Educators ’ perspectives on risk factors for learner-on-learner bullying

Copyright: © 2013. The Authors. Licensee: AOSIS OpenJournals. This work is licensed under the Creative Commons Attribution License. Educators are the key role-players in the successful development and implementation of most anti-bullying programmes and/or intervention strategies. This article sets out to report on findings from a qualitative study on educators’ perspectives of the risk factors for learner-onlearner bullying. Educators who were furthering their studies at the University of the Free State, South Africa, were invited to take part in the study on different types of bullying. This article focuses on the narratives of 91 participants who described the incidence of learneron-learner bullying. The study uses a four-level, social-ecological model as the theoretical framework. Individual, family, institutional and societal risk factors for bullying were identified by means of a qualitative content analysis. It is suggested in the article that although risk-focused prevention strategies may lead to a reduction in bullying, prevention strategies should not target only these (negative) factors, but try to promote protective factors on all four social-ecological levels.


Introduction
Bullying is such a widespread experience that it appears that most children have been bullied at some time or other during the course of their school careers (Berger, Karimpour & Rodkin 2008: 297;Cowie & Jennifer 2008:21;Ladd & Ladd 2001:25).Bullying others and being victimised have negative long-and short-term consequences on the general health and well-being of the learners involved (De Wet 2005:708).Chronic bullying by peers is associated with serious adjustment problems, including anxiety, social withdrawal, low self-esteem, suicidal tendencies, a dislike and avoidance of school, poor academic performance, rejection by mainstream peers, and a lack of friends (Perry, Hodges & Egan 2001:73).Moreover, Rigby (2001:322) found that victims of bullying are more likely than others to experience particularly distressing mental and physical states of being more anxious, more depressed, more socially dysfunctional, less physically well, and more prone to suicidal ideation than other children.Olweus (2001:17) notes that there is consensus, at least in principle, that bullying at school should not be tolerated and that the school has a major responsibility to counteract and prevent bullying.Educators have a singular duty, based on their profession, as well as a delegated duty, based on the authority delegated to them by the parents or guardians of the children enrolled at the school, to act in loco parentis [in the place of a parent].This compels educators not only to take care of their learners, but also to maintain order (Joubert & Prinsloo 2008:145), thus protecting any learner from being bullied while taking the necessary steps to prevent this form of destructive behaviour from taking place at school.Venter and Du Plessis (2012:3) believe that schools which care about their learners in accordance with Christian values should do their utmost to curb or eradicate all forms of violence.According to them, these schools ought to be characterised by a humane and caring Opvoeders se perspektiewe oor risiko-faktore vir leerder-op-leerder afknouery.Opvoeders vervul 'n sleutelrol in die suksesvolle ontwikkeling en implementering van die meeste antiafknoueryprogramme en/of intervensiestrategieë.Die artikel het ten doel om verslag te lewer oor bevindinge van 'n kwalitatiewe studie oor opvoeders se sieninge oor risikofaktore vir leerder-op-leerder afknouery.Opvoeders wat besig was met verdere studie aan die Universiteit van die Vrystaat, Suid-Afrika, is genooi om aan die studie oor verskillende tipes afknouery deel te neem.Hierdie artikel fokus op die narratiewe van 91 deelnemers wat insidente van leerderop-leerder afknouery beskryf het.Die studie gebruik 'n viervlak, sosiaal-ekologiese model as teoretiese raamwerk.Individuele, familie, institusionele en gemeenskapsrisikofaktore is met behulp van kwalitatiewe inhoudanalise geïdentifiseer.In die artikel word gesuggereer dat ofskoon voorkomingstrategieë wat op risiko's fokus, tot die vermindering van afknouery kan lei, voorkomingstrategieë nie net hierdie (negatiewe) faktore moet teiken nie, maar ook moet poog om beskermende faktore op al vier sosiaal-ekologiese vlakke te bevorder.
Scan this QR code with your smart phone or mobile device to read online.

