Download Bullying vs. Cyberbullying: Participant Roles and Intervention Effectiveness and more Lecture notes Communication in PDF only on Docsity! Master Thesis Communication Studies Joanne Amse 1 Evaluation of the effectiveness of the anti-bullying intervention program ‘Survivors!’ The effectiveness of anti-bullying school intervention programs targeting bulling as group process by the adaption of the Participant Role Approach Joanne Amse s1128566 Examination Committee Dr. Joris van Hoof Niels Baas MSc. 25 August 2014 Communication Studies Master Thesis Communication Studies Joanne Amse 2 INDEX Abstract ..................................................................................................................................... 3 Samenvatting ............................................................................................................................ 4 Introduction .............................................................................................................................. 5 Theoretical framework ............................................................................................................ 7 Method ..................................................................................................................................... 17 Results ..................................................................................................................................... 25 Discussion ................................................................................................................................ 40 References ............................................................................................................................... 49 Appendix ................................................................................................................................. 52 Appendix A Information letter parents ................................................................................. 52 Appendix B Information letter teachers ............................................................................... 54 Appendix C Questionnaire ................................................................................................... 56 Appendix D Tables Self-reported and Behavioral Participant Roles Gender ...................... 60 Master Thesis Communication Studies Joanne Amse 5 INTRODUCTION Bullying is a universal phenomenon which has received more attention over the years of researchers in various countries all over the world. According to the widely recognized and cited definition of bullying of Olweus (1997) “a student is being bullied or victimized when he or she is exposed, repeatedly and over time, to negative actions on the part of one or more other students” (p. 496). Recently bullying is often considered as a group process where bullying can be defined as “a subtype of aggressive behavior, in which an individual or a group of individuals repeatedly attacks, humiliates, and/or excludes a relatively powerless person” (Salmivalli, 2010, p. 112). Aggressive behavior of bullying is hereby differentiated from teasing by three key characteristics: the intention to harm, the repetition over time, and the power imbalance between the bully and victim (Olweus, 1997). Thus, although bullying is a form of aggressive behavior, not all aggressive behavior between students can be considered as bullying (i.e. teasing). Bullying can take on many different forms. Generally, bullying is categorized into direct and indirect bullying which respectively takes place in a public, often physical context (e.g. fighting, pushing) or in a relational, often mental context (e.g. social exclusion, gossiping, spreading rumors) (Olweus, 1997). In general, direct bullying with physical aggression and threats is more common among boys (boys: 17.4% vs. girls: 14.6%). Girls however are more often involved in indirect bullying situations with more verbal and relational ways of harassment (boys: 8.2% vs. girls: 10.2%) (Olweus, 1993, 1997; Rivers & Smith, 1994). A relatively new, but rapidly growing phenomenon among young people is cyberbullying, which includes bullying by the use of electronic or digital information and communication technologies (i.e. mobile phones, Internet) (Baas, De Jong, & Drossaert, 2013). Because of the differences between traditional bullying and cyberbullying regarding the potential harmfulness, threatening nature, potential audience, and visibility (Baas et al., 2013), it can be questioned whether cyberbullying has to be considered as a new form of bullying or just as bullying via a new medium. In the review of Slonje, Smith, and Frisén (2013), the gender differences for cyberbullying are inconsistent across the various studies (e.g. Smith, 2012; Tokunaga, 2010). A growing body of research has demonstrated the negative effects of bullying on mental and physical health for both bullies and their victims. The exposure to violence in schools is related to the development of: 1) emotional and psychosomatic problems; 2) low self-esteem, depression and suicidal tendency; 3) antisocial behaviors which lead to legal, economic, and social problems (Jiménez Barbero, Ruiz Hernández, Llor Esteban, & Pérez García, 2012). Peer victimization specifically has a negative effect on anxiety, depression, social dysfunction, physical wellbeing, and suicidal tendency (Rigby, 2001). As people seem to be much more aware of the negative consequences of bullying for children’s development, consequently the need for intervention programs aimed at preventing or reducing bullying at school has increased as well (Jiménez Barbero et al., 2012). The Dutch government acknowledges this urgent need for effective anti-bullying intervention programs in order to create a safe school climate. Therefore, the “Plan against bullying” is presented on 25 March 2013 to the House of Representatives by the Ministry of Education, Culture and Science in intensive cooperation with the National Children’s Ombudsman. This plan is a response to the many different approaches to bullying among schools and the overload of anti-bullying programs, which are important reasons why bullying is so difficult to fight. Master Thesis Communication Studies Joanne Amse 6 Not all anti-bullying programs have been proven to be effective or even could be contra productive. Especially for secondary education, a proven program and the systematic use of programs seem to be missing. With this plan the Dutch government wants to reduce the overload of anti-bullying programs to a limited number of programs that really work and that emphasize the prevention of all kinds of bullying. Through the legal obligation of a scientifically and empirically proven effective anti-bullying program, the government has the ambition to end the taboo of bullying so that bullying can no longer be ignored (Plan against bullying, 2013). One of these anti-bullying programs from which the effectiveness has not yet been scientifically and empirically examined is the anti-bullying program ‘Survivors!’. This classroom anti-bullying program focuses on the group process of bullying, by emphasizing the important role of defending behavior of bystanders to stop the bullying. The objectives of ‘Survivors!’ can be summarized as following: 1) raise awareness of the own and others’ participant role behavior in bullying situations; 2) raise awareness of the personal and shared responsibility for the atmosphere in the classroom and the protection of (online) boundaries; 3) expand knowledge about bullying and what to do against (cyber)bullying situations; 4) increase the anti-bullying attitudes related to bullying; 5) improve the self-efficacy and outcome expectations for defending behavior. Therefore, in this study the intervention effects were examined for several outcome variables indicating the participant role behaviors, awareness, knowledge, anti-bullying attitude, self-efficacy, and outcome expectations related to bullying and defending behavior in bullying situations. The aim of the present study is to expand general knowledge about the effectiveness of anti-bullying school intervention programs targeting bulling as a group process by examining the effects of the anti-bullying intervention program ‘Survivors!’. Master Thesis Communication Studies Joanne Amse 7 THEORETICAL FRAMEWORK In this theoretical framework, the theoretical perspective and most important constructs included in this study are introduced. At first, bullying is described as a group process which highlights the relevance of the current study. Next, defending behavior in bullying situations and the personal and social factors that could influence the defending behavior of bystanders are described in detail. Additionally, important empirical findings from other intervention programs and the use of theatre in educational interventions are illustrated. Finally, the conceptual research model of the present study is introduced. Bullying as Group Process Empirical research to investigate the possible influence of the group in maintaining and reinforcing bullying has started around the 1990s (Salmivalli & Voeten, 2004). The investigation of bullying from a group perspective embraces the involvement of the whole group in the bullying process, including the bystanders who are present in most of the bullying situations. Bystanders can be described as people who are direct or indirect witnessing the bullying process, but don’t do anything to stop it (Salmivalli, Kaukiainen, & Voeten, 2005). The important role of the bystander response in witnessing and encouraging the bullying process and the lack of support of bystanders to the victim is further explored by the different participant roles of the Participant Role Approach (PRA) (Salmivalli, Lagerspetz, Björkqvist, Österman, & Kaukiainen, 1996). Group Perspective Bullying at school rarely happens with the bully and victim being the only parties involved. It is not an isolated process between the bully and victim (Salmivalli et al., 1996). In most cases there are many more students present at the bullying scene, which in one way or another affects the bullying situation. Observational research of bullying at the schoolyard from O’Connell, Pepler, & Craig (1999) has shown that in no less than 85% of the bullying incidents bystanders had a reinforcing effect by joining or encouraging the bullying. Bullying is therefore a group process in a broad social context with more students involved than only the bully and victim. So in recent years there has been a shift in research addressing the bullying problem from viewing bullying as individual differences in dyads to approaching bullying in relation to the social group or context in which it occurs (Rigby & Johnson, 2006; O’Connell et al., 1999). Recently more researchers emphasize the social character of bullying by considering bullying as a complex group phenomenon that involves social roles and relational processes within the group as well (Salmivalli, 1999, 2001; Sutton & Smith, 1999). Salmivalli (1999, 2001) refers in her articles to earlier research of Lagerspetz, Björkqvist, Berts, and King (1982) and Pikas (1975) who already viewed bullying as a primarily social phenomenon within the context of the whole group, for example a classroom. Lagerspetz et al. (1982) highlighted the collective character of bullying and its foundation on social relationships in the group. Pikas (1975) emphasized the reinforcement of each other’s behavior due to their interaction in bullying situations. Master Thesis Communication Studies Joanne Amse 10 The importance of teacher knowledge on bullying has been addressed in bullying research more often (Allen, 2010). Nicolaides, Toda, and Smith (2002) emphasize the value of the incorporation of existing knowledge about bullying into the training programs for teachers. Subjects that should be included are for example the characteristics of bullies and victims, coping strategies, the importance of the issue, and self-confidence in dealing with bullying issues. However, students’ perceptions and understanding of bullying seem to be different than those of teachers. These differences are expressed in discrepancies between prevalence rates of bullying, the perceived impact of different forms of bullying, and the main focus of intervention programs (Allen, 2010). Thus, variables related to knowledge among students should also be taken into account when measuring the effectiveness of intervention programs. Self-efficacy for Defending Behavior One of the individual differences in the behavior of bystanders refers to the sense of self- efficacy for defending behavior. According to the social cognitive theory of Bandura (1997), self-efficacy can be conceptualized as individual’s belief in his/her own ability to reach desired results through their own actions. The influence of self-efficacy beliefs has been proven to have a positive impact on various domains of pro-social functioning, interpersonal relations, and wellbeing (Bandura, 1997). In the context of the present study self-efficacy can best be operationalized as self-efficacy for defending behavior (Salmivalli et al., 1996). In order to help the victim in an often perceived difficult or even dangerous situation as bullying, bystanders need to have a certain degree of confidence in their own ability to successfully intervene in the situation. Bystanders need to have a lot of courage to stand up against bullies as bullies are often perceived as very powerful and popular (Salmivalli, 2010). If bystanders don’t believe in their own capability to defend or support a victimized peer against these bullies, they will be unlikely to do so (Pöyhönen, Juvonen, & Salmivalli, 2012). Salmivalli (2010) points out a negative relationship between self-efficacy for