Addressing bullying in schools: theory and practice

Foreword | Bullying in schools has become an issue of major importance among educators. However, interventions to reduce bullying have enjoyed only modest and limited success. This paper examines five different explanations for bullying. These emphasise, respectively:

  1. developmental features;
  2. individual differences;
  3. a sociocultural perspective;
  4. group and peer pressure; and
  5. the rationale for restorative justice.

Each has had some impact on school policies and practices. The strengths and limitations of the different explanations are examined. It is concluded that none of them provides a comprehensive explanation for school-based bullying and that their value lies especially in suggesting actions that may be taken by schools in addressing particular bully/victim problems.

Adam Graycar


There is now widespread support for the view that schools should take action to counter bullying among students. Social researchers, beginning with Olweus' (1993) pioneering studies among boys in Scandinavia in the 1970s and 80s, led the way. In time, others followed, especially in parts of Europe and the United Kingdom, and then eventually in the 1990s in Australia (Rigby & Slee 1993). Educators began to ask how much bullying is actually going on in our schools, what kinds of bullying are occurring, and what harm is it doing.

Evidence has steadily accumulated about the negative consequences of student involvement in bully/victim problems. On the basis of longitudinal studies, it has been concluded that repeated exposure to being bullied can, and indeed often does, undermine the health and wellbeing of vulnerable students (Egan & Perry 1998; Rigby 1999). It is also known that the perpetrators of bullying not only tend to experience depression and engage in suicidal thinking but also, if not corrected at school, are more likely to act violently as adults in the home and workplace (Farrington 1993). Children who are both bullies and victims are seen as especially prone to mental illness.

Social science research findings often make little impact on what goes on in society, unless there is a social movement that actively seeks to make use of them. In the late twentieth century in Western society in particular, the zeitgeist was one in which people were becoming increasingly attuned to the injustice of social discrimination, especially on gender and racial grounds. Bullying was, and still is in many minds, bracketed with harassment whose roots are in social prejudice. Currently, educational jurisdictions in Australia see their role as providing information to schools about bullying and advice on how they could counter it. Some of them have mandated that all schools have anti-bullying policies.

The impact so far

It has proved difficult to establish whether addressing bullying in schools has, in fact, resulted in actual reductions in bullying incidents. First, there are formidable problems associated with sampling. Selected schools must be representative of the general population of schools and assessments must be made on at least two occasions over a span of years. Identical questions must be asked at different time points. Unfortunately, it is quite possible that the meaning that bullying has for respondents may change over the years, probably in the direction of respondents including more phenomena in their identifying of bullying, especially non-obvious, indirect forms of bullying such as exclusion. Given these difficulties there appears to be only one published study, conducted in England, that has attempted to estimate changes over time, not associated with specified interventions (Smith & Shu 2000). They suggest that there has been a slight tendency towards a reduction in reported bullying.

Where specific intervention programs have been evaluated, the outcomes have been mixed. A program implemented in Norway, devised by Olweus (1993) claimed a 50 per cent reduction in reported bullying in the Bergen area. However, a further investigation conducted by Roland (1993) in Rogaland, another part of Norway where the same program was used, provided negative results. In a meta-evaluation report on 12 intervention programs carried out in a number of countries (namely, Norway, England, Spain, Canada, Belgium, Switzerland, Finland and Australia), it was concluded that in general the programs produced modest reductions in the proportions of children being bullied by peers, and these improvements tended to be limited to kindergartens and primary schools (Rigby 2002a). At the same time, more positive results were obtained in schools where programs had been more thoroughly implemented.

Do we have a road map?

There is now a great deal of instructive material on how to reach the desired goal of no bullying. Unfortunately, it is less of a road map and more like a chart used by the early voyagers, with some areas labelled terra incognito. The instructions often tend to be somewhat vague or capable of various interpretations. They include:

  • adopt "a whole school approach";
  • develop an appropriate anti-bullying policy;
  • ensure that effective strategies are in place applying both preventative and interventive measures.

Schools are enjoined to employ a "systemic" or "holistic" approach and to build a "positive school ethos". At this general level, there appears to be little disagreement. However, when we look closer, there are alternative perspectives on the problem that imply different pathways.