Read online:
ethos free from discrimination, intolerance and violence.Oosthuizen, Roux and Van der Walt (2003:388), moreover, believe that for learner security to prevail, educators should not merely 'clamp-down' on destructive behaviour, but enter 'into loving, caring, serving, guiding and 'disciplining' relationships with learners'.Educators' understanding of the risk factors for bullying, rather than the legal and moral imperatives or research findings, affect their willingness to become part of anti-bullying programmes and/or intervene in incidents of bullying.
Most researchers (e.g.Cowie & Jennifer 2008:25-39;Olweus 2001:16;Rigby 2001:324;Swearer, Peugh, Espelage et al. 2006:258;Yoon & Barton 2008:262) agree that a reduction in bullying is most likely to occur if schools adopt a whole school approach in which educators play a leading role.It is desirable that educators join forces with learners, parents and members of the community to develop, implement and evaluate an anti-bullying programme for their school.It is, moreover, important that all educators act together in a planned and agreed-upon way to counteract bullying.Educators should be well informed about the nature and quality of peer relations in their school.What educators regard as bullying and their perspectives of the risk factors for bullying have an impact on the kind of action and policy they are likely to support (Rigby 2001:324-325;Yoon & Barton 2008:262).
Bullying is in contravention of the Christian values of tolerance, neighbourly love, care and respect for others, patience, kindness and trust (Venter & Du Plessis 2012:2).Bullying behaviour negates these Christian values, being motivated by intentional power abuse to the detriment of the victim (Venter & Du Plessis 2012:3).This view is supported by Myers (2013:3).According to him, a common Hebrew word which overlaps with the English word 'to bully' is 'ashaq' (Myers 2013:1).Myers (2013) uses Ecclesiastes 4:1 as his point of departure to give a biblical perspective of bullying: Again I saw all the oppressions that are done under the sun.And behold, the tears of the oppressed, and they had no one to comfort them!On the one side of their oppressors there was power, and there was no one to comfort them.(p. 1) The emphasis of Ecclesiastes 4:1 on the desolation and loneliness of the victims of bullying highlights the need to acknowledge the role of all stakeholders (victims, bystanders and perpetrators of bullying, the parents or caregivers, educators and the community) when studying the risk factors for bullying (see 'Theoretical framework').In the recommendations emanating from the findings of this study I will emphasise the need to inculcate Christian values at the level of the individual, family, school and community (see 'Conclusion').
Given the negative consequences of bullying by peers, and the importance of the role educators should play in preventing the practice, the aim of this article is to report on the findings of a qualitative study on educators' perspectives of risk factors for learner-on-learner bullying.Olweus (2001), one of the world's leading experts on bullying, defines school bullying in the following general way: … a student is being bullied or victimised when he or she is exposed, repeatedly and over time, to negative action on the part of one or more other students.(p. 5) This definition emphasises the negative (aggressive) acts that are carried out repeatedly over time.The definition further specifies that in bullying, there is a certain imbalance of power or strength: the person who is exposed to negative acts has difficulty defending him-or herself (Olweus 2001:6).Olweus (2001:6) identifies eight specific forms of bullying: physical and verbal (including racial and sexual) harassment; threatening and coercive behaviours; as well as more indirect ways of harassment, including 'relational' victimisation in the form of active social isolation, back talking, having rumours spread, and so on.Friendly and playful teasing, as well as fighting or arguing between two or more people of about the same strength, is not bullying.

Concept clarification
An acknowledgement that many different contextual variables make important contributions to the development and maintenance of bullying is found not only in the theoretical framework that underpins this study, but also in Espelage's (2004) definition of bullying as: … an ecological phenomenon that is established and perpetuated over time as a result of the complex interplay between the individual child, their family, peer group, school and community, as well as their culture.(p.4)

Theoretical framework
The socio-ecological model is a theoretical framework that can be used to examine the multiple effects and interrelatedness of social elements in an environment.In this article, I attempt to make sense of the multiple causes of bullying in terms of the risk factors operating at four different levels.The first level identifies biological and personal factors that influence how individuals behave and thus increase their likelihood of becoming victims or perpetrators of bullying: demographic characteristics, personality disorders, and a history of experiencing, witnessing or engaging in bullying behaviour.The second level focuses on the home environment: modelling of bullying by siblings and/or parents or guardians may influence the development of a bullying and/or victimisation tendency in the individual.The third level focuses on the organisational or institutional factors that shape or structure the environment within which the individual exists and in which interpersonal relations occur.These factors can be the school culture and climate, rules, policies, and acceptable behaviour within the teaching and learning environment.The fourth level looks at the broad societal factors that help to create a climate in which bullying is encouraged or inhibited: the responsiveness of the criminal justice system and/or trade unions; social and cultural norms regarding gender roles; the social acceptability of bullying and violence; and political instability (Berger et al. 2008:300;Cowie & Jennifer 2008:18;Swearer et al. 2006:258).All four levels are explored in this study.

