The aim of this study was to investigate the unique prevalence and impact of cyberbullying in adolescence, in comparison to traditional bullying. In this sample of 11–16 year olds, pure cyber-victimization was very rare at around 1% of the total pupil population and 4% of victims of bullying. Cyber-victimization occurred mostly alongside traditional types of school bullying, such as direct and relational bullying. In terms of outcomes, pure cyber-victims had similar outcomes to pure direct victims and pure relational victims. Those who experienced poly-victimization by different means had the lowest self-esteem and most behavioral difficulties.
The finding of few pure cyber-victims found in this UK sample of adolescents is consistent with the low prevalence rates recently reported by other studies that assessed both traditional and cyber-victimization in the USA [6, 11]. Traditional or ‘in-person’ victimization was most prevalent, with almost all victimization being carried out by using direct or relational means. The majority of adolescents who reported experience of cyber-victimization were also victimized via these traditional means, supporting evidence that cyberbullying creates few new victims . In this respect, these findings provide further evidence that cyberbullying is another tool in the toolbox for bullies. It should be seen as an extension of in-person bullying and not the unique or distinct phenomenon which has been portrayed [29, 30].
Regarding the impact upon psychological and psychosocial outcomes, pure cyber-victimization had similar effects as pure direct and pure relational victimization. Thus, any type of victimization is related to poorer psychological outcomes; namely, more behavioral and emotional difficulties and lower self-esteem. Furthermore, in accordance with other findings [17, 18], those who are victimized via multiple forms, in particular via multiple traditional forms (DV & RV, or DV, RV & CV), have especially low self-esteem and high behavioral difficulties.
Why do our and other recent research findings contradict the headlines of an epidemic of cyberbullying and its particular tragic consequences? Firstly, early research on cyberbullying  failed to assess traditional bullying, so effects on self-esteem or behavior were confounded by the most common types of peer victimization that adolescents experienced: direct and relational bullying. Secondly, the prevalence rates reported often over-shadow other important information regarding the participants and definitions used. For example, 94% of the adolescents assessed by Juvonen and Gross  had access to or use of the internet at home, and the 72% prevalence reported in this study was based upon experiencing one or more incidences of online ‘bullying’ in the past year. However, a single incidence of online harassment in one year should not be considered as bullying according to recognized consensus definitions that the aggressive acts have to be repeated .
It is important to understand that bullying occurs in peer relationships and is not an individual characteristics construct such as conduct disorder. Bullying is about exerting dominance and power to attain access to resources . In adolescence, this includes dating and forming romantic relationships and those who are victimized have less romantic success than the bullies . Indeed, bullying is one strategy to reduce intra-sexual competition, i.e., to defame and exclude competitors . Understanding the evolutionary function of bullying requires that the bully is in the same environment and seen to be dominant to obtain access to the resources. Bullies also like to see the effects of bullying, i.e., the suffering of the victim and social isolation . This is achieved by traditional means as shown here, and by also using new electronic means. Thus, cyberbullying on its own is very rare. It is not surprising that a recent review has shown that the risk factors for becoming a bully or victim of traditional and cyberbullying are very similar or identical .
Finally, our findings are consistent with previous reports that females engage more in relational bullying  and are more likely to be cyber-victims . This may be explained by female adolescents spending more time on social media in contact with peers , meaning there is more opportunity, and that cyberbullying is similar in nature to female-dominated relational bullying, i.e., disrupting social relationships rather than confronting the victim directly. Moreover, those of lower socioeconomic status, indicated by pupil premium, were more likely to be victimized consistent with findings of a recent meta-analysis .
This study has a number of strengths. It involved a large sample of adolescents with experience of victimization and used reliable and valid measures to investigate bullying experiences, emotional and behavioral difficulties and self-esteem. Participants were provided with behavioral descriptions for acts of traditional and cyber-victimization, and a stringent criterion was used of including only those frequently or often victimized.
There are also limitations of the study. Firstly, this study focused on comparative frequency of cyber versus traditional victimization and is not representative of the UK as a whole. However, the prevalence and pattern of associations (such as with sex) are highly consistent with other UK-wide research previously reported . Secondly, the nature of the association between victimization type and self-esteem and behavioral difficulties in this cross-sectional study is correlational and we therefore cannot infer any causation from the findings. However, there are now longitudinal and genetically sensitive studies  that have shown that being victimized by peers has adverse effects that are as detrimental as being abused by adults , get under the skin , and last a lifetime [43, 44]. To ascertain the effects of cyberbullying, in particular, future longitudinal research is needed .
To conclude, traditional types of school victimization remain the most frequent type of peer victimization amongst adolescents. Although pure cyber-victimization had similar psychological outcomes to pure direct and relational victims, poly-victims had the highest risk of poor psychological functioning. From a public health perspective, considering the low prevalence of pure cyber-victimization compared to traditional peer victimization, cyber-victimization has only a small unique impact on adolescent mental health; it is an overrated phenomenon. Cyberbullying is another means for traditional bullies to gain dominance and access to resources. Schools must acknowledge and address this issue, despite incidences often occurring outside of the school grounds. However, any bullying prevention and intervention still needs to be primarily directed at combatting traditional bullying while considering cyberbullying as an extension that reaches victims outside the school gate and 24/7.