قربانی زورگویی و سلامت روان نوجوان: اثرات عمومی و گونه شناختی در میان جنسیته

Abstract Purpose Victims of bullying are susceptible to a variety of detrimental consequences. It remains unclear, however, whether the type of bullying victimization and the gender of the victim matter as they relate to two mental health consequences: (1) depression, and (2) suicide ideation. Methods We examined the effects of the bullying victimization experiences of 1,874 adolescents. Controlling for known predictors of maladaptive mental health, we assessed whether any bullying victimization and any type of bullying victimizations were associated with depression and suicide ideation. Each of these relationships was also compared across gender.

ntroduction Over the past few decades, bullying has been identified as one of the most significant problem behaviors confronting children and adolescents (Olweus, 1991 and Surgeon General Report on Youth Violence, 2001). In fact, the Committee on Injury, Violence, and Poison Prevention of the American Academy of Pediatrics has recently identified bullying as a key area in their revised policy statement on the role of pediatricians in the prevention of youth violence (Committee on Injury, Violence, and Poison Prevention, 2009). Generally identified as the persistent harassment (physical, verbal, emotional, or psychological) of one individual by another and accompanied by a power imbalance, bullying affects approximately thirty percent of sixth to tenth grade students (Nansel et al., 2001). Although many school-aged youths will experience episodic harassment from their peers, what concerns scholars and the public alike is the persistent victimization of individuals' who are most vulnerable to the effects of bullying and the impact these victimizations have on individuals' social, physical, and mental health (Arseneault et al., 2006, Sharp et al., 2000 and Sourander et al., 2000). Bullying incidents can manifest in a variety of different forms (Olweus, 1993). Until recently, bullying was typically perceived as involving either physical aggression (i.e., pushing, shoving, or other forms of physical coercion), verbal abuse (i.e., name calling or teasing in a hurtful manner), or relational behaviors (i.e., spreading rumors or socially excluding individuals) (Wang, Iannotti, & Nansel, 2009). The advancement of technology, however, has ushered a new type of bullying—typically referred to as electronic or cyber bullying—that potentially has more far reaching effects (Hinduja and Patchin, 2008 and Raskauskas and Stoltz, 2007). Individuals perpetrating this form of aggression generally rely on text messages, instant messages, emails, or social networking websites to harass or spread rumors about victims. Cyber bullying incidents may have an especially severe impact since the message or attacks can appear in multiple places on-line and endure over lengthy periods of time (Hinduja and Patchin, 2008, Patchin and Hinduja, 2006 and Ybarra, 2004). An estimated 20 to 30 percent of youth have experienced some form of cyber bullying (Hinduja and Patchin, 2008, Patchin and Hinduja, 2006, Wolak et al., 2006 and Ybarra, 2004), and rates of victimization have increased over the last decade (Jones, Mitchell, & Finkelhor, 2012). Furthermore, there is growing evidence that a small proportion of youth experience bullying victimization on and off-line (Erdur-Baker, 2010, Hinduja and Patchin, 2008, Kowalski and Limber, 2007 and Ybarra and Mitchell, 2004). In light of the extent and nature of bullying, scholars have recently invested significant attention into understanding the effects of being bullied. Studies have shown being a victim of a bullying incident corresponds with deficiencies in academic success and school attendance (Glew, Fan, Katon, Rivara, & Kernic, 2005), emotional well-being (Nansel et al., 2001 and van der Wal et al., 2003), attention-deficit disorder (Kumpulainen, Rasanen, & Puura, 2001), associations with deviant peers (Rodkin & Hodges, 2003), involvement in violence (Nansel, Overpeck, Haynie, Ruan, & Scheidt, 2003), psychiatric symptoms (Kumpulainen & Rasanen, 2000), levels of depression and suicide ideation (Hinduja and Patchin, 2008, Kaltiala-Heino et al., 1999 and Klomek et al., 2008), and attempted suicides (Klomek et al., 2007 and Klomek et al., 2009). Fueled by highly publicized cases in which adolescents committed suicide after repeatedly being bullied, research documenting the effects of bullying has risen to the forefront of public health concerns (Hylton, 2008 and Mayer and Furlong, 2010).

