Abstract
A substantial body of research has developed surrounding the phenomenon of bullying online and off-line among youth populations. These studies demonstrate there are significant psychological and emotional consequences for bullying victims. Researchers have not, however, explored in depth how these outcomes differ across the sexes based on the types of bullying they experience. In addition, few have explored these issues in a cross-national context to consider how variations in technological access and exposure may impact behavior. Thus, this study utilizes a nationally representative sample of youths from Singapore, a country with significant access to technology, to consider how suicidal ideation and school truancy as a consequence of bullying are correlated with bullying victimization online and off-line as well as technology use and various demographic factors. The findings indicate that bullying victimization across virtual and real spaces are significantly related to these outcomes and that there are significant differences between the sexes concerning suicidal ideation. The utility of this study for both criminal justice and mental health practitioners are examined in depth.
Over the past few decades, bullying has been identified as one of the most significant problem behaviors confronting children and adolescents (Olweus, 1991). In fact, some researchers have labeled bullying among school-age youths a major public health concern (Nansel et al., 2001). Bullying behaviors typically involve persistent physical, verbal, or emotional harassment of one individual over another, often accompanied by a power imbalance that negatively impacts the intended target (Dake, Price, & Telljohann, 2003). Bullying usually involves physical or verbal abuse in the real world, though this has changed over the last two decades with the adoption and dissemination of technology (Hinduja & Patchin, 2008; Raskauskas & Stoltz, 2007). The use of cell phones, e-mail, instant messaging (IM), social networking sites, and other forms of computer-mediated communications (CMC) allow individuals to create and send harassing messages or rumors about victims in a distributed fashion (Berson, Berson, & Ferron, 2002; Hinduja & Patchin, 2008; Holt & Bossler, 2009; Twyman, Saylor, Taylor, & Comeaux, 2010; Ybarra & Mitchell, 2004).
As a result, individuals may now be bullied both online and off-line, creating substantive emotional and psychological consequences. Persistent bullying, regardless of environment, makes victims vulnerable to social, physical, and mental health consequences (Arseneault, Walsh, Trzesniewski, Newcombe, Caspi, & Moffitt, 2006; Sourander, Helstela, Helenius, & Piha, 2000). Studies demonstrate that bullying victimization affects victims’ school attendance and academic success (Glew, Fan, Katon, Rivara, & Kernic, 2005), emotional well-being (Nansel et al., 2001; van der Wal, de Wit, & Hirasing, 2003), psychiatric symptoms (Kumpulainen & Rasanen, 2000), and levels of depression and suicidal ideation (Kaltiala-Heino, Rimpela, Marttunen, Rimpela, & Rantanen, 1999; Klomek et al., 2008). There is also a growing body of research examining the impact of both real world, or traditional bullying, and cyberbullying which suggests that males and females differ in their response to repeated victimization (Espelage, Mebane, & Swearer, 2004; Li, 2006; Scheithauer, Hayer, Pertermann, & Jugert, 2006; Smith et al., 2008; Whitney & Smith, 1993).
Although scientific evidence has provided a foundation for which to understand gender differences in responses to bullying victimization, scholars have yet to identify the gendered pathways through which bullying victimizations online and off-line lead to specific concerns like skipping school and suicidal ideation. In addition, most research on cyberbullying utilizes youth samples from the United States (Hinduja & Patchin, 2008; Marcum, 2010; Wolak, Mitchell, & Finkelhor, 2007), Canada (Beran & Li, 2005, 2007), and Turkey (Erdur-Baker, 2010). Few researchers have examined the issue of cyberbullying among Asian youth (Hokoda, Lu, & Angels, 2006; Huang & Chou, 2010; Li, 2006; Wong, Lok, Lo, & Ma, 2008), despite the fact that many Asian nations have significant high-speed Internet connectivity and substantive access to technology.
