Relationship between boys' normative beliefs about aggression and their physical, verbal, and indirect aggressive behaviors.

Lim, Si Huan ; Ang, Rebecca P.

Over the years, a consensus has been established among researchers studying aggression: males tend to be more aggressive than females (Archer, 2004a; Archer, 2004b; Archer & Haigh, 1997; Bettencourt & Miller, 1996; Crane-Ross, Tisak & Tisak, 1998; Eagiy & Steffen, 1986; Tappere & Boulton, 2004). In a meta-analytic review of the social-psychological literature, Eagly and Steffen (1986) found that 89% of social-psychological reports indicated that males were more aggressive than females. Moreover, Bettencourt and Miller (1996) and Archer (2004a) obtained a close to moderate effect size of 0.33 and 0.42, respectively for the finding that males scored higher than females for general aggression.

Other studies examined different aspects of aggression such as indirect, verbal, and physical aggression and the findings suggest that sex differences may differ for each type of aggression. In one study, Crick, Bigbee, and Howes (1996) found that girls cited more indirect aggressive acts to express anger, as compared to boys. However, in a meta-analytic review of sex differences in aggression in real-world settings, measured using self reports, Archer (2004a) reported a negligible effect size of 0.02, illustrating that girls were not higher on indirect aggression compared to boys.

Past research on verbal aggression supported the consensus that males tend to be verbally more aggressive than females (Archer, 2004a; Archer 2004b; Tapper & Boulton, 2004). However, the magnitude of the effect sizes for the differences found in verbal aggression varied from study to study, even when similar methods were used. For example, Archer (2004a) summarized the findings of 40 studies using self-reports in a meta-analysis and reported an effect size of 0.30. In a different study, also using self-reports, Archer and Haigh (1997) reported a much smaller effect size of 0.11. In contrast, the studies on physical aggression yielded relatively conclusive findings, compared to those on indirect and verbal aggression. In these studies, which involved children (Tapper & Boulton, 2004), adolescents (Crane-Ross et al., 1998), and adults (Archer, 2004a; Archer, 2004b; Archer & Haigh, 1997; Bettencourt & Miller, 1996; Eagly & Steffen, 1986), males were shown to have higher scores for physical aggression as compared to females. For most of these studies, moderate effect sizes were found for sex differences in physical aggression. For instance, Tapper and Boulton (2004) calculated the number of aggressive acts per hour for children aged three to six and found that boys engaged in a greater number of physically aggressive acts per hour as compared to girls. This finding had a moderate effect size of 0.63. Using self-reports, Archer (2004b) found similar sex differences among undergraduates. Male undergraduates reported more physical aggression than did female undergraduates, with a moderate effect size of 0.62.

Aggressiveness among males was particularly pronounced when their opponents were males (Hoppe, 1979), probably because males usually have more approving beliefs and intentions about aggression toward other males (Harris, 1994). In one study, using hypothetical events, Archer and Haigh (1997) showed that among students, males scored higher on instrumental beliefs supporting same-sex physical aggression. There was a similar finding by Winstok and Enosh (2007). In their study, they found that among Israeli adolescents, males were most concerned with the gender of the provocateur and had intentions to react more aggressively in terms of physical aggression toward male provocateurs. To explain this phenomenon, past research has provided plausible reasons from evolutionary and social perspectives.

Evolutionary Perspective

According to the evolutionary perspective, males and females had different selection pressures as a result of their different amount of parental investment. For females, parental investment in their offspring is usually great; in order to maximize survival, they had to be very selective, obtaining males with good genes and resources. On the other hand, for males, parental investment in their offspring is usually minimal; in order to maximize survival, they tend to seek to mate with numerous females. To attract numerous females, they had to show that they have good genes and resources, and this is often displayed through being dominant over other males. As a result, inter-male competition was frequent and there were greater risks involved when males engaged in conflicts with other males (Archer, 1996), which might explain why males were often more aggressive, especially toward other male provocateurs.