Research design
This study followed a qualitative and descriptive research design.Qualitative research can be used to provide an understanding of a specific phenomenon.The focus of this study was educators' perspectives on the risk factors of learner-on-learner bullying.The study aimed at providing a description of educators' insights into bullying (Bless, Higson-Smith & Kagee 2006:46).The research was undertaken within an interpretative framework with its emphasis on experience and interpretation.Interpretive research is concerned with meaning and it seeks to understand people's definitions and understanding of situations.Henning, Van Rensburg and Smit (2011:21) emphasise that the interpretive paradigm does not concern itself with the search for broadly applicable laws and rules, but rather seeks to produce a descriptive analysis that emphasises a deep, interpretive understanding of social phenomena.This ties in with the focus of this study, namely to gain an understanding of the risk factors for learner-onlearner bullying.

Data collection
A convenient, voluntary sample of educators who were furthering their studies at the University of the Free State, South Africa, was opportunistically selected.During 2012, I invited educators who were busy with their BEd (Hons) studies to take part in a study on bullying.The following introductory detail was given to the participants in a questionnaire: Bullying includes a variety of behaviours, ranging from psychological acts (e.g.shouting) to physical assaults.Bullying can be either direct (e.g.physical and verbal aggression) or indirect (e.g.threats, insults, name calling, spreading rumours, writing hurtful graffiti, cyber bullying or ignoring the victim).The literature has identified the following types of bullying: learner-on-learner bullying; educator-on-learner bullying; learner-on-educator bullying; and workplace bullying (i.e.workers/educators being bullied by their principals, colleagues or the parents of learners).
The majority of the educators who were invited to take part in the study (181 of 205) completed the questionnaire.More than half of the 181 participants (50.3%) described incidents of learner-on-learner bullying from their perspective as educator onlookers and/or bystanders.Ten (5.5%) participants wrote about their childhood experiences as victims of bullying.The rest of the participants wrote about workplace bullying (32.6%), educator-on-learner bullying (9.9%) and educatortargeted bullying (7.2%).
In line with the aim of this article, only the answers to the open and closed questions of the participants who wrote about learner-on-learner bullying were analysed.Table 1 gives a summary of the demographic details of the 91 participants who described incidents of learner-on-learner bullying from their perspective as educator onlookers and/or bystanders.
A number of open-ended questions were asked in the questionnaire: 1. Please share with me your experience(s) as a victim and/ or an onlooker of bullying (it may be only one or several types of bullying).2. What were/are, according to you, the reasons (risk factors) why you or the other person(s) were/are bullied?If you were not the victim of bullying or don't know of any incidents of bullying at your school, give reasons why you or the other people you know were not bullied.3. What can be done to stop bullying?4. Have you ever bullied someone?Please share this information (who was the victim, what did you do to the person and why?).
A few closed questions and biographical questions were also asked.The data on which this article is based are primarily drawn from educators' answers to question 2 and question 4 of these open-ended questions.

Data analysis
The guidelines for qualitative content analysis drawn up by Henning et al. (2011:104-106) were used to reduce, condense and group the content of the participants' answers to the open-ended questions.A coding frame was drawn up, also providing for verbatim reporting where applicable.I worked through all the data and coded them.Related codes were thereafter organised into categories.After I had completed the categorisation, I re-read the participants' answers to the questions to check whether I had captured all the important insights that had emerged from the data.The categories, patterns and themes that could also be linked to the aim of this article were identified and described.The identification of emergent themes allowed the information to be analysed and related to the literature.The researcher used an independent qualitative researcher to do an independent re-coding of some of the data in order to determine whether the same themes became evident and could be confirmed.Consensus discussions between the researcher and the independent expert were held in order to determine the final findings of the research.

Validation strategies
Validation within an interpretive approach to qualitative research is marked by a focus on the importance of the researcher, as well as on the interpretations that are temporal, located and always open to reinterpretation (Creswell 2007:205).The following two strategies were used to combat threats to the validity of my study: investigator triangulation (the independent expert and the researcher read and coded the transcripts and took part in consensus discussions) and transferability (rich, thick descriptions allow readers to make decisions regarding transferability).The detailed descriptions in this article may enable the readers to transfer information to other settings and therefore determine whether the findings can be applied (Creswell 2007:202-209).

Ethical consideration
The participants' dignity, privacy and interests were respected at all times.The questionnaires did not contain any identifying aspects, names, addresses or code symbols.Before completing the questionnaires, the participants were also informed that the process was completely voluntary and that they could withdraw at any stage during the process.The researcher, who was present during the completion of the questionnaires at all times, was available, if necessary, to support or refer traumatised participants.