In order to address these gaps in the literature, this study will examine the way that skipping school and suicidal ideation as a consequence of bullying are impacted by traditional and cyberbullying experiences, technology use, and demographic characteristics using a nationally representative sample of youth in Singapore. Citizens of this country have a high degree of access to computers, technology, and high-speed connectivity (Central Intelligence Agency World FactBook, 2006; Liau, Khoo, & Hwaang, 2005), making this sample extremely valuable in order to understand variations in the impact of bullying victimization online as well as off-line. The findings will also be segmented by gender to understand sex differences in the consequences of bullying victimization. This study benefits both criminal justice and mental health professionals by identifying variations in the impact of bullying victimization in virtual and real environments for male and female youths.
Exploring the Consequences of Bullying
Recent research has demonstrated a substantial tie between bullying victimization and negative outcomes. In particular, school absence may be a direct consequence of real-world bullying experiences in an attempt to avoid persistent or repeated victimization. There is also evidence that victims of bullying in online environments report increased school truancy (Katzer, Fetchenhauer, & Belschak, 2009; Ybarra, Mitchell, Finkelhor, & Wolak, 2007), diminished academic performance (Beran & Li, 2007), and feelings that school is no longer a safe place (Varjas, Henrich, & Meyers, 2009). Suicidal thoughts also appear to be directly correlated with both real-world (Kim, Koh, & Leventhal, 2005; van der Wal et al., 2003) and cyberbullying experiences (Hinduja & Patchin, 2008; Klomek et al., 2008; Li, 2006). There is also evidence that a small proportion of youth experience bullying victimization online and off-line (Erdur-Baker, 2010; Hinduja & Patchin, 2008; Kowalski & Limber, 2007; Ybarra & Mitchell, 2004). As a consequence, it is likely that children who report negative emotional responses or school absence may do so as a result of bullying in multiple environments.
Given youths’ increased exposure to technology (Hinduja & Patchin, 2008; Smith et al., 2008; Ybarra & Mitchell, 2004) and the prevalence of cyberbullying, it is possible that there is a relationship between the use of CMCs, mobile phones, school absences, and suicide. In fact, there is a direct correlation between time spent online in social networks, chat rooms, and e-mail and experiencing electronic bullying or harassment (Berson et al., 2002; Hinduja & Patchin, 2008; Holt & Bossler, 2009; Twyman et al., 2010; Ybarra & Mitchell, 2004). In addition, there is some evidence to suggest that bullying and harassment is gendered, in that females increase their risk of victimization by spending increased time online (Holt & Bossler, 2009; Ybarra & Mitchell, 2004). Furthermore, there is some evidence to support that victims of bullying in both virtual and real environments utilize technology more frequently, although males engage in more risky information sharing practices generally (Erdur-Baker, 2010).
Finally, there are mixed results concerning the demographic correlates of various bullying outcomes. Specifically, school adjustment, or attitudes toward school, appears to be correlated with both bullying victimization (Beran & Li, 2007; Forero, McLellan, Rissel, & Bauman, 1999; Nansel et al., 2001; Sourander et al., 2000; Wolke, Woods, Bloomfield, & Karstadt, 2000), and suicidal ideation generally (Arseneault et al., 2006). Youths’ attitudes toward school can be negatively impacted by bullying experiences online and off-line, affecting academic performance and fear of school environments (Esbensen & Carson, 2009; Glew, et al., 2005; Sourander et al., 2010). In turn, school affect will most likely be related to both increased levels of truancy (Beran & Li, 2007; Esbensen & Carson, 2009; Nansel et al., 2001; Wolke et al., 2000) and suicidal ideation (van der Wal et al., 2003).
Age may also be related to differential outcomes related to bullying victimization. For instance, multiple studies demonstrate that younger children are more likely to experience bullying in the real world (Olweus, 1993). By contrast, cyberbullying is more likely to be reported by older youth (Tokunaga, 2010). This difference may be a consequence of the age-attenuated use of technology, since very young children are likely to have limited or highly regulated access to computers and mobile phones (Smith et al., 2008). As a consequence, it is unclear how age may affect reported levels of school truancy or suicidal thoughts.