In addition, since most of the females' responsibilities were about nurturing their offspring, males had to take up the remaining responsibilities such as co-opting resources of others for one's offspring, defending against attacks from rivals, and negotiating status and power hierarchies (Buss, 1997). These responsibilities usually involved physical aggression against other male rivals in order to deter future aggression from these rivals. Therefore, from the evolutionary perspective, the aggressive patterns displayed by males today, could be partly due to differential selection pressures on males and females that resulted from the differing amounts of parental investment required by them.

Social Perspective

According to social perspective, greater aggression among males was argued to be due to the traditional masculine socialization of the male gender role. This could have led to the restriction of emotional expression and greater display of toughness, dominance, and aggression (Feder, Levant, & Dean, 2007). This might explain Eagly and Steffen's (1986) finding that men tend to be more aggressive than women when the situation provided an opportunity for physical aggression, as compared to psychological aggression. Probably, overt physical aggression was more obvious than psychological aggression and thus was easily reinforced by socialization.

With this masculine socialization, a social norm was created in which it was more acceptable for males to engage in aggressive behaviors, especially when they were directed toward other males. For example, using hypothetical scenarios, Harris (1994) found that male undergraduates expected more approval of aggression from their friends in scenarios not involving male-female dating; specifically, male undergraduates expected more approval of aggression from their friends when aggression was not directed toward females. More importantly, this social norm would shape individuals' own normative beliefs about aggression which would, in turn, affect their display of aggressive behaviors. This was shown in a study conducted by Henry et al. (2000). Guerra et al.'s (2007) study, showed that injunctive norms, which were classmates' beliefs about the acceptability of aggression, affected individuals' normative beliefs about aggression; in classrooms where teachers and students had salient norms against aggression, aggressive beliefs and behaviors diminished over time. Salmivalli and Voeten (2004) provided similar evidence by showing that in the classroom, a low degree of anti-bullying norms predicted bullying of others, as well as reinforcement of bullies.

Following this line of research, it seemed that one's attitudes and beliefs about the acceptability and legitimacy of aggression would have a direct effect on aggressive behaviors. In fact, this was true across a wide range of individuals, including children, adolescents, and adults. In one study, it was shown that in children, individual differences in aggressive behavior were predicted by preceding differences in normative beliefs about aggression (Huesmann & Guerra, 1997). Also, children who believed strongly in the legitimacy of aggression were less withdrawn, less prosocial, and more aggressive than those who did not hold such beliefs (Erdley & Asher, 1998). For adolescents, convention-violating and aggressive behaviors could also be predicted by normative beliefs about legitimacy of aggression (Bellmore, Witkow, Graham, & Juvonen, 2005; Crane-Ross et al., 1998). For young adults, it has been shown that more approving beliefs about aggression led to higher levels of self-reports regarding one's own aggressive behaviors (Harris, 1994).

Thus, it is not difficult to see that past evolutionary pressures could have yielded a specific pattern of male aggressive behaviors, which were reinforced by present-day social norms. This sustained pattern of male aggression is problematic because what appeared adaptive in our evolutionary past might be maladaptive in today's civilized society, when aggression becomes an externalized behavioral problem. It could be argued that since our evolutionary past is fixed, studies investigating the link between normative beliefs about aggressive behaviors brought some hope to intervention programs aimed at reducing aggression. In particular, intervention should attempt to modify individuals' social-cognitive beliefs (Guerra et al., 2007). Yet, individuals' normative beliefs about aggression are usually complex, often involving the consideration of aggression under provocation (Huesmann & Guerra, 1997), which might result in different patterns of aggressive behaviors in the face of provocation.

Aggression under Provocation

Past research suggests that under provocation, even a weak one, retaliatory aggression becomes more acceptable for adjudicated males (Haft, Floyd, & Shinn, 2006). Another line of research suggested that aggression heightened following increases in provocation (Hoppe, 1979; Stadler, Rohrmann, Steuber, & Poustka, 2006). In an interesting experiment with manipulated provocation, Chermack, Berman, and Taylor (1997) showed that the shock levels set by male participants to a provocative opponent increased significantly as provocation increased. On the other hand, male participants assigned to a low constant provocation group showed highly stable aggressive responding. These studies suggested that individuals tended to be aggressive under provocation. However, only limited research has demonstrated a relation between beliefs about retaliatory aggression and retaliatory aggression under provocation (Zelli, Dodge, Lochman, & Laird, 1999).