Theoretical perspectives on understanding bullying and their implications

1. Developmental theory

Some explanations of bullying draw upon an understanding of child development. They point out that bullying begins in early childhood when individuals begin to assert themselves at the expense of others in order to establish their social dominance. They tend at first to do so crudely, for instance by hitting out at others, especially those less powerful than themselves, in an attempt to intimidate them. But as Hawley (1999) points out, as children develop they begin to employ less socially reprehensible ways of dominating others. Verbal and indirect forms of bullying become more common than physical forms. In time, the kind of behaviour that is generally labelled as "bullying" becomes relatively rare. Consistent with this view is evidence that physical bullying is much more common in early childhood than later, and that what is identified as bullying gradually becomes less and less apparent as children become older (Smith & Sharp 1994). However, as a comprehensive explanation of bullying this view fails to take into account that although there is a general diminution in reported victimisation over time, the trend is temporarily reversed when children move from primary to secondary school and find themselves in a new environment which is less benign (Rigby 1996). Clearly, social environmental factors must also be taken into account. Nevertheless the developmental perspective is useful in providing guidance as to how bully/victim problems can be tackled. For example, older children are thought to be more likely to respond positively to problem-solving approaches which require a more sophisticated appreciation of the options available to them (Stevens et al. 2000).

2. Attributions to individual differences

Broad explanations in terms of developmental processes and environmental influences fail to take into account individual differences between people that may lead to interactions that result in one person bullying another. For example, children who repeatedly bully others at school tend to be low in empathic regard for others and inclined towards psychoticism (Slee & Rigby 1993). Children who are frequently targeted as victims at school are inclined to be psychologically introverted, to have low self-esteem and lack social skills, especially in the area of assertiveness (Rigby 2002b). How such qualities arise has been subject to considerable debate. Currently, it is generally acknowledged that genetic influences play a part and these may interact with adverse social conditions to which children may be exposed. For example, dysfunctional family life in which children do not feel loved and/or feel over-controlled by parents can lead to them acting aggressively at school (Rigby 1994), especially if the school ethos does nor discourage aggressive behaviour.

There are limitations in this approach. In some relatively benign environments introverted children with low self-esteem are not bullied; being aggressive and generally unempathic does not invariably lead a child to bully others. There is, for example, evidence that bullying is relatively rare in Steiner schools, which provide a highly supportive social environment and respect for individual differences (Rivers & Soutter 1996). Moreover, individuals who are dissimilar in personality may belong to the same sociocultural group and seek collectively to impose on those they regard as outsiders.

Acknowledgment of the role of individual differences in making bullying possible has led some schools to introduce programs that can assist vulnerable children to defend themselves more effectively, for instance through developing better social skills and learning how to act more assertively. Anger management programs have also been developed to help children who are prone to act aggressively to control their aggression.

3. Bullying as a sociocultural phenomenon

A further perspective seeks to explain bullying as an outcome of the existence of specified social groups with different levels of power. The focus is typically on differences which have a historical and cultural basis, such as gender, race or ethnicity and social class. Major emphasis has been placed upon differences associated with gender. Society is seen as essentially patriarchal. Males are seen as generally having more power than females as a consequence of societal beliefs that males should be the dominant sex. In order to maintain their dominance, boys feel justified in oppressing girls. Numerous studies have, in fact, indicated that boys are more likely than girls to initiate bullying (Olweus 1993; Smith & Sharp 1994). Moreover, it is clear that boys are more likely to bully girls than vice versa. For example, in a large-scale Australian study of some 38,000 children (Rigby 1997) a much higher proportion of girls claimed to be bullied exclusively by boys (22.1 per cent) than boys reporting being bullied only by girls (3.4 per cent). With cross-gender bullying it is clearly mostly one-way traffic, and this may derive, in part, from the way in which some boys have come to think about how they should behave in the company of girls.

The process according to which boys come to develop characteristics which lead to them engaging in oppressive behaviour is sometimes described as "the construction of hegemonic masculinity" (Connell 1995; Gilbert & Gilbert 1998). This is held not only to account largely for boys bullying girls, but also for boys bullying boys who do not possess stereotypical masculine qualities. Such children are commonly referred to as "gay" and may include children whose sexual orientation is homosexual. The use of language with sexual connotations to insult children regarded as "gay" is certainly widely prevalent in schools (Duncan 1999), although the extent to which it occurs has surprisingly not, as yet, been investigated. Explaining the bullying of girls by girls can invoke the notion of the construction of femininity, with girls deviating from an idealised conception of what it is to be feminine being more readily targeted.