defending and withdrawing or staying out from a bullying situation. Moreover, a positive relationship was found between self-efficacy for defending and standing up for the victim. Andreou, Vlachou, and Didaskalou (2005) indicated specifically higher levels of self-efficacy for students in the defender role opposed to the students in the role of outsider. Outcome expectations for Defending Behavior Besides the individuals’ believe in their own ability to take actions for defending behavior, the belief in the desired outcome of their defending behavior could play a role as well. The outcome expectations of defending behavior has been associated with specific outcomes concerning the frequency of bullying, the victim’s wellbeing, and one’s personal status (Pöyhönen et al., 2012). Pöyhönen et al. (2012) argue that it is important to investigate these outcome expectations for defending behavior separately from efficacy beliefs because these expectations can hold back bystanders from intervening even if they have a high sense of self-efficacy. The defending strategies can then be targeted to either the stronger belief of bystanders in themselves or in the difference that they can make by defending behavior. The research of Pöyhönen et al. (2012) regarding bystander responses in bullying situations shows various motivational underpinnings for defending, remaining passive, and reinforcing the bully. Defending was linked to the expectation that the victim will feel better and one’s own status will improve. Remaining passive was associated with conflicting expectations and values. These bystanders may value the outcome expectations but don’t trust it will happen which results in withdrawing from the bullying situation. Reinforcing the bully was related to negative outcome expectations. These students didn’t care whether the bullying decreased or expected that defending the victim would be bad for their own status. Master Thesis Communication Studies Joanne Amse 11 Attitude Another factor that could play an important role in relation to bystander responses is the attitude towards bullying. Rigby and Slee (1991, 1993) and Rigby (1997) examined the attitudes and beliefs of Australian school children towards bullying in general and the victims of bullying. They used items of the Pro-victim Scale related to three factors: the tendency to despise the victims of bullies, general approval for school bullies, and avowed support for intervention to assist the victim. Results have shown that a positive attitude towards victims was negatively correlated with supporting bullying behavior. In contrast to this, the positive attitude towards victims was positively correlated with expressing approval of others who intervened to put an end to the bullying (for example teachers, mentors, and other students). This has led to the expectation that the positive attitude towards victims would also correlate positively with the willingness of bystanders to help victims in a bullying situation. This belief is, however, based on the assumption that bystanders are more willing to help the victim if they like the victim or have some feelings of empathy for him/her (Rigby & Johnson, 2006). Additionally, the attitude towards victims in general or towards specific victims in the own school class could be different. The social role of the victim within the class effects how others see the victim and thus the possibilities to connect with peers (Salmivalli, 2010). Specific research on participant roles of Salmivalli and Voeten (2004) indicated more approving attitudes with regard to bullying among children in pro-bullying roles such as bullies, reinforcers and assistants. Strong attitudes against bullying among children were indicated among children in the anti-bullying role of defender. Awareness The influence of the group context at school, i.e. the classroom, does often have a bigger impact than other social groups due to its involuntary membership. This means that the victim and other group members cannot easily escape from bullying situations at school, because they are classmates or go to the same school (Salmivalli, 2010). The fact that bullying situations at school do often have multiple witnesses reduces the likelihood of bystander intervention. This phenomenon is also referred to as the “bystander effect” (Darley & Latane, 1968, cited by Salmivalli, 2010). It might be a consequence of the bystanders’ diffusion of responsibility or their incorrect interpretation of the bullying situation (Salmivalli, 2010). Bystanders seem to be unaware of their personal responsibility to stop the bullying. Even if they are aware of their own responsibility to do something against it they seem to ignore it. They expect others to take action instead of themselves. This lack of intervention by other bystanders leads also to a misjudgment of the seriousness of the situation and suffering of the victim, especially with indirect bullying. If bystanders monitor the behavior of other bystanders who don’t intervene, they often think it is unnecessary to intervene. This copying behavior of not intervening gives wrong signs to the bully, victim, and other bystanders about the personal and shared responsibility to stop the bullying. Master Thesis Communication Studies Joanne Amse 12 Anti-bullying Intervention Programs The shift in recent research to emphasize the critical role of bystanders in the interaction between the bully and victim in the bullying process has influenced the approach of anti- bullying interventions as well. A number of anti-bullying intervention programs which take on this group approach to bullying are described by Olweus (1997), Salmivalli et al. (2005), Kärna et al. (2011), and Van der Meer (2013). Additionally, the use of (participatory) theatre in intervention programs is also discussed. Better understanding of the key principles and effectiveness of these intervention programs at class-, group, and school level will provide more information for examining the effectiveness of ‘Survivors!’. Group Approach Intervention Programs It has been shown that attempts to change the behavior of bullies do rarely have long-lasting effects (Salmivalli et al., 2005). It might be easier to express the already existing anti-bullying attitudes of bystanders and transform these attitudes into actual behavior than to influence both the attitudes and behavior of active, initiative taking bullies (Salmivalli, 2010). Therefore the focus of interventions has shifted to influence the behavior of bystanders in the bullying process. This new focus has led to the notion of researchers and policy makers that interventions should no longer only be directed to the individual and dyadic characteristics of the bully and victim, but should recognize bullying as a group process. Many anti-bullying interventions now target the group as a whole by trying to influence the behavior of bystanders in the bullying process (Pepler, Craig, Ziegler, & Charach, 1994; Salmivalli, 1999; Salmivalli et al., 1996; Stevens, Bourdeaudhuij, & Oost, 2000; Sutton & Smith, 1999). Olweus Bullying Prevention Program (OBPP) As the majority of the intervention research has been carried out in school context, most of the anti-bullying intervention programs have been inspired and/or modeled after the whole-school approach by Dan Olweus (Bauer, Lozano, & Rivara, 2007; Smith, Schneider, Smith, & Ananiadou, 2004; Merrell, Gueldner, Ross, & Isava, 2008; Salmivalli et al., 2005). Olweus (1994, 1997) argues that a suitable anti-bullying intervention program can reduce bully/victim problems by focusing primarily on changing attitudes, behavior, and routines around bullying. The Olweus Bullying Prevention Program (OBPP) is therefore developed to improve peer relations and restructure the social school environment into a safe and positive place. By the promotion of school-wide awareness of bullying the OBPP aims to increase awareness and knowledge about the bully/victim problem, achieve active involvement of teachers and parents, develop clear rules against bullying, and provide support and protection for the victims (Olweus, 1994, 1997). The intervention program of Olweus (1997) is based upon four key principles, ideally carried out by teachers at school and parents at home: 1) the creation of an environment with warmth, positive interest and involvement from adults; 2) clear limits and rules to unacceptable behavior; 3) consistent application of non-hostile, non-physical sanctions by rule violation; 4) the behavior of adults as authorities. These basic principles are translated into explicit measures targeted at three systematic levels: school level, class level, and individual level. Master Thesis Communication Studies Joanne Amse 15 The Current Study The goal of the present study is to expand the knowledge about the effectiveness of anti- bullying school intervention programs targeting bulling as group process by examining the effects of the anti-bullying intervention program ‘Survivors!’. The intervention effects were examined for several outcome variables indicating the participant role behaviors, awareness, knowledge, anti-bullying attitude, self-efficacy, and outcome expectations related to bullying and defending behavior in bullying situations. By examining bullying at these aspects from a group perspective, insight will be obtained into how and what should be changed at peer- group level (Salmivalli, 2010). In the present study the effectiveness of the anti-bullying program ‘Survivors’ is examined. The aim of the anti-bullying program ‘Survivors!’ is to encourage the discussion about (cyber)bullying and to reduce actual bullying behavior among first year students of secondary school. The students are also encouraged to report bullying incidents. The focus is on bullying as a group process, whereby the important role of bystanders by reinforcing or stopping the bullying is emphasized. The intervention program of “Survivors!” is developed on the basis of the so called “five-track approach” from psychologist and bullying expert Bob van der Meer. The five- track approach is based upon three psychological mechanisms: the conspiracy to keep silence (the fear to be seen as betrayer), the bystanders’ dilemma (the diffusion of responsibility to intervene), and blaming the victim (the justification of bullying). The philosophy behind this approach and these mechanisms is the belief that the bullying will stop when the group that supports the victim is larger than the group behind the bully (Van der Meer, 2013). The approach is targeted to the ‘silent majority’ which consists of five subgroups which are comparable with the participant roles of Salmivalli et al. (1996): students who join the bully because they are afraid of the bully (reinforcer or assistant), students who join the bully because of their own benefit (reinforcer or assistant), students who don’t join or stop the bully (outsider), students who don’t know or see the bullying (outsider), and students with a high social status who sometimes helps the victim (defender). The other parties involved in the bullying process are the victim, the bully, the parents and the teachers. According to Van der Meer (2013), the mobilization of this silent majority forms the key to raise awareness of the bully that his/her bullying behavior is unacceptable. When the silent majority speaks up against bullying, possibly supported by parents or teachers, the situation to stop the bullying in the long run is optimal. Subsequently, he defined the following key points: the overall responsibility of the school; provide support to the child who is being bullied; provide support to the child who is bullying; involve the middle group (the rest of the class) by the solutions for the bullying problem; provide support to the parents of the child who is being bullied and the child who is bullying. Van der Meer (2013) claims that a good intervention approach against bullying encounters three requirements. First, the approach is integral in two ways: the involvement of all parties involved (five-track approach) and the approach of bullying within the explanatory model of violence. Second, the problem is approached structurally. Third, the approach leads to a long-term change in attitude. Master Thesis Communication Studies Joanne Amse 16 According to Salmivalli et al. (2005), the possible influence of the initial participant roles before the intervention should be taken into account by further interpretation of the changes due to the intervention. In order to assess this possible influence of the initial self- reported participant roles on the effects of the intervention, the moderation effect of the self- reported participant roles is also examined in this research (see Figure 1). Merrell et al. (2008) pointed out that future intervention studies should measure the knowledge and perception on bullying as well as the actual bullying behavior. Therefore the association between the awareness, knowledge, anti-bullying attitude, self-efficacy, outcome expectations and the participants’ reported defending behavior in bullying situations is also studied. The effects of the intervention were examined by comparing the results at two time points: before the intervention and after the intervention. The results were based on self- reports of the participants on the various outcome variables. It was hypothesized that the intervention would increase the awareness, knowledge, anti-bullying attitude, self-efficacy, outcome expectations related to bullying. Additionally, it was expected that improvement of these outcome variables would increase the participants’ defending behavior in bullying situations. Furthermore it was hypothesized that through the intervention the level of agreement between the self-reported and behavioral participant roles would increase. Conceptual Research Model Figure 1 Conceptual Research Model Intervention Effects and Moderator Effect Self-reported Participant Roles Master Thesis Communication Studies Joanne Amse 17 METHOD In the present study the effectiveness of the anti-bullying intervention program ‘Survivors!’ is examined in a within-subject research design with a pretest and a posttest for an experimental group. Next to the background variables and the bullying experiences, the following six dependent variables were measured: Self-reported and Behavioral Participant Roles, Awareness, Knowledge, Attitude, Self-efficacy for Defending Behavior, and Outcome Expectations for Defending Behavior. Anti-bullying Intervention Program ‘Survivors!’ ‘Survivors!’ is the anti-bullying intervention program of the organization Switch, which is part of Youth for Christ (YfC). YfC is an organization that focuses on helping young people between 10 and 23 years old with social issues at school, on the streets and in the church. Switch provides theater, community internships and youth workers at school and focuses on the following themes: bullying, social media, identity, justice and poverty, and religion. The role of religion does not explicitly comes forward in all theater shows, including the intervention program ‘Survivors!’. The intervention program ‘Survivors!’ is focused on the theme bullying and consists of a theater show, talk show and lesson. During the theater show diverse direct, indirect and cyberbullying situations at school are enacted by one of the two groups of actors of Switch (Group Blue/Group Red). After that the actors of the theater show are the guests of the talk show. During the talk show the bullying situations of the theater show are discussed and several statements are presented to the students of the school who serve as the audience of the talk show, for example “How do you keep the atmosphere in your class enjoyable for everyone?”. The students have the opportunity to react and give their opinion about the statements. After the theater and talk show the students have a lesson with their own class led by one of the actors of the theater show. The goal of the lesson is to expand the knowledge and skills of students on how to behave in (cyber)bullying situations and how to stop it. The participant roles, causes and solutions in bullying situations are discussed. Therefore the content and characters of the theater show are deliberated and linked to the own experiences of the students. A platform is provided for students to share their personal bullying experiences. Through exercises the students then get to practice on how to stand confident, say “stop” to bullies, ask for help, and protect their privacy on the Internet. The objectives of the anti-bullying intervention program of ‘Survivors!’ can be summarized as followed: 1) raise awareness of the own and others’ participant role behavior in bullying situations; 2) raise awareness of the personal and shared responsibility for the atmosphere in the classroom and the protection of (online) boundaries; 3) expand knowledge about bullying and what to do against (cyber)bullying situations; 4) increase the anti-bullying attitudes related to bullying; 5) improve the self-efficacy and outcome expectations for defending behavior. Master Thesis Communication Studies Joanne Amse 20 Regardless the different roles in which participants can experience bullying, the bullying experiences of the participants can also be compared on the basis of the different forms of bullying. In general, 79.4% of the participants (N = 932) have experienced any form of indirect bullying, 67.6% of the participants (N = 794) have experienced direct bullying, and 50.4% of the participants (N = 592) have experienced cyberbullying. There were only significant sex differences for the experience of direct bullying (t (784) = 7.971; p < .001), whereby boys experience more direct bullying than girls (M = 1.70 vs. M = 1.44). There were no significant sex differences found for the experience of indirect bullying (M = 1.57 vs. M = 1.53) (t (930) = 1.379; p = .168) and cyberbullying (M = 1.47 vs. M = 1.50) (t (590) = - .850; p = .396). The significant higher number of boys who experience direct bullying is consistent with previous research. In general direct bullying is more common among boys, while girls are more often involved in indirect bullying situations (Olweus, 2010). This difference was however not confirmed in this research. The gender differences for cyberbullying are however inconsistent across various studies, so these results cannot be verified with previous research (Slonje, Smith, & Frisén, 2012). Of the participants who have indicated that they have experienced bullying as an outsider, the participants most often witnessed a bullying situation in which other students left someone out on purpose, called mean names or spread rumors. This kind of indirect bullying was also experienced the most by the participants who have experienced bullying as a victim. The most frequent form of experienced bullying by the participants who have experienced bullying in the role of the bully was direct bullying like pushing, kicking or hitting someone. Measurements Self-reported Bullying, Self-reported Victimization, and Self-reported Observed Bullying To measure the experience of bullying, victimization and observed bullying from different points of view, the same three questions about the bullying experiences were formulated from the perspective of an outsider, bully or victim. Questions for direct bullying, indirect bullying, and cyberbullying from the revised Bully/Victim Questionnaire (Olweus, 1996) were used. E.g. for the point of view of the victim, the questions were formulated as followed: ‘I was deliberately left out, called mean names, or there were spread rumors about me’; ‘I was pushed, kicked or beaten up’; ‘I received unpleasant phone calls, messages or pictures via my phone or the Internet’. The participants answered on a 5-point Likert scale (0 = never, 5 = very often). Self-reported and Behavioral Participant Roles The participants were asked to categorize themselves into one of the bullying participant roles they thought applied best to their behavior (Salmivalli & Voeten, 2004). The participants were presented with the following bullying participant roles and corresponding descriptions: victim (‘I was bullied’), bully (‘I was bullying others’), passive reinforcer (‘I was present when someone else was being bullied; laughed and teased sometimes with the bully, but never started the bullying myself’), active reinforcer (‘I joined in the bullying’), defender (‘I tried to stop the bullying’), outsider (‘I saw the bullying happening, but didn’t interfered in the situation’). Master Thesis Communication Studies Joanne Amse 21 Furthermore, the revised 15-item version of the Participant Role Questionnaire (PRQ) (Salmivalli & Voeten, 2004) was used. Participants were asked to think back of a previous bullying experience in school and evaluate their own behavior in such situations on a 5-point Likert scale (0 = never, 5 = very often). For every participant role, except for the role of victim, three different items from the PRQ were used. Dependent on the internal consistency of these three items, these items were combined into scales reflecting the different participant roles. Originally, the PRQ consists of five scales: the bully scale, the assistant scale, the reinforcer scale, the defender scale, and the outsider scale (Salmivalli & Voeten, 2004). In the present study, the internal consistency of these existing scales was however insufficient. The Cronbach’s α coefficients for these scales in the present study were respectively .67 for the bully scale, .57 for the assistant scale, .63 for the reinforcer scale, .81 for the defender scale, and .38 for the outsider scale. A factor analysis was performed in order to determine which combination of items and scales had a higher internal consistency. By the use of factor analysis, the following three scales could be distinguished: the bully scale, the reinforcer scale, and the defender scale. For this study, the scales for the participant role of assistant (in this research referred to as active reinforcer) and reinforcer (in this research referred to as passive reinforcer) were combined into one reinforcer scale. Sutton and Smith (1999) and Goossens et al. (2006) used this classification of the reinforcer scale as well to make a clear distinction between active, initiative taking bullies and others who join the bullying in a later stadium. The bully scale (α = .67) consists of items that describe active, initiative-taking, leadership behavior to start the bullying: ‘I start the bullying’; ‘I always find new ways of bullying the victim’; ‘I make the others join in the bullying’. The reinforcer scale (α = 0.73) consists of items that describe behavior which actively or passively reinforce the bullying: ‘I come around to watch when someone is being bullied’; ‘I come around to laugh when someone is being bullied’; ‘I come around to scream or shout when someone is being bullied, i.e. Come to see! Someone is beaten up here!’; ‘I join in the bullying, when someone else has started it’; ‘I help the bully’; ‘I hold the victim, so that he/she can be bullied’. The defender scale (α = .82) consists of items that describe behavior that supports and defends the victim in bullying situations: ‘I comfort the victim or encourages him/her to tell others about the bullying, i.e. teachers, parents’; ‘I try to stop bullying myself’; ‘I convince the others to stop bullying’. The eventual Cronbach’s α coefficients for the bully, reinforcer, and defender scale in the present sample were considered to be satisfactory based on previous research of Salmivalli et al. (2005). In their research, all outcomes on the revised 15-item version of the PRQ were based on self-report as well with similar internal consistencies for the bully scale (α = .68), assistant scale (α = .67), reinforcer scale (α = .67), defender scale (α = .79), and outsider scale (α = .60). Next to the bully, reinforcer, and defender scale, there were three items in the questionnaire that described behavior which can be considered as remaining passive or withdrawing in bullying situations: ‘I don’t take sides with anyone during the bullying situation’; ‘I ignore the bullying/I pretend not to notice that someone is being bullied’; ‘I am usually bot present in the bullying situation’. Because of the low internal consistency of these three items together (α = .38), these items were not put together into one outsider scale. Instead, the separate item ‘I ignore the bullying/I pretend not to notice that someone is being bullied’ was used to describe the role of outsider in bullying situations. This item had the most comprehensive description of the application of the outsider strategy in bullying situations. Furthermore, this item corresponded most with the general description of the outsider at the beginning of the questionnaire (‘I saw the bullying happening, but didn’t interfered in the situation’). Master Thesis Communication Studies Joanne Amse 22 Awareness To examine how aware participants were of their own and shared responsibility for the atmosphere in the classroom and the protection of (online) boundaries, they were presented with eight questions derived from the four central objectives of Survivors! (Switch, 2013). For each objective two questions were formulated: 1) awareness of the shared responsibility for the atmosphere in the classroom (‘It’s my task to keep the atmosphere in the classroom well’; ‘It’s the task of the whole class together to keep the atmosphere in the classroom well’); 2) awareness of the personal responsibility to set your own boundaries (‘It’s my task to say stop’; ‘It’s my task to set my own boundaries’); 3) awareness of the personal responsibility to respect the boundaries of others (‘I have to stop with what I was doing when the other says stop’; ‘I have to respect other people’s boundaries and take them seriously, also on the Internet’); and 4) awareness of the possibilities to protect your privacy at the Internet (‘It’s my task to protect my own privacy on the Internet’; ‘It’s my task to set my own boundaries on the Internet’). The participants had to evaluate on a 5-point Likert scale (0 = strongly disagree, 5 = strongly agree) to what extent they agreed or disagreed on the statements. Mean values for all the separate statements were used in the analysis. Knowledge The knowledge about bullying of the participants was measured in the form of a quiz consisting of 10 MC-questions with three answer options per question. The questions were derived from the three knowledge objectives of Survivors! (Switch, 2013): 1) the participants know what (cyber)bullying is (e.g. ‘The three main forms of bullying are…’; ‘Cyberbullying is…’); 2) the participants know the difference between teasing and bullying (e.g. ‘The difference between teasing and bullying is determined by…’); 3) the students know what to do against (cyber)bullying situations (e.g. ‘If other students don’t stop with bullying, even if I said stop, the best thing that I can do is…’; ‘When I am being bullied at the Internet, the best thing that I can do is…’). A total score of right answers was calculated to create a final grade between 0 and 10. Attitude The revised Pro-victim Scale (Rigby, 1997) was used to measure the student’s positive and negative attitude towards bullying. Based on factor analysis, two items from the revised Pro- victim Scale were excluded from analysis (‘Kids who bully others weaker than themselves should be bullied themselves’; ‘Kids should not bully others who are weaker than themselves’). Furthermore, the questionnaire was expanded with an item regarding the attitude towards own defending behavior in bullying situations to better fit the present study (‘I have to try to help the one who is being bullied’). The final overall attitude scale (α = .72) consists of 9 items with a 5-point Likert scale (0 = strongly disagree, 5 = strongly agree). Factor analysis of all items of the attitude scale revealed a pro-bullying scale and anti-bullying scale. The pro-bullying scale (α = .69) consists of the items: Soft kids are stupid (*); ‘Kids should not complain about being bullied’ (*); ‘Nobody likes a wimp’ (*); ‘It’s funny to see kids get upset when they are bullied’ (*); ‘Kids who are being bullied usually deserve it’ (*). The anti-bullying scale (α = .66) consists of the items: ‘A bully is really a coward’; ‘I like it when someone sticks up for kids who are being bullied’; ‘It makes me angry when a kid is bullied without reason’; ‘I have to try to help the one who is being bullied’. Master Thesis Communication Studies Joanne Amse 25 RESULTS The main goal of the present study was to expand the knowledge about the effectiveness of anti-bullying school intervention programs targeting bulling as group process by examining the effects of the anti-bullying intervention program ‘Survivors!’