Specificity of Normative Beliefs in Predicting Aggression

It is important to establish the link between beliefs about retaliatory aggression and retaliatory aggression under provocation in order to refine intervention programs, as there is evidence suggesting that beliefs are specific in terms of predicting behaviors. Werner and Nixon's (2005) study revealed that normative beliefs about relational aggression were uniquely associated with engagement in relationally aggressive behaviors but not physically aggressive behaviors, while normative beliefs about physical aggression were uniquely associated with engagement in physically aggressive behaviors but not relationally aggressive behaviors. This implied that the prediction of aggressive behaviors based on normative beliefs about aggression was highly specific. Thus, past research was inadequate in showing whether specific normative beliefs about retaliatory aggression could predict retaliatory aggressive behaviors.

PRESENT STUDY

This study aimed to examine the separate contribution of general normative beliefs about aggression and specific normative beliefs about retaliatory aggression in predicting aggressive behaviors, using a sample of elementary school boys. We sought to identify the contribution of specific normative beliefs about retaliatory aggression in predicting aggressive behaviors, controlling for the contribution of general normative beliefs about aggression. Based on past studies, it was believed that general normative beliefs about aggression would contribute significantly in predicting aggressive behaviors (Crane-Ross et al., 1998; Erdley & Asher, 1998; Huesmann & Guerra, 1997). Recent but limited evidence suggests that specific beliefs about retaliatory aggression could contribute significantly in predicting aggressive behaviors; it is plausible that as individuals become more approving of aggressive behaviors under provocation (Felix & McMahon, 2007; Haft et al., 2006), this could contribute to use of aggressive behaviors (Crane-Ross et al., 1998; Erdley & Asher, 1998). Therefore, we hypothesized that specific normative beliefs about retaliatory aggression would also contribute significantly in predicting aggressive behaviors.

We were also interested in how specific normative beliefs about retaliatory aggression against males and against females differ in predicting aggressive behaviors. We expected that specific normative beliefs about retaliatory aggression against males, being part of specific normative beliefs about retaliatory aggression, would contribute significantly in predicting aggressive behaviors. On the other hand, although also part of specific normative beliefs about retaliatory aggression, we expected that specific normative beliefs about retaliatory aggression against females would not contribute significantly in predicting aggressive behaviors. This is because males' evolutionary past supported inter-male aggression (Archer, 1996) and social conventions were strongly against males' aggression toward females (Feder et al., 2007; Harris, 1994; Hilton, Harris, & Rice, 2000). Therefore, even though males might have specific beliefs about retaliatory aggression against females, they might not externalize these beliefs.

Instead of simply assessing general aggressive behaviors, we chose to use three distinct types: verbal, physical, and indirect. While much has been written about males' physical aggression, males' engagement in verbal and indirect aggression has not been as frequently researched. Physical aggression alone is probably not an accurate reflection of males' aggressive behavior as it is not uncommon for individuals usually to engage in more than one type of aggressive behavior (Salmi valli, Kaukiainen, & Lagerspetz, 2000). Moreover, male Spanish adolescents were shown to be more aggressive than female Spanish adolescents: physical, verbal, and indirect aggressive behaviors (Toldos, 2005). Additionally, it has been shown that there were moderate correlations between physical, verbal, and indirect aggression (Hayward & Fletcher, 2003; Salmivalli et al., 2000). This finding suggested that although the different types of aggression were related, they were nonetheless distinct. Furthermore, children as young as those in Grade 3 developed different coping responses for the different types of aggression (Phelps, 2001), which further supported the view that the various types of aggression were qualitatively different. Hence, it is essential that each type of aggressive behavior be studied independently so that the contribution of normative beliefs about aggression in predicting aggressive behaviors would not be obscured by the interacting effects of the different types of behaviors.