Finally, gender appears to have a mixed effect on bullying victimization outcomes. While many studies find a relationship between bullying victimization and school absences (e.g., Glew et al., 2005; Nansel et al., 2001; Olweus, 1993), these studies suggest males are only slightly more likely to skip school than females. By contrast, females appear to be much more likely to report depression and suicidal ideation as a consequence of psychological forms of bullying, particularly when it is constant over time (Kim et al., 2005; Klomek et al., 2009; van der Wal et al., 2003). Males are also likely to report suicidal ideation as a result of direct or traditionally physical forms of bullying, though not at the same rate as females generally (Kim et al., 2005; Klomek et al., 2009; van der Wal et al., 2003).
Method
Taken as a whole, it is clear that young people face a substantive risk of bullying online and off-line, which can lead to significant negative emotional and physical consequences. Although there are well-recognized relationships between suicidal ideation, truancy, and both physical and cyberbullying victimization, it is unclear how technology use and demographic factors may be related to specific bullying outcomes. Additionally, limited research has considered how these factors may vary across international samples of youth. Thus, this study provides an analysis of the variables influencing both truancy and suicidal thoughts stemming from bullying are correlated with bullying experiences online and off-line, routine computer activities, and demographic characteristics in a nationally representative sample of Singapore youth.
Participants
These analyses utilized data from a self-report survey collected from two primary (n = 933) and eight secondary schools (n = 3,382) in Singapore in 2006. These schools were located across the country, and incorporated students from all social, economic, and cultural backgrounds. This sample was developed to understand the prevalence and incidence of both real-world and cyberbullying across Singapore. All survey instruments were administered by classroom instructors during school hours, and required about 20–25 min to complete. Consent for participation in the study was acquired at the beginning of each school year as parents and/or guardians are informed that their students will be asked to participate in educational and psychological assessments during the academic term. Those who are willing to allow their child to participate are required to sign a consent form. Students whose parents did not give consent or those who did not wish to participate were not required to complete the survey.
The survey administration process began with an interactive discussion session with students managed by the research team. A series of slides describing how the research team defined bullying and what harms result from these behaviors were explained. Specifically, the study defined bullying as any treatment that a person received that was deliberate, intended to hurt, unjustified, repeated, and carried out by a more powerful person or group in keeping with the larger research literature on this issue (e.g., Olweus, 1993; Nansel et al., 2001). Students were allowed to ask questions during this period to clarify any concerns that they had or confusion over what constitutes bullying. Directions were provided on survey completion and the protections afforded to students. Participation in the study was voluntary, and students were informed that they could leave questions blank if they felt uncomfortable providing a response. No questions concerning personal information, such as name or address, were included in the survey to ensure anonymity.
A total of 2,303 males (55.5%) and 1,844 females (44.5%) participated in the study, which was slightly overrepresentative of male youths in Singapore in 2006 (CIA World FactBook, 2007). In 2007, there were a total of 362,329 males under the age of 15, and 337,964 females, creating a sex ratio under the age of 15 of 1.07 males for every female (CIA World Fact Book 2007). Thus, this sample appears similar to that of the larger population.
The total sample size for the analyses was 3,096 respondents. The analyzed sample was smaller than the sample collected in part to the insertion of a cyberbullying questionnaire administered after data collection began at one of the two primary schools. No attempts were made to collect this additional data due to the difficulty in reassembling the respondent population (N = 253) at that institution. Additionally, the majority of missing data come from no responses for the cyberbullying and mobile phone bullying victimization items. This may be a result of the voluntary nature of the survey and the fact that these items appeared on a separate survey instrument completed after questions related to physical bullying. The sample analyzed, however, is similar to that of the initial sample (see Table 1).
Descriptive Statistics.
Note. IM = instant messaging; MMS = multimedia messaging service.
