| | Examining Ethnic, Gender, and Developmental Differences in the Way Children Report Being a Victim of “Bullying” on Self-Report MeasuresReceived 3 August 2007; accepted 21 December 2007. published online 02 May 2008.
|
Refers to article:
|
|
What Is Bullying?
Matthew C. Aalsma, James R. Brown
Journal of Adolescent Health
August 2008 (Vol. 43, Issue 2, Pages 101-102)
Full Text |
Full-Text PDF (66 KB)
|
Abstract PurposeRacial/ethnic differences in children's self-reports of being a frequent victim of bullying were assessed via two commonly used strategies: a definition-based single-item measure and behavior-based multiresponse measure. MethodsLogistic regression analyses were conducted on survey data from 24,345 youth to examine ethnic differences in youths' responses to definition-based and behavior-based measures of victimization. Separate analyses were conducted for boys and girls at different school levels. ResultsPrevalence estimates were higher using the behavior-based measure than definition-based measure. Several ethnic differences emerged, such that African American youth tended to be less likely than their white counterparts to indicate that they were bullied using the definition-based measure. African American girls and African American and Asian middle-school boys who reported being a victim via the behavior-based measure were less likely to report being a frequent victim of “bullying” via the definition-based measure. ConclusionPrevalence estimates vary considerably by the way in which victimization is assessed. African American youth who were victimized tended to under-report being a victim of “bullying.” Self-report studies of bullying should carefully consider the measures used to assess victimization. Increased awareness of the detrimental effects of school bullying among public health researchers, practitioners, and policymakers has led to a growing interest in cost-effective strategies for assessing the prevalence of bullying. Several studies have indicated that involvement in bullying is associated with short- and long-term social–emotional problems [1], [2], as well as an increased risk for involvement in aggressive acts and criminal behavior [3], [4], [5]. Despite the growing consensus regarding the social–emotional problems associated with bullying, the assessment methods used to examine bullying have come under question, particularly with regard to the use of self-report measures of victimization [6]. Self-reports are the most commonly used method for assessing the prevalence of bullying and peer victimization, in part because of their efficiency and minimal cost [7]. However, there is limited information regarding the reliability and validity of self-report measures of victimization [6], [7], [8]. In addition, the issue of social desirability is a common concern with self-reports, as it increases children's likelihood of underreporting their involvement in bullying [9], [10]. Self-report measures of bullying have also been criticized for the definition of bullying provided on the survey [6], [11]. The most often cited definition comes from Olweus [12], which describes bullying as a repeated and intentional aggressive act, which typically involves an imbalance of power. Despite the definition's frequent use, it remains unclear whether children are able to effectively discriminate between “bullying” and other aggressive acts (e.g., fighting) [13]. Some researchers have suggested that this discrepancy may be in part because of cultural or developmental differences in children's understanding of the term bullying or the vocabulary used to define it [11]. There is also an inconsistency in the format of self-report measures. Surveys using the definition-based self-report format typically provide a definition of bullying that participants are expected to use when completing the survey [7]. This type of self-report measure usually inquires about the frequency with which the respondent has been victimized through single-item measures (e.g., “How often have you been bullied in the past month?”), with response options ranging from “Not at all” to “Several times a week” [14], [15], [16]. Conversely, behavior-based self-report measures include multiple questions or offer several responses to one question regarding the disparate forms of bullying (e.g., “How often in the past month have you been … hit, kicked or threatened; the subject of rumors or lies; left out of social events?”) [17], [18], [19], allowing participants to select more than one form of victimization. Behavior-based self-report surveys of bullying have tended to receive more support from researchers because of their ability to differentiate among different forms of bullying and victimization [7], as well as avoid youth's preconceived notions of what constitutes “bullying” as opposed to other forms of victimization. However, it is unclear whether youth are reporting only bullying events via behavior-based questions or if they are including other forms of aggression (e.g., fights). Racial/ethnic, gender, and developmental differences in the risk for victimization  Relatively few large-scale studies of bullying have been conducted using ethnically diverse samples or samples of boys and girls across all school levels (i.e., elementary, middle, and high school). Yet, prevalence estimates for frequent victimization across race/ethnicity, gender, and developmental level vary considerably by the way in which bullying is assessed. With regard to ethnicity, although findings from the Bureau of Justice Statistics indicate that the rates of violent victimization at school do not vary by ethnicity, African American youth are more likely to be victims of