Introduction

Bullying is a serious and highly prevalent problem worldwide. It is estimated that more than one in three students aged 13 to 15 report bullying involvement (Fry et al. 2018). Bullying occurrence at school-age represents a serious threat to the well-being and healthy development of adolescents, as it has been associated with low school performance (Currie et al. 2012), relational problems (Smith et al. 2002), and suicide-related behaviours (Arango et al. 2016).

Bullying is an intentional aggressive and negative behaviour, repeated over time, that involves a power imbalance favouring the aggressor, with victims having no means to defend themselves (Craig and Harel 2004). The most common forms of bullying behaviours among adolescents are name-calling, teasing, making threats, spreading rumours, taking of personal belongings, and rejection by excluding someone from a group on purpose (Smith et al. 2002). Most of the bullying situations tend to start in school, and sometimes they are not taken as serious but instead, as a normal interaction between peers (Committee on the Biological and Psychosocial Effects of Peer Victimization et al. 2016).

According to the World Health Organization (WHO) in 2015, one in four adolescents was bullied in the previous month (World Health Organization 2015). However, estimates of bullying prevalence vary greatly across surveys. The Health Behaviour in School-Aged Children (HBSC) study, in a survey conducted in 42 countries and regions across the WHO European Region and North America, showed a worldwide prevalence ranging between 3% and 35% of young people reporting the involvement in bullying during the past 2 months (Craig and Harel 2004). The involvement in bullying can be as victim-only, when the child suffers from bullying but is not an aggressor; as aggressor-only, when the involvement is solely as aggressor, but not as victim-only; and as victim-aggressor when the child is involved as victim and aggressor simultaneously. Also, according to the WHO European Health Information Gateway from 2017, 11% of girls and 17% of boys aged 11 years old, and 14% of girls and 16% of boys aged 13 have reported being victims of bullying at least twice in the previous 2 months (World Health Organization 2017). In addition to age, gender seems to play an important role in bullying involvement (Steinfeldt et al. 2011). Gender constitutes a structure of social practice that establishes relations of power, attitudes, and hierarchies among people, groups, and institutions (Steinfeldt et al. 2011). Research suggests that boys are more prone to be victims-only and aggressors-only of bullying, especially in its physical expression (Craig and Harel 2004), while girls are more likely to engage in situations of indirect bullying, such as teasing or gossiping (Committee on the Biological and Psychosocial Effects of Peer Victimization et al. 2016; Finkelhor et al. 2015).

Additionally, family socioeconomic circumstances are expected to have a significant influence in the involvement in bullying, and literature states that children from low socioeconomic backgrounds and living with a single parent are more likely to be involved in bullying behaviours (Nordhagen et al. 2005). However, some inconsistent findings are being observed regarding the different roles (victim-only, aggressor-only, or victim-aggressors).

This study aims to assess the prevalence of bullying among Portuguese middle and high public school adolescents during the 2014/2015 academic year, identifying social and behavioural factors and well-being related feelings, and also the tactics mostly used according to gender.

Materials and methods

This was a cross-sectional study conducted in public schools from Porto, Portugal, during the 2014/2015 academic year. The study protocol was approved by the Authority of Portuguese School (Direção Geral dos Estabelecimentos Escolares — DGEstE), and ethical approval was granted by the Ethics Commission from the Instituto de Saúde Pública da Universidade do Porto.

The four largest public school groups in Porto were contacted, and the Executive Board of each school in the school groups designated a professor who would be the contact person. Schools received a covering letter explaining the study and attaching the informed consent form.

Written informed consent was obtained from both the adolescents and their legal guardians, but kept separated from the questionnaire and placed at each school in order to ensure anonymity.

Participants

The sample included 2623 (1186 boys and 1437 girls) enrolled in public schools, from the 7th to 12th grade. The overall participation rate was 60%. For the analysis, participants who did not fulfil the bullying section of the questionnaire were excluded. No significant differences were found between included and excluded participants with regard to school grade, age, and parents’ occupation; however, boys were more frequently excluded (data not shown).

Data collection

From January to April of 2015, professors delivered the questionnaire to students who had returned a signed informed consent by their legal guardians. The questionnaire assessed information on social and demographic characteristics, school life, emotional well-being, and bullying experiences. Information was collected anonymously, and the self-administered structured questionnaire took approximately 20 minutes to be completed.

Mother and father occupation were recorded and then classified into major professional groups, according to the Portuguese Classification of Occupations - version 2010 (INE, IP 2011). In the analysis we classified parental occupational level based on the parent with the highest occupational level. Then, parental occupation was categorized as “high” (professional and managerial occupations), “middle” (non-manual and manual skilled occupations) and “low” (semi-skilled and unskilled occupations).

