Kirsti M. J. Lagerspetz

Bullying as a group process: Participant roles and their relations to social status within the group

Bullying was investigated as a group process, asocial phenomenon taking place in a school setting among 573 Finnish sixth-grade children (286 girls, 287 boys) aged 12-13 years. Different Participant Roles taken by individual children in the bullying process were examined and related to a) self-estimated behavior in bullying situations, b) social acceptance and social rejection, and c) belongingness to one of the five sociometric status groups (popular, rejected, neglected, controversial, and average). The Participant Roles assigned to the subjects were Victim, Bully, Reinforcer of the bully, Assistant of the bully, Defender of the victim, and Outsider. There were significant sex differences in the distribution of Participant Roles. Boys were more frequently in the roles of Bully, Reinforcer and Assistant, while the most frequent roles of the girls were those of Defender and Outsider. The subjects were moderately well aware of their Participant Roles, although they underestimated their participation in active bullying behavior and emphasized that they acted as Defenders and Outsiders. The sociometric status of the children was found to be connected to their Participant Roles..

Do girls manipulate and boys fight? developmental trends in regard to direct and indirect aggression

Gender differences in regard to aggressive behaviour were investigated in a series of studies of schoolchildren of different age cohorts: 8-year-olds (N = 85), 11-year-olds (N = 167), and 15-year-olds (N = 127). Different types of aggressive behaviour were measured with peer nomination techniques, supported by self-ratings. The social structure of the peer groups were also studied. The results of the 11-year-old cohort were previously presented by Lagerspetz et al. [1988; Aggressive Behavior 14:403-4141, but they are compared here with the other age groups. The principal finding was that girls of the two older cohorts overall make greater use of indirect means of aggression, whereas boys tend to employ direct means. Previously, the main difference between the genders has been thought to be that boys use physical aggressive strategies, while girls prefer verbal ones. Our studies suggest that the differentiation between direct and indirect strategies of aggression presents a more exact picture. Indirect aggressive strategies were not yet fully developed among the 8-year-old girls, but they were already prominent among the 11-year-old girls. Aggressive behaviour was assessed overall by the children themselves to be the highest in this age group.

Is indirect aggression typical of females? gender differences in aggressiveness in 11‐ to 12‐year‐old children

Gender differences regarding aggressive behaviour were investigated in 167 school children, 11 to 12 years of age, through peer-rating techniques supported by self-ratings and interviews. The social structure of the peer groups also was studied. The principal finding was that girls made greater use of indirect means of aggression, whereas the boys tended to employ direct means. Gender differences in verbal aggression were less pronounced. The social structure of peer groups was found to be tighter among girls, making it easier for them to exploit relationships and harm their victims by indirect manipulative aggression. Because indirect aggression has rarely been satisfactorily studied with tests of aggression, this finding may help to explain 1) the generally lower correlation found between peer-rated and self-rated aggression in girls than among boys (indirect means not being so readily recognized by the subject as a kind of aggression) and 2) the low stability of aggressiveness in girls often found in developmental studies.

The relationships between social intelligence, empathy, and three types of aggression

Relationships between social intelligence, empathy, and three types of aggressive behavior were studied. Peer-estimation techniques were used to measure all major variables. Altogether, 526 Finnish schoolchildren from three age groups (10, 12, and 14 years old) participated in the study. As was hypothesized, indirect aggression correlated positively and significantly with social intelligence in every age group studied. Physical and verbal forms of aggression had almost zero correlation to social intelligence. Empathy correlated negatively and significantly with every type of aggression except indirect aggression in 12-year-old children. The major findings are in Line with the developmental theory by Bjorkqvist et al. [1992. Aggr Behav 18:117-127] suggesting that indirect aggression requires more social intelligence than direct forms of aggression.

Sex differences in covert aggression among adults

In previous studies [Aggressive Behavior 14:403-414, 1988; Aggressive Behavior 18:117-127, 1992] it was found that adolescent girls used indirect means of aggression more than adolescent boys, whereas physical aggression was used more by boys, and direct verbal aggression was used equally by both sexes. The present study investigated whether males, as adults, start to employ indirect aggression to the same extent is females. Three hundred thirty-three university employees (162 males, 176 females) filled in the Work Harassment Stale by Bjorkqvist et al. [1994]. Special attention was drawn to two subscales: rational-appearing aggression and social manipulation. It was found that males used the former type of aggression significantly more often than females, while females used the latter more than males. Both are variants of covert aggression, in which the perpetrator tries to disguise his/her aggressive intentions, in order to avoid retaliation and/or social condemnation.