Since males usually engage in multiple aggressive behaviors (Salmivalli et al., 2000) and were more aggressive in all three types of behaviors (Toldos, 2005), we would expect similar patterns to emerge for these behaviors when applying general and specific normative beliefs about aggression to predict physical, verbal, and indirect aggressive behaviors.

Hence, the present study had two hypotheses. First, that general normative beliefs about aggression would contribute significantly in predicting self-reported physical, verbal, and indirect aggressive behaviors. Second, that specific normative beliefs about retaliatory aggression would contribute significantly in predicting self-reported physical, verbal, and indirect aggressive behaviors even after controlling for general normative beliefs about aggression. In particular, specific normative beliefs about retaliatory aggression against males but not specific normative beliefs about retaliatory aggression against females would contribute significantly in predicting self-reported physical, verbal, and indirect aggressive behaviors.

METHOD

Participants

Two hundred and forty-nine Grade 4 (N = 113) and Grade 5 (N = 136 boys), ranging in age from 9 to 14 years (M = 10.80, SD = .91) were sampled from two elementary schools (N = 85; N = 164). Self-reported ethnic identification of these 249 boys were as follows: 142 Chinese, 26 Indian, 67 Malay, 1 Eurasian, and 13 Other.

Measures

Normative Beliefs about Aggression Scale (NOBAGS). This is a 20item scale (Huesmann & Guerra, 1997) consisting of three subscales measuring children's beliefs about the acceptability of aggressive behaviors. For each of the items, participants were required to respond on a 4-point scale, indicating if the particular aggressive behavior was very wrong (0), a bit wrong (1), a bit OK(3). The first subscale contained 8 items measuring children's general beliefs about the acceptability of aggressive behaviors; half of these items were negatively worded (e.g., "It is wrong to take it out on others by saying bad things when you are angry") while the other half of these items were positively worded (e.g., "It is OK to push other people around if you are angry"). A total score was obtained by totaling up the scores on these 8 items, with a higher score indicating greater acceptability of general aggression. In the current study Cronbach's alpha for this subscale was 0.78.

The second subscale contained 6 items measuring children's specified beliefs about the acceptability of retaliatory aggressive behaviors toward males, and the third subscale contained 6 items measuring children's specified beliefs about the acceptability of retaliatory aggressive behaviors toward females. Of the 6 items that measured children's beliefs about the acceptability of retaliatory aggressive behaviors toward males, 3 items were negatively worded (e.g., "If a boy says something bad to a girl, do you think it's wrong for the girl to shout at him?") and 3 items were positively worded (e.g., "If a boy hits a girl, do you think it's OK for the girl to hit him back?"). A total score was obtained by adding up these scores, with a higher score indicating greater acceptability of retaliatory aggression toward males. In the current study, Cronbach's alpha for this subscale was 0.76.

Of the 6 items that measured children's specified beliefs about the acceptability of retaliatory aggressive behaviors toward females, 4 items were negatively worded (e.g., "If a girl says something bad to a boy, do you think it's wrong for the boy to shout at her?") and 2 items were positively worded (e.g., "If a girl says something bad to another girl, Mary, do you think it's OK for Mary to shout at her?"). A total score was obtained by adding up these responses, with higher scores indicating acceptability of retaliatory aggression toward females. In the current study, Cronbach's alpha for this subscale was 0.83.

Assessment of aggressive behaviors. Three single items were used to assess how often children engage in different types of aggressive behaviors. To assess the frequency of physical aggression, the item "I physically hurt someone (e.g., hit, kick)" was used. To assess the frequency of verbal aggression, the item "I called someone a bad name because of how he or she looks like" was used. To assess the frequency of indirect aggression, the item "I spread rumors about someone" was used. For these three items, participants were required to respond on a 5-point scale, 1 being never, 2 being one time or two times this year, 3 being a few times this year, 4 being about one time every week, and 5 being about a few times every week.