Findings and discussion
An analysis of the verbatim responses of the participants revealed that, in accordance with Bronfenbrenner's socioecological systems theory, the risk factors of learner-onlearner bullying may be found in the individual and at family and school level.

Theme 1: Individual risk factors: Bullies' and victims' characteristics and behaviour
The bully Researchers (Baldry & Farrington 2000:18;Berger et al. 2008:302;Meland et al. 2010:365) generally describe bullies as aggressive, tough, confident, impulsive, of average popularity, but below average school attainment.Meland et al. (2010:365), moreover, found that bullies are more depressed and pessimistic than their peers not involved in bullying.A study by Meland et al. (2010:365) also showed that bullies have relational problems with parents and educators.
The educators who took part in this study also described bullies in negative terms.The bullies were, for example, depicted as deviant (e.g.'that boy has low morals' and 'they are full of hatred'), disrespectful (e.g.'I don't think they respect humans even though they are humans themselves'), mentally disturbed (e.g.'unnoticed psychopathic behaviour'), troubled youths (e.g.'cry for help' and 'some are going through other forms of trauma') who may have an inferiority complex (e.g.'bullies try to impress somebody else' and 'they want to prove to others that they are grownups') and lack social skills ('he is very silly').A number of participants suggested that bullies are underachievers (e.g.'the bully is often someone who has failed his grade' and 'in most cases bullies are slow learners').Participants suggested that bullies are cowards (e.g.'they bully smaller boys because they know the boys cannot defend themselves').The bullies were furthermore typified as conceited: 'The bully knew that he can bully a smaller boy to do his homework and do menial tasks for him' and 'The big boys do not like to carry lunch boxes, but during lunch time or break, they take the smaller children's lunch boxes.' An important feature of bullying, namely the imbalance of power or strength, is underlined in the previous two quotations and Olweus's definition (2001:6).The view that bullying is an expression of hegemonic masculinity is clearly illustrated by the following extract from one of the narratives: Bigger boys want to show that they are in control.Smaller boys are bullied by these macho figures if they believe that smaller, shy boys are not the embodiment of what the bullies perceive to be masculine traits ('Vrystaat boere').
A number of participants suggested that bullying behaviour was an expression of the bullies' frustrations (e.g.'vent his anger' and 'the learner who has a speech defect will express her anger and frustration by fighting with her classmates').
Participants mentioned that some of the bullies at their respective schools were known to abuse alcohol and drugs (dagga), and smoke cigarettes.It was bluntly stated by three of the participants that these learners took money from fellow-learners to pay for their habits.Two of the participants suggested that the bullies were known drug dealers.It was furthermore suggested by these two participants that learner gang members bully fellow learners to gain control of the drug dealings in the school.Turf wars between different gangs are therefore not uncommon.Participants who focused on drugs and gangs as possible risk factors for bullying noted that learners are often bullied to become members of a specific gang.Learners succumb to the pressure in order to get 'protection'.
In the foregoing exposition the gender of the bullies, as illustrated by the personal pronouns used in the quotations of the aggressors, is predominately male.The presumption that the bullies are mostly boys is supported by data from a question in the questionnaire in which the participants had to indicate the gender of the bully: 75 of the 91 (82.4%) participants specified that the bullies are boys.This finding resonates well with research by Andreou and Bonoti (2010:173), Berger et al. (2008:300), as well as Berger and Rodkin (2009:79).Andreou and Bonoti (2010:173) and Berger et al. (2008:300), however, stress that differences regarding gender go deeper than prevalence rates: girls display more relational bullying, whereas boys are more inclined to physical aggression.
The foregoing exposition demonstrates that the bully is mostly stereotypically portrayed as a tough, arrogant, powerful, underachieving boy who bullies his weaker peers (Berger et al. 2008:302).Bullies were also typified as deviant, gutless, mentally disturbed, drug and alcohol abusers and delinquents.Some bullies were, however, less harshly described as attention seeking, socially awkward, troubled youths whose bullying behaviour is (merely) an expression of their frustrations.

Victims of bullying
Some of the reasons why learners are bullied are encapsulated in the following statement: 'Bullying in my school is almost always aimed at learners who are perceived to be "different", "weaker" and "socially awkward"' (Female, 38, Head of Department).
In the ensuing discussion attention will be given to victims of bulling who are perceived to be 'different', 'weaker' and 'socially awkward'.Thereafter, findings on the passive and provocative victims, as well as on the gender of the victims, are presented.

