رفتار اطرافیان در مواقع زورگویی: حساسیت عمومی اخلاقی، متارکه اخلاقی و خودکارامدی مدافع

Abstract The aim of the present study was to investigate how basic moral sensitivity in bullying, moral disengagement in bullying and defender self-efficacy were related to different bystander behaviors in bullying. Therefore, we examined pathways that linked students' basic moral sensitivity, moral disengagement, and defender self-efficacy to different bystander behaviors in bullying situations. Three hundred and forty-seven teenagers completed a bullying survey. Findings indicated that compared with boys, girls expressed higher basic moral sensitivity in bullying, lower defender self-efficacy and moral disengagement in bullying. Results from the SEM showed that basic moral sensitivity in bullying was negatively related to pro-bully behavior and positively related to outsider and defender behavior, mediated by moral disengagement in bullying, which in turn was positively related to pro-bully behavior and negatively related to outsider and defender behavior. What differed in the relations between outsider and defender behaviors was the degree of defender self-efficacy.

Results Descriptive results The means and standard deviations and the matrix of correlations between all variables are presented in Table 2. Correlations among the different bystander behaviors ranged from −.17 to −.38. Basic moral sensitivity in bullying situations was negatively associated with pro-bully behavior and positively associated with defender behavior and outsider behavior. In contrast, moral disengagement in bullying situations was positively associated with pro-bully behavior and negatively associated with defender behavior and outsider behavior. In addition, basic moral sensitivity and moral disengagement in bullying situations were negatively inter-correlated. Moreover, defender self-efficacy was positively associated with defender behavior and negatively correlated with outsider behavior. Because gender was associated with all other variables, mean differences based on gender were further explored. Table 2. Means, standard deviations and inter-correlations between all variables. M (SD) 1 2 3 4 5 6 7 1. Gender 1 2. Basic moral Sensitivity 5.77 (1.32) .39*** 1 3. Moral disengament 1.72 (.99) −.39*** −.53*** 1 4. Defender self-efficacy 3.96 (1.52) −.14** .06 .03 1 5. Pro-bully behavior 1.41 (.54) −.34*** −.37*** .56*** −.10 1 6. Outsider behavior 3.21 (.91) .23*** .22*** −.26*** −.44*** −.17** 1 7. Defender behavior 2.50 (.95) .16** .31*** −.29*** .47*** −.38*** −.28*** 1 Note. Gender (boys = 0, girls = 1); **p < .01; ***p < .001. Table options Six one-way ANOVAs showed significant differences in all measures. Compared with boys (m = 5.16, sd = 1.53), girls expressed significantly higher basic moral sensitivity in bullying (m = 6.20, sd = .95, F1, 346 = 60.90, p < .001, partial η2 = .15). In contrast, boys displayed significantly higher moral disengagement in bullying (m = 2.19, sd = 1.24, F1, 346 = 63.04, p < .001, partial η2 = .15) as compared with girls (m = 1.40, sd = .59). Defender self-efficacy was found to be significantly higher among boys (m = 4.22, sd = 1.59) than girls (m = 3.78, sd = 1.45, F1, 346 = 7.21, p < .01, partial η2 = .02). As bystanders in bullying, girls were significantly more prone to act as outsiders (m = 3.38, sd = .84) than boys (m = 2.96, sd = .96, F1, 346 = 18.65, p < .001, partial η2 = .05) as well as to act as defenders (m = 2.63, sd = .91) than boys (m = 2.31, sd = .98, F1, 346 = 9.31, p < .01, partial η2 = .03). In contrast, boys were significantly more prone to act as pro-bullies (m = 1.63, sd = .65) when encountering bullying situations as bystanders as compared with girls (m = 1.26, sd = .38, F1, 346 = 44.18, p < .001, partial η2 = .11). The SEM analysis The structural model was tested using the EQS 6.1 (Bentler, 1995). The scores for the three bystander behaviors (pro-bully, outsider, and defender) were included in the model as dependents. The score for basic moral sensitivity was introduced as an independent that was mediated by the score for moral disengagement. The score of defender self-efficacy was also introduced as an independent. Results of the SEM are shown in Fig. 1, which includes all the standardized coefficients that are significant at and below the .05 level. The model-fit statistics for the model (N = 347; CFI = .92; S-BSS = 258.44, df = 144; RMSEA = .048, 90% CI [.04, .06]) indicate an acceptable fit. The Wald test suggested no further modifications of the model, which allowed us to conclude that the model fits the sample. R2 values indicate that this model explains 51% of the variance for pro-bully behavior, 51% of the variance for outsider behavior, and 70% of the variance for defender behavior. Fig. 1 shows all path coefficients. Contribution of basic moral sensitivity, moral disengagement and defender ... Fig. 1. Contribution of basic moral sensitivity, moral disengagement and defender self-efficacy to different bystander behavior in bullying. Note: *p < .05. Figure options The overall model showed a negative relation between basic moral sensitivity in bullying and moral disengagement in bullying. Basic moral sensitivity in bullying was indirectly and negatively related to pro-bully behavior, negatively mediated by moral disengagement in bullying, which in turn was directly and positively related to pro-bully behavior. Furthermore, basic moral sensitivity in bullying was indirectly and positively related to outsider and defender behavior, negatively mediated by moral disengagement in bullying, which in turn was directly and negatively related to both these behaviors. What differed in the relations between outsider and defender behaviors was defender self-efficacy. A strong negative relationship between defender self-efficacy and outsider behavior and a strong positive relationship between defender self-efficacy and defender behavior appear to explain, at least in part, why students who display high basic moral sensitivity and low moral disengagement in bullying decide either to intervene and help a victim or to remain passive. Because we found significant gender differences in all measures, we also tested the structural model for both boys and girls. To carry out this step, the sample was divided into two samples: females and males. When we examined the parameters, the Wald test indicated that the parameter between moral disengagement and defender behavior in the first model for the female sample should be dropped in order to have a model with better fit to the data. Thus, this parameter was removed from further analyses. The model-fit statistics for the second model for the female sample (N = 206; CFI = .89; S-BSS = 211.61, df = 145; RMSEA = .047, 90% CI [.03, .06]) and the male calibration sample (N = 141; CFI = .91; S-BSS = 212.59, df = 144; RMSEA = .058, 90% CI [.04, .07]) indicated models with an acceptable fit. R2 values for the female sample model indicate that this model explains 14% of the variance for pro-bully behavior, 54% of the variance for outsider behavior, and 69% of the variance for defender behavior. R2 values for the male sample model indicate that this model explains 45% of the variance for pro-bully behavior, 46% of the variance for outsider behavior, and 69% of the variance for defender behavior. Fig. 2 shows all path coefficients. Contribution of basic moral sensitivity, moral disengagement and defender ... Fig. 2. Contribution of basic moral sensitivity, moral disengagement and defender self-efficacy to different bystander behavior in bullying for boys and girls. Note: *p < .05; gender = boys/girls. Figure options As in the overall model, in both the male model and female model, there was a significant and negative relationship between basic moral sensitivity and moral disengagement. Basic moral sensitivity was indirectly and negatively related to pro-bully behavior, negatively mediated by moral disengagement in bullying, which in turn was directly and positively related to pro-bully behavior. Although significant, these associations were weaker among girls than boys. Furthermore, moral disengagement was negatively related to outsider behavior and defender behavior among boys and not at all related to these bystander behaviors among girls. As in the overall model, there was also a strong negative relationship between defender self-efficacy and outsider behavior and a strong positive relationship between defender self-efficacy and defender behavior for both boys and girls. In addition, a negative association between defender self-efficacy and pro-bully behavior were found only in the female sample.