Making an Impact on School Bullying

Interventions and Recommendations

Peter K. Smith

• Publisher: Routledge
• Year: 2019
• ISBN:
  9781351201933
• Edition: 1
• Topic: Psychologie
• Subtopic: Histoire et théorie en psychologie

Exploring international and intercultural perspectives, Making an Impact on School Bullying presents a much-needed insight into the serious problem of bullying in schools. As the effect of bullying on victims can be devastating, and bystanders and even perpetrators are often also negatively affected by the experience, finding successful solutions to the problem of bullying is crucial for improving school life around the world.

This invaluable book looks at a range of practical interventions that have addressed the problem of school bullying. Peter Smith presents a curated collection of seven examples of successful anti-bullying procedures from around the world - including the US, Europe and Asia - and an exploration of cyberbullying. Each chapter examines the context in which the interventions took place, how theoretical knowledge transferred into practice, and the impact and legacy of the work. Covering the most important and widely-used strategies to combat bullying, the book provides readers with a roadmap to developing practical and impactful interventions.

Ideal reading for students and researchers of education and developmental psychology, Making an Impact on School Bullying is also useful for school counsellors and education authorities.

1

INTRODUCTION

Peter K. Smith

Bullying in school certainly has a long history in terms of occurrence, but a relatively short history of research. It has been recognized as an issue since the nineteenth century, as for example in Tom Brown’s Schooldays (Hughes, 1857) in England. But it has only really been studied and better understood since the 1970s. This chapter gives a background to the nature of school bullying, and its causes and consequences. It then surveys the kinds of interventions made to reduce school bullying, and some of the main issues thrown up by these, as a prelude to the eight chapters on specific programmes or topics which follow. A final concluding chapter examines some of the themes coming from these various intervention efforts, and makes some recommendations for future work.

What is bullying?

Bullying is one form of aggressive behaviour. As such, it is taken as intentionally done to hurt another person. The standard definition of bullying, especially in the school context, comes from the work of Dan Olweus, and is discussed by him in Chapter 2. As he makes clear, the distinctive features of bullying are that it is repetitive, and that there is a power imbalance such that it is difficult for the victim to defend himself or herself.

These criteria are widely accepted, but there remain some issues. The definition agreed by the US Centers for Disease Control and Prevention (Gladden et al., 2014; and see Chapter 5) is:

Bullying is any unwanted aggressive behaviour(s) by another youth or group of youths who are not siblings or current dating partners that involves an observed or perceived power imbalance and is repeated multiple times or is highly likely to be repeated.

This is close to the Olweus criteria, but does limit the concept to youth, and would exclude sibling bullying or dating violence. It would exclude workplace bullying, prison bullying, family abuse, teacher-pupil bullying, and pupil-teacher bullying – all of which have been areas of study! In fact an edited collection by Monks and Coyne (2011), called Bullying in Different Contexts, looked at exactly these different kinds of behaviours under the common framework of bullying. A succinct definition that would cover these different contexts is that bullying is the ‘systematic abuse of power’ (Smith & Sharp, 1994).

Another challenge to the classic definition of bullying has come with online or cyberbullying. This issue is discussed by Campbell in Chapter 9. The most common issue here has concerned that of repetition, as just one aggressive act by a perpetrator could be repeated (passed on, commented on negatively) by many others – so does this single act by the perpetrator count as bullying? Here, the phrase ‘is highly likely to be repeated’ from Gladden et al. (2014), above, may be helpful, as someone acting in this way – for example, posting a very negative comment on a social network site – can reasonably expect that it will be seen and commented on. As regards imbalance of power, the anonymity often conferred by online attacks can itself be a power imbalance; and if there is no anonymity, the more standard criteria such as status in the peer group may become relevant.

Even if (depending on definition) bullying can occur in many contexts, the earliest sustained work, and the largest volume of research, has concerned school bullying. This is the topic of this book, and the next section overviews how this research programme has developed.

History of research

Figure 1.1 shows the number of publications on bullying, with the keyword ‘bully’, from ISI Web of Science, by 5 year periods (Smith, 2016). This would include all kinds of contexts for bullying, but the majority are on school bullying. As can be seen, there was very little research activity up to the late 1970s. There was a moderate amount up to the 1990s, and then an exponential growth in research activity in this century.