Discourse of otherness
Three categories of otherness, or being different, were identified by the participants, namely ethnic or racial, physical and intellectual.Juvonen, Nishina and Graham (2001:115) argue that the ethnic makeup of a school can signal an imbalance of power.Numeric ethnic minority groups within a particular school are thus likely to be viewed to be weaker or subordinate as compared with majority groups.This study found that learners are intolerant of those who are not proficient in the dominant language(s) or racial group of the particular school.Whereas one of the participants wrote about a Chinese learner who was relentlessly bullied by fellow learners 'because he looks different and speaks different', another noted that a learner from Lesotho was ridiculed because he could not speak Afrikaans.Findings from Peterson and Ray's (2006:258) study on the bullying of gifted children are corroborated by findings from the current study: participants mentioned that learners who were bullied were, among other things, 'very clever' and 'more intelligent than the other girls in her class'.Mention was, however, also made of a boy who was bullied because 'he was a slow learner'.The stigmatisation of an 'other' was also found in the victimisation of those who were 'too fat' or wear glasses ('They called her "four eyes"').Taylor (2011:178) likewise found that overweight children are likely to be victims of bullying.

Victims are weaker and younger than their bullies
In my exposition of participants' depiction of the bullies, emphasis was placed on the stereotypical portrayal of the bully as physically strong, older and conceited.It is therefore not surprising that participants and researchers (Baldry & Farrington 2000:18;Meland et al. 2010:365;Perry et al. 2001:75) place the victims at the opposite range of the continuum.Perry et al. (2001:75) found that physical weakness contributes directly to bullying.The following quotations from the participants' narratives will suffice: 'they are young, afraid to fight for their belongings and … soft', 'small, young and afraid to speak up for themselves' and 'the younger pupils are bullied because they are powerless and afraid of the senior learners'.

Victims are socially awkward loners
The narratives abound with examples of introverts who were struggling to survive because of a lack of social skills and friends (e.g.'they are usually bullied if they don't have any friends' and '… they have no one to talk to').Perry et al. (2001:80) call these loners 'rejected children'.The claim by Perry et al. (2001:80) that these children are easy targets because bullies believe that attacking them will 'go unpunished by mainstream peers' is reiterated by a participant who wrote: 'these children don't have many friends.This makes it easy for bullies to victimise them.' These lone victims may find themselves in a downward spiral, because relational problems may increase with increased bullying and vice versa (Meland et al. 2010:365).
The vulnerability of those who do not have friends, siblings or 'protectors' is highlighted by the following two quotations: 'He is new at school and has no friends' and 'they don't have older brothers and sisters at the school to protect them'.In their study on contextual antecedents of bullying, Juvonen et al. (2001:115) also found that children who transfer to a new school are 'likely to be prime targets' of bullying.

Passive victims of bullying
Participants suggested that those learners with a gentle demeanour (e.g.'soft-spoken, well-mannered learners' and 'he is a sensitive boy who is inclined to shed tears easily') and who unassertively shy away from confrontation (e.g.'he just let it slide') were often bullied.These children may be labelled 'passive victims' (Olweus 2001:12;Perry et al. 2001:76).They do little to provoke their bullies directly; rather, they are socially withdrawn and may appear anxious and even depressed to their peers.Whereas Perry et al. (2001:76) write that bullies often 'sense and exploit these vulnerabilities', Olweus (2001:11) notes that passive victims 'simply fell easy prey to aggressive, powerful bullies'.

Provocative victims
Provocative victims do not necessarily fit the stereotypical picture of a weeping, maladjusted and isolated child (Berger et al. 2008:302;Olweus 2001:12).Some of the educators blamed the victims for provoking their aggressors, with one of the participants describing the victims as 'annoying and restless children'.Victims' perceived arrogance (e.g.'… he had a loud mouth.He tried to appear tough by talking big and for that the bigger guys bullied him'), as well as wrong choices were highlighted by participants as reasons why children were bullied.A participant wrote that one of the girls at her school 'allowed' her friends to take 'inappropriate' photos of her.These photos were distributed via SMSs to children of 'questionable moral values'.This girl was consequently relentlessly bullied by her peers.The suggestion by the participants that these victims actually deserved the treatment they got is not uncommon in bullying research.Olweus (2001:12) found that educators often focus on the provocative victims.According to him, educators argue that the behaviour of these victims is irritating and creates tension, which explains why these learners are disliked and bullied by their peers.