Procedure

Following the procedure for ethical clearance and data collection from schools in Singapore, permission was sought and approval obtained from the Ministry of Education in Singapore and both schools prior to conducting the research. A passive consent procedure was used to obtain children's participation approval from parents. All parents involved were informed about the nature of the study well in advance of questionnaire administration. Parents were requested to contact the school if they did not want their child to participate. None of the parents withheld consent. Assent was also required from the children. Participation was voluntary and participants were informed that they could refuse or discontinue the study at any time without penalty. The questionnaire was administered in English, and participants completed the questionnaires in their respective classrooms. The classroom teacher and one to two research assistants were present in each classroom to answer children's questions regarding the procedure. No translation was needed since English is the main language of instruction for schools in Singapore.

RESULTS

Table 1 presents the means and standard deviations of the predictors (general normative beliefs about aggression, specific normative beliefs about retaliatory aggression against males, and specific normative beliefs about retaliatory aggression against females), and dependent variables (verbal, physical, and indirect aggression). A series of hierarchical multiple regression analyses investigated whether specific normative beliefs about retaliatory aggression accounted for physical, verbal, and indirect aggressive behaviors, controlling for general normative beliefs about aggression, in a sample of elementary school boys from Singapore.

The hypotheses were tested using three separate runs of hierarchical multiple regression using physical, verbal, and indirect aggressive behaviors as dependent variables, respectively. Physical, verbal, and indirect aggressive behaviors were regressed onto specific normative beliefs about retaliatory aggression against males and specific normative beliefs about retaliatory aggression against females, controlling for general normative beliefs about aggression. In each analysis, general normative beliefs about aggression was entered at Step 1 of the analysis while specific normative beliefs about retaliatory aggression against males and specific normative beliefs about retaliatory aggression against females were entered at Step 2. These results are reported in Table 2.

General normative beliefs about aggression made a statistically significant contribution to the prediction of physical aggressive behavior [[DELTA][R.sup.2] = .06, [DELTA]F(1, 247) = 16.29, p < .01], verbal aggression behavior [[DELTA][R.sup.2] = .05, [DELTA]F(1,247) = 14.02, p < .01], and indirect aggressive behavior [[DELTA][R.sup.2] = .13, [DELTA]F(1, 247) = 35.98, p < .01]. The first hypothesis was supported. When general normative beliefs about aggression was controlled for, specific normative beliefs about retaliatory aggression still made a statistically significant contribution to the prediction of physical aggressive behavior [[DELTA][F.sup.2] = .02, [DELTA]F(1, 245) = 3.26, p < .05], verbal aggressive behavior [[DELTA][R.sup.2] = .04, [DELTA]F(1, 245) = 5.32, p < .01], and indirect aggressive behavior [[DELTA][R.sup.2] = .06, [DELTA]F(1, 245) = 5.734, p < .01]. Standardized beta weights indicated that specific normative beliefs about retaliatory aggression against males was responsible for the additional variance explained for physical aggressive behavior ([beta] = .19, p < .05), verbal aggressive behavior ([beta] = .22, p < .01), and indirect aggressive behavior ([beta] = .18, p < .05). In contrast, standardized beta weights indicated that specific normative beliefs about retaliatory aggression against females was not responsible for the additional variance explained for physical aggressive behavior ([beta] = -.02, ns), verbal aggressive behavior ([beta] = .01, ns), and indirect aggressive behavior ([beta] = .07, ns). The second hypothesis was supported.

DISCUSSION

Consistent with previous research indicating that general normative beliefs about aggression predicted greater aggressive behaviors (Bellmore et al., 2005; Crane-Ross et al., 1998; Erdley & Asher, 1998; Haroris, 1994; Huesmann & Guerra, 1997), this study found that among elementary school boys, general normative beliefs about aggression contributed significantly in predicting physical, verbal, and indirect aggressive behaviors. More importantly, this study added to past research in establishing the link between specific normative beliefs about retaliatory aggression and aggressive behaviors, controlling for the contribution of general normative beliefs about aggression. Specifically, the finding in that controlling for general normative beliefs about aggression, specific normative beliefs about retaliatory aggression against males but not specific normative beliefs about retaliatory aggression against females, was the significant predictor of physical, verbal, and indirect aggressive behaviors among elementary school boys.