FIGURE 1.1 Number of publications in five-year intervals, up to 2015, from ISI Web of Knowledge searching for the term ‘bully’.

The systematic study of bullying in schools in the west started in Scandinavia. The first important publication was Olweus’ book Hackkycklingar och översittare: Forskning om skolmobbning (1973; expanded version in English translation 1978). Olweus developed a self-report questionnaire (in which pupils could report on their own experiences) to assess bullying. There were also studies in Japan and in South Korea at least from the 1980s (Morita, 1985; Koo, 2007). At that time these research traditions were quite separate.

An important development in the 1990s was a change in researcher’s definition of aggression, and hence of bullying. Aggression had previously been thought of as just being physical (e.g. hitting, perhaps damaging belongings) and verbal (e.g. threats, insults). However, researchers in Finland (Björkqvist, Lagerspetz, & Kaukiainen, 1992) and the US (Crick & Grotpeter, 1995) now broadened the concept to include indirect and relational bullying, such as rumour spreading, and social exclusion. This approach became widely accepted.

Another important methodological step during the 1990s was the introduction of participant roles in bullying by Christina Salmivalli and colleagues (Salmivalli et al., 1996). Previously, research had focused on bullies, and victims. The participant role approach divides the bullying role into ringleaders (who initiate and lead the bullying), assistants (who join in) and reinforcers (who laugh or encourage the bullying). Victims can be considered as passive or non-provocative victims, or effectively as bully/victims if they are in both bully and victim roles, someone who annoys others and is in turn attacked (Pikas, 1989). Other roles are defenders (who help the victim in some way), bystanders (who are aware of the bullying but ignore it), and outsiders (who are truly unaware of the bullying).

From the mid-1990s to the mid-2000s, research on traditional or offline bullying became a major international research programme. Researchers from Europe, North America and Japan shared experiences through a UNESCO-supported meeting, and surveys and interventions took place in many countries (Smith et al., 1999).

This century has seen awareness of cyberbullying, using mobile phones and the Internet. Appearing first as text message and email bullying, this has increased and diversified from the mid-2000s (Rivers & Noret, 2010). Since around 2008 the development of smart phones and social networking sites have offered many new tools for cyberbullies. Cyberbullying is considered in some detail in Chapter 9. Compared to traditional bullying, it is much more likely to be perpetrated and experienced outside school; but is still often between class- or school-mates. It is characterized by increased possibilities for anonymity of the perpetrator, a potentially much greater audience for the attacks, and the comparative lack of respite for the victim (Kowalski et al., 2014).

The decade since the mid-2000s has seen a massive increase in publications (Figure 1.1). Many of these describe characteristics of bullying, and many look at predictors of involvement, and effects of this; some of these findings are summarized in the next sections. There have also been more intervention studies; and more use of mixed-method or qualitative studies (Patton et al., 2015; Smith, 2014; Zych et al., 2015). As the number of relevant studies has increased, recent years have seen been more meta-analyses on various topics, including interventions (to be considered later).

Measurement and prevalence of bullying

There are various ways of measuring bullying, but for prevalence rates, self-report questionnaires given to children, usually at school, have been by far the predominant method. The Olweus Bully/Victim Questionnaire was a pioneering exemplar of this approach; it is filled in anonymously, and asks about the frequency with which someone has been bullied, or taken part in bullying others, over a specified time period (often, one school term). There are many other questionnaires available, especially since cyberbullying became common in the last decade.

Such questionnaires are quick and easy to administer, but some researchers question how valid they are given issues of over-sensitivity or denial (in admitting being bullied) or social desirability (in admitting bullying others) (Cornell & Brockenbrough, 2004). Alternatives to self-report questionnaires include peer nominations, widely used on a class basis; teacher nominations, which are used more with younger children; interviews and focus groups; direct observations; and incident reports. Each method has its own advantages and disadvantages, but pupil self-report remains the most feasible for obtaining detailed data on a large scale.