. In this results section, the intervention effects are described for respectively the participant roles, awareness, knowledge, attitude, self-efficacy for defending behavior, and outcome expectations for defending behavior. At the end, the moderator analysis of the self-reported participant roles and the regression analysis of the reported defending behavior in bullying situations are illustrated. Participant Roles Self-reported Participant Roles Table 2 shows the self-reported categorization of the participants into one of the participant roles in the bullying process. The participants who filled out more than one participant role or didn’t filled out the self-reported participant role in the pretest or posttest were excluded from analysis (N = 296). Results show that the most common participant roles were respectively outsider, defender, and victim. There were significant sex differences in the distribution of the participant roles for the pretest (χ 2 (5) = 43.356, p < .001) and the posttest (χ 2 (5) = 20.735, p = .001) (Appendix D Table 1). The sex differences show a similar pattern for the pretest and posttest. Only the percentage of active reinforcers changed between the sexes for the pretest (0.9% of the boys, 1.2% of the girls) and the posttest (1.3% of the boys, 1.2% of the girls). The participant roles of bully, passive reinforcer, and outsider were more frequent among boys, while among girls there were more defenders. The number of boys and girls who identified themselves as victims in the bullying process was also different for the pretest (13.5% of the boys, 14.6% of the girls) and the posttest (14.3% of the boys, 16.5% of the girls). Significant differences were found for the distribution of the self-reported participant roles between the pretest and the posttest (χ 2 (25) = 1458.288, p < .001) (Table 3). Overall 74% (N = 649) of the participants chose the same participant role for the pretest and posttest to describe their behavior in the bullying process. If we further analyze this level of agreement within the self-participant roles indicated by the pretest, the following distribution can be made: victim (82.9%), bully (80.0%), passive reinforcer (48.1%), active reinforcer (22.2%), defender (75.2%), outsider (74.6%). The biggest difference is shown within the relatively small group of active reinforcers (N = 9). Except the role of victim, these participants have shifted to all other participant roles of bully (N = 2), passive reinforcer (N = 3), defender (N = 1), and outsider (N = 1). Furthermore, most of the shifted self-reported victims, passive reinforcers, and defenders assessed their behavior the second time as an outsider (victims: 10.6%, passive reinforcer: 27.8%, defenders: 16.9%). Master Thesis Communication Studies Joanne Amse 26 Table 2 Number and Percentage of Self-reported Participant Roles Pretest and Posttest for Boys and Girls (% within Gender) Participant Role Pretest Posttest Victim 123 (14.0%) 135 (15.4%) Boys Girls 62 (13.5%) 61 (14.6%) 66 (14.3%) 69 (16.5%) Bully 15 (1.7%) 24 (2.7%) Boys Girls 12 (2.6%) 3 (0.7%) 15 (3.3%) 9 (2.2%) Passive reinforcer 54 (6.2%) 54 (6.2%) Boys Girls 42 (9.1%) 12 (2.9%) 36 (7.8%) 18 (4.3%) Active reinforcer 9 (1.0%) 11 (1.3%) Boys Girls 4 (0.9%) 5 (1.2%) 6 (1.3%) 5 (1.2%) Defender 331 (37.7%) 311 (35.4%) Boys Girls 135 (29.3%) 196 (46.9%) 136 (29.6%) 175 (41.9%) Outsider 364 (39.4%) 343 (39.1%) Boys Girls 205 (44.6%) 141 (33.7%) 201 (43.7%) 142 (34.0%) Total 878 (100.0%) 878 (100.0%) Boys Girls 460 (52.4%) 418 (47.6%) 460 (52.4%) 418 (47.6%) Table 3 Self-reported Participant Roles Pretest and Posttest (% within Self-reported Participant Roles Pretest and % of Total) Self-reported Participant Role (Pretest) vs. Self-reported Participant Role (Posttest) Self-reported Participant Role (Posttest) Self-reported Participant Role (Pretest)* Victim Bully Passive reinforcer Active reinforcer Defender Outsider Total Victim N % of PRpre % of Total 102 (82.9%) (11.6%) 0 (0.0%) (0.0%) 3 (5.6%) (0.3%) 0 (0.0%) (0.0%) 14 (4.2%) (1.6%) 16 (4.6%) (1.8%) 135 (15.4%) (15.4%) Bully N % of PRpre % of Total 0 (0.0%) (0.0%) 12 (80.0%) (1.4%) 3 (5.6%) (0.3%) 2 (22.2%) (0.2%) 4 (1.2%) (0.5%) 3 (0.9%) (0.3%) 24 (2.7%) (2.7%) Passive reinforcer N % of PRpre % of Total 1 (0.8%) (0.1%) 1 (6.7%) (0.1%) 26 (48.1%) (3.0%) 3 (33.3%) (0.3%) 7 (2.1%) (0.8%) 16 (4.6%) (1.8%) 154 (6.2%) (6.2%) Active reinforcer N % of PRpre % of Total 0 (0.0%) (0.0%) 2 (13.3%) (0.2%) 4 (7.4%) (0.5%) 2 (22.2%) (0.2%) 1 (0.3%) (0.1%) 2 (0.6%) (0.2%) 11 (1.3%) (1.3%) Defender N % of PRpre % of Total 7 (5.7%) (0.8%) 0 (0.0%) (0.0%) 3 (5.6%) (0.3%) 1 (11.1%) (0.1%) 249 (75.2%) (28.4%) 51 (14.7%) (5.8%) 311 (35.4%) (35.4%) Outsider N % of PRpre % of Total 13 (10.6%) (1.5%) 0 (0.0%) (0.0%) 15 (27.8%) (1.7%) 1 (11.1%) (0.1%) 56 (16.9%) (6.4%) 258 (74.6%) (29.4%) 343 (39.1%) (39.1%) Total N % of PRpre % of Total 123 (100.0%) (14.0%) 15 (100.0%) (1.7%) 54 (100.0%) (6.2%) 9 (100.0%) (1.0%) 331 (100.0%) (37.7%) 346 (100.0%) (39.4%) 878 (100.0%) (100.0%) *Self-reported Participant Role Pretest (PRpre) Master Thesis Communication Studies Joanne Amse 27 Behavioral Participant Roles To assign participants to a behavioral participant role based on the 15 items of the PRQ, a participant was considered to have a certain behavioral participant role if 1) he/she scored above the mean on the specified scale, and 2) he/she scored higher on that scale than on any of the other scales. If a participant did not score above average on any of the scales, so if a participant had a negative mean difference score on all the scales, he/she was considered not to have a clearly definable behavioral participant role. For the determination of the behavioral participant roles, the participants who characterized themselves as self-reported victims in the pretest or posttest were excluded from analysis (N = 156). The total number of participants left was 722 participants. Conform the procedure described above, it was possible to assign 71.2% (N = 514) of these participants to one of the following participant roles: bully, reinforcer (passive reinforcer and active reinforcer), defender or outsider. For the determination of the behavioral participant roles, the participant role of the passive and active reinforcer was combined into one participant role of reinforcer. Sutton and Smith (1999) and Goossens et al. (2006) used this clarification as well to make a clear distinction between active, initiative taking bullies and others who join the bullying in a later stadium. 28.8% (N = 208) of the participants had a negative difference score compared to the mean score on all participant role scales, so they were considered to have no particular behavioral participant role. Using the role selection criteria for the behavior of participants in a bullying situation, the distribution of the behavioral participant roles for the pretest and posttest is presented in Table 4. The most common behavioral participant roles for the pretest were respectively defender (N = 265), outsider (N = 121), reinforcer (N = 80), and bully (N = 48). For the posttest there were less bullies (N = 45) as well as outsiders (N = 98), while more participants were labeled as reinforcer (N = 106). The frequency of the behavioral participant role of defender (N = 265) was equal for the pretest and posttest. The sex differences between the behavioral participant roles show a similar pattern as the distribution of the self-reported participant roles for the pretest (χ 2 (4) = 72.054, p < .001) and the posttest (χ 2 (4) = 51.194, p < .001) (Appendix D Table 1). The behavioral participant roles of bully, reinforcer, and outsider were more frequent among the boys, while among the girls there were more defenders. There seemed to be significant differences in how the participants described their behavior in bullying situations between the pretest and posttest (χ 2 (16) = 484.618, p < .001) (Table 5). The behavior of 56.4% (N = 407) of the participants was assessed to the same role for the pretest and posttest. If we take a closer look at the levels of agreement within the different behavioral participant roles categorized by the pretest, the percentages of agreement were as follows: bully (35.4%), reinforcer (51.2%), defender (71.7%), outsider (41.3%), no participant role (54.4%). The biggest difference is shown within the group of outsiders. The behavior of these participants for the posttest was evaluated as a bully (5.0%), reinforcer (14.0%), defender (15.7%), and no participant role (24.0%). There were 109 participants who couldn’t be categorized in one of the behavioral participant roles for the pretest and posttest. The other 99 participant who weren’t categorized in one of the behavioral participant role the first time were characterized as a bully (N = 5), reinforcer (N = 23), defender (N = 48) or outsider (N = 23) the second time. Master Thesis Communication Studies Joanne Amse 30 Posttest Behavioral Participant Role* Self-reported Participant Role Bully Reinforcer Defender Outsider Total Bully N % of BPR % of Total 11 24.4% 2.1% 7 6.6% 1.4% 2 0.8% 0.4% 1 1.0% 0.2% 21 4.1% 4.1% Reinforcer N % of BPR % of Total 8 17.8% 1.6% 23 21.7% 4.5% 15 5.7% 2.9% 8 8.2% 1.6% 54 10.5% 10.5% Defender N % of BPR % of Total 9 20.0% 1.8% 18 17.0% 3.5% 194 73.2% 37.7% 12 12.2% 2.3% 233 45.3% 45.3% Outsider N % of BPR % of Total 17 37.8% 3.3% 58 54.7% 11.3% 54 20.4% 10.5% 77 78.6% 15.0% 206 40.1% 40.1% Total N % of BPR % of Total 45 100.0% 8.8% 106 100.0% 20.6% 265 100.0% 51.6% 98 100.0% 19.1% 514 100.0% 100.0% * Behavioral Participant Role (BPR) Secondary Roles of Victims To know more about how victims behaved in a bullying situation in which someone else was bullied, the secondary role of the victims was examined. Therefore the participants who identified themselves as a victim in the pretest or the posttest were selected (N = 156). To assign the secondary role of these self-reported victims, the same role selection criteria for the behavioral participant roles described earlier were used. Conform this procedure, it was possible to assign a behavioral participant role to 70.5% (N = 110) of the victims for the pretest and 66.7% (N = 104) of the victims for the posttest. The results show that the behavior of the self-reported victims in a bullying situation most corresponded with the behavioral role of defender (70.0% vs. 67.3%) and outsider (18.2% vs. 18.3%). The behavior of six participants of the pretest and two participants of the posttest described the secondary role of a bully. These participants can be characterized as the so called “bully/victims” as referred to in the anti-bullying literature. Master Thesis Communication Studies Joanne Amse 31 Awareness Awareness of the shared responsibility for the atmosphere in the classroom After the intervention the students are more convinced of their personal responsibility to keep the atmosphere in the classroom well than before the intervention (M = 2.53; M = 2.97). This difference is significant (t (1126) = 11.921, p < .001). However, this overall mean for all students after the intervention lies close to 3. This means that the average answer to the statement is ‘neither agree nor disagree’. There is a small negative significant difference (t (1127) = -2.291, p = .011) between the shared feeling of responsibility for a good atmosphere in the classroom before and after the intervention (M = 4.42; M = 4.33). In contradiction to their personal responsibility, the students are more convinced that it’s the task of the whole class together to keep the atmosphere in the classroom well. Awareness of the personal responsibility to set your own boundaries After the intervention the students are more aware of their personal responsibility to say ‘stop’ against bullying (M = 2.78; M = 3.46) (t (1129) = 16.203; p < .001). With an overall mean score of 2.78 before the intervention and an overall mean score of 3.46 after the intervention, the students strengthened their awareness on this matter. Before the intervention most students tended to react negative or neutral to the statement, while after the intervention more students agreed on the statement. There is no significant difference between the pretest and the posttest regarding the clarification of own boundaries (t (1122) = .906; p = .365). Before and after the intervention the students are well aware of their own responsibility to carry out their own boundaries (M = 4.47; M = 4.51). Awareness of the personal responsibility to respect the boundaries of others After the intervention the students agree more on the statement that they have to stop with what they were doing when other students say ‘stop’ (M = 4.37; M = 4.47) (t (1124) = 2.785; p = .0025). In other words, the students are more willing to react positively on other students in a bullying situation if others express their own boundaries by saying ‘stop’. The students argue that they have gotten more respect for other people’s boundaries and take them more seriously after the intervention, also on the Internet (t (1126) = 6.466; p < .001). The overall mean score of the students on this statement was respectively 4.22 and 4.47. Awareness of the possibilities to protect your privacy at the Internet After the intervention the students are less aware of their own task to protect their privacy on the Internet (t (1119) = -2.177, p = .015). This is shown by the negative difference in the overall mean score between the pretest (M = 4.37) and the posttest (M = 4.29). The students find it their task to set their own boundaries on the Internet (t (1130) = 3.915; p < .001). With an average mean score of 4.29 before the intervention and an average mean score of 4.44 after the intervention, the students seem to be well aware of the importance of protection of their online privacy by setting own boundaries on the Internet. Master Thesis Communication Studies Joanne Amse 32 Table 7 Means, Standard Deviations, and Confidence Intervals Awareness Awareness Pretest M (SD) Posttest M (SD) 95% CI Awareness of the shared responsibility for the atmosphere in the classroom It’s my task to keep the atmosphere in the classroom well. 2.53 (1.21) 2.97 (1.31)** [.368, .513] It’s the task of the whole class together to keep the atmosphere in the classroom well. 4.42 (1.10) 4.33 (1.16)* [-.165, -.013] Awareness of the personal responsibility to set your own Boundaries It’s my task to say ‘stop’. 