Results Table 1 summarizes the mean, standard deviations, and ranges for each variable in the analyses. Findings are presented for the full sample and by sex. As seen in the table, females have significantly higher grades (t = 5.77) and are more involved in school (t = 2.29) than males. They also consume marijuana less frequently than males (t = 3.41). In terms of the frequency of bullying victimizations, there are no significant differences between males and females in the frequency of cyber bullying victimization, although males had a slightly higher frequency score. However, significant sex differences did exist for verbal (t = − 3.38), physical (t = − 4.48), and the general bullying (t = − 3.36) victimization measure, with males exceeding females on each of these measures (e.g. Nabuzoka, 2003). In terms of mental health outcomes, female respondents reported significantly higher depression (t = 7.82) and suicidal ideation (t = 4.20) scores than male students. Combined, these data suggest that although males are more frequently bullied, females report higher levels of mental health problems. Table 1. Descriptive Statistics of the Sample by Gender Full Sample Females Males Variables Mean SD Range Mean SD Range Mean SD Range t Age 13.77 1.92 11–18 13.71 1.88 11–18 13.84 1.97 11–18 − 1.65 Race (1 = non-white) 0.30 0.46 0–1 0.31 0.46 0–1 0.30 0.46 0–1 0.34 Sex (1 = male) 0.49 0.50 0–1 ---- ---- ---- ---- ---- ---- ---- Grades 3.46 1.18 1–5 3.60 1.16 1–5 3.33 1.19 1–5 5.77** Neighborhood Disorder 9.67 3.59 7–21 9.61 3.46 7–21 9.72 3.71 7–21 − 0.75 Parental Investment 6.12 2.56 3–12 6.13 2.57 3–12 6.12 2.56 3–12 0.11 School Involvement 8.65 1.81 3–12 8.73 1.75 3–12 8.56 1.88 3–12 2.29* Alcohol Frequency 0.22 0.41 0–1 0.20 0.40 0–1 0.23 0.42 0–1 − 1.65 Marijuana Frequency 0.10 0.30 0–1 0.08 0.28 0–1 0.13 0.33 0–1 − 3.41** Other Drug Frequency 0.13 0.34 0–1 0.13 0.34 0–1 0.14 0.35 0–1 − 0.70 Negative Consequences 0.66 1.63 0–11 0.62 1.55 0–11 0.69 1.71 0–11 − 0.98 Any Bullying 2.57 4.12 0–24 2.30 3.55 0–24 2.87 4.64 0–24 − 3.36** Cyber Bullying 0.34 1.14 0–8 0.33 1.02 0–8 0.35 1.26 0–8 − 0.46 Verbal Bullying 1.45 2.06 0–8 1.31 1.84 0–8 1.60 2.26 0–8 − 3.38** Physical Bullying 0.82 1.70 0–8 0.67 1.44 0–8 0.98 1.91 0–8 − 4.48** Depression 2.59 3.48 0–15 3.15 3.57 0–15 2.00 3.28 0–15 7.82** Suicide Ideation 0.66 1.49 0–6 0.96 1.98 0–6 0.67 1.79 0–6 4.20** *p < .05; **p < .01. Note: t-tests were conducted on the mean differences for females and males. Table options Table 2 presents the results of the mean differences on depression and suicide ideation scale scores across bullying victims and non-victims. Note that these analyses utilize the dichotomous indicator of bullying prevalence (as opposed to frequency) in order to identify those who have experienced a specific type of bullying (cyber, verbal, or physical), as well as those who have experienced bullying in general (any type). Across each type of bullying (both general and typological), individuals who were victimized had significantly greater levels of depression and suicide ideation. That is, those experiencing physical bullying (t = − 7.71), verbal bullying (t = − 9.50), cyber bullying (t = − 8.48), or any bullying (t = − 9.39) reported significantly higher depression scores. Correspondingly, those experiencing physical bullying (t = − 6.86), verbal bullying (t = − 8.56), cyber bullying (t = − 6.97), or any bullying (t = − 8.43) also reported significantly higher suicide ideation scores. These findings held true across categories of sex (analyses available upon request). Consistent with past research (see Roland, 2002), these results suggest, at least at the bivariate level of analysis, that individuals who are bullied—regardless of its form—experience significantly greater mental health consequences related to depression and suicide ideation. Table 2. Mean Differences on Outcome Measures Across