Given the link between general normative beliefs about aggression and aggressive behaviors, researchers have suggested intervention programs that change social-cognitive beliefs of children (Guerra et al., 2007). Altering the social-cognitive beliefs of aggressive children to a belief system that does not support legitimacy of aggression could help reduce their aggressive behavior. The importance of the present findings lies in the implication that normative beliefs about aggression were specific in predicting aggressive behaviors. In particular, specific normative beliefs about retaliatory aggression against males contributed uniquely and significantly over and above general normative beliefs that aggression predicts physical, verbal, and indirect aggressive behaviors. Therefore, when intervention programs are designed to change children's beliefs about aggression, specific procedures have to be implemented to change both general and specific normative beliefs in order to be effective in reducing both general aggression and retaliatory aggression.

Thus, our findings could contribute to the design and development of prevention programs. For example, such programs could target existing social norms. As previously mentioned, it was often more acceptable for males to be aggressive toward or retaliate against males (Archer & Haigh, 1997; Hope, 1979; Winstok & Enosh, 2007). Furthermore, Kerestes and Milanovic (2006) showed that direct and indirect aggression were both related to peer rejection and, in particular, aggressive boys were rejected more by their male classmates than by their female classmates. Modifying norms for male-to-male aggression may reduce the threat of peer rejection felt by boys.

Additionally, social norms that encourage prosocial behaviors of males can be established. It has been shown that men tend to help women more (Eagly & Crowley, 1986), which might explain why, in this study, specific normative beliefs about retaliatory aggression against females did not predict aggressive behaviors since males were socialized to help women. In the same way that social norms that encourage males to help females could reduce aggression by males against females, social norms that encourage males to help males might also reduce this type of aggression.

One limitation of this study was that the assessment of aggressive behaviors did not specify the gender of the opponents even though it could be argued that due to greater sensitivity to the gender of the provocateur (Winstok & Enosh, 2007) and approval of aggression toward and retaliation against males (Archer & Haigh, 1997; Hoppe, 1979; Winstok & Enosh, 2007), it was highly likely that in the absence of a specific gender of opponent in the study, participants had a male opponent in mind. Furthermore, single items were used to measure physical, verbal, and indirect aggressive behaviors. Lastly, both normative beliefs about aggression and aggressive behavior assessments were based on self-report. Thus, common method variance may result in spuriously high correlations among constructs. Additionally, the use of self-reports without behavioral assessments provided by others could be inaccurate. In addition to specifying the gender of opponents and the use of multiple items in the measurement of aggressive behavior, future research should consider a multiple informant and multiple method strategy to overcome the problem of common method variance.

CONCLUSION

Findings from this study suggest that general normative beliefs about aggression predict aggressive behaviors. When general normative beliefs were controlled for, specific normative beliefs about retaliatory aggression against males contributed uniquely and significantly in predicting aggressive behaviors. This implies that future intervention programs should be designed to change specific normative beliefs about retaliatory aggression, as well as general normative beliefs in order to be more effective in reducing aggressive behaviors among children.

REFERENCES

Archer, J. (1996). Sex differences in social behavior: Are the social role and evolutionary explanations compatible? American Psychologist, 51, 909-917.

Archer, J. (2004a). Sex differences in aggression in real-world settings: A meta-analytic review. Review of General Psychology, 8, 291-322.

Archer, J. (2004b). Which attitudinal measures predict trait aggression? Personality and Individual Differences, 36, 47-60.

Archer, J., & Haigh, A. (1997). Sex differences in beliefs about aggression. Opponent's sex and the form of aggression. The British Journal of Social Psychology, 38, 71-84.