2.78 (1.28) 3.46 (1.38)** [.602, .768] It’s my task to set my own boundaries 4.47 (1.01) 4.51 (1.02) [-.038, .104] Awareness of the personal responsibility to respect the boundaries of others I have to stop with what I was doing when the other says ‘stop’. 4.37 (1.05) 4.47 (1.03)* [.030, .174] I have to respect other people’s boundaries and take them seriously, also on the Internet. 4.22 (1.14) 4.47 (1.05)** [.172, .323] Awareness of the possibilities to protect your privacy at the Internet It’s my task to protect my own privacy on the Internet. 4.37 (1.01) 4.29 (1.10)* [-.156, -.008] It’s my task to set my own boundaries on the Internet. 4.29 (1.13 ) 4.44 (1.05)** [.076, .230] Overall Awareness Overall Mean Score Awareness 3.92 (.69) 4.11 (.84) [.143, .233] Notes: CI = Confidence Interval of the Difference; Awareness is based on a 5-point Likert scale (0 = strongly disagree, 5 = strongly agree). * p < .05 ** p < .001 Knowledge Total score Results show that after the intervention the overall knowledge about bullying has increased significantly (t (987) = 18.177, p < .001). The overall grade, on a scale from 0 to 10, increased from an 8.0 before the intervention to an 8.9 after the intervention. In general, girls performed better on the quiz than boys. The boys improved their grade from a 7.7 to an 8.5, whereas the girls got respectively an 8.3 and a 9.3. After the intervention 45.2% of the students answered all the questions of the quiz correctly, while before the intervention no more than 15.1% had a perfect score. Besides these low percentages of perfect scores, still 4.6% (22 persons) did not ‘pass’ the quiz after the intervention, because they answered less than six questions correct. Specific questions Further analyses of the answers on the specific questions of the quiz are presented in Table 8. Some remarkable improvements occurred for the three goals of knowledge described earlier. After the intervention more students know what (cyber)bullying is, as 89% of the students distinguishes physical bullying, mental bullying, and cyberbullying as the three main forms of bullying (question 5) and 97% of the students defines cyberbullying as bullying on the Internet (question 4). Master Thesis Communication Studies Joanne Amse 35 Self-efficacy for Defending Behavior Student’s overall beliefs about their efficacy to take action and stand up for the victim in a bullying situation seem to be increased after the intervention (M = 2.96; M = 3.15) (t (1140) = 8.341; p < .001). The student’s perception about their ability to mediate between the people involved in the bullying process (M = 3.28; M = 3.48) (t (1140) = 7.737; p < .001) and to attack the bully (M = 2.52; M = 2.69) (t (1139) = 5.592; p < .001) were both perceived as less difficult after the intervention. Defending behavior that focuses on mediation was for both the pretest and posttest believed to be less difficult than defending behavior that focuses on attacking the bully (Table 10). Efficacy beliefs for defending behavior increased significantly after the intervention for all forms of mediation. When students try to stop the bullying through mediation, they seemed to find it least difficult to tell others that bullying is stupid (M = 3.59; M = 3.73) or that it doesn’t pay off to join in the bullying (M = 3.59; M = 3.67). The most direct form of mediation by talking to the bully and victim involved in the bullying situation was perceived as most difficult (M = 2.63; M = 2.93). Looking for help by involving an adult or telling an adult about the bullying was indicated as a relatively easy thing to do to stop the bullying (M = 3.35; M = 3.61). Efficacy beliefs for defending behavior in several forms of attacking the bully were increased significantly, except for calling the bully names (t (1121) = 1.437; p = .0755). This was however already perceived as the easiest action to take within this self-efficacy category (M = 2.82; M = 2.88). Additionally, most students find it rather difficult to attack the bully (M = 2.33; M = 2.60) or take revenge on the bully (M = 2.42; M = 2.60) in order to stand up for the victim. Table 10 Means, Standard Deviations and Number of Participants Self-efficacy for Defending Behavior Pretest and Posttest Items Pretest M (SD) Posttest M (SD) N Self-efficacy Mediation (α = .72) 3.28 (.87) 3.48 (.98)** 1141 I am looking for help by an adult, i.e. a teacher or mentor, by involving an adult or telling an adult about the bullying. 3.35 (1.21) 3.61 (1.26)** 1132 I tell the others that it doesn’t pay off to join in the bullying. 3.59 (1.23) 3.67 (1.29)* 1110 I try to mediate between the bully and the victim by talking with each other. 2.63 (1.29) 2.93 (1.33)** 1114 I say to the others that bullying is stupid. 3.59 (1.25) 3.73 (1.26)** 1117 Self-efficacy Attacking Bully (α = .82) 2.52 (1.15) 2.69 (1.24) 1140 I call the bully names in order to defend the victim. 2.82 (1.38) 2.88 (1.47) 1122 I attack the bully in order to defend the victim. 2.33 (1.35) 2.60 (1.44)** 1117 I take revenge on the bully for the victim. 2.42 (1.32) 2.60 (1.39)** 1130 Self-efficacy Scale (α = .77) 2.96 (.78) 3.15 (.91)** 1141 Note: Self-efficacy for Defending Behavior is based on a 5-point Likert scale (1= very difficult, 5 = very easy). * p < .05 ** p < .001 Master Thesis Communication Studies Joanne Amse 36 Outcome Expectations for Defending Behavior After the intervention the students had significantly stronger expectations that the bullying would decrease (M = 3.39; M = 3.69) (t (1136) = 7.171; p < .001), one’s own status would improve (M = 2.75; M = 2.89) (t (1117) = 3.587; p < .001), and the atmosphere in the classroom would improve (M = 3.57; M = 3.92) (t (1128) = 8.842; p < .001) if they would defend the victim and/or try to stop the bullying. The expectations concerning the victim’s wellbeing didn’t increased significantly (t (1134) = -.270; p = .787). However, this consequence of defending behavior was already perceived the strongest (M = 4.27; M = 4.27) compared to the other possible consequences (Table 11). Besides the victim’s wellbeing, the students have stronger expectations of the possible positive effects of defending behavior after the intervention (M = 3.50; M = 3.69). Before the intervention most students neither agreed nor disagreed on the statements about the frequency of bullying (N = 403; 34.5%) and one’s personal status (N = 445; 38.4%). Most students agreed on the expected outcome that the victim would feel better when he/she would stand up for the victim (N = 935; 80.1%) and that the atmosphere in the classroom would improve when he/she would try to convince others to stop the bullying (N = 629; 54.0%). After the intervention most students were convinced of the positive consequence of defending behavior on the frequency of bullying (N = 662; 58%), the victim’s wellbeing (N = 902; 79.1%), and the atmosphere in the classroom (N = 756; 66.4%). Though less students believed that defending behavior would improve one’s personal status or popularity in the classroom (N = 293; 25.9%). Table 11 Means, Standard Deviations and Number of Participants Outcome Expectations for Defending Behavior Pretest and Posttest Items Pretest M (SD) Posttest M (SD) N The bullying will decrease or stop if I stand up for someone who is being bullied. 3.39 (1.20) 3.69 (1.29)** 1137 The person who is being bullied will feel better when I stand up for him or her. 4.27 (.99) 4.27 (1.10) 1135 My status/popularity in the classroom will improve if I try to convince others to stop the bullying. 2.75 (.04) 2.89 (.04)** 1118 The atmosphere in the classroom will improve when I try to convince others to stop the bullying 3.57 (1.27) 3.92 (1.22)** 1129 Outcome Expectations (α = .74) 3.50 (.83) 3.69 (.95)** 1142 Note: Outcome Expectations for Defending Behavior are based on a 5-point Likert scale (0 = strongly disagree, 5 = strongly agree). * p < .05 ** p < .001 Moderator Effect Self-reported Participant Roles For this research the possible moderator effect of the self-reported participant roles is analyzed to examine whether the image that the participants have of their own role in the bullying process had an effect on answering the questions about awareness, knowledge, attitude, outcome expectations and self-efficacy. Table 12 and Table 13 show the output from the MANOVA analysis including the MANOVA and Univariate test statistics. To determine the nature of the effect, Univariate tests were performed. Master Thesis Communication Studies Joanne Amse 37 Using Pillai’s Trace, there was a significant moderator effect of the self-reported participant roles on at least one of the outcome variables, V = .067, F (15, 1788) = 2.716, p < .001. Separate ANOVAs on the outcome variables revealed significant effects on awareness (F (3, 598) = 4.499, p = .004), knowledge (F (3, 598) = 2.736, p = .043), and attitude (F (3, 598) = 4.567, p = .004). No effects were shown on outcome expectations (F (3, 598) = .822, p = .482), and self-efficacy (F (3, 598) = .092, p = .965). Table 12 Multivariate Tests Difference Total Score Dependent Variables Test Value F df Error df p Pilai’s trace .067 2.716 15 1788 .000** Wilks’ lambda .934 2.733 15 1640 .000** Hotelling’s trace .070 2.746 15 1778 .000** Roy’s largest Root .050 6.005 5 596 .000** * p < .05 ** p < .001 Table 13 Univariate Tests Difference Total Score Dependent Variables Outcome Variable Mean Difference F df Error df p Awareness 1.7326 4.499 3 598 .004* Knowledge .9419 2.736 3 598 .043* Attitude .3272 4.567 3 598 .004* Outcome Expectations .8488 .822 3 598 .482 Self-efficacy 1.3472 .092 3 598 .965 * p < .05 ** p < .001 Table 14 Means and Standard Deviations Outcome Variables Pretest and Posttest vs. Self-reported Participant Roles Pretest Outcome Variables vs. Self-reported Participant Roles (Pretest) Self-reported Participant Role (Pretest) Outcome variable Bully Reinforcer Defender Outsider Pretest M (SD) Posttest M (SD) Pretest M (SD) Posttest M (SD) Pretest M (SD) Posttest M (SD) Pretest M (SD) Posttest M (SD) Awareness 3.33 (1.21) 3.87 (1.07) 3.69 (.74) 3.64 (1.06) 4.12 (.61) 4.35 (.66) 3.77 (.71) 3.97 (.86) Knowledge 6.72 (1.27) 8.19 (1.52) 7.30 (1.78) 8.32 (2.16) 8.43 (1.32) 9.17 (1.29) 7.77 (1.62) 8.72 (1.59) Attitude 3.83 (.62) 4.03 (.64) 3.87 (.71) 4.00 (.75) 4.63 (.40) 4.60 (.46) 4.24 (.54) 4.35 (.57) Self-efficacy 3.42 (1.16) 2.10 (.93) 3.10 (.77) 2.62 (.94) 3.11 (.73) 2.73 (.84) 2.84 (.73) 2.95 (.92) Outcome Expectations 3.16 (1.16) 3.44 (1.14) 3.32 (.90) 3.36 (1.12) 3.79 (.72) 3.92 (.82) 3.31 (.78) 3.56 (.95) Notes: Awareness, Attitude, and Outcome Expectations for Defending Behavior are based on a 5-point Likert scale (0 = strongly disagree, 5 = strongly agree); Self-efficacy for Defending Behavior is based on a 5-point Likert scale (1= very difficult, 5 = very easy); Knowledge is based on a 10-point scale representing the mean grade for the quiz. Master Thesis Communication Studies Joanne Amse 40 DISCUSSION The present study was designed to evaluate the effectiveness of a classroom based anti- bullying intervention program aimed at the group process of bullying. In the present study, the effectiveness of the anti-bullying program ‘Survivors!’ was examined among first year students of secondary schools. This intervention program aims at creating awareness among all students about their own responsibilities and possibilities within the group to prevent or stop the bullying. The intervention effects were examined for several outcome variables indicating the participant role behaviors, awareness, knowledge, attitude, self-efficacy, and outcome expectations related to bullying and defending behavior. In addition, the moderator effect of the self-reported participant roles on the intervention effects and the association between the outcome variables and the participants’ reported defending behavior were examined. In general, the results give clear evidence to support the effectiveness of the intervention program for all outcome variables assessed in this study. This is in contrast with numerous previous studies who studied the effectiveness of school anti-bullying programs. The majority of these studies have shown inconsistencies in evaluation findings as these results were often non-significant or modest, with only a few positive outcomes or even negative effects (Smith et al., 2004). One of the few evaluation studies who have shown effective results is the KiVa Antibullying Program of Kärna et al. (2011), who also used the Participant Role Questionnaire of Salmivalli et al. (1996). Participant Roles The own categorization of participant roles have shown that the most common self-reported participant roles for the pretest and posttest were respectively outsider (39.4%; 39.1%), defender (37.7%; 35.4%), victim (14.0%; 15.4%), passive reinforcer (6.2%; 6.2%%), active reinforcer (1.0%; 1.3%), and bully (1.7%; 2.7%). By the use of the introduced role selection criteria it was possible to assign 71.2% of the participant to a behavioral participant role. In earlier research of Salmivalli et al. (1996), 83.7% of the participants were assigned to a participant role. They used similar role selection criteria on the basis of self- and peer report of the PRQ. Given the fact that the percentage of the research of Salmivalli et al. (1996) is based on self- and peer report and that the PRQ is originally a peer-reported measurement instrument, the percentage of 71.2% of the present study is considered to be satisfactory. Of the students for whom it was possible to assign them to a behavioral participant role, the behavioral participant roles for the pretest and posttest were respectively defender (51.6%; 51.6%), outsider (23.5%; 19.1%), reinforcer (15.6%; 20.6%) and bully (9.3%; 8.8%). Compared to recent national research on the participant role approach, a remarkable lower percentage of students was categorized in pro-bullying roles (bullies, reinforcers), while a higher percentage of students was identified in one of the anti-bullying roles (victims, defenders, outsiders). A possible explanation relates to the data gathering as, in contrary to other evaluation studies, for this study only self-reports of bullying were used as outcome measure instead of self-reports and peer reports. Salmivalli et al. (1999) and Sutton & Smith (1999) pointed out these differences between self- and peer reports in anti-bullying research and awareness of their participant roles. The underestimation of participants of their active bullying behavior and their emphasis on their role as defender or outsider is more common with the use of self- report. According to Juvonen, Nishina, and Graham (2000) students are more likely to report their own involvement in bullying situations in a more favorable or social desirable light. Master Thesis Communication Studies Joanne Amse 41 This may withhold students from revealing their active involvement in bullying situations as a bully or reinforcer or exaggerate their involvement in the role of victim, defender, or outsider. In consistency with previous studies, there was a significant difference of engagement in self-reported and behavioral participant roles between the sexes. The role of bully and reinforcer were more common among boys, while the role of defender was more typical for girls. However, contrary to the existing literature, the role of outsider was more frequent among boys instead of girls. Sex differences in judgment and context awareness can explain this higher number of outsiders among boys. Salmivalli et al. (1999) found that personality factors (e.g. status, self- esteem) were important predictors for the participant role behavior of boys, whereas girls seemed to be more influenced by group context (e.g. the behavior of peers). This combined with the research of Juvonen et al. (2000) and Sutton and Smith (1999), boys may find it more important to present a more attractive image of their participant role behavior in bullying, for example because of their status. Additionally, due to the focus of boys on their individual behavior instead of the group context, boys may seem to underestimate their involvement in the group process of bullying more often than girls. It was hypothesized that through the intervention the level of agreement between the self-reported and behavioral participant roles would increase. Unfortunately, there was no improvement after the intervention in the correct assessment of the self-reported participant roles and the behavioral participant roles, for both girls (73.0%; 67.1%) and boys (59.0%; 52.5%). Before the intervention 65.6% of all students were able to match their self-reported and behavioral participant roles, while after the intervention 59.3% made the correct classification of their participant role. This absence of improvement could be caused by a lack of self-reflection of students. These ideas and feelings that students have of their own participant role behavior could possibly be influenced by their self-awareness (the awareness of their own behavior in relation to other group members) or their self-esteem (the judgment of their own behavior) in bullying situations. Awareness The results of the present study show a raising awareness of the responsibility for the atmosphere in the classroom and the protection of (online) boundaries. However, the discrepancy between the attitudes and behavior (Rigby and Slee, 1991; Salmivalli & Voeten, 2004; Salmivalli 2010) is also visible in the awareness of students in this research. In contradiction with their attitude towards their shared responsibility, students are still not sufficiently aware of their personal contribution to the atmosphere in the classroom and their personal task to say ‘stop’ in bullying situations. Salmivalli (2010) assigns this contradiction to their diffusion of responsibility. Even if bystanders are aware of their responsibility to stop the bullying, they don’t seem to act on it. In contrast to their lack of their personal responsibility to stop the bullying themselves, most students do indicate that they will stop bullying if others take the initiative. So if the victimized peer or bystanders do intervene during a bullying situation by expressing their boundaries, the students will respect these boundaries and react to it. Their own reaction is thus related to the interpretation of other bystanders in the bullying situation. In other words, bystanders monitor the behavior of other bystanders and copy this behavior, both positive and negative (Salmivalli, 2010). Master Thesis Communication Studies Joanne Amse 42 In contradiction with offline bullying situations, students do embrace their personal responsibility when it comes down to the protection of their own online boundaries. This contradiction might be explained by the perceived higher negative impact of cyberbullying compared to traditional bullying, especially for picture/video clip bullying (Slonje & Smith, 2008). Due to the potential harmfulness, the threatening nature, potential audience, and visibility of cyberbullying (Baas et al., 2013), students may be more aware of the negative consequences of cyberbullying and thus the importance to protect their online boundaries. Knowledge The knowledge of students about bullying has improved significantly after the intervention. The overall mean grade for the knowledge quiz, on a scale from 0 to 10, increased from an 8.0 before the intervention to an 8.9 after the intervention. The students know better what (cyber)bullying is, what determines the difference between teasing and bullying, and what to do against (cyber)bullying situations. A concern lies however in the primary focus of this research on the students’ self- report of bullying behavior. Merrell e al. (2008) remarked that the positive effects of the intervention mainly concerned indirect measures of bullying behavior. This emphasizes the difference between what the students know or belief about the right behavior in bullying situations versus the students’ actual behavior in bullying situations. So the fact that the students have theoretically improved their knowledge about bullying and defending behavior does not mean that they will actually bring this into practice. Attitude In general, the students did already have a strong attitude against bullying which became even stronger after the intervention. These results are consistent with the stronger anti-bullying attitudes for the students after the intervention of the studies of Kärna et al. (2011) and Salmivalli et al. (2005). The overall improvement in this study is mainly caused by weakening the pro-bullying attitude of students. After the intervention the students expressed less tendency to reject other students because of their perceived weakness. Additionally, the students were less likely to justify or support the bullying. The results on the anti-bullying attitude scale showed mixed findings. The students seemed to show more approval for the bully and less support for defending behavior of others, while at the same time they seemed to value their own defending behavior in bullying situations more. Future research is necessary to examine why the students scored different on these subcategories of the anti-bullying scale. Self-efficacy for Defending Behavior Results suggest that, consistent with the recent evaluation study of Kärna et al. (2011) and Salmivalli et al. (2005), student’s beliefs about their efficacy to take action and stand up for the victim in a bullying situation has increased significantly after the intervention. Defending behavior that focuses on mediation was believed to be less difficult than defending by attacking the bully. When students tried to stop the bullying through mediation, more indirect forms of mediation (i.e. telling others about the bullying) was perceived as relatively easy compared to direct mediation between the bully and victim. Efficacy beliefs for defending behavior in several forms of attacking the bully increased significantly, although it was still difficult. Within the self-efficacy category of attacking the bully, the students found it most easy to verbally attack the bully by calling him/her names. Master Thesis Communication Studies Joanne Amse 45 Practical Implications The practical question of importance raised by the present study is: How can defending behavior among students be encouraged through interventions that focus on the group process of bullying? To put an end to the inconsistency between attitudes and actual defending behavior in bullying situations, the intervention program should focus more on self-reflection and commitment to anti-bullying behavior within the group involved. At this moment, the intervention program mainly focuses on raising general awareness by offering information about the group mechanisms in bullying situations, such as the participant roles. However, providing information and developing feelings of responsibility doesn’t mean that the students will actually do the things that they know they are supposed to do in bullying situations. In the present study attitude accounted only for 20.2% of the variance in reported defending behavior, which means that almost 80% of reported defending behavior is determined by other factors. This was already mentioned by Jiménez Barbero et al. (2012), who argued that the most effective interventions were aimed at changing attitudes and beliefs as well as improving social and interpersonal skills. For the transformation process of attitudes into actual behavior, self-reflection on own bullying behavior is the first thing required as it appears that most students are not able to make a correct estimation of their own participant role behavior in bullying situations. This limited ability of students to reflect on their own bullying behavior could be improved by the use of discussions or activities in small groups. Following the rules of constructive criticism, the students can give each other feedback on their behavior in recent bullying situations from their own observations. By the use of peer feedback, students can get a clear view on the possible discrepancy between their own perception and the perception of others on their role in the bullying process. Besides the need for self-reflection, students should be taught more explicitly how to take the initiative to stop the bullying themselves instead of monitoring and copying the passive behavior of other bystanders. Therefore the students should be offered the opportunity to rehearse anti-bullying defending behavior in person in a safe, fictive context. Drama and role-play exercises can be used to explore feelings and actions associated with different participant role behaviors, for example: ‘Why is it so difficult to support the victim?’ or ‘How does it feel to be a defender or assistant?’ Experiences in fictive situations could result in more motivation and self-confidence to execute this practiced defending behavior in real life bullying situations. So the students should not only be aware of their personal responsibility to stop the bullying, but act on it as well. Individual motivation and interpersonal skills to display defending behavior are however not always sufficient if other group members involved in the bullying process don’t accept or encourage this behavior. Therefore interventions should also pay more attention to the rewarding structure of defending behavior in the classroom. The development of shared class rules with the input from all students of the class could increase this commitment of the whole class to anti-bullying behavior (Olweus, 1991). These classroom rules should include rules that reinforce and reward supporting and defending behavior as well as the bystander’s reactions to the bullying situation. Master Thesis Communication Studies Joanne Amse 46 Another important implication that comes forward in the present study entails the involvement of the group context in interventions. This means that interventions should not only be targeted to the individual bullies and victims, but target the group as a whole. The involvement of the group context in interventions can however be expanded to a broader level than only the group members involved in the bullying situation. There are many more factors on multiple levels that could influence the involvement in the bullying process: individual level (e.g. social status, self-esteem, and assertiveness), group level (e.g. group pressure), family level (e.g. role model of parents), classroom level (e.g. attention of the teacher for bullying), school level (e.g. anti-bullying school policy), neighborhood or society (e.g. acceptance of violence and aggression). These different levels don’t stand alone, but influence each other as well. According to Jiménez Barbero et al. (2012), this multidisciplinary perspective would increase the chance of success of interventions through the involvement of all professional disciplines of the educational center, including the parents. Practically, this means that instead of only offering a show, lesson and possible returnlesson once a year, the intervention program has to be developed into a program that will be part of the schools’ continuous anti-bullying work to ensure the long-term effectiveness. To integrate the intervention program more into whole school policy, commitment to this implementation in the curriculum is necessary (Jiménez Barbero et al., 2012). The teachers have to be motivated and provided with resources to work with in the classroom. Therefore the materials of the intervention program could be extended with specific teacher guidance and program content which elaborates on the content of the program. This would provide the teachers and mentors with support to adapt the program into their own lessons as well. Another kind of support could be given by visiting the schools before, after or in-between the intervention and returnlessons. By doing so, the organization will be able to protect the continuity of the program content and monitor the intervention effects over a longer period of time. By doing this, the intervention program could possibly be adapted to the social and cultural characteristics of the specific school population, i.e. gender, age, education level (Jiménez Barbero et al., 2012). The government could play an important role by the implementation of this multidisciplinary approach in the form of whole-school policies. With the Plan against bullying, the government already legally obligates schools to use a scientifically and empirically proven effective anti-bullying program. Additionally, the government could recommend intervention programs which embrace this whole-school approach. The government can set requirements regarding essential measures of the intervention program at at least three levels: individual level, classroom level, and school level. These above mentioned implications for anti-bullying interventions concerning the emphasis on self-reflection, commitment, class rules, and the multidisciplinary perspective could also be applied to other interventions among students aimed at physical, mental, or social health issues such as drug and alcohol use. Future Research The recent study provides several useful insights for anti-bullying intervention programs and anti-bullying research in general. It provides schools and other stakeholders with practical guidance on how to reduce bullying and victimization effectively by the use of a participatory theatre, classroom intervention. Furthermore, it gives scientific and empirical support for the effectiveness of anti-bullying program targeting bullying as a group process. Nonetheless, future research is needed to expand the knowledge about the characteristics of intervention components, the interaction effects at multiple levels and the long term effectiveness. Master Thesis Communication Studies Joanne Amse 47 The intervention examined in the present study consists of a theater show, a talk show, and a lesson. The intervention effects measured in this study thus indicate the effect of the combination of these three components of intervention (theater show, talk show, and lesson). Consequently, nothing can be said about the effectiveness of these intervention components separately. Additionally, the precise effect of the separate components of intervention is not clear. Does the theater show has a different effect than the lesson? Is the theater show more effective with the talk show than without the talk show? Furthermore, it is interesting to examine if the different forms of bullying are equally susceptible for the interventions. In summary, future research should find out which (combination of) interventions are most effective for which form of bullying. So why are (combinations of) interventions effective, for whom and under which conditions? Furthermore, the intervention effects for the present study were only analyzed at general level. However, there could be a lot of variation between the findings at educational, school-, class-, or individual level. Factors that could have an impact on the intervention effects are for example the anti-bullying policy per school, recent bullying incidents in the class, and the process of implementation of the intervention. As it was shown there were some differences between the recent sample and other samples used in anti-bullying research regarding the comparability of Group Red/Blue, the higher bullying experiences, and the higher number of outsiders among boys. Future research should thus explore the intervention effects at multiple levels between various representative samples. Moreover, the present study only examined the results of the intervention program at short term. Future research should evaluate the effects of the intervention on the long term, especially because change in behavior becomes only visible after a period of time. Special attention in the long run effectiveness should be given to the nature and reasons of change in participant role behavior (Salmivalli, 2001). The evaluation of participant role behavior on the long term would give more insight in the influence of the group context and the specific bullying situations on the stability and/or change of participant role behavior. Conclusion The main conclusions of the present study can be summarized as followed: The classroom anti-bullying intervention program ‘Survivors!’, which focuses on the role of bystanders in the group process of bullying, has been proven to be effective among first year students of secondary schools. Improvements has been shown for the awareness, knowledge, attitude, self-efficacy for defending behavior, and outcome expectations for defending behavior. The most common participant roles were defender and outsider. There was a significant difference of engagement in participant role behavior between the sexes. Unfortunately, there was no improvement in the correct assessment of the self-reported participant roles and the behavioral participant roles after the intervention. The results show an increasing awareness of the personal and shared responsibility for the atmosphere in the classroom and the protection of (online) boundaries. However, the students are still not sufficiently aware of their personal contribution to the atmosphere in the classroom and their personal task to say ‘stop’ in bullying situations. The knowledge of students about bullying has improved significantly after the intervention. The students know now better what (cyber)bullying is, what determines the difference between teasing and bullying, and what to do against (cyber)bullying situations. The overall mean grade for the knowledge quiz, on a scale from 0 to 10, increased from an 8.0 before the intervention to an 8.9 after the intervention. Master Thesis Communication Studies Joanne Amse 50 Meer, B. van der (2013). Vijfsporenaanpak. Retrieved November 19, 2013, from http://www.bobvandermeer.info/index.php?option=com_content&task=view&id=213 &Itemid=78. Merrell, K. W., Gueldner, B. A., Ross, S. W., & Isava, D. M. (2008). How effective are school bullying intervention programs? A meta-analysis of intervention research. School Psychology Quarterly, 23(1), 26. Nicolaides, S., Toda, Y., & Smith, P. K. (2002). Knowledge and attitudes about school bullying in trainee teachers. British Journal of Educational Psychology, 72(1), 105-118. O'Connell, P., Pepler, D., & Craig, W. (1999). Peer involvement in bullying: Insights and challenges for intervention. Journal of adolescence, 22(4), 437-452. Oesterman, K., Bjoerkqvist, K., Lagerspetz, K. M., Kaukiainen, A., Huesmann, L. R., & Fra czek, A. (1994). Peer and self‐estimated aggression and victimization in 8‐year‐old children from five ethnic groups. Aggressive Behavior, 20(6), 411-428. Olthof, T., Goossens, F. A., Vermande, M. M., Aleva, E. A., & Van der Meulen, M. (2011). Bullying as strategic behavior: Relations with desired and acquired dominance in the peer group. Journal of school psychology, 49(3), 339-359. Olweus, D. (1993). Bullying at school: What we know and what we can do. Oxford, UK, and Cambridge, MA, USA: Blackwell Publishers. Olweus, D. (1994). Bullying at school: basic facts and effects of a school based intervention program. Journal of child psychology and psychiatry, 35(7), 1171-1190. Olweus, D. (1997). Bully/victim problems in school: Facts and intervention. European Journal of Psychology of Education, 12(4), 495-510. Pepler, D. J., Craig, W. M., Ziegler, S., & Charach, A. (1994). An evaluation of an anti- Bullying intervention in Toronto schools. Canadian Journal of Community Mental Health (Revue canadienne de santé mentale communautaire), 13(2), 95-110. Pikas, A. (1975): “SB stoppar vi mobbning.” Stockholm: Prisma. Ponzetti Jr, J. J., Selman, J., Munro, B., Esmail, S., & Adams, G. (2009). The effectiveness of participatory theatre with early adolescents in school‐based sexuality education. Sex Education, 9(1), 93-103. Pöyhönen, V., Juvonen, J., & Salmivalli, C. (2012). Standing up for the victim, siding with the bully or standing by? Bystander responses in bullying situations. Social Development, 21(4), 722-741. Rigby, K., & Johnson, B. (2006). Expressed readiness of Australian schoolchildren to act as bystanders in support of children who are being bullied. Educational psychology, 26(3), 425-440. Rigby, K., & Slee, P. T. (1991). Bullying among Australian school children: Reported behavior and attitudes toward victims. The Journal of Social Psychology, 131(5), 615-627. Rigby, K., & Slee, P. T. (1993). Dimensions of interpersonal relation among Australian children and implications for psychological well-being. The Journal of Social Psychology, 133(1), 33-42. Rigby, K. (1997) Attitudes and beliefs about bullying among Australian school children. Irish Journal of Psychology, 18(2), 202 - 220. Rigby, K. (2001). Health consequences of bullying and its prevention in schools (Doctoral dissertation, Guilford Press). Rivers, I., & Smith, P. K. (1994). Types of bullying behaviour and their correlates. Aggressive Behavior, 20(5), 359-368. Master Thesis Communication Studies Joanne Amse 51 Salmivalli, C., Lagerspetz, K., Björkqvist, K., Österman, K., & Kaukiainen, A. (1996). Bullying as a group process: Participant roles and their relations to social status within the group. Aggressive behavior, 22(1), 1-15. Salmivalli, C. (1999). Participant role approach to school bullying: Implications for interventions. Journal of adolescence, 22(4), 453-459. Salmivalli, C. (2001). Group view on victimization — Empirical findings and their implications. In J. Juvonen & S. Graham (Eds.), Peer harassment in school: The plight of the vulnerable and victimized (pp. 398–419). New York: Guilford Press. Salmivalli, C., & Voeten, M. (2004). Connections between attitudes, group norms, and behaviour in bullying situations. International Journal of Behavioral Development, 28(3), 246-258. Salmivalli, C., Kaukiainen, A., & Voeten, M. (2005). Anti‐bullying intervention: Implementation and outcome. British Journal of Educational Psychology, 75(3), 465-487. Salmivalli, C. (2010). Bullying and the peer group: A review. Aggression and violent behavior, 15(2), 112-120. Smith, J. D., Schneider, B. H., Smith, P. K., & Ananiadou, K. (2004). The effectiveness of whole-school antibullying programs: A synthesis of evaluation research. School Psychology Review, 33(4), 547-560. Stevens, V., Bourdeaudhuij, I., & Oost, P. (2000). Bullying in Flemish schools: An evaluation of anti‐bullying intervention in primary and secondary schools. British Journal of Educational Psychology, 70(2), 195-210. Slonje, R., Smith, P. K., & Frisén, A. (2013). The nature of cyberbullying, and strategies for prevention. Computers in Human Behavior, 29(1), 26-32. Sutton, J., & Smith, P. K. (1999). Bullying as a group process: An adaptation of the participant role approach. Aggressive Behavior, 25(2), 97-111. Master Thesis Communication Studies Joanne Amse 52 APPENDIX APPENDIX A Information letter parents © Individueel: De vrageniijst dient individueel ingevuld te worden. Er zijn geen goede of foute
antwoorden. Het gaat om de eigen mening van de leerlingen, dus ze mogen niet afkijken bij hun
buurman of buurvrouw.
© Anonimiteit: De vragenlijst wordt anoniem ingevuld en verwerkt. Dit betekent dat er geen namen aan
de ingevulde vragenlijsten worden gekoppeld en er dus enkel een algemeen beeld geschetst kan
worden. De leerlingen hoeven dus niet hun volledige naam op de vragenlijst te noteren. Om de
antwoorden goed te kunnen verwerken zijn hiervoor alleen de eerste letter van de voormaam, de
eerste letter van de achternaam en de geboortedatum nodig.