Victim Status Any Bullying Victim Cyber Bully Victim Verbal Bully Victim Physical Bully Victim Yes No t Yes No t Yes No t Yes No t Depression 3.21 1.81 4.62 2.28 3.27 1.87 3.57 2.20 (3.69) (3.06) − 9.39* (4.41) (3.21) − 8.48* (3.71) (3.06) − 9.50* (3.89) (3.22) − 7.71* Suicide Ideation 0.89 0.37 1.39 0.55 0.91 0.39 1.05 0.51 (1.68) (1.14) − 8.43* (1.98) (1.36) − 6.97* (1.70) (1.15) − 8.56* (1.80) (1.32) − 6.86* *p < .01. Note: Means are reported with standard deviations in parentheses. Table options The next set of analyses focus on investigating the effects of the bullying victimization experience on students' mental health while controlling for a number of covariates. Table 3 presents the results from a tobit regression where depression scale scores were regressed on a set of covariates for the full sample, and then separately for females and males. The purpose of this analysis was to determine if the frequency of any type of bullying corresponded with respondents reporting higher levels of depression. The results indicate that older individuals, females, those living in disorganized neighborhoods, with higher parental investment, consuming marijuana less frequently but other drugs more frequently, and experiencing a greater number of negative consequences from alcohol use reported higher levels of depression. Pertinent to the research question, it is also notable that those experiencing a bullying victimization reported higher depression scores. The gender specific analysis revealed identical findings with the exception of the negative consequences of alcohol consumption variable becoming non-significant for males. A coefficient comparison test revealed no significant differences in the bullying parameter across sex, suggesting bully victimizations impact levels of depression similarly for males and females. Table 3. Tobit Regression Predicting Depression – General Bullying Victimization Measure Full Sample Female Model Male Model Variables Estimate 95% CI p Estimate 95% CI p Estimate 95% CI p Age .342 (.213 to .470) < .0001 .413 (.248 to .578) < .0001 .253 (.048 to .458) .0156 Race (1 = non-white) − .315 (-.852 to .222) NS − .429 (− 1.100 to .241) NS − .172 (− 1.057 to .713) NS Sex (1 = male) − 2.340 (− 2.809 to − 1.872) < .0001 ----- ----- ----- ----- ----- ----- Grades − .048 (− .249 to .152) NS − .056 (− .305 to .194) NS − .027 (− .360 to .306) NS Neighborhood Disorder .164 (.096 to .233) < .0001 .122 (.035 to .210) .0063 .220 (.110 to .329) < .0001 Parental Investment .317 (.223 to .410) < .0001 .306 (.191 to .422) < .0001 .338 (.181 to .495) < .0001 School Involvement − .085 (− .222 to .052) NS − .152 (− .325 to .022) NS .017 (− .206 to .240) NS Alcohol Frequency .442 (− .203 to 1.088) NS .685 (− .132 to 1.502) NS .127 (− .927 to 1.180) NS Marijuana Frequency − 1.072 (− 1.958 to − .186) .0177 − 1.002 (− 2.164 to .160) NS − .935 (− 2.343 to .473) NS Other Drug Frequency 2.518 (1.775 to 3.261) < .0001 2.533 (1.613 to 3.454) < .0001 2.557 (1.320 to 3.795) < .0001 Negative Consequences .220 (.038 to .403) .0180 .275 (.047 to .503) .0183 .116 (− .186 to .419) NS Bullying Victimization .211 (.151 to .271) < .0001 .243 (.158 to .328) < .0001 .193 (.105 to .281) < .0001 Table options Table 4 presents the results from a logistic regression where suicide ideation scale scores were regressed on a set of covariates for the full sample, and then separately for females and males. Similar to the previous analysis, the purpose of this regression was to determine if respondents who experienced the effects of bullying victimization frequency reported higher levels of suicidal ideation. The results indicate that females, those living in disorganized neighborhoods, with higher parental investment, who were involved less in school, consumed marijuana less often but consumed other drugs more often, and experienced a greater number of negative consequences from