Social Interaction of Bullies and Victims
The first research question was whether bullies and victims would demonstrate different interactional styles when participating in two different games (one competitive and one cooperative).
Effects of Context on Interactions of Bullies and Victims
The second research question concerned the possible effects that the social and physical setting, in terms of opportunities and constraints for behavior, may have on the way bullies and victims interact with each other and with other partners. Specifically, we asked children to interact during a competitive game and a cooperative game.
As social-psychological studies have shown, cooperation and competition are two opposite ends of a single dimension (Deutsch, 1949; Pepitone, 1980). On one hand, cooperative interdependence involves a positive correlation for goal achievement for the people involved: Both partners work to obtain a shared goal. On the other hand, competitive interdependence involves a negative correlation for goal achievement for the participants. The game goal is exclusive, and each partner can interfere with the other's actions.
Effects of Partner on Interactions of Bullies and Victims
The third research question addressed how bullies' and victims' behaviors are differently affected by an opposite-status partner (a bully with a victim) compared with a control-status partner (a bully or a victim interacting with a child identified as neither bully nor victim).
The competitive game was used and standardized in a previous study by Fonzi, Menesini, and Gentili (1994). It resembles the game "Non t'arrabbiare" ("Take It Easy!") published in Italy by Schmidt Spiel & Freizeit (Eching, W. Germany). This game is similar to the one known as "Parcheesi" in the United States. The game consists of a path along which the two players move four pieces. The goal is to move all four pieces from one's house toward the competitor's house.
The two participants were given the following rules:
--They had to agree on the choice of houses, because the two houses had to be on the opposite sides of the track.
--They had to agree on who would throw the dice.
--Each of the four pieces had to be advanced during the first four throws of the dice.
-Two pieces could not stop in the same box, so the player who arrived second was to go ahead to the next box.
--When one of the players arrived at his or her own "Box A," he or she had to choose among three different competitive strategies:
a. move one of the other player's pieces three spaces backward,
b. move one of his or her own pieces three spaces forward,
c. move his or her own piece six spaces forward and move the other player's piece three spaces forward.
6. When a player arrived at a "Box A" of the other player, he or she had to choose one of the following:
a. add a piece to the other player,
b. eliminate one of his or her own pieces,
c. eliminate two of his or her own pieces and one piece of the other player.
This task was a cooperative dyadic game similar to that of Doise and Mugny (1981) and used by one of the authors in previous studies (Fonzi & Menesini, 1993; Menesini & Fonzi, 1991). The game involved motor coordination between the two partners; therefore, cooperation was necessary for any successful outcome.
The task was as follows: A mobile object was linked to three pulleys by wires. The three pulleys were fixed to a table top (82 x 95 cm). One of the pulleys was stationary, and the others were movable. The two children operated one pulley each, in order to make the object move along a track drawn on a sheet of paper.
Social Interaction of Bullies and Victims
With regard to the first aim of the study, which was to examine the different interactional styles of bullies and victims, significant differences between the two partners were found in relation to regulative and collaborative behaviors. Bullies used a higher percentage of regulation than victims did, whereas victims used more collaborative behaviors than bullies did.
For the more analytic categories of the children's behavior, bullies showed a higher level of commands and victims showed a higher level of compliant behavior. Interaction effects revealed that these differences were greater during the cooperation task when the two partners interacted more frequently. Bullies were more dominant, and they regulated and gave commands to the partner more frequently than victims did, whereas victims complied and acquiesced to the partner more than bullies did.
The hypothesis that bullies would demonstrate higher levels of aggressive behavior was not supported. Interestingly, the frequency of aggressive behavior was generally low in both games and in relation to both partners. One might think that structured settings like the games used in this study can reduce the rate of aggressive exchanges among children. This result has relevant implications in terms of intervention and suggests that a large number of aggressive episodes take place during informal activities or during recess (Genta et al., 1996; Whitney & Smith, 1993). Further studies comparing structured tasks with informal settings may provide support for this result.
Effects of Context on Bullies' and Victims' Interactions
The second aim of the study was to study the effects of the game setting on children's interactions. The results indicated that the two games elicited different patterns of behavior in the participants. Both bullies and victims made use of more collaboration and regulation in the cooperative setting than in the competitive setting and, in general, the absolute level of interaction was higher during cooperation than during competition. The cooperative task demanded a higher level of interdependence between the two partners to reach the common goal; therefore, the children could profit from interacting with each other.
An unexpected result was the effect that game type had on aggression. Although the absolute value of aggression was relatively low, the comparison between the two settings revealed a higher level of aggression during the cooperative task. Within the aggressive category, the most frequent behaviors were angry gestures, like shaking fists, grinding teeth, and snorting. These behaviors were directed to the context, not to the partner, and were related to the children's efforts to perform well and complete the task properly. In this context, aggressive behaviors did not appear to represent a direct attack toward the partner or a type of proactive bullying (Coie & Dodge, 1998; Coie et al., 1991); rather, they appeared to express difficulties the children were having in coordinating their actions and controlling their emotions.
The results also demonstrated a higher level of cohesive behaviors during cooperation than during competition. Cohesion may be used to counterbalance negative effects within the dyad and help the two partners reduce emotional conflict. Although the general level of social interaction was higher in the cooperative task than in the competitive task, these higher levels of both aggression and cohesion may suggest both interactional difficulties and efforts aimed to maintain a cooperative relationship between the two partners.
During the competitive game, behaviors such as play activity and verbal comments on the game occurred more frequently, whereas collaboration and regulation occurred less frequently. The structure of the competitive setting was regulated by game rules and, although the goal was interdependent, the game did not demand a high level of interaction between the two partners. The differences between the two settings confirm the roles of physical and social constraints imposed on behavior by the settings in which interaction takes' place.