Measures
Dependent variables
There are two dependent variables used in this study, each relating to the impact of bullying on individual behavior. First, respondents were asked: “Have you ever stayed away from school because of bullying?” This item was written to encompass the individual experience of bullying victimization, as well as prospective concerns over bullying within their school generally. Four response categories were provided, including: (1) No, I’ve never thought of doing so; (2) No, but I’ve thought of doing so; (3) Yes, I have once or twice; and (4) Yes, more than twice. The modal response category was no (78.5%), though 12% thought of skipping school, 3.4% had skipped once or twice, and 1.6% had done so more than twice over the last year. Due to the limited variation, the responses were dichotomized into whether the respondents had never thought of skipping school (option 1 = 0) and whether they had either contemplated or actually did skip school (options 2–4 = 1; see Table 1).
Second, respondents were asked: “Have you ever thought of killing yourself because you were bullied?” Response categories included: (1) No, I’ve never thought of doing so; (2) Yes, I have once or twice; and (3) Yes, more than twice. The primary response was no (77%), though 16% had thought of doing so once or twice, and 2.4% had done so more than twice. These items were dichotomized (0 = no; 1 = yes) in order to assess the factors associated with any reported thoughts of suicide.
Bullying Victimization
Scholars have found that there are substantive differences in the experiences of bullying victimization in virtual and real environments. Bullying in the real world may take multiple forms and range in severity from name calling to physical violence (Wang, Iannotti, & Nansel, 2009). Cyberbullying, however, can occur across multiple forms of CMCs and involve teasing or more serious actions such as masquerading as or posting threats to the victim (Willard, 2004). Mobile phones can also be used to send text messages to a victim independent of their laptop or desktop computer (Willard, 2004).
Three binary measures were created to identify what impact personal experiences with traditional, cyber, and mobile phone bullying victimization have on bully-driven truancy and suicidal ideation. In order to assess traditional bullying victimization, respondents were asked: “Did any of these things happen to you while you were being bullied”: (1) being teased; (2) called names; (3) threatened; (4) left out; and (5) being hit or kicked (options: never, sometimes, and often). These 5 items created a reliable scale (α = .814) for the frequency of bullying experiences in the real world. The bullying victimization literature primarily dichotomizes victimization due to relatively low reported rates (Hinduja & Patchin, 2008; Ybarra & Mitchell, 2004). Thus, this scale was collapsed into a binary measure (0 = No; 1 = Yes) in order to be able to more directly compare this measure with cyber and mobile bullying victimization measures which were dichotomized due to limited variation.
To assess cyber and mobile phone bullying victimization, respondents were presented with an item assessing their experiences in various communications venues. Specifically, students were presented with the following item: “If you have been cyber bullied, how often would you say you have been bullied this way: 1) chat rooms, 2) email, 3) computer IM, 4) bulletin board systems, 5) newsgroups, 6) mobile phone text messages, and 7) mobile multi-media (MMS) messages.” Four responses were provided: (a) rarely or never, (b) occasionally, (c) fairly often, and (d) very often. Many researchers suggest that bullying experiences via CMC, like chat rooms, and mobile phones are equivalent behaviors that constitute cyberbullying (Hinduja & Patchin, 2008). The fact that cell phones are constantly on an individual’s person and usually turned on make it easier for bullying messages sent via text messages to be received immediately. Additionally, traditional text and multimedia messaging service (MMS) messages cannot be received on a computer unless the recipient specifically forwards the content via e-mail. Therefore, there are distinct differences between these communication methods and the prospective experience of bullying in each.
In order to examine possible differences in the impact of cyber and mobile bullying victimization, separate measures were created. The 5 items for bullying via chat rooms, e-mail, computer Instant Messaging, bulletin board systems, and newsgroups were summed to assess cyberbullying victimization (α = .921). The 2 items regarding mobile phone text messages and mobile multimedia messages were combined to create a scale for the prevalence of mobile phone bullying victimization (α = .806). Each of these scales were collapsed into binary measures (0 = No; 1 = Yes) due to limited variation.