The gender of the victims of bullying
In their response to a survey item in which the participants had to specify the gender of the victims, 69.2% indicated that the victims were boys.Studies by Berger and Rodkin (2009:79) as well as by Andreou and Bonoti (2010:173), support this view that boys are more often the victims of bullying than girls.Perry et al. (2001:75), on the other hand, found that the likelihood of being a victim of bullying does not differ much by gender, but that boys and girls are bullied in different ways.Boys are more often victims of physical bullying, whereas girls are more often subjected to acts of relational aggression, such as gossip and social exclusion from groups.
The contradiction in findings may be ascribed to the unseen nature of emotional bullying.Educators are often unaware of the victimisation of girls.
In the preceding account, the victims were largely portrayed as physically and emotionally weak, young, socially awkward male loners.Children were also stigmatised and victimised for being different.Whereas unassertive, gentle children easily fall prey to bullies, others provoke their bullies through their irritating or improper behaviour.

Reasons why some learners are not targeted by bullies
Research suggests that bullying is rife in schools (Berger et al. 2008:295).Yet, not all learners are victims or bullies.A quarter of the participants wrote notes on the following question: 'If you were not the victim of bullying or don't know of any incident[s] of bullying at your school, give reasons why you or the people you know were not bullied.' A fair number of those who answered this question mentioned that children with strong personalities, self-assurance and positive self-images tend to be excluded from bullying as they are able to defend themselves against bullies.Two of the participants, moreover, wrote that learners who know their 'place' in the learners' pecking order do not get bullied.
A study by Perry et al. (2001:77) also found that assertive children who are perceived to be friendly, likely to share and cooperate, skilled in joining the play of other children and possess a sense of humour were not victimised.One of the findings of this study, namely that physical strength is an essential deterrent (e.g.'because of their physical power -other people are afraid of them'), resonates well with findings by Perry et al. (2001:75).According to these researchers, physical strength gives children the ability and confidence to ward off attacks from other children assertively and effectively.The importance of having 'protection' against bullies was stressed by four of the participants.The protectors may be older siblings looking after the best interests of the younger one, or older children being paid 'protection fees' by younger learners.This study suggests that learners who are not targeted possess 'protective factors' (Cowie & Jennifer 2008:18): they are physically and emotionally strong, socially skilled and resilient individuals.These characteristics contrast sharply with the risk factors of victims of bullying.
Theme 2: The home environment as a risk factor for learner-on-learner bullying Whereas the preceding discussion focused on the traits of the bully and the victims, attention will now be given to the participants' suggestion that the roots of school bullying can be found in the destructive, violent behaviour of siblings, parents and other family members.Three participants, for example, made mention of bullies who were believed to be subjected to cruel chastisement by an uncle, molestation by a father or harassment by older siblings.Baldry and Farrington (2000:18), Duncan (2004:235) and Swearer et al. (2006:261) support the notion that maltreatment by parents, including physical, sexual and emotional abuse and neglect, can be linked to bullying.Baldry and Farrington (2000:18) as well as Swearer et al. (2006:261) found moreover that the parents or guardians of bullies are authoritarian, lack warmth and involvement, and use power assertively in their disciplinary practices, for example through physical punishment and violent emotional outbursts, thus demonstrating a permissive attitude towards the child's bullying behaviour.Participants also argued that a lack of positive role models at home, 'problems at home', a 'lack of attention from their parents' and 'absent' parents may lead to the belief among learners that aggressive behaviour guarantees the admiration, acceptance and support -which they lack at home -of their peers.
Several participants suggested that the economic status of parents may have an impact on bullying.These participants simply stated that bullies generally came from poor households.Three of the participants noted that bullies in their schools were over-age learners who attend school to get food (from the daily lunches provided by the government's feeding scheme).Whereas some of the participants showed sympathy for poverty-stricken bullies who took their peers' lunchboxes because they were hungry, others condemned this kind of behaviour and insinuated that these bullies were thieves.
Whereas researchers such as Duncan (2004:232) studied the possible influence of parent-child relations and the influence of the parenting style on the bully and the victim, educators who took part in this study focused mainly on the impact of such influences on the bullies.Nevertheless, there is a relatively large body of research (Duncan 2004:233;Perry et al. 2001:86) which finds that the relationship between the victims and their parents is overly intense, with parents being over involved in their children's lives.Perry et al. (2001:86) believe that 'negative family experiences' inadvertently increase the probability of children displaying behaviours during peer interaction that 'invite' abuse from aggressive peers.The only evidence of participants' insight into the role parents might play in 'inviting abuse' centres on victims' allegedly flaunting their parents' wealth ('M bullied B, because B used to take huge amounts of money to school' and 'T always carried money and sweets with him.Senior boys … bullied T').It may thus be argued that children of rich, overindulgent parents are sought-after targets for bullies.
Learners coming from households that do not conform to prevailing social norms were also bullied.One of the participants noted, for example, that a Grade 7 girl's mother was apparently a prostitute; thus the girl was relentlessly ridiculed.Findings by Ray and Gregory (2001:30) on bullying experiences of children with gay and lesbian parents confirm the findings in this study.An educator indicated that a girl at his school was bullied because her guardian is supposedly gay.
In the foregoing discussion, participants and the literature highlighted the home environment as a risk factor.Physical and emotional abuse and neglect, as well as poverty are, among other things, identified as risk factors for becoming a bully.The victims of bullying, on the other hand, grow up with overindulgent, rich parents or guardians, as well as with people who do not conform to prevailing social norms.

