January 21, 2015

Kohm, A. (2015, in press). Childhood bullying and social dilemmas. Aggressive Behaviour.

This is the first blog post of the new year, and comes from one of my current undergraduate dissertation students Jennifer Burns. 

Research shows that as well as the bully and victim in a bullying scenario, other children will also be present, who are known as ‘bystanders’. These bystanders, despite sometimes wanting to, rarely step in and defend the victim of bullying behaviours. Rather, they often withdraw from the scene or assist in bullying. Therefore, this study set out to identify factors which may explain these behaviours of bystander children in bullying contexts.

Previous research has shown that both individual and group factors will influence how children behave in bullying situations, and so this study looked at both individual attitudes to bullying and group norms regarding bullying. Group norms are the expected standard of behaviour within a group, whilst attitudes are students moral beliefs regarding bullying.

Additionally, this study applied Social Dilemma theory to school based bullying. Social Dilemmas are considered situations where an individual must make a decision based on the outcomes for themselves, as they do not believe that others will support or join them in their choice of behaviour. So, a child may not want to defend another if they feel that, unless other children help, it is a waste of time.

This study therefore predicted that children with anti-bullying attitudes and anti-bullying group norms would be less likely to bully and more likely to defend. Conversely, the greater the social dilemma reported, the more likely they would be to withdraw or assist bullying.

A total of 292 students aged 11-14 years, from a residential school in the USA participated. There were roughly equal numbers of male and female participating students. All children stayed in residences with approximately 10-12 other children of the same sex.

The children completed an online survey that related to bullying experiences within their school residences. The children accessed these online at school. The measure first asked them to report how their fellow housemates usually behave in bullying situations. They then completed sections which asked about their attitudes towards bullying, revealing either anti-bullying or pro-bullying attitudes, and subsequently about the group norms regarding bullying within their residence. The final section asked about the social dilemmas that the children face for physical, verbal and relational bullying. 

Nearly all students reported witnessing bullying in their residences; most common was verbal or relational bullying. Over half of students said that standing up for the victim by themselves would be dangerous and ineffective. In contrast, nearly half of students reported that group efforts to defend the victim are safer and more effective. Very few students believed their housemates would support them if they defended a victim. Individual’s attitudes, reported group norms, and perception of social dilemma each contributed to predicting how children react to bullying. Higher perception of a social dilemma was related to higher bullying and withdrawing behaviour. This study therefore advanced insight into group factors that are associated with, and appear important to bystander behaviour.

January 30, 2014

Do peer- and self-reports of bullying predict the same forms of later adjustment?

Scholte, R. H. J., Burk, W. J., & Overbeek, G. (2013). Divergence in self- and peer-reported victimization and its association to concurrent and prospective adjustment. Journal of Youth and Adolescence, 42, 1789-1800. 

Two methods commonly used by researchers for measuring bullying in schools are self-reports and peer-reports. Self-reports ask a child a series of questions about their bullying experiences. Peer-reports instead ask children nominate members of their peer group who they think are bullied. While both measure are often used in research they are not often used together. The information each measure generates can contradict the other, with each measure often pointing to separate individuals as victims of bullying. Both measures have benefits and drawbacks. Self-reports allow researchers a better insight into bullying activities that may occur outside of school, and therefore are not being measured by peer-reports, however they are also more sensitive to lying, bias and misinformation. Peer-reports may give a more realistic and objective account of bulling activities within school, however they are less sensitive and may miss information that could identify a child as bullied.

This study combined the strengths of peer- and self-reports to generate more information on bullying victims. Using these measures the authors wanted to see if they could find different kinds of bullying victims, and how well these different victim types adapted emotionally and socially during, and after, being bullied.

A total of 1,346 adolescents from 23 different schools took part in the experiment. Just over half were female and the majority were of Dutch ethnicity. The average age was 14.2 years at the beginning of the year-long study.

Students were given questionnaires to measure self- and peer-reports of victimization. A year later these measure were given to the same participants again, this allowed the experimenters to see how the participants were adapting to bullying after a period of time.

Identification of victim TypesFour different types of victims were found by using self- and peer-reports. Self-peer victims, identified by high scores on both self- and peer-measures of bullying; self-identified victims, identified by high self-report scores of victimization; peer-identified victims, characterized by high scores on peer-measures; and non-victims, individuals who had low peer- and self-report measures.