violence outside of school. White students are, however, more likely than African American or Hispanic students to be victims of theft at school [20]. The World Health Organization's 1998 nationally representative study of bullying [14] used a definition-based self-report format and found no significant difference in the likelihood of being frequently bullied among Hispanic, African American, and white youth. Yet, in a more recent nationally representative data collection using this same definition-based measure, African American youth were significantly less likely than white and Hispanic youth to report frequent victimization on the definition-based measure [21]. Similarly, Juvonen et al [22] assessed bullying through self-report, peer nominations, and teacher questionnaires and found that minority youth were less likely than white children to report being victimized. Seals and Young [23] used a behavior-based self-report and found no significant difference between white and African American youths' likelihood of frequent victimization. Finally, using peer nominations, Hanish and Guerra [24] found that Hispanic students were less likely to be victimized than African American and white youth. However, they also found that these ethnic differences varied over time and by grade level. These inconsistencies in the racial/ethnic trends for victimization appear to be because of, at least in part, the way in which bullying is measured. When focusing specifically on discrepant trends in self-report measures of bullying, it is possible that ethnic/cultural factors may influence the way children perceive the term “bullying” [25], [26], which contributes to variation in their response on behavior-based compared to definition-based measures. For example, specifying the word “bullying” in the definition-based measure may conjure up preconceived notions about bullying and stereotypical views of victims, which may or may not be congruent with the child's self-image [7]. With regard to gender differences in frequent victimization, most studies using the definition-based measure of bullying show that boys are more likely to report frequent victimization [14], [21], [27]. However, there do appear to be some gender differences in the specific types of bullying experienced, such that girls are more likely than boys to report indirect forms (e.g., leaving out, rumor spreading) and boys are more likely to report direct forms (e.g., pushing, hitting) of victimization [28], [29]. This, in turn, may contribute to discrepancies in self-reports of bullying across behavior-based measures (that specify relational forms) and definition-based measures (that include the word “bullying,” which may be traditionally viewed as pertaining to physical types of victimization rather than a broader range of victimization forms) [12]. Although there are relatively few studies contrasting elementary, middle, and high school youth, there is a general consensus in the literature that the prevalence of frequent bullying tends to increase in late elementary school, peak during middle school, and decline in high school [12], [14], [30]. However, there may be developmental differences in the way younger children conceptualize the term “bullying” that influence the way they respond to definition-based questions compared to behavior-based questions. For example, elementary school children may have difficulty understanding and consistently applying the specified definition of bullying when completing a definition-based measure. Overview of the current study The aforementioned discrepancies in students' self-reports of involvement in bullying suggest a need for further research to better understand how these inconsistencies in self-reports of victimization may disproportionately affect minority youth, girls, and younger children. The present study examines variation in children's self-reports of bullying when assessed using two different strategies: (1) a definition-based single-item measure [14], [31], and (2) a behavior-based self-report format that inquires about the occurrence of different forms of victimization within a specified timeframe [18], [19]. The data come from an anonymous school-based survey of bullying administered by a public school district. The large, diverse sample of 24,345 children (grades 4–12) allows us to examine possible age, racial/ethnic, and gender differences in the youths' likelihood of reporting victimization via the two self-report measures. Method Participants Data were collected in December 2006 from 24,345 students in grades 4–12 from 107 elementary, middle, and high schools in a large Maryland public school district. To ensure anonymity among participants, students were only asked select demographic questions, including their race, sex, and grade level. Student and school characteristics are reported in Table 1. | | |  | | Elementary schools | Middle schools | High schools | Total | |
|---|
| Students (n) | 7,147 | 11,408 | 5,790 | 24,345 | | | Sex | %(n) | %(n) | %(n) | %(n) | | | Female | 48.71(3,481) | 50.14(5,720) | 49.93(2,891) | 49.67(12,092) | | | Male | 51.29(3,666) | 49.86(5,688) | 50.07(2,899) | 50.33(12,253) | | | Race/ethnicity | | | | | | | White | 63.93(4,569) | 60.77(6,933) | 69.88(4,046) | 63.87(15,548) | | | Black/African American | 15.78(1,128) | 19.42(2,215) | 14.73(853) | 17.24(4,196) | | | Hispanic | 4.32(309) | 4.53(517) | 3.99(231) | 4.34(1,057) | | | Asian/Pacific Islander | 2.43(174) | 3.52(401) | 3.58(207) | 3.21(782) | | | Other | 13.53(967) | 11.76(1,342) | 7.82(453) | 11.35(2,762) | | | Schools(n) | 76M(SD) | 19M(SD) | 12M(SD) | 107M(SD) | | | Enrollment | 429(15.69) | 885(58.98) | 1929(105.87) | 678(50.04) | | | % Minority | 35.48(2.90) | 35.65(5.15) | 29.87(18.15) | 34.88(2.34) | | | Student–teacher ratio | 23.86(0.41) | 19.27(0.38) | 22.18(0.36) | 22.86(0.34) | | | FARMS | 26.31(2.34) | 24.01(3.50) | 14.33(2.95) | 24.55(1.83) | | | Urbanicity | %(n) | %(n) | %(n) | %(n) | | | Urban | 59.21(45) | 57.89(11) | 50.00(6) | 57.94(62) | | | Suburban | 26.32(20) | 31.58(6) | 33.33(4) | 28.04(30) | | | Rural | 14.47(11) | 10.53(2) | 16.67(2) | 14.02(15) | | | | |