Theme 3: Risk factors for learneron-learner bullying within the school environment
The literature emphasises the impact of school climate (Berger et al. 2008:308;Swearer et al. 2006:260;Yoon & Barton 2008:250), teaching styles (Rigby 2001:325), disciplinary practices (Rigby 2001:326;Yoon & Barton 2008:262) and educator-relationships (Berger et al. 2008:308) on bullying.Rigby (2001:236) found that educators often emulate power imbalances, which is typical of bullying, through their dominating and authoritarian behaviour.In their answers to the following question the participants made no reference to the role they or other educators might play in promoting bullying: 'What were/are, according to you, the reasons why you or the other person(s) were/are bullied?' Yet in their responses to a question on their own bullying behaviour ('Have you ever bullied someone?Please share this information (who was the victim, what did you do to the person and why?'), 15 (16.5%) of the participants described how they bullied their learners and abused their power.This included corporal punishment ('I took a duster and gave him a hiding on the fingers' and 'I beat them harshly … I am protector and judge'), verbal abuse ('I shouted at them' and 'I insulted them'), forcing learners to do manual labour ('I forced him to sweep the classroom') and emotional bullying ('I forced her to report to me every day').In the next quotation, a participant describes how she assaulted a learner.Although she acknowledges that her conduct was wrong, it brought about the 'desired results': Researchers' (Swearer et al. 2006:261;Yoon & Barton 2008:250) finding that a school's traditions have an influence on school bullying confirms findings from this study.Two of the participants identified their respective schools' initiation practices as a risk factor: 'The so-called initiation habit or ritual is horrendous and inhuman.It degrades newcomers...' 'The initiation practices of newcomers at my school are appalling.
Those who can't stomach it are labelled as wimps and losers'.
The foregoing discussion focused first on educators' harsh and degrading treatment of learners.Learners were subjected to verbal, physical and emotional abuse by those who were supposed to act as caring parents.Secondly, attention was given to violent initiation practices.If those in positions of power act aggressively, and the school's initiation traditions sanction unabated violence, learners get the message that bullying is acceptable behaviour.
Theme 4: Societal risk factors for learner-on-learner bullying Whilst Swearer et al. (2006:262) found that the characteristics of the community in which children live and go to school may have an influence on bullying, participants paid little attention to the impact of societal factors on bullying.De Klerk and Rens's (2003:353) argument that even though the South African Constitution uses language that can be described as 'value language', that the country is experiencing a 'moral crisis' is supported by findings from this study.The participants' negligible awareness of the influence of societal factors on school bullying focused on the lack of positive values in the community (e.g.'the value system has changed', 'positive values are not put into practice' and 'there are no role models in our community') and the total disregard for those in positions of authority.Mention was also made of turf wars among opposing gangs and drug dealers spilling over into the school grounds (see Theme 1: 'The characteristics and behaviour of the bully').

Strengths and limitations of the study
The greatest strength of this study lies in its qualitative approach, which allows for a rich exploration of educators' perceptions on different types of bullying.The decision to use questionnaires with mainly open-ended questions gave participants the opportunity to reflect on their answers.The use of questionnaires rather than focus group or individual interviews allowed me to get information-rich data from a relative large number of participants within a short period of time.
There are two noteworthy limitations of this study: generalisability and a lack of in-depth follow-up interviews with some of the participants.The generalisabilty of the research findings is limited because findings were generated in a qualitative inquiry.The research design was not intended to produce results that account for or predict the behaviour of bullies, victims and/or bystanders.This liability was clear from the outset.However, because the inquiry generated new insights into educators' understanding of the risk factors for bullying, it could be used to generate recommendations on the reduction of bullying.This study should have been complemented by in-depth semi-structured interviews, which would have allowed (some of) the participants to 'tell their stories' as bystanders and onlookers of learner-onlearner bullying.