Inleveren vragenlijst
‘Wanner de vrageniijsten volledig zijn ingevuld wil ik u verzoeken alle vragenlijsten per klas en per
afnamemoment (voor en na de voorstelling) in een envelop te verzamelen. Dit zijn dus twee enveloppen per
klas. Deze kunt u bij het bezoek van Switch aan één van de acteurs meegeven.
‘Mocht u nog vragen en/of opmerkingen hebben, dan kunt u gerust contact opnemen met ondergetekende.
Nogmaals hartelijk dank voor uw deeIname en uiteraard ontvangt uw school na verwerking van het onderzoek
een overzicht van de gevonden resultaten.
Met vriendelijke groet,
Joanne Bijleveld
Telefoon: 0343 515 757,
E-mail:
[email protected]
=
cS
=
=
oA
Youth for Christ —_
Hapidstraat 260
3972 LL Driebergen
aaa 515 744 ki ari77018. genet FOR CHRIST
info@youthlorebvitnl —Giten =: NLOBRABO ONT 4448 a8 NBI
weenyouthforchist.ni —Balaingen: NL27 RABOONI1 181111 _—
Master Thesis Communication Studies 55
Joanne Amse
Master Thesis Communication Studies Joanne Amse 56 APPENDIX C Questionnaire
Kruis het juiste antwoord aan op onderstaande meerkeuzevragen.
1. Alsik een pestsituatie in de klas wil stoppen, kan ik het beste...
(© Ersamen met klasgenoten iets van zeggen en anders een docent inschakelen.
© Met een groep vrienden de pester opwachten en hem/haar een lesje leren.
©. Niets doen. Dan gaat het het snelst over.
2. _Alsikiets wat anderen bij me doen niet leuk vind kan ik het beste...
© Hard meelachen, zodat ze denken dat het me toch niets kan schelen.
Q Helemaal niet reageren, zodat de lol er voor hun ook af is.
Q Duidelik zeggen dat ik het niet leuk vind en dat ze normal moeten doen,
3. Als anderen niet stoppen met pesten, zelfs als ik “stop” gezegd heb, kan ik het beste...
Q Eropslaan. Wie niet horen wil, moet maar voelen.
Mezetf zo stil mogelk houden en hopen dat het vanzelf over gaat.
O Hulp zoeken bij klasgenoten en/of een mentor of docent.
4, Cyberpestenis...
O Pesten op Internet
O Geestelik pesten (schelden, roddelen, negeren),
© Alles op Intemet wat ik niet leuk vind
5. De drie hoofdvormen van pesten zijn...
O Lichamelijk pesten, geestelijk pesten, pesten op Intemet.
© Slaan, schelden en roddelen
O Pesten op Internet, reddelen en negeren.
6. Het verschil tussen pesten en plagen wordt bepaald door...
O Degene die het doet.
O Degene die het overkornt.
© Deomstanders.
7. Drie rollen die je altijd ziet als er gepest wordt zijn...
© Pester, docente, conciérge:
© Meeloper, gepeste, docent.
© Pester, gepeste, toeschouwer.
8. Pestenis oké...
© Alsdegene die gepest wordt initant is
© Als degene die pest een vervelende thuissituatie heeft.
© Nooit!
9. Ikan zelf de kans dat ik gepest word zo klein mogelijk maken door...
© Stevig te staan, duidelijk min grenzen aan te geven en voor mezelf op te komen.
O. Zelf te gaan pesten.
© Tezorgen dat ik 20 weinig mogelijk opval.
10. Als ik word gepest op Internet kan ik het beste...
© Gewoon terugschelden
© Bewijzen verzamelen en het melden bij een docent of de webbeheerder.
© Almin social mediaprofielen van Intemet halen.
[PESTERVARINGEN
Geef aan hoe vaak je de volgende dingen afgelopen maand hebt gezien of gehoord op school (toeschouwer).
Nooit Soms _ Regelmatig
1. lemand werd expres buitengesloten, uitgescholden of er werden roddels ° ° fe}
over iemand verspreid
2. lemand werd geduwd, geschopt of geslagen. ° ° °
3. lemand ontving vervelende telefoontjes, berichten of afbeeldingen via zijn/ oO ° °
haar telefon of het Internet.
Geef aan hoe vaak de volgende dingen afgelopen maand bij jou zelf zijn gebeurd (gepeste).
Nooit Soms —Regelmatig
1. Ikwerd expres buitengesloten, uitgescholden of er werden roddels over mi =O ° °
verspreid
2. tkwerd geduwd, geschopt of geslagen. ° ° oO
3. Ikontving vervelende telefoontjes, berichten of afbeeldingen via mijn ° ° °
telefoon of het Internet.
Vaak Heel vaak
UNIVERSITEIT TWENTE.
° °
° Q
° °
Vaak — Heel vaak
QO °
° °
° °
2
Master Thesis Communication Studies
Joanne Amse
57
Master Thesis Communication Studies Joanne Amse 60 APPENDIX D Tables Self-reported and Behavioral Participant Roles Gender Table A Self-reported and Behavioral Participant Roles Pretest and Posttest for Boys and Girls (% within Gender) Self-reported Participant Role * Gender Pretest Posttest Self-reported Participant Role Boy Girl Total Boy Girl Total Victim N % of Gender 62 13.5% 61 14.0% 123 14.0% 66 14.3% 69 16.5% 135 15.4% Bully N % of Gender 12 2.6% 3 0.7% 15 1.7% 15 3.3% 9 2.2% 24 2.7% Passive reinforcer N % of Gender 42 9.1% 12 2.9% 54 6.2% 36 7.8% 18 4.3% 54 6.2% Active reinforcer N % of Gender 4 0.9% 5 1.2% 9 1.0% 6 1.3% 5 1.2% 11 1.3% Defender N % of Gender 135 29.3% 196 46.9% 331 37.7% 136 29.6% 175 41.9% 311 35.4% Outsider N % of Gender 205 44.6% 141 33.7% 346 39.4% 201 43.7% 142 34.0% 343 39.1% Total N % of Gender 460 100.0% 418 100.0% 878 100.0% 460 100.0% 418 100.0% 878 100.0% Behavioral Participant Role * Gender Pretest Posttest Behavioral Participant Role Boy Girl Total Boy Girl Total Bully N % of Gender 35 9.2% 13 3.8% 48 6.6% 35 9.2% 10 2.9% 45 6.2% Reinforcer N % of Gender 67 17.7% 13 3.8% 80 11.1% 77 20.3% 29 8.5% 106 14.7% Defender N % of Gender 96 25.3% 169 49.3% 265 36.7% 103 27.2% 162 47.2% 265 36.7% Outsider N % of Gender 75 19.8% 46 13.4% 121 16.8% 59 15.6% 39 11.4% 98 13.6% No participant role N % of Gender 106 28.0% 102 29.7% 208 28.8% 105 27.7% 103 30.0% 208 28.8% Total N % of Gender 379 100.0% 343 100.0% 722 100.0% 379 100.0% 343 100.0% 722 100.0% Master Thesis Communication Studies Joanne Amse 61 Table B Self-reported Participant Roles vs. Behavioral Participant Roles Pretest for Boys and Girls (% within Behavioral Participant Roles and % of Total) Boys Behavioral Participant Role* Self-reported Participant Role Bully Reinforcer Defender Outsider Total Bully N % of BPR % of Total 7 20.0% 2.6% 4 6.0% 1.5% 0 0.0% 0.0% 1 1.3% 0.4% 12 4.4% 4.4% Reinforcer N % of BPR % of Total 7 20.0% 2.6% 21 31.3% 7.7% 2 2.1% 0.7% 9 12.0% 3.3% 39 14.3% 14.3% Defender N % of BPR % of Total 6 17.1% 2.2% 5 7.5% 1.8% 76 79.2% 27.8% 8 10.7% 2.9% 95 34.8% 34.8% Outsider N % of BPR % of Total 15 42.9% 5.5% 37 55.2% 13.6% 18 18.8% 6.6% 57 76.0% 20.9% 127 46.5% 46.5% Total N % of BPR % of Total 35 100.0% 12.8% 67 100.0% 24.5% 96 100.0% 35.2% 75 100.0% 27.5% 273 100.0% 100.0% Girls Behavioral Participant Role* Self-reported Participant Role Bully Reinforcer Defender Outsider Total Bully N % of BPR % of Total 3 23.1% 1.2% 0 0.0% 0.0% 0 0.0% 0.0% 0 0.0% 0.0% 3 1.2% 1.2% Reinforcer N % of BPR % of Total 4 30.8% 1.7% 6 46.2% 2.5% 2 1.2% 0.8% 3 6.5% 1.2% 15 6.2% 6.2% Defender N % of BPR % of Total 3 23.1% 1.2% 2 15.4% 0.8% 132 78.1% 54.8% 8 17.4% 3.3% 145 60.2% 60.2% Outsider N % of BPR % of Total 3 23.1% 1.2% 5 38.5% 2.1% 35 20.7% 14.5% 35 76.1% 14.5% 78 32.4% 32.4% Total N % of BPR % of Total 13 100.0% 5.4% 13 100.0% 5.4% 169 100.0% 70.1% 46 100.0% 19.1% 241 100.0% 100.0% * Behavioral Participant Role (BPR) Master Thesis Communication Studies Joanne Amse 62 Table C Self-reported Participant Roles vs. Behavioral Participant Roles Posttest for Boys and Girls (% within Behavioral Participant Roles and % of Total) Boys Behavioral Participant Role* Self-reported Participant Role Bully Reinforcer Defender Outsider Total Bully N % of BPR % of Total 9 25.7% 3.3% 4 5.2% 1.5% 0 0.0% 0.0% 1 1.7% 0.4% 14 5.1% 5.1% Reinforcer N % of BPR % of Total 6 17.1% 2.2% 16 20.8% 5.8% 7 6.8% 2.6% 7 11.9% 2.6% 36 13.1% 13.1% Defender N % of BPR % of Total 6 17.1% 2.2% 11 14.3% 4.0% 76 73.8% 27.7% 8 13.6% 2.9% 101 36.9% 36.9% Outsider N % of BPR % of Total 14 40.0% 5.1% 46 59.7% 16.8% 20 19.4% 7.3% 43 72.9% 15.7% 123 44.9% 44.9% Total N % of BPR % of Total 35 100.0% 12.8% 77 100.0% 28.1% 103 100.0% 37.6% 59 100.0% 21.5% 274 100.0% 100.0% Girls Behavioral Participant Role* Self-reported Participant Role Bully Reinforcer Defender Outsider Total Bully N % of BPR % of Total 2 20.0% 0.8% 3 10.3% 1.2% 2 1.2% 0.8% 0 0.0% 0.0% 7 2.9% 2.9% Reinforcer N % of BPR % of Total 2 20.0% 0.8% 7 24.1% 2.9% 8 4.9% 3.3% 1 2.6% 0.4% 18 7.5% 7.5% Defender N % of BPR % of Total 3 30.0% 1.2% 7 24.1% 2.9% 118 72.8% 49.2% 4 10.3% 1.7% 132 55.0% 55.0% Outsider N % of BPR % of Total 3 30.0% 1.2% 12 41.4% 5.0% 34 21.0% 14.2% 34 87.2% 14.2% 83 34.6% 34.6% Total N % of BPR % of Total 10 100.0% 4.2% 29 100.0% 12.1% 162 100.0% 67.5% 39 100.0% 16.2% 240 100.0% 100.0% * Behavioral Participant Role (BPR)