Procedure All students completed an anonymous Web-based survey, which was available through a password-protected Web-site at school during school hours. To assure comprehension of the survey items by the younger participants, the elementary school teachers were instructed to read the questions and response options out loud as the students completed the survey. The survey required a mean of 10.26 minutes to complete (Median = 9.0), and approximately 74% of the students in the targeted grades participated. The survey was conducted by the school district using passive parental consent and youth assent procedures. Parents were informed by the school that the district was conducting the district-wide anonymous survey of bullying behavior. The existing nonidentifiable data were obtained from the school district, and were approved for analysis by the Committee on Human Subjects Research at the researchers' institution. Measures Demographic characteristics Participants selected one of the following options to indicate their race/ethnicity: white, black/African American, Hispanic/Latino, Asian/Pacific Islander, and Other (see Table 1). Students reported their grade level and the school level (elementary, middle, or high school) was based on students' enrollment at a particular school. Definition-based measure of frequent victimization Consistent with previous studies [14], [15], [31], the survey defined bullying as occurring “when a person or group of people repeatedly say or do mean or hurtful things to someone on purpose. Bullying includes things like teasing, hitting, threatening, name-calling, ignoring, and leaving someone out on purpose.” The survey included one item to assess the frequency with which the participant had been bullied by others (how often have you been bullied during the last month?). Response options were not at all, once a month, two to three times during the month, once a week, and several times a week [15], [21], [31]. Prior research indicates that being a victim two or more times a month is an important threshold for frequent victimization [31]; therefore, the participants' original responses were dichotomized into frequently victimized (two or more times in the last month) or not frequently victimized. Behavior-based victimization measure Participants also responded to a multiresponse format question regarding the frequency with which they had experienced different forms of bullying within the last month (within the last month, has someone repeatedly tried to hurt you or make you feel bad by … ). Students were able to check multiple response options concerning ten possible forms of bullying (name-calling, threatening, teasing, making sexual comments or gestures, pushing/shoving, hitting/slapping/kicking, stealing, emailing/blogging, leaving out/ignoring, or spreading rumors/lies) [18]. A dichotomous variable was created for each of the ten response options, and a single variable was created to signify whether or not the participant had experienced any of the ten forms of bullying. Statistical analyses Logistic regressions were performed using Stata 9 to explore differences between racial/ethnic groups in their odds of reporting frequent victimization through either the definition-based single-item victimization measure or the behavior-based measure, with white as the comparison group for all analyses. Odds ratios for reporting being victimized in at least one of any of the 10 forms included in the behavior-based measure were obtained. In addition, the 10 forms of victimization were divided into three types: direct physical (pushing/shoving, hitting/slapping/kicking, stealing), direct verbal (name calling, teasing, threatening, making sexual comments or gestures), and indirect (leaving out on purpose, spreading rumors or lies, emailing/blogging). Odds ratios (OR) were obtained for each of the three victimization types. We then examined the odds of reporting being victimized via the definition-based single-item victimization measure among the youth who reported being victimized in at least one of the forms included in the behavior-based measure. Because the students were clustered in 107 schools, Huber-White robust standard errors were estimated. Both our preliminary analyses and prior research on gender differences in the rates of victimization indicated stratifying by gender was necessary [14], [32]. We also stratified by school type (elementary [ES], middle [MS], and high school [HS]) to explore possible developmental differences [30]. Results Frequent bullying reported via the definition-based single-item question Results of the logistic regressions stratified by sex and school type comparing the odds of reporting being frequently bullied via the definition-based single-item measure (with two or more times during the past month as the threshold) are presented in the far left column of Table 2 for boys and Table 3 for girls. The regression analyses indicated that among both ES boys and girls, minority students were no more likely than white students to report being frequently bullied through the definition-based single-item measure. However, among MS youth, both African American boys and girls had decreased odds of reporting frequent bullying. Among MS girls, no other racial/ethnic groups differed significantly from whites, whereas among MS boys, Asians had decreased odds. Among HS girls, African Americans had decreased odds, whereas girls characterized as “Other” had increased odds of reporting being frequently bullied compared to white girls. Among HS boys, both Hispanics and Asians were more likely than their white counterparts to indicate that they were frequently bullied. | | | | | Definition-based single-item measurea | Behavior-based multiresponse measureb | Contrasting response to definition-based and behavior-based measures | |