It is sometimes claimed that bullying tends to be associated with racial or ethnic divides. It is argued that some ethnic groups are more powerful than others whom they seek to dominate. Typically, the less powerful are the victims of colonialism. For example, Indigenous communities in Australia in the late eighteenth century were subjected to British colonialism. Aboriginal people were seen by many as inferior - and this perception still lingers in the minds of people who retain racist beliefs. Through a process of cultural transmission, non-Indigenous children may feel justified in bullying their Aboriginal peers. Evidence from Australian studies suggests that indeed Aboriginal students are more likely than other students to be the recipients of verbal abuse (Rigby 2002b). However, some studies conducted outside Australia have not found that race or ethnicity is significantly associated with peer victimisation (for example, Junger-Tas 1999; Losel & Bliesener 1999). Despite claims that children are at risk of being bullied at school by peers of a higher social class, research evidence is not supportive (Duyme 1990; Olweus 1993; Ortega & Mora-Merchan 1999; Almeida 1999).

The sociocultural perspective on bullying can have striking implications for how a school approaches the problem of bullying. Attention is directed towards how the school curriculum in its broadest sense can influence children to accept and respect sociocultural differences. It is suggested that not only should the school curriculum explicitly and directly address issues related to differences in gender, race or ethnicity and social class in order to counter prejudice and discrimination, but importantly the mode of delivery of the curricula should indirectly address bullying through the stimulus it provides to cooperative problem-solving, emotional sensitivity and independent critical thinking. The Australian national web site on bullying in schools (, based mainly on a sociocultural approach to bullying, has placed primary emphasis upon this approach. Some writers embracing a sociocultural perspective in which gender considerations are pre-eminent, have suggested that schools need to abandon their current emphases upon "rationality", which is characteristic of masculinity, in favour of exploring with students their expressive and emotional worlds (Kenway & Fitzclarence 1997). The use of strict codes of behaviour governing bullying and the use of counselling methods to deal with individual cases are equally abhorred. Both are seen as based on an underlying faith in rationality and, as such, essentially counterproductive. This view emphasises the use of the school curriculum as a means of developing emotional understanding and positive interpersonal relations rather than controlling undesirable behaviour through the use of negative sanctions and/or counselling methods that impose authoritarian solutions to bully/victim problems.

4. Bullying as a response to group and peer pressures within the school

This approach has something in common with the sociocultural approach in that it conceives bullying as understandable in a social context. However, the context is not defined according to sociocultural categories such as gender, race and class. There is first a broad social context consisting of the behaviours and attitudes of members of the entire school community. Individuals are seen as influenced to a degree by their perceptions of what may be called the school ethos, and student welfare polices may be systematically directed towards its improvement (Soutter & McKenzie 2000). Secondly, students are powerfully influenced by a smaller group of peers with whom they have relatively close association. Such groups are typically formed within a school on the basis of common interests and purposes, and provide support for group members. They may also constitute a threat to outsiders, sometimes to ex-members, whom they may bully. Situations commonly arise in a school whereby children are members of, and supported by, a group that is, in some situations, more powerful than an individual or smaller group that they wish to bully in some way. The motive may be a grievance or imagined grievance, a prejudice (explicable in sociocultural terms) or simply a desire to have fun at the expense of another person. Importantly, the acts of bullying are seen as typically sustained by a connection with a group rather than by individual motives such as personal malevolence. This view presupposes that bullying is typically a group phenomenon. Early studies of bullying in Scandinavia adopted the term "mobbing" suggesting that children are bullied by mobs (Olweus 1993). While this may sometimes occur, more commonly the bullying is carried out by one or two people with the passive support of others (Pepler & Craig 1995). When students are asked whether they have bullied others as individuals or as members of a group, among those who have bullied others about half admit to bullying alone; others say they have acted as part of a group (Rigby 2002b).

The implications for schools are that they must be aware of the roles played by groups as distinct from individuals. They need to identify groups and work with them. Several methods have been devised for working with groups of children who have bullied or are suspected of bullying others. One, the "no blame approach" (Maines & Robinson 1992), involves a teacher or counsellor meeting with the group of children identified as having bullied someone, in the company of some other children. The teacher describes to the group the suffering that has been endured by the victim, and the group is expected to consider ways in which the situation can be improved. The "non-bullies" in the group are expected to exert positive peer pressure, that is, influence the "bullies" to act more benevolently towards the victim.

An alternative method, generally used with older children, called the "method of shared concern" (Pikas 2002) involves working initially with individuals suspected of being in a group that is bullying someone. The teacher's aim here is to communicate his/her concern for the victim and invite (and then monitor) responsible individual action - and in so doing to lessen the influence the group may have on each individual's actions.