Technology Use
Nine measures for technology use were included to understand the impact that Internet and cell phones may have on skipping school and suicidal ideation. Specifically, two questions were asked concerning Internet connectivity. Respondents were asked, “Can you use the Internet to ‘chat’, send e-mails, instant messages, or post on web pages in school?” and “Can you use the Internet to ‘chat’, send e-mails, instant messages, or post on web pages at home? (0 = no; 1 = yes). The proportion of respondents with Internet connectivity at home (70.5%) is somewhat larger than the number of individuals with Internet connectivity generally (53.9%) in Singapore (CIA World FactBook, 2007). Respondents were also asked, “Do you have your own mobile phone?” (0 = no; 1 = yes). The proportion of respondents with cell phones (76.3%) was slightly lower than the national percentage of cell phone ownership (94.7%), though it is unclear how many juveniles own these devices in Singapore (CIA World FactBook, 2007). Thus, the sample appears to reflect similar technology adoption trends within Singapore.
In order to assess technology use, respondents were also asked how frequently they engaged in various tasks while online and through various media. Specifically, respondents were asked, “How often do you do these?” (1) enter chat rooms, (2) send e-mails, (3) send computer instant messages, (4) post on bulletin boards, (5) write on blogs, and (6) send mobile phone multimedia messages. There were four response categories for each form of CMC, including (1) rarely or never, (2) 1 to 2 times per week, (3) 3 to 4 times per week, and (4) almost everyday. Each of these 6 items was included to capture variations in basic Internet and cell phone usage within this sample of youth.
Demographics
Finally, several demographic characteristics were included to examine their relationships with the effects of bullying based on correlates identified in existing research. School adjustment was assessed using a 7-item Likert-type scale question with pictorial responses for the following question: “Now look at these pictures and place a circle around the letter under the face which is most like you when you are at school?” The seven available response options ranged from (1) very happy to (7) very sad. Though the literature is mixed as to the validity of such visual measures (Chambers & Craig, 1998; Davies & Brember, 1994; Reynolds-Keefer, Johnson, Dickenson, & McFadden, 2009), the range in ages in the sample and variation in reading comprehension across the respondents made this a more useful measure to assess school adjustment. Level in school was included as a proxy for age and ranged from 3 to 11 in keeping with the sample of primary and secondary schools. Finally, gender was included as a binary measure (0 = male; 1 = female), and slightly overrepresents females (54.7%) relative to the composition of female youth (48.3%) in Singapore (CIA World FactBook, 2007).
Findings
In order to examine how different forms of reported bullying victimization were related to school truancy and suicidal ideation as a consequence of bullying, we first ran crosstabs. These models indicated that all three bullying victimization measures were significantly related to both outcomes (see Table 2). Specifically, 22% of those students who experienced traditional bullying victimization reported contemplating or skipping school relative to the 6% of students who had no traditional bullying victimization experiences. Similarly, 22.4% of students who experienced traditional bullying reported suicidal ideation. Of the youth who were cyberbullied, 26.7% contemplated skipping school (as compared to 14.8% who were not victimized) and 28.1% considered suicide (compared with 16.1%). In addition, 27.8% of Singapore youth who were bullied via their mobile phones considered skipping school and 26% thought about suicide. Thus, these models indicated a significant link between all three forms of bullying victimization and both school truancy and suicide ideation and support further univariate and multivariate analyses to understand the influence of bullying victimization experiences relative to other behavioral and attitudinal factors.
Crosstabs for Skipping School and Suicidal Ideation and Bullying Victimization.
The correlation matrix presented in Table 3 demonstrates that there are significant correlations between all forms of bullying victimization, suicidal ideation, and skipping school due to bullying. School truancy had significant, but weak, relationships with traditional, cyber, and mobile bullying victimization. Similarly, suicidal ideation had significant but weak correlations with traditional, cyber, and mobile bullying victimization. In addition, traditional bullying victimization was weakly correlated with cyber and mobile bullying victimization, while cyber and mobile bullying victimization were moderately correlated with each other.