Studies on school bullying (Genta et al., 1996; Olweus, 1993a; Whitney & Smith, 1993) have documented that most bullying occurs on the playground. In the school context, we can hypothesize that the constraints and the opportunities for aggressive behavior in the classroom differ from those on the playground. These constraints may explain the observed differences across settings (playground vs. class). However, as suggested by the present study, differences may exist between cooperative and interdependent tasks compared with more individual and parallel activities.
In addition, the low level of aggression we found in both games can partly reduce the validity of the study, because our aim was to observe aggressive and victimized children. At the same time, a context eliciting too high a level of aggression can create ethical problems for researchers. In relation to this issue, our decisions on the use of contrived play groups for this study took account primarily of participants' rights.
Effects of the Partner on Bullies and Victims' Interactions
Regarding the question of possible effects of the partner (opposite status or control status) on social interactions of bullies and victims, our hypotheses were partially supported. Significant effects were found only at a second level of analysis, when subcategories of regulation and collaboration were used. Significant differences emerged for the categories asking for explanations and opposition to the other's initiatives.
Asking for explanations occurred more frequently when a bully or a victim interacted with a control child than when a bully and a victim interacted with each other. The opposite trend was found for opposition, which occurred more frequently in a mixed-status setting (when bullies and victims interacted) than in a control setting. Exchanges between bullies and victims were characterized by higher levels of opposition and conflict, whereas interactions between a bully or a victim and a control-status partner showed a higher level of cooperation with the partner, expressed by behaviors like asking for help and explanations. According to the Status x Game x Partner interaction found for complying behavior, victims were more submissive than bullies, specifically during the cooperative task and in interaction with a bully partner.
In the structured context used in this study, bullies showed a bossy style toward the partner's behavior, possibly aimed at status seeking and dominance, often regulating and opposing the other's initiative. Victims, on the other hand, complied with the partner and used a more submissive style of interaction. These behavioral characteristics of the two partners are reinforced when bullies and victims interact.
Previous studies have shown that aggressive children place more value than nonaggressive children on achieving control over their victims (Boldizar et al., 1989). Thus, through submission, the victimized child reinforces the aggressor's motivation. The victim's submission and the aggressor's success, in turn, serve to increase the chances of the victim's being repeatedly abused (Pierce & Cohen, 1995). In the present study, a dynamic of dominance-submission appeared to develop between the bully-victim partners, specifically during the cooperative game when interdependence between the two partners was required.
Interaction with a control partner, on the other hand, resulted in a scenario in which bullies and victims seemed more able to cooperate with the partner, by opposing the partner's initiatives less frequently (bullies) and showing a tendency to affirm themselves in a positive way by asking the partner for help and explanations (victims). The probability that a negative dynamic between bullies and victims will lead to negative and stable patterns of interaction was confirmed also by the comparison between reciprocal- and nonreciprocal-nominations dyads.
All of the dyads observed in this study were in the same classes and were familiar with each other, but those who recognized reciprocally the role of the partner showed higher levels of dominant-submissive behaviors. In the dynamic between bullies and victims, a vicious cycle seems to occur. Bullies and victims collude in their complementary behaviors, and the dynamic within the dyad reinforces them to repeat these styles of interaction.
This behavior, in turn, exposes bullies to more opportunities for using violent and antisocial behavior (Loeber & Hay, 1997; Olweus, 1979) and victims to the risk of stable victimization, fear, anxiety, loss of self-esteem, and depression (Kochenderfer & Ladd, 1997). A way to interrupt this vicious cycle may be to facilitate interaction with a control partner who is external to the problem and able to elicit more positive behaviors from both bullies and victims.
Finally, the strengths and limitations of this study deserve attention. A strength is the observational approach and the use of contrived settings to observe bullies and victims in interaction. The findings support the idea that there are both individual and contextual characteristics that can explain the dynamic between bullies and victims. On a theoretical level, the findings underline the need to consider the individual as well as the social context of bullying in terms of environmental and social context and type of partner (Sutton, Smith, & Swettenham, 1998).
On a practical level, there are direct implications for interventions. Task characteristics and types of partner seem to be relevant in planning and implementing interventions in schools. More specifically, insights into the dynamics between bullies and victims may yield more innovative intervention strategies with both bullies and victims. For example, an approach focused on individual children can give results that may not be maintained when bullies and victims interact together or are in groups in which the level of interdependence among the partners is high.
On the other hand, results of studies that evaluated how effective anti-bullying interventions can be have revealed that the most promising interventions are those based on relational and systemic approaches (Olweus, 1993a; Smith & Sharp, 1994). A focus on the social contexts and the cultural climate leading to bullying and victimization is necessary for any intervention aimed at reducing the problem. Limits of the present study are the small sample size and the lack of control for gender effects. Thus, the results of the present investigation can be considered to be preliminary, resulting from a direct observational approach.
Further studies may contribute by replicating the data and disentangling the role of gender from bully--victim status. We have made a first attempt to control for gender effects with a subsample of 4 boys from the present study. That analysis revealed the expected trend we had found for the whole sample.
- Menesini, Ersilia; Melan, Elena; Pignatti, Barbara; Interactional Styles of Bullies and Victims Observed in a Competitive and a Cooperative Setting; Journal of Genetic Psychology, Sep2000, Vol. 161 Issue 3, p261
Reflection Exercise #12
The preceding section contained information regarding bullying behavior in cooperative vs. competitive tasks. Write three
case study examples regarding how you might use the content of this section in
In Menesini’s study, what was behavioral difference was noticed between the competitive and cooperative settings? Record the letter of the correct answer the .