Emotional well-being was assessed using four general questions on the frequency of feeling “anxious or stressed”, “angry”, “sad and desperate” and “suicidal ideation”. Possible answers were “never”, “almost never”, “a few times”, “sometimes” and “a lot of times”.

The Bully Scale Survey developed by the Centers for Disease Control and Prevention (CDC) (Hamburger et al. 2011) was administered to assess bullying involvement.

Children with negative answers to all the scales situations (type of victimization and type of aggression) were categorized as “not involved in bullying”; children were considered to be “victim-only” if they answered positively to at least one type of victimization in the frequency “sometimes” and “a lot of times” and “never” in the aggressors scale. They were classified as “aggressor-only” if they reported at least one type of aggression in the frequency “sometimes” and “a lot of times” and “never” in the victims scale; finally, they were categorized as “victim-aggressor” if they reported the occurrence of both situations of being victims and aggressors. This scale allowed us to understand the involvement in bullying, either as only-victim, as only-aggressor or as victim-aggressor, and has been used in other similar studies, showing high consistency between the answers (Cronbach’s alpha = 0.877).

Data analysis

The chi-square test was used to compare proportions. Logistic regression models were fitted to estimate the association of bullying involvement as dependent variable with parental occupation level, school grade retention, missing school, and emotional well-being related feelings, adjusting for the covariates such as gender, parental cohabitation, and grade. A descriptive analysis was conducted using the Statistical Package for the Social Sciences 22.0 software (SPSS®).

Results

This study shows that 34.2% of adolescents reported being involved in bullying (31.1% girls and 37.9% boys); of these, 14.3% of adolescents reported being involved in bullying as victims-only, 8.3% as aggressors-only, and 11.7% as victim-aggressor. Girls reported more frequently being victims of bullying (15.1% vs 13.5%), and boys tended to be more frequently involved as aggressors (11.3% vs 5.9%) and also as victim-aggressor (13.3% vs 10.2%). Sample characteristics are presented in Supplementary Table 1.

Table 1 shows that the frequency of being a victim decreases with age (16.9% vs 16.7% vs 11.9%) while being involved as aggressor (3.6% vs 5.9% vs 7.4%) or victim-aggressor (6.5% vs 12.6% vs 10.6%) increases with age among girls. Girls who lived with only the mother reported being involved more frequently as victims of bullying compared with other forms of involvement (18.4% vs 4.7% vs 11.9%). No significant statistical differences were found in the prevalence of bullying among girls according to grade, parental occupation, school grade retention, or missing school. In boys, while the prevalence of victimization decreases with age (14.8% vs 13.3% vs 12.6%) and school grade (14.3% vs 12.3%), the prevalence of aggression increases with age (6.2% vs 11.1% vs 15.2%) and school grade (9.4% vs 13.9%), and the same trend is observed for involvement as victim-aggressor, for age (9.2% vs 13.3% vs 16.7%) and school grade (11.4% vs 16.4%) (Table 1).

Table 1 Prevalence of bullying according to sociodemographic and school characteristics, by gender
Full size table

Table 2 shows the prevalence of bullying according to well-being related feelings in girls and boys. The frequency of being involved in bullying as victim-only increased, with increasing frequency of feeling “nervous or stressed” (7.6% vs 13.6% vs 22.2%), “angry” (8.5% vs 18.2% vs 21.8%), “sad and desperate” (10.6% vs 25.4% vs 25.0%) and “having suicidal thoughts” (13.8% vs 31.7% vs 26.3%); the same trend is observed for aggressor-only (“nervous or stressed”: 5.8% vs 5.1% vs 6.5%, “angry”: 4.2% vs 4.5% vs 11.1%, “sad and desperate”: 5.2% vs 6.9% vs 8.3%; and “having suicidal thoughts”: 5.8% vs 7.3% vs. 5.3%) and for victim-aggressor (“nervous or stressed”: 3.0% vs 7.3% vs 18.6%, “angry”: 3.9% vs 10.8% vs 20.6%, “sad and desperate”: 8.0% vs 8.3% vs 31.1%; and “having suicidal thoughts”: 8.2% vs 25.6% vs 39.5%), in girls. In boys, the frequency of well-being related feelings increases when they report being victim-only or victim-aggressors, although this pattern is not observed for those involved as aggressor-only (Table 2).

Table 2 Prevalence of bullying according to emotional well-being related feelings, by gender
Full size table

Table 3 shows that independently of age, parental cohabitation, and school grade, involvement in bullying is statistically significantly associated with adverse well-being related feelings, particularly for victims-only and victim-aggressors, in both girls and boys.