|---|
| Frequently bullied | % Victimized by type | Experienced any type of victimization | Odds of reporting being bullied among youth who reported any type of victimization | |
|---|
| OR(95% CI) | %(n) | Dir. physical %(n) | Dir. verbal %(n) | Indirect %(n) | OR(95% CI) | %(n) | OR(95% CI) | %(n) | |
|---|
| Elementary school | | | Black/AA | 0.90(0.70–1.15) | 28.92(155) | 51.68(277) | 55.97(300) | 48.13(258) | 1.26(0.96–1.66) | 72.20(387) | 0.78(0.58–1.03) | 38.50(149) | | | Hispanic | 0.87(0.60–1.27) | 28.30(45) | 44.65(71) | 50.94(81) | 42.77(68) | 0.92(0.65–1.30) | 65.41(104) | 0.77(0.51–1.17) | 38.46(40) | | | Asian/PI | 0.95(0.52–1.71) | 30.00(24) | 37.50(30) | 56.25(45) | 33.75(27) | 1.01(0.63–1.60) | 67.50(54) | 0.85(0.46–1.59) | 40.74(22) | | | Other | 1.04(0.84–1.27) | 31.95(170) | 50.56(269) | 57.14(304) | 47.93(255) | 1.50(1.19–1.88) | 75.56(402) | 0.87(0.70–1.08) | 41.29(166) | | | White(ref) | 1.00 | 31.20(736) | 44.81(1057) | 50.83(1199) | 40.86(964) | 1.00 | 67.32(1588) | 1.00 | 44.65(709) | | | Middle school | | | Black/AA | 0.59(0.50–0.69) | 19.98(227) | 42.78(486) | 52.73(599) | 35.83(407) | 0.97(0.74–1.27) | 65.32(742) | 0.53(0.44–0.64) | 29.61(219) | | | Hispanic | 0.73(0.52–1.04) | 23.79(64) | 44.61(120) | 47.96(129) | 31.97(86) | 0.88(0.68–1.15) | 63.20(170) | 0.73(0.49–1.09) | 36.47(62) | | | Asian/PI | 0.70(0.50–0.96) | 22.84(45) | 43.65(86) | 51.78(102) | 36.55(72) | 1.02(0.70–1.50) | 66.50(131) | 0.64(0.47–0.88) | 33.59(44) | | | Other | 0.96(0.78–1.20) | 29.12(196) | 51.71(348) | 58.40(393) | 39.38(265) | 1.25(0.93–1.69) | 70.88(477) | 0.81(0.66–1.00) | 38.99(186) | | | White(ref) | 1.00 | 29.86(1019) | 45.15(1541) | 54.38(1856) | 34.16(1166) | 1.00 | 66.04(2254) | 1.00 | 44.06(993) | | | High school | | | Black/AA | 0.81(0.54–1.21) | 20.26(94) | 34.48(160) | 42.89(199) | 31.68(147) | 0.97(0.77–1.22) | 54.09(251) | 0.80(0.55–1.16) | 36.25(91) | | | Hispanic | 1.67(1.17–2.38) | 34.45(41) | 42.02(50) | 56.30(67) | 41.18(49) | 1.46(1.07–1.98) | 63.87(76) | 1.40(0.88–2.22) | 50.00(38) | | | Asian/PI | 1.66(1.01–2.76) | 34.29(36) | 38.10(40) | 49.52(52) | 31.43(33) | 1.02(0.70–1.48) | 55.24(58) | 1.84(0.92–3.72) | 56.90(33) | | | Other | 1.40(0.86–2.30) | 30.61(75) | 39.59(97) | 55.10(135) | 41.63(102) | 1.55(1.11–2.16) | 65.31(160) | 1.12(0.71–1.75) | 44.38(71) | | | White(ref) | 1.00 | 23.91(470) | 34.64(681) | 44.76(880) | 30.87(607) | 1.00 | 54.83(1078) | 1.00 | 41.65(449) | | | | |
| a Definition-based single-item measure of frequent bullying(How often have you been bullied during the last month?, with two or more times in the last month as the threshold for frequent victimization). bBehavior-based measure of bullying(Within the last month, has someone repeatedly tried to hurt you or make you feel bad by … ). For ease of reporting, the 10 forms of victimization were grouped into three dichotomous variables representing direct verbal(name-calling, threatening, teasing, or making sexual comments or gestures), direct physical(pushing/shoving, hitting/slapping/kicking, or stealing), and indirect bullying (emailing/blogging, leaving out/ignoring, or spreading rumors/lies). |
| | | | | Definition-based single-item measurea | Behavior-based multiresponse measureb | Contrasting response to definition-based and behavior-based measures | |
|---|
| Frequently bullied | % Victimized by type | Experienced any type of victimization | Odds of reporting being bullied among youth who reported any type of victimization | |
|---|
| OR(95% CI) | %(n) | Dir. physical %(n) | Dir. Verbal %(n) | Indirect %(n) | OR(95% CI) | %(n) | OR(95% CI) | %(n) | |
|---|
| Elementary school | | | Black/AA | 0.95(0.78–1.16) | 30.74(182) | 52.87(313) | 62.16(368) | 59.29(351) | 1.63(1.24–2.15) | 79.56(471) | 0.78(0.63–0.96) | 37.79(178) | | | Hispanic | 0.84(0.59–1.19) | 28.00(42) | 43.33(65) | 62.67(94) | 56.00(84) | 1.38(0.90–2.12) | 76.67(115) | 0.71(0.51–0.99) | 35.65(41) | | | Asian/PI | 1.16(0.69–1.97) | 35.11(33) | 45.74(43) | 61.70(58) | 54.26(51) | 1.66(0.99–2.78) | 79.79(75) | 0.90(0.51–1.60) | 41.33(31) | | | Other | 0.88(0.69–1.12) | 28.97(126) | 49.20(214) | 57.49(250) | 54.02(235) | 1.24(0.95–1.16) | 74.71(325) | 0.75(0.59–0.96) | 36.92(120) | | | White(ref) | 1.00 | 31.76(702) | 40.27(890) | 51.49(1138) | 51.40(1136) | 1.00 | 70.45(1557) | 1.00 | 43.80(682) | | | Middle school | | | Black/AA | 0.70(0.51–0.94) | 23.17(250) | 43.47(469) | 59.59(643) | 45.32(489) | 1.18(0.82–1.68) | 73.12(789) | 0.60(0.47–0.78) | 30.93(244) | | | Hispanic | 1.14(0.92–1.41) | 33.06(82) | 45.16(112) | 61.69(153) | 51.61(128) | 1.24(0.91–1.70) | 74.19(184) | 1.04(0.81–1.33) | 43.48(80) | | | Asian/PI | 0.94(0.71–1.24) | 28.92(59) | 45.59(93) | 57.84(118) | 49.02(100) | 1.30(0.90–1.86) | 75.00(153) | 0.84(0.60–1.13) | 37.91(58) | | | Other | 0.97(0.76–1.24) | 29.60(198) | 45.74(306) | 62.78(420) | 51.27(343) | 1.41(1.11–1.79) | 76.53(512) | 0.82(0.65–1.05) | 37.89(194) | | | White(ref) | 1.00 | 30.23(1064) | 36.65(1290) | 56.31(1982) | 46.42(1634) | 1.00 | 69.83(2458) | 1.00 | 42.55(1046) | | | High school | | | Black/AA | 0.60(0.40–0.89) | 12.34(48) | 26.22(102) | 48.59(189) | 41.90(163) | 1.07(0.80–1.42) | 62.21(242) | 0.57(0.40–0.83) | 19.83(48) | | | Hispanic | 1.29(0.96–1.72) | 23.21(26) | 31.25(35) | 51.79(58) | 52.68(59) | 1.26(0.96–1.66) | 66.07(74) | 1.25(0.87–1.79) | 35.14(26) | | | Asian/PI | 0.91(0.59–1.42) | 17.65(18) | 33.33(34) | 53.92(55) | 40.20(41) | 1.01(0.77–1.32) | 60.78(62) | 0.87(0.59–1.30) | 27.42(17) | | | Other | 1.64(1.08–2.51) | 27.88(58) | 35.58(74) | 55.29(115) | 43.75(91) | 1.23(0.92–1.63) | 65.38(136) | 1.62(1.06–2.47) | 41.18(56) | | | White(ref) | 1.00 | 19.04(396) | 24.04(500) | 48.55(1010) | 40.24(837) | 1.00 | 60.62(1261) | 1.00 | 30.21(381) | | | | |