5. Bullying from the perspective of restorative justice

This perspective recognises that some children are more likely than others to be involved in bully/victim problems as a consequence of the kind of character they have developed. Children who bully others typically feel little or no pride in their school and are not well integrated into the community (Morrison 2002). They mishandle their emotional reactions to the distress they cause by not experiencing appropriate feelings of shame; in fact, they tend to attribute unworthy characteristics to those they victimise. By contrast, victims are prone to experience too much inappropriate shame. To some extent, this perspective is one that emphasises individual differences, as in (2) above. But in addition, an important role is ascribed to the school community and to significant people who are implicated in the problem. These can include family and friends of both bullies and victims; that is, significant others who care about them. It is believed that appropriate feelings of shame can and should be engendered in those who bully others through exposing them to condemnation by those they have offended. This, it is thought, can be done constructively in the presence of those whom they care about and who care for them. Success is seen as greatly dependent on the support provided by those who care about the perpetrator as a person and the readiness of the community to forgive and provide sincere acceptance (Morrison 2002). This approach is concerned with "violations against people" and the restoration of positive relationships rather than applying punishment for breaking rules (Cameron & Thorsborne 2001).

Some schools have applied the ideas of restorative justice in a preventative way through a Responsible Citizens Program that encourages students to develop relationships with their peers that are characterised by respect and consideration (Ahmed et al. 2001). Students have been helped through role-playing to resolve conflicts with peers and identify and manage inappropriate feelings of shame. There is some evidence that the program can increase students' feelings of safety and the use of more adaptive means of shame management (Morrison 2002). No reports, however, have yet been received on whether the incidence of bullying has been reduced using this program. When serious cases of bullying occur, they may be resolved through the use of a community conference in which victims are encouraged to express their sense of hurt while perpetrators listen, become contrite, and agree to compensate the victim.


Given that programs designed to reduce peer victimisation have met with only limited success, it appropriate to examine the range of explanations offered to account for bullying in schools and their implications for school policies and practices. Five such explanations have been examined and each shown to make some contribution to our understanding of bully/victim problems in schools. Each has had some impact on what Australian schools are doing to counter bullying in Australia. (Rigby 2001; Morrison 2002; Rigby & Thomas 2002). However, no single view is sufficiently comprehensive in providing a definitive answer as to what is "best practice". Therefore, in applying anti-bullying policies and procedures, schools should consider the strengths and limitations of each suggested approach and the appropriateness of its application to particular bully/victim problems.