Bellmore, A. D., Witkow, M. R., Graham, S., & Juvonen, J. (2005). From beliefs to behavior: The mediating role of hostile response selection in predicting aggression. Aggressive Behavior, 31, 453-472.

Bettencourt, B. A., & Miller, N. (1996). Gender differences in aggression as a function of provocation: A meta-analysis. Psychological Bulletin, 119, 422-447.

Buss, D. M. (1997). Human aggression in evolutionary psychological perspective. Clinical Psychology Review, 17, 605-619.

Chermack, S. T., Berman, M., & Taylor, S. P. (1997). Effects of provocation on emotions and aggression in males. Aggressive Behavior, 23, 1-10.

Crane-Ross, D., Tisak, M. S., & Tisak, J. (1998). Aggression and conventional rule violation among adolescents: Social-reasoning predictors of social behavior. Aggressive Behavior, 24, 347-465.

Crick, N. R., Bigbee, M. A., & Howes, C. (1996). Gender differences in children's normative beliefs about aggression: How do I hurt thee? Let me count the ways. Child Development, 67, 1003-1014.

Eagly, A. H., & Crowley, M. (1986). Gender and helping behavior: A meta-analytic review of the social-psychological literature. Psychological Bulletin, 100, 283-308.

Eagly, A. H., & Steffen, V. J. (1986). Gender and aggressive behavior: A meta-analytic review of the social-psychological literature. Psychological Bulletin, 100, 309-330.

Erdley, C. A., & Asher, S. R. (1998). Linkages between children's beliefs about the legitimacy of aggression and their behavior. Social Development, 7, 321-339.

Feder, J., Levant, R. F., & Dean, J. (2007). Boys and violence: A gender-informed analysis. Professional Psychology: Research and Practice, 38, 385-391.

Felix, E. D., & McMahon, S. D. (2007). The role of gender in peer victimization among youth: A study of incidence, interrelations, and social cognitive correlates. Journal of School Violence, 6, 27-44.

Guerra, N., Henry, D., Huesmann, L. R., & Tolan, P. (2007). Changing the way children "think" about aggression: Social-cognitive effects of preventive intervention. Journal of Consulting and Clinical Psychology, 75, 160-167.

Haff, D. W., Floyd, H. H., & Shinn, L. K. (2006). Acceptance patterns of retaliatory aggression among adjudicated youth by sex and race. Psychological Reports, 99, 335-342.

Harris, M. B. (1994). Gender of subject and target as mediators of aggression. Journal of Applied Social Psychology, 24, 453-471.

Hayward, S. M., & Fletcher, J. (2003). Relational aggression in an Australian sample: Gender and age differences. Australian Journal of Psychology, 55, 129-134.

Henry, D., Guerra, N., Huesmann, R., Tolan, P., VanAcker, R., & Eron, L. (2000). Normative influences on aggression in urban elementary school classrooms. American Journal of Community Psychology, 28, 59-81.

Hilton, N. Z., Harris, G. T., & Rice, M. E. (2000). The functions of aggression by male teenagers. Journal of Personality and Social Psychology, 79, 988-994.

Hoppe, C. M. (1979). Interpersonal aggression as a function of subject's sex, subject's sex-role identification, opponent's sex, and degree of provocation. Journal of Personality, 47, 317-329.

Huesmann, L. R., & Guerra, N. G. (1997). Children's normative beliefs about aggression and aggressive behavior. Journal of Personality and Social Psychology, 72, 408-419.

Keresteg, G., & Milanovic, A. (2006). Relations between different types of children's aggressive behavior and sociometric status among peers of the same and opposite gender. Scandinavian Journal of Psychology, 47, 477-483.

Phelps, C. E. R. (2001). Children's responses to overt and relational aggression. Journal of Clinical Child Psychology, 30, 24.0-252.

Salmivalli, C., Kaukiainen, A., & Lagerspez, K. (2000). Aggression and sociometric status among peers: Do gender and type of aggression matter? Scandinavian Journal of Psychology, 41, 17-24.

Salmivalli, C., & Voeten, M. (2004). Connections between attitudes, group norms, and behavior in bullying situations. International Journal of Behavioral Development, 28, 246-258.