| a Definition-based single-item measure of frequent bullying(How often have you been bullied during the last month?, with two or more times in the last month as the threshold for frequent victimization). bBehavior-based measure of bullying(Within the last month, has someone repeatedly tried to hurt you or make you feel bad by … ). For ease of reporting, the 10 forms of victimization were grouped into three dichotomous variables representing direct verbal(name-calling, threatening, teasing, or making sexual comments or gestures), direct physical(pushing/shoving, hitting/slapping/kicking, or stealing), and indirect bullying(emailing/blogging, leaving out/ignoring, or spreading rumors/lies). |
Victimization reported via the behavior-based question Report of any form of victimization Students' odds of reporting being victimized through at least one of the 10 forms included in the behavior-based measure were computed. Among ES students, African American girls and other boys had increased odds of reporting that they experienced all forms of victimization surveyed compared to white students. No statistically significant differences were found between racial/ethnic groups among MS students. Similarly, there were no differences among HS girls, although Hispanic and Other boys were more likely than white boys in HS to report experiencing any form of victimization. Type of victimization Frequencies are reported in the middle columns on Table 2, Table 3. Compared to white girls in ES, African American girls had increased odds of experiencing direct physical (OR = 1.66, p < .001), direct verbal (OR = 1.55, p < .001), and indirect (OR = 1.38, p < .001) forms of victimization. African American boys in ES had increased odds of experiencing direct physical (OR = 1.32, p = .039) and indirect (OR = 1.34, p = .007) forms of victimization. Compared to white ES girls, Hispanic (OR = 1.58, p = .022), Asian (OR = 1.52, p = .035), and Other girls (OR = 1.27, p = .034) were more likely to report direct verbal victimization. Among MS girls, all minority groups were more likely than their white counterparts to experience direct physical victimization (African American: OR = 1.55, p < .001, Hispanic: OR = 1.58, p = .022; Asian: OR = 1.52, p = .034; Other: OR = 1.28, p = .034). Among MS boys, only youth characterized as Other differed from their white peers (OR = 1.29, p = .008) with regard to direct physical victimization. Among HS students, Hispanic girls were more likely to experience direct physical (OR = 1.44, p = .033) and indirect (OR = 1.65, p < .001) victimization, whereas Hispanic boys were more likely to experience all three types of victimization (direct physical: OR = 1.37, p = .003; direct verbal: OR = 1.59, p = .002; indirect: OR = 1.57, p = .017). Report of frequent bullying among students who were victimized in any form The odds of reporting being victimized via the definition-based single-item measure among youth who reported being victimized in at least one of the forms included in the behavior-based measure were calculated (see far right column on Table 2, Table 3). Among ES boys who experienced any form of victimization, there were no significant differences between any minority group and their white counterparts. Among ES girls who reported any form of victimization, however, African Americans, Hispanics, and Others were less likely to report that they had been bullied frequently via the definition-based single-item measure. Among MS students victimized in any form surveyed, both African American boys and girls, as well as Asian boys had decreased odds of reporting being bullied at least twice during the previous month compared to whites. Among girls in HS who experienced at least one form of victimization, African Americans had decreased odds and Others had increased odds of reporting that they were bullied frequently. There were no statistically significant racial/ethnic differences among HS boys who had been victimized. Discussion The current paper examined racial/ethnic, gender, and developmental differences in the way children self-report being a frequent victim of bullying. Inspection of the frequencies in Table 2 for boys and Table 3 for girls shows that approximately 20% to 30% of youth surveyed reported being frequently bullied when responding to the definition-based single-item question, whereas approximately 55% to 80% reported being the victim of at least one specific form of bullying in the last month via the behavior-based measure. These findings suggest a clear discrepancy in the way the children interpreted the single-item bullying question and the behavior-based question, such that the behavior-based measure was a more sensitive measure of bullying. With regard to ethnicity, the logistic regression analyses revealed a trend whereby minority youth (i.e., African American MS boys and girls, Asian MS boys, and African American HS girls) were less likely than their white counterparts to report being the victim of frequent bullying via the definition-based single-item measure. In contrast, minority