Differences between victim types on present future adjustmentSelf-peer victims were found to have the highest levels of peer rejection and the highest amounts of loneliness. Peer-identified victims had the highest levels of peer rejection, but also one of the highest levels of self-esteem along with non-victims.Both self-victims and non-victims had more reciprocal friendships compared with peer-victims and self-peer victims, and the quality of these friendships did not make a difference in distinguishing the different groups.

Looking at these results it can be seen that those who were identified by their peers as being bullied had more difficulty adapting socially, where as victims identified by only themselves struggled to adapt during, and after the bullying, emotionally. All four victim types were found to be stable over a year, meaning if a participant was classified as a self-peer identified victim they were likely to still be classified as a self-peer identified victim a year later.

Scholte and Burk’s research demonstrates how by combining old knowledge, and creating newer measures, we are better able to form a deeper and more detailed level of understanding what bullying is and how its use affects others.

November 05, 2013

Relational aggression, victimization, adjustment and the hostile attribution bias

This blog post was written by Jenna Anderson, and is (possibly) the final one to be posted from my current group of dissertation students. Thanks to Jenna for her hard work, and thanks to all the other students who have done similar posts recently.

Ostrov,J. M., & Godleski, S. A. (2013). Relational aggression, victimization, and adjustment during middle childhood. Development and Psychopathology, 25, 801 – 815. [[NB web link here is to full-text of the article]]

This study looked at the direct and indirect links between relational aggression and relational victimisation. Relational aggression refers to behaviours such as excluding individuals or spreading rumours, therefore providing a direct link to similar behaviours in victims of bullying.

Social Process Model
In order to establish the link between relational aggression and relational victimisation, Ostrov and Godleski focused on the Social Process Model. This outlines that aggressive behaviour can lead on to victimisation (this is the direct link), however, it can also be indirectly linked, in which aggression is said to lead to peer rejection, which then leads on to bullying.

Predicted Mediators
Within this study, the authors predicted several factors which they hypothesised would have an indirect effect on the link between relational aggression and relational victimisation.

The first of these factors was loneliness. They highlighted through the use of previous research that loneliness during adolescence can lead to a variety of different problems, such as aggression. Research was also presented which highlighted the link between loneliness and peer victimisation, suggesting that individuals who experience loneliness may be more likely to become victims of bullying. They suggested that they had found enough evidence to justify testing for an indirect link between relational aggression, loneliness and relational victimisation.

The second mediator suggested by the authors was the role of Hostile Attribution Biases (HAB). This relates to when an individual interprets cues from their environment in a negative, hostile way. The research presented suggests that individuals who display relational aggression are more likely to display HAB towards relational provocation, such as excluding someone. This suggests that there may be an indirect link between HAB, aggression and victimisation.

The final mediator suggested was depressive symptoms. They found research which suggests that internalising problems (This refers to when an individual conveys their problems through processes which affect themselves but do not affect others), a core feature of depression, can lead to relational aggression in adolescents. They hypothesised that children who experience symptoms of depression may be more likely to be victimised, which then highlights the link between relational aggression, relational victimisation and depressive symptoms.

Role of Gender
The key gender difference highlighted within this genre of research is that boys are more likely to display physical aggression than relational aggression in comparison with girls, who are more likely to display relational aggression than physical aggression.

There were 1035 participants in the stud, of whom 522 were girls. The average age of the participants was 8 years and 4 months old.

Several methods were used to collect the data. Teacher reports were used to collect data for relational aggression and relational victimisation. A self-report method was used to collect data relating to HAB. A loneliness and dissatisfaction questionnaire was used to measure the loneliness mediator. In order to measure depressive symptoms, a self-report measure was used, in which the children report how they are feeling and in order to obtain data relating to physical aggression, teacher reports were used.


The results of this study highlighted a direct link between relational aggression, and the likelihood that this would lead on to future relational victimisation. It was found that, of the three mediating factors, only loneliness was said to affect the direct link between aggression and victimisation. Despite this, all three mediators were found to be associated with each other, supporting previous research. Interestingly, the researchers also found that the link between relational aggression and relational victimisation could be reversed, suggesting that not only is aggression likely to lead to future victimisation, but victimisation is also likely to lead to future aggression. This study provided support for the direct and indirect aspects of the Social Process Model when applied to relational aggression.