alcohol use reported higher levels of suicide ideation. The bullying victimization measure indicated that those who have been bullied were more likely to have thoughts of suicide. The gender specific analysis revealed identical findings with the exception of the school involvement measure becoming non-significant for females and the marijuana use measure and the negative consequences from alcohol use measure becoming non-significant for males. A coefficient comparison test again uncovered no significant differences in the bullying victimization parameter across sex. Table 4. Logistic Regression Predicting Suicide Ideation – General Bullying Victimization Measure Full Sample Female Model Male Model Variables Estimate 95% CI p Estimate 95% CI p Estimate 95% CI p Age .052 (.984 to 1.128) NS .067 (.977 to 1.172) NS .034 (.931 to 1.150) NS Race (1 = non-white) − .051 (.717 to 1.260) NS − .132 (.606 to 1.268) NS .068 (.687 to 1.670) NS Sex (1 = male) − .822 (.340 to .568) .000 ----- ----- ----- ----- ----- ----- Grades − .048 (.854 to 1.064) NS − .093 (.790 to 1.052) NS .011 (.849 to 1.204) NS Neighborhood Disorder .099 (1.068 to 1.142) .000 .093 (1.049 to 1.149) .000 .107 (1.059 to 1.171) .000 Parental Investment .142 (1.098 to 1.209) .000 .147 (1.089 to 1.233) .000 .133 (1.056 to 1.235) .001 School Involvement − .099 (.844 to .972) .006 − .067 (.850 to 1.030) NS − .135 (.786 to .972) .013 Alcohol Frequency .169 (.853 to 1.645) NS .067 (.690 to 1.658) NS .246 (.767 to 2.133) NS Marijuana Frequency − .611 (.342 to .863) .010 − .810 (.231 to .858) .016 − .404 (.340 to 1.309) NS Other Drug Frequency 1.208 (2.362 to 4.741) .000 1.034 (1.756 to 4.503) .000 1.392 (2.395 to 6.764) .000 Negative Consequences .181 (1.098 to 1.308) .000 .266 (1.144 to 1.488) .000 .098 (.970 to 1.254) NS Bullying Victimization .070 (1.041 to 1.104) .000 .064 (1.020 to 1.113) .004 .075 (1.036 to 1.121) .000 Table options Table 5 presents the results from a tobit regression where depression scale scores were regressed on a set of covariates for the full sample, and then separately for females and males. Unlike the previous analyses that examined bullying in general, the present analysis unpacked the types of bullying victimizations and examined the impact of specific types of bully victimization reported on levels of depression. The results for the model examining the full sample suggest that older individuals, females, those living in a more disorganized neighborhood, who have more parental involvement, who use marijuana less frequently but use other drugs more frequently, and who experienced a greater number of negative consequences from alcohol consumption reported higher levels of depression. In terms of the three types of bullying victimizations, only those individuals reporting more frequent verbal bullying experienced higher levels of depression. Table 5. Tobit Regression Predicting Depression – Typological Bullying Victimization Measure Full Sample Female Model Male Model Variables Estimate 95% CI p Estimate 95% CI p Estimate 95% CI p Age .343 (.214 to .471) < .0001 .401 (.236 to .566) < .0001 .257 (.052 to .461) .0138 Race (1 = non-white) − .282 (− .821 to .257) NS − .438 (− 1.110 to .235) NS − .124 (− 1.007 to .759) NS Sex (1 = male) − 2.357 (− 2.828 to − 1.886) < .0001 ----- ----- ----- ----- ----- ----- Grades − .047 (− .247 to .154) NS − .052 (− .301 to .197) NS − .027 (− .359 to .306) NS Neighborhood Disorder .164 (.095 to .232) < .0001 .123 (.036 to .211) .0059 .224 (.115 to .334) < .0001 Parental Investment .315 (.222 to .409) < .0001 .306 (.191 to .421) < .0001 .328 (.171 to .484) < .0001 School Involvement − .085 (− .222 to .053) NS − .157 (− .330 to .016) NS .012 (− .211 to .235) NS Alcohol Frequency .428 (− .219 to 1.074) NS .681 (− .134 to 1.496) NS .106 (− .945 to 1.157) NS Marijuana Frequency − 1.028 (− 1.917 to − .139) .0234 − 1.096 (− 2.261 to .069) NS − .811 (− 2.216 to .593) NS Other Drug Frequency 2.516 (1.771 to 3.261) < .0001 2.483 (1.562 to 3.403) < .0001 2.582 (1.346 to 3.818) < .0001 Negative Consequences .226 (.044 to .409) .0152 .273 (.046 to .501) .0187 .144 (− .158 to .445) NS Cyber Bullying .077 (− .161 to .315) NS .489 (.178 to .801) .0021 − .373 (− .757 to .012) NS Verbal Bullying .336 (.173 to .499) < .0001 .220 (.007 to .434) .0434 .401 (.144 to .658) .0022 Physical Bullying .116 (− .087 to .320) NS .140 (− .128 to .407) NS .175 (− .146 to .497) NS Table options Turning to the gender specific analyses, the results are generally consistent with those reported for the full sample with a few exceptions. First, the negative consequences from alcohol consumption significantly affects depression levels for only females such that females reporting more consequences also reported higher levels of depression. Second, only females' levels of depression were significantly and positively affected by cyber bullying (Hinduja and Patchin, 2008 and Ybarra and Mitchell, 2004). A coefficient comparison test indicated significant sex differences for the cyber bullying coefficient (z = 3.415), indicating the effect of the frequency of cyber bullying on depression is sex-specific. However, the coefficient comparison tests for verbal and physical bullying revealed no significant differences across sex, suggesting the frequency of verbal and physical victimizations impact depression levels comparably across sex. Finally, Table 6 presents the results from a logistic regression where suicide ideation was regressed on a set of covariates for the full sample, and then separately for females and males. The results indicate that females, those living in disorganized neighborhoods, with higher parental investment, who were involved less in school, consumed marijuana less often but other drugs more often, and experienced a greater number of negative consequences from alcohol use reported higher levels of suicide ideation. It is notable, however, that none of the bullying victimization measures were significantly related to raising the odds of suicide ideation. The findings related to bullying victimizations in the gender specific models replicated those in the full sample model suggesting that no singular type of frequency of bullying victimization is predictive of suicide ideation either in the full sample or either of the sex specific samples. Table 6. Logistic Regression Predicting Suicide Ideation – Typological Bullying Victimization Measure Full Sample Female Model Male Model Variables Estimate 95% CI p Estimate 95% CI p Estimate 95% CI p Age .053 (.984 to 1.129) NS .065 (.974 to 1.169) NS .036 (.933 to 1.153) NS Race (1 = non-white) − .048 (.718 to 1.265) NS − .125 (.609 to 1.278) NS .068 (.686 to 1.671) NS Sex (1 = male) − .830 (.337 to .564) .000 ----- ----- ----- ----- ----- ----- Grades − .049 (.853 to 1.063) NS − .091 (.791 to 1.054) NS .008 (.846 to 1.201) NS Neighborhood Disorder .099 (1.067 to 1.142) .000 .093 (1.048 to 1.149) .000 .107 (1.058 to 1.170) .000 Parental Investment .142 (1.098 to 1.209) .000 .147 (1.089 to 1.233) .000 .131 (1.054 to 1.234) .001 School Involvement − .098 (.844 to .973) .007 − .068 (.848 to 1.028) NS − .133 (.787 to .974) .014 Alcohol Frequency .168 (.852 to 1.644) NS .065 (.688 to 1.654) NS .248 (.768 to 2.138) NS Marijuana Frequency − .598 (.346 to .875) .012 − .823 (.227 to .850) .015 − .367 (.354 to 1.355) NS Other Drug Frequency 1.210 (2.366 to 4.749) .000 1.026 (1.741 to 4.471) .000 1.400 (2.413 to 6.821) .000 Negative Consequences .182 (1.099 to 1.309) .000 .265 (1.143 to 1.486) .000 .101 (.974 to 1.257) NS Cyber Bullying .035 (.924 to 1.161) NS .095 (.937 to 1.289) NS − .037 (.817 to 1.138) NS Verbal Bullying .079 (.997 to 1.175) NS .083 (.971 to 1.217) NS .059 (.937 to 1.200) NS Physical Bullying .075 (.973 to 1.194) NS .020 (.884 to 1.177) NS .146 (.994 to 1.347) NS