youth tended to be more likely to report experiencing at least one form of bullying through the behavior-based measure (i.e., Other ES and HS boys, Hispanic HS boys, African American ES girls, and Other MS girls). Of particular interest was the finding that African American youth who reported victimization by endorsing at least 1 of the 10 forms surveyed were less likely than white youth to indicate they had been bullied via the single-item measure (i.e., African American girls across all school levels, and African American MS boys). This suggests that among youth who are victimized, African American youth tend to be less likely than white youth to report that they had experienced “bullying.” Furthermore, at all school levels African American girls, regardless of whether they reported at least one form of victimization, were less likely than white youth to report being frequently bullied via the definition-based single-item measure. This trend was particularly robust for African American girls, and a similar pattern emerged for other minority groups at certain school levels, such that their likelihood of reporting victimization differed from White students'. It appears there may be cultural differences either in the way certain victimization experiences are perceived or in normative beliefs about being a victim of bullying that affect youths' willingness to report the experience. Although it is difficult to determine why these discrepancies occurred for some ethnic groups but not others, theoretical work by Triandis [33] on subjective culture, which purports that different cultural groups perceive their social environments and experiences in different ways, suggests that race/ethnicity may play a role in how children perceive the term bullying or conceptualize their victimization experiences [26]. It is possible, therefore, that the African American youth perceived greater stigma associated with being victimized through bullying. This, in turn, may have led the African American youth to be less likely than their white counterparts to endorse the definition-based statement, which contained the word “bullying” than the specific items regarding various forms of frequent victimization. With regard to the robust finding for African American girls, there is research that suggests that African American families may socialize their children in ways which are more gender neutral [29], [34]; therefore, the African American girls, like boys in general, may be more sensitive than other girls to the stigma of being victimized through bullying. Finally, because the data are cross-sectional, it is difficult to draw conclusions regarding developmental differences. However, as anticipated, the ES children who had been victimized (particularly the girls) tended to be less likely to report that they were bullied via the definition-based measure. It is possible that these younger children may not have consistently applied the definition of bullying when completing the two measures. Future research, perhaps of a qualitative nature, should further explore potential cultural, gender, and developmental differences in children's conceptualization of or willingness to report being a victim of bullying. Limitations Some potential limitations should be considered when interpreting the current findings. Although the sample is large and diverse, all the data come from a single Maryland school district; thus, the generalizability of the findings is unknown. The data are cross-sectional and were collected using an anonymous self-report measure; therefore, social desirability may have influenced students' responses. However, a recent study by Wang et al [35] found that adolescents reported higher and perhaps more valid rates of sensitive information (e.g., drug use) on a Web-based survey format than on a written survey. Accordingly, the Web-based administration of the survey may have resulted in more accurate data regarding the participants' experiences. We also used children's self-identification of their racial/ethnic groups. It is possible that some of the younger participants may not have understood the question or known their ethnicity [36], [37], whereas other children may have responded inaccurately to disguise their ethnicity [36]. In addition, the definition of bullying used in the current study specified only two of the three typical features of bullying (repeated and intentional, but not the power imbalance) [12]. Although the power imbalance is an important feature of bullying, the notion is challenging to convey to younger children. In fact, Nansel and Overpeck [38] have noted that “most measures of bullying probably fall short in fully delineating one or more of these elements” (p. 1135). We applied the recommended threshold of two or more times a month as a cut point for determining frequent bullying [31]. Thus, we may have observed a different pattern of results had we used a different threshold. Similarly, we only required one form of bullying for inclusion in the behavior-based dichotomous grouping. Because a student may experience multiple forms of bullying, it is difficult to determine which forms or combinations are associated with an increased likelihood of considering this experience to be bullying. Future research should examine combinations of victimization that increase the likelihood that a youth reports being frequently bullied. It is also possible that unmeasured contextual factors could have influenced the rates of bullying. For example, research by Graham and colleagues [39], [40] suggests that students' likelihood of being victimized is greater among ethnic minority youth in classrooms that are disproportionately