Stadler, C., Rohrmann, S., Steuber, S., & Poustka, F. (2006). Effects of provocation on emotions and aggression: An experimental study with aggressive children. Swiss Journal of Psychology, 65, 117-124.

Tapper, K., & Boulton, M. J. (2004). Sex differences in levels of physical, verbal, and indirect aggression amongst primary school children and their associations with beliefs about aggression. Aggressive Behavior, 30, 123-145.

Toldos, M. P. (2005). Sex and age differences in self-estimated physical, verbal, and indirect aggression in Spanish adolescents. Aggressive Behavior, 31, 13-23.

Werner, N. E., & Nixon, C. L. (2005). Normative beliefs and relational aggression: An investigation of the cognitive bases of adolescent aggressive behavior. Journal of Youth and Adolescence, 34, 229-243.

Winstok, Z., & Enosh, G. (2007). Exploring the intention to react to aggressive action among Israeli adolescents. Journal of Research on Adolescence, 17, 507-524.

Zelli, A., Dode, K. A., Lochman, J. E., & Laird, R. D. (1999). The distinction between beliefs legitimizing aggression and deviant processing of social cues: Testing measurement validity and the hypothesis that biased processing mediates the effects on beliefs on aggression. Journal of Personality and Social Psychology, 77, 150-166.

This project was conducted by Si Huan Lim and supervised by Rebecca P. Ang, as part of the Undergraduate Research Experience on Campus (URECA) program from Nanyang Technological University, Singapore. The project was supported by the URECA program and an Academic Research Fund Tier 1 grant (RG96/07) awarded to Rebecca P. Ang.

Si Huan Lim and Rebecca P. Ang, Division of Psychology, School of Humanities and Social Sciences, Nanyang Technological University, Singapore.

Requests for reprints should be sent to Rebecca P. Ang, Division of Psychology, School of Humanities and Social Sciences, Nanyang Technological University, Nanyang Avenue, Singapore 639798. E-mail: rpang@ntu.edu.sg Table 1 Means and Standard Deviations for Predictors and Dependent Variables Standard Variables Mean Deviation General 4.43 4.56 NOBAGS Specific 4.80 4.05 NOBAGS(M) Specific 4.81 4.42 NOBAGS(F) Physical Aggression 1.69 1.02 Verbal Aggression 1.94 1.19 Indirect Aggression 1.58 0.96 Note. N = 249. General NOBAGS = General normative beliefs about aggression. Specific NOBAGS(M) = Specific normative beliefs about retaliatory aggression against males. Specific NOBAGS(F) = Specific normative beliefs about retaliatory aggression against females. Table 2 Hierarchical Regression Analysis for Predicting Physical, Verbal and Indirect Aggressive Behaviors from General Normative Beliefs about Aggression and Specific Normative Beliefs about Retaliatory Aggression [DELTA] [DELTA] Predictor Measure [beta] [R.sup.2] [R.sup.2] F and Step Predicting PA Step 1 General NOBAGS .25 ** .06 .06 16.29 ** Step 2 Specific NOBAGS(M) .19 * .09 .02 3.26 * Specific NOBAGS(F) -.02 Predicting VA Step 1 General NOBAGS .23 ** .05 .05 14.02 ** Step 2 Specific NOBAGS(M) .22 ** .09 .04 5.32 ** Specific NOBAGS(F) .01 Predicting IA Step 1 General NOBAGS .36 ** .13 .13 35.98 ** Step 2 Specific NOBAGS(M) .18 * .17 .04 5.73 ** Specific NOBAGS(F) .07 Note. PA = Physical aggressive behavior. VA = Verbal aggressive behavior. IA = Indirect aggressive behavior. General NOBAGS = General normative beliefs about aggression. Specific NOBAGS(M) = Specific normative beliefs about retaliatory aggression against males. Specific NOBAGS(F) = Specific normative beliefs about retaliatory aggression against females. * p <.05. ** p <.01.