Pearson Correlation Matrix.
Note. IM = instant messaging; MMS = multimedia messaging service.
*p < .05. **p < .01. (two-tailed); All Ns presented below the r value.
Many of the technology measures were not correlated with skipping school, though home Internet use, bulletin board use, and MMS texting were related. By contrast, those with no Internet access at home and school and no access to mobile phones were correlated with suicidal ideation. The use of IM programs and blogging were also negatively related to suicidal ideation. School adjustment was positively correlated with both skipping school and suicidal ideation, indicating that being sad at school was linked with both outcomes. School level was negatively correlated with both outcomes, suggesting younger individuals were more likely to experience and report these behaviors. Finally, being male was weakly associated with skipping school while being female was correlated with suicidal ideation. Thus, the significant correlations in the correlation matrix provide ample support to conduct multivariate analyses.
Binary logistic regression models were estimated for skipping school and suicidal ideation respectively due to the use of dichotomous dependent variables with limited variation (see Tables 4 and 5). In addition, victimization is viewed as a probabilistic event by many researchers (e.g., Holt & Bossler, 2009; Nansel, Overpeck, Haynie, Ruan, & Scheidt, 2003), making the use of binary logistic regression models invaluable to identify the significant factors that increase the odds of a specific outcome. In this case, binary logistic regression techniques allow for the identification of the factors that increase the odds of reporting skipping school or suicidal ideation as a result of bullying.
Skipping School Due to Bullying.
Note. IM = instant messaging; MMS = multimedia messaging service.
*p < .05. **p < .01. Model 1: Chi-square = 287.440**; −2LL = 2,486.283. Model 2: Male model: Chi-square = 171.373**; −2LL = 1,358.733. Female model: Chi-square = 123.400**; −2LL = 1,120.145. There were no significant differences (z > 1.96) between the partitioned models.
Suicidal Ideation Due to Bullying.
Note. *p < .05. **p < .01. Model 1: Chi-square = 567.241**; −2LL = 2,314.054. Model 2: Male model: Chi-square = 340.810**; −2LL = 1,156.679. Female model: Chi-square = 243.825**; −2LL = 1,133.969. Shaded cells illustrate significant differences (z > 1.96) between partitioned models.
Multicollinearity was not an issue for these analyses. Although cyber and mobile bullying victimization were moderately correlated, multicollinearity diagnostics indicated that no variance inflation factor (VIF) exceeded 10 and no tolerance level dipped below 0.2. In fact, the cyberbullying victimization measure had the lowest tolerance (0.690) and highest VIF (1.448) within the models, clearly within acceptable levels. 1
Table 4 presents the effects of the various factors on school truancy. Model 1 provides the initial regression results which indicate that all forms of bullying victimization were significant correlates of truancy. In fact, traditional bullying victimization had an odds ratio of 3.568, which was larger than that of cyberbullying (1.488) and mobile phone bullying (1.420). All technology use measures, with the exception of MMS texting, were nonsignificant. Student school adjustment was also significant, indicating that those with negative attitudes were more likely to be absent (Beran & Li, 2007; Esbensen & Carson, 2009; Nansel et al., 2001). Gender was also nonsignificant, which was surprising given the evidence supporting gender differences in bullying victimization outcomes (Kim et al., 2005; Klomek et al., 2009; van der Wal et al., 2003).
Since the literature implies that there are distinct differences in bullying experiences and outcomes based on the gender of bullying victims, we partitioned the model by gender and ran equality of coefficient tests (Paternoster, Brame, Mazerolle, & Piquero, 1998) to examine any variation in the correlates of truancy between the sexes. Though cyberbullying victimization was significant for males, and mobile phone bullying victimization was significant for females, there were no statistically significant differences in the correlates of skipping school for boys and girls. In fact, traditional bullying and school adjustment were significant correlates for both genders; those who experience greater instances of traditional bullying and have a negative affect were likely to skip school.