nonminority. However, it is not clear whether contextual factors, such as class composition, would account for the divergent responses on two different measures of bullying. Future research should examine these and other possible contextual influence, such as school indicators of disorder, as possible predictors of discrepancies in self-reports of bullying. Conclusions These findings highlight the significance of cultural, gender, and developmental factors that may distinguish these groups in terms of their conceptualization of bullying, as it appears that some youth who are experiencing bullying are less likely than others to label those experiences as bullying. The inconsistencies in reports of bullying evidenced when assessing bullying through different question formats suggest that greater attention should be paid to the issue of measurement in the research on bullying [7]. Of particular concern is the finding that these inconsistencies tend to be greater for minority youth, especially African American girls. The analyses of the current data suggest that assessing the separate behaviors using a behavior-based measure would provide considerably larger prevalence estimates than a definition-based measure; however, it is unclear which estimates are most accurate. The precision of bullying measurement should be improved so that prevalence rates can be estimated more accurately and preventive interventions can target at-risk youth more effectively [7]. In conclusion, when assessing the prevalence of bullying, researchers should be aware that minority students may be less likely to report that they were “bullied” frequently through a definition-based single-item measure, although they may report frequently experiencing specific forms of victimization when the various forms of victimization are assessed discretely. Consequently, it would be advantageous for future research on bullying to use multiple self-report formats to better estimate bullying prevalence rates among minority populations. Acknowledgments The authors would like to thank Dr. Rhonda Gill and Mrs. Lucia Martin from the Maryland public school system for providing access to the data. This project was funded by the Centers for Disease Control and Prevention (1U49CE000728 and 1K01CE00133-01) and the Hamilton Fish Institute on School and Community Violence, as part of its Field-Initiated Studies Program, under Grant No. 2005-JL-FX-0157 awarded by the Office of Juvenile Justice and Delinquency Prevention, Office of Justice Programs, U.S. Department of Justice. Points of view or opinions in this article are those of the authors and do not necessarily represent the Centers for Disease Control and Prevention, the U.S. Department of Justice, or the Hamilton Fish Institute. References [1]. [1]Gladstone GL, Parker GB, Malhi GS. Do bullied children become anxious and depressed adults? (A cross-sectional investigation of the correlates of bullying and anxious depression). J Nerv Ment Dis. 2006;194(3):201–208. MEDLINE |
CrossRef
[2]. [2]Hawker DS, Boulton MJ. Twenty years' research on peer victimization and psychosocial maladjustment: a meta-analytic review of cross-sectional studies. J Child Psychiatry. 2000;41:441–455. [3]. [3]Brockenbrough K, Cornell DG, Loper AB. Aggressive attitudes among victims of violence at school. Educ Treat Child. 2002;25(3):273–287. [4]. [4]Leary MR, Kowalski RM, Smith L, Phillips S. Teasing, rejection, and violence: case studies of the school shootings. Aggress Behav. 2003;29(3):202–214. [5]. [5]Verlinden S, Hersen M, Thomas J. Risk factors in school shootings. Clin Psychol Rev. 2000;20(1):3–56. MEDLINE |
CrossRef
[6]. [6]Greif JL, Furlong MJ. The assessment of school bullying: using theory to inform practice. J Sch Violence. 2006;5(3):33–50. [7]. [7]Furlong MJ, Sharkey JD, Felix E, et al. Bullying assessment: a call for increased precision of self-reporting procedures. In: Jimerson SR, Swearer SM, Espelage DL, eds. The International Handbook of School Bullying. New York: Routledge, in press. [8]. [8]Cornell DG, Sheras PL, Cole JC. Assessment of bullying. In: Jimerson S, Furlong MJ editor. Handbook of School Violence and School Safety. Mahwah, NJ: Lawrence Erlbaum; 2006;p. 191–210. [9]. [9]Crothers LM, Levinson EM. Assessment of bullying: a review of methods and instruments. J Counsel Dev. 2004;82:496–503. [10]. [10]Hazler RJ, Carney JV, Granger DA. Integrating biological measures into the study of bullying. J Counsel Dev. 2006;84:298–307. [11]. [11]Elinoff MJ, Chafouleas SM, Sassu KA. Bullying: considerations for defining and intervening in school settings. Psychol Sch. 2004;41(8):887–897. [12]. [12]Olweus D. Bullying at School. Oxford, England: Blackwell; 1993;. [13]. [13]Land D. Teasing apart secondary students' conceptualizations of peer teasing, bullying and sexual harassment. Sch Psychol Int. 2003;24:147–165. [14]. [14]Nansel TR, Overpeck M, Pilla RS, et al. Bullying behaviors among US youth: prevalence and associations with psychosocial adjustment. JAMA. 2001;285(16):2094–2100. MEDLINE |
CrossRef
[15]. [15]Olweus D. The Revised Olweus Bully/Victim Questionnaire. Bergen, Norway: University of Bergen; 1996;. [16]. [16]Swearer SM. Bully Survey. Lincoln, NE: University of Nebraska; 2001;. [17]. [17]Espelage DL, Holt MK. Bullying and victimization during early adolescence: peer influences and psychosocial correlates. In: Geffner RA, Loring M, Young C editor. Bullying Behavior: Current Issues, Research, and Interventions. Binghamton, NY: Haworth Press; 2001;p. 123–142. [18]. [18]Rigby K. The relationship between reported health and involvement in bully/victim problems among male and female secondary schoolchildren. J Health Psychol. 1998;3:465–476.