  • Ahmed, E., Harris, N., Braithwaite, J. & Braithwaite, V. 2001, Shame Management Through Reintegration, Cambridge University Press, Melbourne.
  • Almeida, A. 1999, "Portugal", in P.K. Smith, Y. Morita, J. Junger-Tas, D. Olweus, R. Catalano & P.T. Slee (eds), The Nature of School Bullying: A Cross-national Perspective, Routledge, London, pp. 174-86.
  • Cameron, L. & Thorsborne, M. 2001, "Restorative justice and school discipline: Mutually exclusive?" in J. Braithwaite & H. Strang (eds), Restorative Justice and Civil Society, Cambridge University Press, Cambridge, pp. 180-94.
  • Connell, R 1995, Masculinities, Allen and Unwin, Sydney.
  • Duncan, N. 1999, Sexual Bullying: Gender Conflict and Pupil Culture in Secondary Schools, Routledge, London.
  • Duyme, M.D. 1990, "Antisocial behaviour and postnatal environment: A French adoption study", Journal of Child Psychology and Psychiatry and Allied Disciplines, vol. 31, no. 5, July, pp. 699-71.
  • Egan, S.K. & Perry, D.G. 1998, "Does low self-regard invite victimisation?" Developmental Psychology, vol. 34, pp. 199-309.
  • Farrington, D.P. 1993, "Understanding and preventing bullying", in M. Tonny & N. Morris (eds), Crime and Justice, vol. 17, University of Chicago Press, Chicago.
  • Gilbert, R. & Gilbert, P. 1998, Masculinity Goes to School, Allen and Unwin, Sydney.
  • Hawley, P.H. 1999, "The ontogenesis of social dominance: A strategy-based evolutionary perspective", Developmental Review, vol. 19, pp. 97-132.
  • Junger-Tas, J. 1999, "The Netherlands", in P.K. Smith, Y. Morita, J. Junger-Tas, D. Olweus, R. Catalano & P.T. Slee (eds), The Nature of School Bullying: A Cross-national Perspective, Routledge, London, pp. 205-23.
  • Kenway, J. & Fitzclarence, L. 1997, Gender and Education, vol. 9, no. 1, pp. 117-33.
  • Losel, F. & Bliesener, T. 1999, "Germany", in P.K. Smith, Y. Morita, J. Junger-Tas, D. Olweus, R. Catalano & P.T. Slee (eds), The Nature of School Bullying: A Cross-national Perspective, Routledge, London, pp. 224-49.
  • Maines, B. & Robinson, G. 1992, Michael's Story: The "No Blame" Approach, Lame Duck Publishing, Bristol.
  • Morrison, B. 2002, "Bullying and victimisation in schools: A restorative justice approach", Trends and Issues in Crime and Criminal Justice, no. 219, Australian Institute of Criminology, Canberra.
  • Olweus, D. 1993, Bullying in Schools, Blackwell, Cambridge, Massachusetts.
  • Ortega, R. & Mora-Merchan, J.A. 1999, "Spain", in P.K. Smith, Y. Morita, J. Junger-Tas, D. Olweus, R. Catalano & P.T. Slee (eds), The Nature of School Bullying: A Cross-national Perspective, Routledge, London, pp. 157-73.
  • Pepler, D.J. & Craig, W.M. 1995, "A peek behind the fence: Naturalistic observations of aggressive children with remote audiovisual recording", Developmental Psychology, vol. 31, no. 4, pp. 548-53.
  • Pikas, A. 2002, "New developments of Shared Concern Method", School Psychology International, vol. 23, pp. 307-26.
  • Rigby, K. 2002a, A Meta-evaluation of Methods and Approaches to Reducing Bullying in Preschools and in Early Primary School in Australia, Commonwealth Attorney-General's Department, Canberra.
  • --- 2002b, New Perspectives on Bullying, Jessica Kingsley, London.
  • --- 2001, Stop the Bullying: A Handbook for Schools, Australian Council for Educational Research, Melbourne.
  • --- 1999, "Peer victimisation at school and the health of secondary students", British Journal of Educational Psychology, vol. 22, no. 2, pp. 28-34.
  • --- 1997, Manual for the Peer Relations Questionnaire, The Professional Reading Guide, PRQ, Point Lonsdale, Victoria.
  • --- 1996, "Peer victimisation and the structure of primary and secondary schooling", Primary Focus, vol. 10, no. 7, October, pp. 4-5.
  • --- 1994, "Psycho-social functioning in families of Australian adolescent schoolchildren involved in bully/victim problems", Journal of Family Therapy, vol. 16, no. 2, pp. 173-89.
  • Rigby, K. & Barrington Thomas, E. 2002, How Australian Schools are Responding to the Problem of Peer Victimisation in Schools, Criminology Research Council, Canberra.
  • Rigby, K. & Slee, P.T. 1993, "Dimensions of interpersonal relating among Australian school children and their implications for psychological wellbeing", Journal of Social Psychology, vol. 1331, pp. 33-42.
  • Rivers, I. & Soutter, A. 1996, "Bullying and the Steiner school ethos", School Psychology International, vol. 17, pp. 359-77.
  • Roland, E. 1993, "A developing tradition in research and management", in D.P. Tattum (ed.), Understanding and Managing Bullying, Heinemann Education, Oxford, pp. 15-30.
  • Slee, P.T. & Rigby, K. 1993, "The relationship of Eysenck's personality factors and self-esteem to bully/victim behaviour in Australian school boys", Personality and Individual Differences, vol. 14, pp. 371-73.
  • Smith, P.K. & Sharp, S. (eds) 1994, School Bullying: Insights and Perspectives, Routledge, London.
  • Smith, P.K. & Shu, S. 2000, "What good schools can do about bullying: Findings from a survey in English schools after a decade of research and action", Childhood, vol. 7, pp. 193-212.
  • Soutter, A. & McKenzie, A. 2000, "The use and effects of anti-bullying and anti-harassment policies in Australian schools", School Psychology International, vol. 21, pp. 96-105.
  • Stevens, V., de Bourdeaudhuij, I. & Van Oost, P. 2000, "Bullying in Flemish schools: An evaluation of anti-bullying interventions in primary and secondary schools", British Journal of Educational Psychology, vol. 70, pp. 195-210.

About the authors

Dr Ken Rigby is an Adjunct Associate Professor in the Division of Education, Arts and Social Sciences at the University of South Australia


This paper is taken from the report of research undertaken with the assistance of a grant from the Criminology Research Council.

Download final report: How Australian schools are responding to the problem of peer victimisation in schools