The models for suicidal ideation are presented in Table 5. Model 1 illustrated that all forms of bullying victimization were significant correlates of suicidal ideation (Hinduja & Patchin, 2008; Kim et al., 2005; Li, 2006). Traditional bullying had the largest odds ratio of the three forms of bullying victimization in this model as well at 1.943. Though individual forms of technology use were nonsignificant, both home Internet use and mobile phone ownership reduced the odds of suicidal ideation. Finally, females, those in lower grades, and those with negative school adjustments were all more likely to report suicidal thoughts.
The model was again parsed by gender to consider any differences in the correlates of suicidal ideation across the sexes. Traditional bullying victimization was significant for both genders. Though cyberbullying victimization significantly increased the odds of suicidal ideation for females, the difference in coefficients between the male and female models was not significant. The effect of mobile bullying victimization on suicidal ideation was, however, significantly stronger for males than females. Males experience of mobile phone bullying victimization increased the odds of suicidal ideation by 2.206 times, relative to 1.145 times for females. In addition, the effect of IM usage was significantly different between the two sexes as the model indicated that spending more time writing IM decreased the odds of suicidal ideation for females. Finally, the effect of school level on suicidal ideation significantly differed between male and female youths as an increase in school levels served as a greater protective factor for males than it did for females.
Discussion and Conclusions
This study sought to understand the factors most likely to influence school truancy and suicidal ideation as a consequence of bullying victimization in a sample of youth from Singapore. The findings indicated that all forms of bullying victimization were significantly related to both skipping school and suicidal thoughts. In fact, traditional bullying victimization experiences had the largest overall impact on school truancy due to bullying. This is sensible given school absence enables youths to escape the bullying experience (Glew et al., 2005). At the same time, the fact that cyber and mobile phone bullying victimization were significant correlates for skipping school suggests that virtual experiences have strong influences on youth behavior in the real world as well (e.g. Katzer et al., 2009; Ybarra et al., 2007). The same relationships are also evident concerning suicidal ideation from bullying, supporting the growing body of literature connecting suicidal thoughts and victimization online and off-line (Hinduja & Patchin, 2008; Kim et al., 2005; Klomek et al., 2008; van der Wal et al., 2003). Thus, these findings demonstrate that the experience of bullying victimization in the real and virtual world have a significant impact on individual behavior with potentially serious, if not fatal, consequences.
Differences in the influence of technology on both truancy and suicidal ideation due to bullying were also present in this study. Individuals who more frequently engage in MMS texting were more likely to skip school. This may be a reflection of the relationship between the increased use of technology and a generally higher risk of online bullying victimization (Berson et al., 2002; Erdur-Baker, 2010; Hinduja & Patchin, 2008), which may in turn increase the perceived value in school truancy for a victim. At the same time, those who do not have home Internet access or a mobile phone were more likely to report suicidal ideation. This may be related to the fact that those in earlier school levels with negative affect were more likely to report suicidal ideation. Those youth who report both real and virtual bullying victimization but do not have access to this technology given their age (see Smith et al., 2008) may feel isolated and therefore more likely to have suicidal thoughts. In fact, this finding is supported by the fact that younger females were more likely to report suicidal ideation (see Table 5). Since access to technology appears to be age-attenuated, older youth have the ability to find and connect with peers for social support (Beran & Li, 2007; Erdur-Baker, 2010; Hinduja & Patchin, 2008), or even engage and bully their tormenters (Holt & Bossler, 2009; Liau et al., 2005; Marcum, 2010). Further research is needed to better explore this issue and identify whether technology truly insulates or aggravates negative outcomes from bullying (Beran & Li, 2007; Erdur-Baker, 2010; Hinduja & Patchin, 2008; Smith et al., 2008).