Table 3 Adjusted odds ratios* (AOR*) and 95% confidence intervals (95% CI) for the association of bullying involvement with parental occupation level, school grade retention, missing school, and emotional well-being related feelings, by gender
Full size table

Involvement as victims was associated with feeling “nervous or stressed” (AOR = 3.14, 95% CI: 2.01-4.89), “angry” (AOR = 3.46, 95% CI: 2.43–4.92), “sad and desperate” (AOR = 3.74, 95% CI: 2.72–5.15), and “having suicidal thoughts” (AOR = 4.72, 95% CI: 2.91–7.66) in girls. In boys the same trend is found: (AOR = 2.26, 95% CI: 1.57–43.25), “angry” (AOR = 2.77, 95% CI: 1.94–3.96), “sad and desperate” (AOR = 3.81, 95% CI: 2.51–5.79), and “having suicidal thoughts” (AOR = 3.03, 95% CI: 1.39–6.60). Similarly, involvement as victim-aggressors was also associated with feeling “angry” (AOR = 2.18, 95% CI: 1.32–3.58), “sad and desperate” (AOR = 2.06, 95% CI: 1.27–3.35), and “having suicidal thoughts” (AOR = 2.46, 95% CI: 1.09–5.50) in girls. In boys, associations were found for feelings of “angry” (AOR = 1.88, 95% CI: 1.28–2.76) and “having suicidal thoughts” (AOR = 2.73, 95% CI: 1.18–6.30). However, for aggressors-only the point estimates tended to be attenuated when compared to pattern observed for victims-only and victim-aggressors (Table 3).

Table 4 shows the frequency of tactics of victimization and aggression disclosed by adolescents. Some differences were observed between girls and boys in the use of the tactics included in the operationalization of bullying. Boys reported more frequently being victims of tactics such as “called me names” (69.8%), “made fun of me” (61.1%), “said they will do bad things to me” (22.8%), “played jokes on me” (57.1%), “broke my things” (16.1%), “attacked me” (14.4%), and “pushed or shoved me” (30.9%); while girls were more likely to report being victims of tactics as “nobody would talk to me” (13.5%) or “said mean things behind my back” (60.6%), or even “wrote bad things about me” (14.7%). With regard to tactics of aggression, girls reported to use more frequently tactics as “didn’t talk to them” (43.8%) or “said mean things behind their back” (40.8%) (Table 4).

Table 4 Victimization and aggression tactics reported according to gender
Full size table

In our sample, 61.5% of the adolescents involved in bullying have disclosed the episode to someone. Of those, 43.2% reported the episodes to a sibling or other adolescent, 30.2% reported it to their parents, 5.7% to another adult, 5.2% to a teacher, and 2.6% to a school staff.

Discussion

This study showed that bullying behaviours are common among more than one third of Portuguese public school-going adolescents, and main differences among girls and boys are found with regard to type of involvement and forms of aggression used.

Girls were more frequently involved in bullying as victims-only, while boys tended to be more frequently aggressors-only or be involved both as victim and aggressor, which is in line with previous research reporting gender differences in the involvement in bullying (Committee on the Biological and Psychosocial Effects of Peer Victimization et al. 2016; Currie et al. 2012; Finkelhor et al. 2015; Moon et al. 2016). Boys are not necessarily more aggressive but instead, girls are more subtle and more prone to report being victims of relational aggression than boys (Committee on the Biological and Psychosocial Effects of Peer Victimization et al. 2016; Finkelhor et al. 2015), which seems to make this behaviour more visible among boys (Craig and Harel 2004).

With regard to differences in the tactics used, boys report using more direct forms of aggression, expressing their behaviour in more physical ways such as attacking, threatening, or ruining things, while girls report more indirect forms of bullying such as saying mean things behind others’ backs. Additionally, behaviours reported by aggressors-only can be motivated by the desire to increase their higher status and power among peers (Currie et al. 2012) or with stereotypical characteristics of masculinity (Navarro et al. 2011). Traditionally, boys’ aggressive behaviour and violence seems to be reinforced, whereas indirect involvement in bullying is more consistent with traditional roles associated with femininity, justifying the social roots of stereotypical participation of boys and girls in bullying (Silva et al. 2013). These stereotypes reveal the strategies used by girls and boys to achieve an important role in the group and in peer relations (Carrera Fernandez et al. 2013).

According to the literature, bullying behaviours are expected to decrease with age (Moon et al. 2016), and our results show a trend of decreasing in victimization associated with the increase of age in both boys and girls. However, as age increases, the number of aggressors also increases in boys. The transition between school grades contributes to an increase in victims among the younger ones once they are exposed to bullying by the older students (Dijkstra et al. 2008). This might be explained by the sense of power from older boys towards younger peers or even an academic pressure to show dominance, as described in a previous study (Moon et al. 2016).

