying in the last year (13%).
From the statistical analysis carried out, it was possible to observe a significant association between having lived in a family context characterized by aggression, or having lived parental divorce/separation, or having parents with alcohol or drug problems, and a greater probability of suffering cyberbullying victimization experiences.
These results are in accordance with those found by other research on the subject, which underlines how these traumatic events, particularly the experience of hostility, litigiousness and violence between caregivers, represent a potentially predictive element of developmental difficulties and psychological discomfort in adolescence and adulthood (Marín-Cortés et al., 2019;Reiter et al., 2013).
However, there was no statistically significant association between the death of a parent and a greater probability of being a victim of cyberbullying.
A possible interpretation of these results may be that the perception of a poor emotional support by the caregiver, due to his absence caused by the disease, can determine in the child a state of insecurity and less assertiveness, which can determine a greater propensity to suffer abuse, to remain in a role of victim in the relational dynamics with peers, a greater reluctance to report the wrongs suffered to the parent precisely because of the difficult condition experienced in the family unit. This is confirmed by some international studies on the subject (Garaigordobil & Machimbarrena, 2017;Méndez et al., 2017;Mishna & Alaggia, 2005;Raskauskas & Stoltz, 2007).
Subsequently, the authors divided the sample 'victims of cyberbullying in the last year' with respect to the gender variable to understand if the distribution of the variable of potentially traumatic intrafamilial events suffered changes.
It has been observed that with the same traumatic event, females have a greater probability of being victims of cyberbullying. Significant association, on the other hand, is found in all traumatic events taken into consideration except for the death of a parent, regardless of being male or female.
This disagrees with the results of the Italian study, carried out by Gualco et al. (2021), on cyberbullying, which analysed the data of the Italian sample who participated in the ISRD3. Indeed, there appears to be a significant association for the female group between having suffered in childhood events such as severe illness of a parent, parental separation/divorce, severe conflict between parents and parental alcohol or drug problems. For the group of males, on the other hand, the significance of the association is only in terms of having lived within a family context characterized by high aggression and physical violence between parents.

Conclusions
In the light of the results obtained, it is clear that any intervention in cyberbullying must take into account the importance of the family context, hoping for the involvement of caregivers so that they can understand the decisive role they can play in preventing certain behaviours associated with cyberbullying. These observations also make it possible to broaden the field of research in a preventive and predictive way where there is the possibility of observing children and pre-adolescents knowing their family situation in order to be able to carry out early interventions in situations of high risk. Notes 1. https://www.unicef.org/press-releases/unicef-poll-more-third-young-people-30-countries-report-being-victimonline-bullying?fbclid=IwAR07vqG-LvlZHKohSWmSlhROgyyl3BhFiXoXPE9ToVLyiEfotsXcDSvamEc 2. The Self-Report Delinquency Study (ISRD) is an international collaborative research enterprise with a crossnational description and explanation of juvenile delinquency as its main objective. The ISRD project commenced in the early 1990s with the ISRD 1. It began with a number of pilot studies, workshops and working papers exploring the possibility of conducting a truly standardized international self-report study of delinquency. The impetus for the ISRD-1 was the conviction that research on the prevalence of juvenile delinquency would be highly relevant for criminal policy as well as for criminological theory. This expectation helped to start the project in spite of the then unsolved methodological and organizational difficulties. Thirteen countries, most of which belonging to the European Union, collaborated in the first study. They were Finland, Great Britain, The Netherlands, Germany, Belgium, Spain, Italy, Portugal, Switzerland, Northern Ireland, Greece, New Zealand and the US (Nebraska). The target group for the ISRD-1 was 12-18 years. Most countries used selfadministered pencil and paper questionnaires and a few used personal interviews. The premises and results of the first 'sweep' of the ISRD study have been presented in great detail in previous publications (Junger-Tas & Marshall, 1999;Junger-Tas et al., 2003, 2004, 1994. In 2005, the second ISRD commenced. The ISRD-2 design was a major improvement over ISRD-1, in particular with respect to focus on the importance of developing and enforcing a research protocol that was to be followed by all participants. The ISRD-2 was conducted in 25 European and 6 American countries. The target age group was 12-17 years. The tool used was the self-report questionnaire. The results have been presented in previous publications (Gatti et al., 2015(Gatti et al., , 2010(Gatti et al., , 2013Junger-Tas, 2012;Junger-Tas et al., 2010;Steketee et al., 2013).
The ISRD-3 commenced in 2014. Data were collected between 2014 and 2015.The project involved the same countries as for the ISRD-2.