One well-cited source of prevalence data comes from the Health Behaviour in School-aged Children (HBSC) surveys, of about 1,500 pupils each at 11, 13 and 15 years, from around 40 countries, mostly European and North American. Starting in 1993/1994, this data is gathered every four years. The surveys are anonymous self-report, and ask about experiences over the past couple of months. Victim or bully rates are calculated from reporting this happening ‘at least two or three times in the past couple of months’ or more (ignoring ‘it only happened once or twice’). The 2009/2010 HBSC survey (Currie et al., 2012) found rates for bullying others averaged out at 10.3%, and for being bullied at 11.3%. In the most recent 2013/14 survey (Inchley et al., 2016), the corresponding figures were 8.3% and 11.0%.

The prevalence rates reported in different studies vary greatly. Cook et al. (2010a) examined 82 studies published from 1999 to 2006, of which 61 used self-report data. Prevalence rates averaged around 20% for bullies, 23% for victims, and 8% for bully/victims, but with a high variability across studies. These are higher figures than HBSC, probably because bullying that only happened ‘once or twice’ in a time referent period was often included.

Prevalence data are influenced by many factors: what definition is used (e.g. whether it includes indirect as well as direct forms, and cyberbullying); what frequency is regarded as bullying (e.g. once/twice a term; once a month, once a week or more); and what time span is being asked about (e.g. last month, last term, last year, ever at school). If a short time span (last month, or term) is taken, then the time of giving a questionnaire in the school or calendar year can be important. These issues make it difficult to compare across different studies. It also means that absolute prevalence figures are rather meaningless, when presented on their own. However, they are vital for looking at effectiveness of interventions, where the same methodology will be used for pre-, post- and perhaps follow-up assessments.

Attitudes to bullying and reasons why some children bully others

Although most school pupils say they do not like bullying, a significant minority do say they could join in bullying. These ‘pro-bullying’ or ‘anti-victim’ attitudes have been found to increase with age up to around 14 or 15 years, after which they may start to decline (Rigby, 1997).

One explanation for this age change comes from the dominance hypothesis (Salmivalli, 2010). This argues that many children who bully do so because of a desire for dominant status in the peer group. Ringleader bullies especially may feel rewarded if assistants and reinforcers support their bullying actions, and if many bystanders remain passive. Concerns about status in the peer group and worries about peer rejection are known to peak during the adolescent years, especially soon after puberty (Ellis et al., 2012). At this mid-adolescent period there may therefore be more motivation to initiate bullying (to exhibit dominance) and to assist or reinforce bullying (to avoid peer rejection or being bullied oneself) (Pouwels, Lansu, & Cillessen, 2018).

Volk et al. (2012) have argued that bullying can be considered as an evolutionary adaptation, in the sense that a perpetrator may get some (at least short-term) benefits from bullying others. Some bullies are popular in the peer group, do not have low self-esteem, and are socially skilled in manipulating others. Several studies have found that bullies, especially ringleader bullies, do well on theory of mind tasks (Smith, 2017); they understand how to hurt a victim effectively and with some peer support, while avoiding adult detection – so-called ‘cold cognition’. This may be true of many bullies, but not all; Peeters, Cillessen, and Scholte (2010), in a Dutch sample of 13-year-olds, distinguished three kinds of pupils nominated as bullies by peers: one group was popular and socially intelligent; a second group was relatively popular and with average social intelligence scores; a third group, the smallest numerically, was unpopular and had lower than average scores on social intelligence.

The dominance hypothesis predicts that the likelihood of a child taking part in bullying another will be influenced by peer group norms and expectations about bullying and defending, as well as what support the victim might have. There is considerable evidence supporting this. For example, Kärnä et...

Table of contents

• Cover
• Half Title
• Series Page
• Title Page
• Copyright Page
• Table of Contents
• List of illustrations
• List of contributors
• Series Editor Foreword: Routledge Psychological Impacts. Impact on Bullying
• 1. Introduction
• 2. The Olweus Bullying Prevention Program (OBPP): New evaluations and current status
• 3. KiVa anti-bullying programme
• 4. Lessons learned from the national implementation and international dissemination of the ViSC social competence programme
• 5. Anti-bullying programmes in the United States: What works and what doesn’t?
• 6. The Friendly Schools initiative: Evidence-based bullying prevention in Australian schools
• 7. Ijime prevention programmes in Japan
• 8. Peer support and the pupil’s voice: The NoTrap! programme in Italy
• 9. Specific interventions against cyberbullying
• 10. Conclusions: Different levels of challenge in tackling school bullying
• Index