This study also found differences between the sexes regarding suicidal ideation due to bullying. Specifically, females were more likely to report these behaviors in keeping with previous research on the gendered nature of the psychological consequences of bullying victimization (Hinduja & Patchin, 2008; Kim et al., 2005; Klomek et al., 2009; van der Wal et al., 2003). In addition, mobile phone bullying victimization increased the odds of suicidal thoughts for males, which is in contrast to the larger literature suggesting males report negative consequences from traditional bullying experiences (Erdur-Baker, 2010; Tokunaga, 2010). This may, however, reflect the consequence of reporting multiple bullying victimization experiences generally. In addition, increased use of IM systems decreased the odds of suicidal ideation for females, suggesting that the social connectivity afforded by IM may help to decrease feelings of social isolation for females but not for males. Thus, additional research is needed to disentangle the relationship between technology use and suicidal ideation generally (Erdur-Baker, 2010).
Taken as a whole, this study suggests that the negative consequences of bullying do not differ substantively between this population of youth in Singapore and research samples from across the world. As a consequence, this has direct implications for school administrators, mental health professionals, and parents in Singapore and elsewhere. Specifically, the direct effect of traditional bullying victimization on both skipping school and suicidal thoughts indicate that attempts to diminish bullying in school environments may help reduce multiple negative outcomes for school performance and mental health overall (Nansel et al., 2001, 2003; Hinduja & Patchin, 2008). At the same time, there is a need to identify strategic interventions that extend beyond school boundaries due to the influence of cyber and mobile phone bullying on both truancy and suicide (Hinduja & Patchin, 2008; Marcum, 2010). There are, however, many challenges to such strategies due to questions over the responsibility of school administrators to manage youth behaviors outside of campus settings (Marcum, 2010).
To that end, parents must play a pivotal role in managing their child’s use of technology (Hinduja & Patchin, 2008; Marcum, 2010). Careful supervision of youth activity while online, including the use of filtering software, can help reduce the likelihood that their child is targeted by bullies via the web. Similarly, mobile phone management must be encouraged and enforced despite the portability and lack of interception capabilities afforded by these devices. Additionally, there is some evidence that children are less likely to report bullying victimization via mobile devices for fear that they may lose the phone altogether (Hinduja & Patchin, 2008; Marcum, 2010). Thus, parents must carefully educate their children on the risk of bullying victimization via mobile phones and ensure that they can speak to one or both parents about negative experiences via text to increase the likelihood of reporting. Finally, parents must pay attention to traditional and emotional warning signs of victimization, such as changes in affect or mood, sadness, and a lack of appetite which can indicate a child may be experiencing bullying online or off-line (Hinduja & Patchin, 2008; Nansel et al., 2001, 2003; Ybarra & Mitchell, 2004).
There are, however, several limitations of this study that must be addressed. First, the size of the missing data in these analyses, coupled with the use of cross sectional data, makes it difficult to identify the causal order of the relationships identified in this study. In particular, the relationships identified between school attachment, bullying victimization, truancy, and suicidal ideation must be further considered using longitudinal data in order to disentangle these effects over time (Esbensen & Carson, 2009). Second, this study assessed the impact of experiencing bullying victimization online and off-line rather than the severity of bullying victimization experienced in each environment. Given the variation in bullying that can occur, this does not allow for an understanding of distinctions in the serious nature of bullying victimization. For instance, it is not clear whether being physically hit or kicked may have a greater impact on youth experiences than repeated text-based harassment and teasing via a chat room or text message. Further research is needed to assess how the relationships identified in this study may be related to the bullying experiences reported.
Finally, the emergence of smart phone technology enabling individuals to access the Internet through their phone blurs the capacity to distinguish between mobile phone and cyberspace-specific bullying victimization as captured in this study (Smith et al., 2008). Additional research is needed to understand variations in smart phone and traditional cell phone ownership among youth and their relationship to cyberbullying victimization and negative outcomes. Such examinations can improve our general understanding of the impact of bullying victimization for youth around the world.
Footnotes
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.