CrossRef
[19]. [19]Rigby K, Slee PT. Dimensions of interpersonal relating among Australian school children and their implications for psychological well-being. J Social Psychol. 1993;133(1):33–42. MEDLINE [20]. [20]Snyder HN, Sickmund M. Juvenile Offenders and Victims: 2006 National Report. Washington, DC: U.S. Department of Justice, Office of Justice Programs, Office of Juvenile Justice and Delinquency Prevention; 2006;. [21]. [21]Spriggs AL, Iannotti RJ, Nansel TR, Haynie DL. Adolescent bullying involvement and perceived family, peer and school relations: commonalities and differences across race/ethnicity. J Adolesc Health. 2007;41:283–293. |
CrossRef
[22]. [22]Juvonen J, Graham S, Schuster MA. Bullying among young adolescents: the strong, the weak, and the troubled. Pediatrics. 2003;112:1231–1237. [23]. [23]Seals D, Young J. Bullying and victimization: prevalence and relationship to gender, grade level, ethnicity, self-esteem, and depression. Adolescence. 2003;38(152):735–747. MEDLINE [24]. [24]Hanish LD, Guerra NG. The roles of ethnicity and school context in predicting children's victimization by peers. Am J Community Psychol. 2000;28(2):201–223. MEDLINE |
CrossRef
[25]. [25]Bierman KL. Peer Rejection: Developmental Processes and Intervention Strategies. New York: Guilford Press; 2004;. [26]. [26]Phelps RE, Meara NM, Davis KL, Patton MJ. Blacks' and Whites' perceptions of verbal aggression. J Counsel Dev. 1991;69:345–350. [27]. [27]Nansel TR, Overpeck MD, Haynie DL, et al. Relationships between bullying and violence among US youth. Arch Pediatr Adolesc Med. 2003;157:348–353. MEDLINE |
CrossRef
[28]. [28]Underwood MK. Social Aggression among Girls. New York: Guilford Press; 2000;. [29]. [29]Putallaz M, Grimes CL, Foster KF, et al. Overt and relational aggression and victimization: multiple perspectives within the school setting. J Sch Psychol. 2007;45:523–547. [30]. [30]Bradshaw CP, Sawyer AL, O'Brennan LM. Bullying and peer victimization at school: perceptual differences between students and school staff. Sch Psychol Rev. 2007;36:361–382. [31]. [31]Solberg M, Olweus D. Prevalence estimation of school bullying with the Olweus/Victim Questionnaire. Aggress Behav. 2003;29(3):239–268. [32]. [32]Craig WM, Pepler DJ. Identifying and targeting risk for involvement in bullying and victimization. Can J Psychiatry. 2003;48(9):577–582. MEDLINE [33]. [33]Triandis HS. Variations in Black and White Perceptions of the Social Environment. Urbana, IL: University of Illinois Press; 1976;. [34]. [34]Hill SA, Sprague J. Parenting in Black and White families: the interaction of gender with race and class. Gend Soc. 1999;13:480–502. [35]. [35]Wang Y, Lee C, Lew-Ting C, et al. Survey of substance use among high school students in Taipei: web-based questionnaire versus paper-and-pencil questionnaire. J Adolesc Health. 2005;37:289–295. |
CrossRef
[36]. [36]Entwisle DR, Astone NM. Some practical guidelines for measuring youth's race/ethnicity and socioeconomic status. Child Dev. 1994;65(6):1521–1540.
CrossRef
[37]. [37]Hughes D. Correlates of African American and Latino parents' messages to children about ethnicity and race: a comparative study of racial socialization. Am J Community Psychol. 2003;31(1/2):15–33. MEDLINE |
CrossRef
[38]. [38]Nansel TR, Overpeck MD. Operationally defining “bullying”: in reply. Arch Pediatr Adolesc Med. 2003;157:1135.
CrossRef
[39]. [39]Graham S. Peer victimization in school. Curr Direct Psychol Sci. 2006;15(6):317–321. [40]. [40]Graham S, Juvonen J. Ethnicity, peer harassment, and adjustment in middle school: an exploratory study. J Early Adolesc. 2002;22:173–199. Department of Mental Health, Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland  Address correspondence to: Anne L. Sawyer, Ph.D. Student, Department of Mental Health, Johns Hopkins Bloomberg School of Public Health, 625 N. Broadway, Baltimore, MD 21205.
PII: S1054-139X(08)00105-5 doi:10.1016/j.jadohealth.2007.12.011 © 2008 Society for Adolescent Medicine. Published by Elsevier Inc. All rights reserved. | |
|
FULL TEXT