Maya Zayat, Beth A. Kotchick, Ph.D., Alison A. Papadakis, Ph.D., Mary Jo Coiro, Ph.D.
Associations among Gender, Ethnicity, Attitudes, and Aggression: A Mediation Model
Variations in aggressive behavior and attitudes are often linked to gender and ethnicity (Cullerton-Sen & Crick, 2005; Funk et al., 2003). This literature suggests that boys typically engage in more physical aggression (PA) and hold more favorable attitudes about PA than girls, while girls typically engage in and have more favorable attitudes about relational aggression (RA) than boys. Gender socialization theory (Block, 1983) provides a framework for understanding these differences by proposing that boys are socialized to value physical dominance, while girls are socialized to value social dominance and relationships. Likewise, racial socialization may shape attitudes toward aggression and predict engagement in aggressive behavior (e.g., Putallaz et al., 2007). In addition, attitudes toward aggression have also been related to engagement in aggressive behavior (e.g., McConville & Cornell, 2003). The purpose of the current study was to examine whether the associations between both gender and ethnicity and engagement in PA and RA among adolescents were mediated by aggression-specific attitudes.
The sample consisted of 275 adolescents (50.9% male), grades six through eight (Mage = 12.3 years), recruited from parochial schools. The sample was predominately (55.3%) Caucasian. Participants completed the Children’s Social Behavior Scale – Self Report (CSBS-S; Crick & Grotpeter, 1995) and the Attitudes toward Aggression Scale (McConville & Cornell, 2003; Van Schoiack-Edstrom et al., 2002). The participant’s teachers completed the Children’s Social Behavior Scale –Teacher Report (CSBS-T; Crick, 1996).
Significant gender differences were found for self and teacher reported PA, with boys engaging in more PA than girls (Table 1). Significant gender differences only emerged for RA once PA was controlled for, with girls self-reporting significantly more RA than boys. Boys also endorsed more positive attitudes toward PA than girls. Significant gender differences for RA attitudes were only found when controlling for PA attitudes, with girls endorsing more RA attitudes that boys. Significant ethnic differences were found for self-reported PA and RA, with youth identifying as non-Caucasian engaging in more aggression than youth identifying as Caucasian. No significant differences were found between ethnicities for teacher-reported PA; however, Caucasian adolescents were rated by teachers as engaging in more RA than non-Caucasian youth. Non-Caucasian youth also endorsed significantly more positive attitudes toward PA than Caucasian youth. Significant correlations were found between positive attitudes toward PA and RA and rates of PA and RA, respectively (Table 2).
Mediation was tested following the process proposed by Baron and Kenny (1986). Regression results and the Sobel test indicated that attitudes toward PA partially mediated the relation between both gender (Sobel test: z = -2.85, p= .004) and PA and between ethnicity (Sobel test: z = 2.52, p = .012) and PA. Attitudes toward RA were not found to mediate the relation between either gender or ethnicity and RA, even when PA was controlled. Overall, these results suggest that attitudes toward aggression should be examined when implementing interventions to reduce aggressive behavior.
Table 1. Means and Standard Deviations of Variables
Variable Male Female Caucasian Non-Caucasian
M (SD) M (SD) t M (SD) M (SD) t
______________________________________________________________________________
Self
PA 4.79 (2.19) 3.71 (1.84) 3.91** 3.71 (1.42) 4.94 (2.32) -4.48**
RA 16.60 (5.80) 15.88 (6.31) 0.88 15.26 (5.85) 17.45 (6.10) -2.72**
Teacher
PA 6.49 (3.37) 4.64 (1.51) 4.69** 5.90 (3.21) 5.35 (2.28) 1.32
RA 9.39 (4.22) 10.62 (5.03) -1.81 10.88 (5.00) 8.80 (3.86) 3.12**
Attitudes
PA 14.28 (4.98) 12.14 (4.55) 3.27** 12.44 (4.36) 14.29 (5.36) -2.77**
RA 10.89 (3.54) 11.15 (4.06) -0.49 10.89 (3.58) 11.19 (4.10) -0.58
** p < .01 level, two tailed
* p < .05 level, two tailed
Table 2. Correlations
1 2 3 4 5 6 7 8
______________________________________________________________________________
1. Gender 1.00 -.013 -.258** -.059 -.325** .133 -.220** .033
2. Ethnicity 1.00 .292** .180** -.096 -.225** .188** .040
3. Self-PA 1.00 .627** .210** .063 .418** .250**
4. Self-RA 1.00 .079 .144 .347** .259**
5. T-PA 1.00 .471** .192* .073
6. T-RA 1.00 .082 .098
7. ATT-PA 1.00 .460**
8. ATT-RA 1.00
Notes:
Self-PA = Self report of physical aggression, Self-RA = Self report of relational aggression, T-PA = Teacher report of physical aggression, T-RA = Teacher report of relational aggression, ATT-PA = Self report of attitudes toward physical aggression, ATT-RA = Self report of attitudes towards relational aggression
** p < .01 level, two tailed
* p < .05 level, two tailed