Peer networks and bullying in schools

This study looked at how the social constellations in school classes relate to bullying problems. Using peer-evaluation questionnaires, the peer networks of children with different participant roles (such as victim, bully, assistant of bully, reinforcer of bully, defender of victim, outsider) were explored. The subjects were 459 sixth-grade-children (218 girls, 241 boys), aged 11 to 12 years, in Finland. The main findings were: 1) Children who tended to behave in either similar or complementary participant roles in situations of bullying formed networks with each other. The individual child’s behavior in bullying situations was strongly connected to how the members of his:her network behaved in such situations. 2) Bullies, assistants, and reinforcers belonged to larger networks than did defenders, outsiders and victims. 3) Children outside the networks were most often victims. It was concluded that behavior in bullying situations can be said to be one feature around which the peer networks in school classes are organized. Thus prevention, as well as intervention strategies against bullying should focus not only on individual children, but also on the wider social context of the class.

Self-Evaluated Self-Esteem, Peer-Evaluated Self-Esteem, and Defensive Egotism as Predictors of Adolescents’ Participation in Bullying Situations

Three dimensions of self-esteem (SE) (self-evaluated, peer-evaluated, defensive egotism) were related to each other and to adolescents’ social behavior. In addition to exploring links between single variables, five SE profiles were formed by means of a cluster analysis and connected to adolescents’ behavior in situations of bullying. Self-and peer-evaluated SEs were significantly correlated, whereas defensive egotism was not connected to either self-or peer-evaluated SE. Adolescents’ SE profiles were associated with their behavior in bullying situations; these connections were stronger among boys than among girls. Bullying others and assisting or reinforcing the bully were typical of adolescents with so-called defensive SE. Defending the victims of bullying was typical of adolescents with genuine high SE. Being victimized by peers was most typical of adolescents with low SE and, among girls, of those in the cluster thed authors named “humble pride.” The social behavior of so-called self-belittlers did not clearly distinguish them from the other groups.

Cross-cultural evidence of female indirect aggression.

Three types of aggressive behavior (physical, verbal, and indirect) were investigated by help of peer estimations based on The Direct and Indirect Aggression Scales (DIAS) (Finland: Abo Akademi University) [Bjorkqvist et al., 1992b]. Aggressive behavior of adolescents of three age groups (8, 11, and 15 years old), in Finland, (Finnish and Swedish speakers),Israel (secular and religious Israelis),Italy,and Poland were studied (n=2094). Indirect aggression was, in proportional terms, the aggressive style mostly used by girls, across nations, ethnic groups, and age groups studied. Verbal aggression was their second most used: style, and physical aggression was applied least often by girls. Among boys, indirect aggression was, in all ages, the least used aggressive style. Physical and verbal aggression was, by boys, used equally often at ages 8 and 11, while, at the age of 15, verbal aggression had surpassed physical aggression and was the most used style. Scores of victimization to others aggression showed somewhat similar trends.

Learning difficulties, social intelligence, and self–concept: Connections to bully–victim problems

Learning skills, social intelligence, and self-concept were related to each other and to bully-victim problems among fifth-grade children (79 boys and 62 girls, aged 11-12 years). In addition to exploring connections between single variables, a person-oriented approach was applied in order to analyze childrens value patterns with respect to learning skills, self-concept, and social intelligence, and how these value patterns are related to bully-victim problems. Social intelligence was found to be positively correlated with learning skills, but negatively related to victimization. Bullying was positively correlated with self-concept scores. However, this was true only of boys. According to cross-tabulations, there were significantly more bullies among children with learning difficulties (LD) than would have been expected by chance. Victimization, on the other hand, was not related to LD. LD childrens proposed victim status was in some degree supported by cluster analysis: a group of LD children emerged, who not only scored high on bullying, but also tended to be victimized by others. In addition, two groups of bullies appeared: one whose members could be interpreted as socially unskilled and another as socially skilled. This finding is in line with recent theoretical reasoning, which calls into question the idea of bullies as a unified group, lacking in social skills.

Peer and Self‐Estimated Aggression and Victimization in 8‐Year‐Old Children From Five Ethnic Groups

The Direct and Indirect Aggression Scales (DIAS) (Bjorkqvist et al. [1992b] Finland: Abo Akademi University) were applied in order to investigate the perception of aggression among 8-year-old children (n=404, girls 200, boys 204) in a cross-cultural comparison. Two samples from Finland (Finnish and Swedish-speaking children), two from Chicago, IL (African Americans and Caucasians), and one from Warsaw; Poland, were included in the study. These types of aggressive behavior and victimization of aggression (physical, verbal, and indirect) were measured using both peer and self estimations. Peer estimations were internally more consistent than self estimations. Children rated themselves as significantly less aggressive than their peers rated them. The opposite was true of victimization. An attributional discrepancy index (ADI) was calculated as the difference between self and peer estimated aggression sc ores. The index may be seen as an indicator of norms pertaining to aggression in different ethnic groups. ADI stores of girls, but not for boys, showed significant variance over culture on all three types of aggression. This indicates greater cultural variation in norms pertaining to aggression for girls than for boys. Cultural variation and sex differences in the patterning of aggressive behavior were analysed using multidimensional scaling (ALSCAL). Cultural variation appeared as distribution along the x-axis (dimension 1) in a two-dimensional solution, and sex differences along the y-axis (dimension 2), indicating that patterns of aggressive behavior are dependent on both culture and sex.