Why do bullies bully? If we knew the answer to this question, we’d be in a better position to prevent or reduce bullying.  But despite the best efforts of parents, educators, and even members of Congress, bullying is disturbingly common. It occurs not only in schools and on playgrounds but also at work, in politics, and on social media platforms.  And, the negative effects of bullying on victims can last for decades.

A longstanding and highly popular explanation of bullying is known as the compensation model.  According to this model, bullying is motivated by low self-esteem. In other words, bullies make up for their own underlying sense of inadequacy by threatening and intimidating other people. But, there is surprisingly little support for the idea that bullies bully to compensate for low self-esteem.  In fact, bullies tend to score very high on measures of self-esteem. But if bullies have high self-esteem, why do they bully? We’re learning that the answer to this question depends on how one measures self-esteem.

In an earlier study, Christina Salmivalli and her collaborators examined the link between self-esteem and bullying in 8th graders. In this study, they measured self-esteem using both  traditional, self-report measures of self-esteem that were completed by the participants themselves and measures of self-esteem that were completed by other students who knew the participants.  In addition, the researchers measured a construct known as “defensive egotism” by asking other students to report whether fellow students had an inflated sense of themselves, always wanted to be the center of attention, and had trouble accepting criticism from others.

The results showed that defensive egotism, as rated by other students, was a good predictor of bullying. In contrast, both self-rated self-esteem and peer-rated self-esteem were poor predictors of bullying. In other words, bullying was related to being egotistical, hungry for attention, and defensive but not to having low self-esteem.

Recently, my colleagues and I examined whether this pattern of results applies equally to different kinds of bullies. Research has revealed that there are at least two kinds of bullies – “pure bullies” and “bully-victims.” Pure bullies are rated high in bullying by their classmates but are rated low in victimization. That is, they bully other students but are not themselves the victim of bullying. These pure bullies tend to be socially skilled and use bullying to get what they want.

In contrast, bully-victims are rated high in both bullying and victimization. They bully other students and are also bullied themselves. Unlike pure bullies, bully-victims tend to have social difficulties, and they seem to bully other students when they feel that they are being bullied or mistreated. In our research, we reasoned that because the bullying of pure bullies is proactive, their bullying should be related to having an inflated sense of self, wanting to be the center of attention, and defensive reactions to criticism. That is, pure bullies should be driven by defensive egotism.  However, we thought that bully-victims would bully other people because they had been bullied.   

Our study involved kids from two countries – a group of 8th and 9th graders at a private school in urban India, and two groups of 5th graders at a public school in a rural area of the southern United States  (Arkansas). In all three groups, we found that both self-rated and peer-rated self-esteem were only weakly related to bullying. In contrast, defensive egotism was strongly related to bullying in all three samples.  Defensive egotism – not low self-esteem – seems to underlie bullying.

To our surprise, defensive egotism predicted both kinds of bullying in all three samples.  Thus, our results showed that defensive egotism contributes to bullying, and this finding holds up well for both kinds of bullies.

So does self-esteem have nothing to do with bullying? This might be a premature conclusion.  In our study, self-esteem, whether rated by students themselves or by others who knew them, was negatively associated with being victimized. Kids who were high in self-esteem got picked on less than kids low in self-esteem. Furthermore, kids who were high in self-esteem reported that they defended  the victims of bullying more than kids low in self-esteem did. So, self-esteem does seem to play a role in bullying, but not in the way that people often assume.  

When we combine our own findings with other research on effective interventions for bullying, we propose a three-pronged approach to bullying prevention. We propose that parents and educators should:

1. be attuned to the defensive egotism of bullies rather than their level of self-esteem. Knowing who is at risk to bully is a first step in many interventions, and the risk factor of defensive egotism is apparent to their peers (though it might be denied by bullies themselves).

2. increase the self-esteem of victims of bullying (and avoid “blaming the victim”).

3. promote and maintain norms that increase the willingness and ability of bystanders to speak up, if not intervene,  when others are bullied.

Finally, interventions for bullying might work best if they are implemented at both the social level (creating new norms and rules) and at the individual level (offering help to specific kids). This is equally true for bullies, victims, and bystanders. Many unanswered questions about bullying remain.  But in the past decade or two, a growing army of researchers has begun to explore creative ways to combat this social problem.


For Further Reading

Salmivalli, C., Kaukiainen, A., Kaistaniemi, L., & Lagerspetz, K. (1999). Self-evaluated self-esteem, peer-evaluated self-esteem, and defensive egotism as predictors of adolescents’ participation in bullying situations. Personality and Social Psychology Bulletin, 25, 1268-1278.

Simon, J.B., Nail, P.R., Swindle, T., Bihm, E.M., & Joshi, K. (2017). Defensive egotism and self-esteem: A cross-cultural examination of the dynamics of bullying in middle school. Self and Identity, 16:3, 270-297, DOI: 10.1080/15298868.2016.1232660

The authors include a social psychologist, a school psychologist, and a counseling psychologist. All are faculty members in the Department of Psychology & Counseling at the University of Central Arkansas.