Girls living only with their mothers reported being victims more frequently, and those who did not live with any parent reported more acts of bullying aggression. Among the hypotheses described in the literature to explain the association between living only with mother and being victim of bullying, it was found that victims of bullying are more frequently girls who develop a negative relationship with the mother (Rigby 1993), adolescents developing emotional problems due to maternal anger, maternal depression, low parent–child involvement, and authoritarian parenting (Curtner-Smith 2000). In fact, family functioning has been linked to different roles in violence (Cunningham et al. 2004). Several studies have reported that participants tend to develop roles as victims or aggressors when they perceive that their family is not well functioning, with a higher level of criticism, fewer rules, greater exposure to intimate partner violence, and more child maltreatment (Holt et al. 2008). Although in our study we did not assess the parent–child relationship, results showed that living only with the mother increased vulnerability to being bullied, while living only with the father decreased the involvement in bullying as victim-only and aggressor-only. However, our sample has a higher number of adolescents who are living with the mother-only (n = 705) compared to those adolescents who are living with the father-only (n = 78), which limits the ability to better explore and understand this last group. The low number of adolescents living with the father-only probably reflects the Portuguese parental regulation, where usually the father only has the right to custody of his children in very specific circumstances.

We found that reporting feeling “nervous or stressed”, “angry”, “sad and desperate” and also “suicidal ideations” increases with the involvement in bullying as victim-aggressor, among both girls and boys and as previously described (Heikkila et al. 2013). In fact, in addition to the physical consequences of the aggression, bullying victimization is also related to somatic symptoms and disturbances, reflecting the emotional effects of aggression. The most common stress-related symptoms include sleep disorders (Hertz et al. 2015), gastrointestinal complaints, headaches, and chronic pain (Beran et al. 2012; Committee on the Biological and Psychosocial Effects of Peer Victimization et al. 2016), and also bed wetting and tummy ache (Kshirsagar et al. 2007). Additionally, a girl victim of bullying is at higher risk of suicidal ideation, feeling more nervous or stressed and angry than a boy (Barzilay et al. 2017). One possible explanation is the fact that girls seem to externalize their emotions better and disclose their depressive feelings more easily than boys (Committee on the Biological and Psychosocial Effects of Peer Victimization et al. 2016; Holt et al. 2008; Renouf and Harter 1990).

Although it is not so clear, adolescents who reported being aggressors-only tended to present a different pattern when compared to the victims-only or those who reported being victim-aggressor. Lower odds of suicidal ideation were observed for those who reported to be aggressors-only, in particular girls. On the oher hand, girls present higher risk of suicidal ideation when involved in bullying as victims-only and as victim-aggressors. Still, being involved in bullying regardless of the type of involvement is a risk factor for suicidal ideation poor mental health (Yen et al. 2015).

Our study showed that friends are the ones with whom adolescents share the involvement in bullying. A small proportion of participants disclosed episodes of bullying to a teacher, and this is in accordance with previous studies that show that more than one-half of victims do not report it to a teacher or their parents, and instead choose their peers to share their experience (Committee on the Biological and Psychosocial Effects of Peer Victimization et al. 2016; Finkelhor et al. 2015). Strategies should be implemented in schools by integrating families and the community to be more alert to signs of bullying episodes, and to create environments of confidence for adolescents to talk about their relational problems.

Study limitations and strengths

This study showed a significant and strong association of bullying tactics and well-being feelings such as suicidal ideation. However due to the cross-sectional design of the study, a temporal sequence and causation cannot be established. Nevertheless, and based on literature, we believe that bullying might be a precursor of mental health consequences such as anxiety, depression, and suicidal ideation (Menesini and Salmivalli 2017).

A significant proportion of adolescents did not participate in this study because they were not at the school on the day of the questionnaire or refused to participate. Differences between excluded and included participants were only observed for gender, with lower participation among boys. Thus, we may expect that the prevalence may be underestimated because boys tend to be more involved in bullying behaviours than girls. Nevertheless, this study provides a picture of bullying involvement in school-going Portuguese adolescents.

Conclusion

A high prevalence of bullying was found in Portuguese public school-going adolescents. Our results showed gender differences in bullying involvement; while girls reported more frequently experiences as victims-only, boys reported more frequently experiences as aggressors-only. Gender differences were also found in the tactics of bullying reported, with boys being more likely to use direct forms of aggression, while girls tend to use more frequently indirect forms.

Adolescents involved in bullying as victim-only or both victim and aggressor were more likely to report suicidal ideation feelings. This result emphasizes the importance of being aware of bullying behaviours that sometimes might be dismissed as normal among school-going adolescents.

Efforts to prevent violent behaviours among adolescents should involve schools, families, and the community. Schools are a privileged setting in which to provide a supportive environment where these adolescents may disclose their problems, and should be involved with families in an early detection of bullying involvement among students and combined efforts to help end these behaviours.