Conclusion
Educators are key role-players in the successful development and implementation of most anti-bullying programmes and/or intervention strategies.This article therefore set out to investigate educators' perspectives of the risk factors for learner-on-learner bullying.The study used a fourlevel, social-ecological model as a theoretical framework.Individual, home environment, school and societal risk factors for learner-on-learner bullying were identified by means of qualitative content analysis.
Risk-focused prevention may lead to a reduction in bullying (Baldry & Farrington 2000:18).Strategies to prevent bullying should not only target these (negative) factors, but try to promote 'protective factors' (Cowie & Jennifer 2008:18).In the subsequent discussion the need to instil protective factors will be highlighted.

Individual level
Participants' perspectives of risk factors correspond, to a large extent, with the clichéd image of bullies as tough, arrogant, underachieving boys who vent their anger and frustration on their weaker peers.Victims were stereotypically portrayed as socially awkward, physically weak loners who stand out in a crowd.Protecting factors, on the other hand, were found to be a resilient temperament, a sense of self-sufficiency, a positive, outgoing disposition and high intelligence (Cowie & Jennifer 2008:18).Schools should therefore try to create a restorative forum for victims and bullies.This forum should create opportunities for open-ended, honest and respectful dialogue between victims and bullies.Respectful dialogue will provide opportunities to connect people, model acceptable social behaviour and maintain caring relations.
Such a forum will also give learners the chance to gain skills in care giving and the capacity to care for those who are 'different'.Weeks (2008:124) rightly emphasises that learners should also be given the opportunity to encourage the best in others and 'improve the self'.Venter and Du Plessis (2012) believe that learners should be guided to treat their fellow human beings humanely and with respect, because:

Family dynamics
This study identified the destructive and violent behaviour of family members, an authoritarian parenting style, apathy and negative role models as factors perpetuating aggressive behaviour.Victims, on the other hand, grow up with overindulgent parents as well as with people who are seen as 'different'.To counteract learner-on-learner bullying, schools need involved parents or guardians who are willing to question and, if necessary, change their own parenting styles and behaviour and strive for the creation of households rooted in protective factors (Cowie & Jennifer 2008:18).Protective factors can be found in, among other things, a strong sense of attachment to one or both parents, characterised by a stable, warm and affectionate relationship; parents who maintain a strong interest in their children's education; open dialogue; parents who provide effective supervision, clear rules and consistent discipline; parents who model values that are embedded in the ethics of caring (honesty, fairness, respect for others, responsibility, kindness and modesty) (Weeks 2008:126); and also in recognition and due praise.

Teaching and learning milieu
Some of the educators who took part in this study described cruel disciplinary practices that may promote bullying.Educators, who are supposed to play a key role in preventing bullying, are often the instigators -albeit unintentionally in some instances -of bullying.Oosthuizen et al. (2003:388) correctly advise educators to enter into loving, caring, serving, guiding and 'disciplining' relationships with learners, rather than resorting to malicious chastisement.This means serving learners in a spirit of self-sacrifice to become 'well-rounded and fully educated individuals capable of responding adequately to their callings in life and able to be true disciples, i.e. followers of Jesus Christ'.
Every school should, therefore, be transformed into a caring community that actively develops protective factors, such as opportunities for learners to seek advice from their educators; learners who are given ample opportunities to be involved in school activities; educators who provide efficient supervision, unambiguous rules and consistent discipline; educators who model positive, caring behaviour, and who acknowledge learners' achievements (Cowie & Jennifer 2008:18).

Community level
Although the participants paid little attention to the impact of community factors on bullying, they highlighted the lack of positive values in the community as a risk factor for bullying.
In a country that is experiencing 'an intense moral crisis' (De Klerk & Rens 2003:353), it is important that community leaders lead by example and exhibit healthy social attitudes towards anti-social and criminal behaviour.

TABLE 1 :
Demographic details of participants who described incidence of learner-on-learner bullying (n = 91).
n, number of participants.
Jesus came into this world, not as an imposter of humanity, nor as a dim reflection of humanity … but as a living, fully functional and holistic representative of humanity.(Iselin& Meteyard, in  Venter & Du Plessis 2012:2)