An analysis of multiple factors of cyberbullying among junior high school
students in Taiwan
Yun-yin Huang
*, Chien Chou
Institute of Education, National Chiao Tung University, 1001 Ta-Hsueh Rd., Hsinchu 30010, Taiwan, ROC
a r t i c l e
i n f o
Article history:
Available online 17 July 2010 Keywords: Cyberbullying Technology Anonymity Academic achievement Peer support Instant messenger
a b s t r a c t
Cyberbullying, as a serious kind of repeated, intentional, and harmful aggressive behavior, cannot be ignored. In light of the limited studies and inconsistent findings on the matter, this study explores cyber-bullying’s frequency and other factors (gender, academic achievement, types of technologies used, and anonymity) relevant to both the issue itself and the East Asian context. The interrelationship of different roles (bullies, victims, and bystanders) in cyberbullying is also examined. A survey was conducted with 545 Taiwan junior high school students. The results indicate that male students were more likely to bully others in cyberspace and that cyberbullying was not affected by one’s level of academic achievement. Regarding the various technologies and various country-specific cyberbullying forms pertinent to tech-nology users, instant messenger (IM) users experienced significantly more cyberbullying than users of other technologies. The survey results also indicate that the anonymity of cyberbullying was not a per-tinent factor. The study found that the dominant attitude toward cyberbullying was indifference, raising alarms about the lack of cyberbullying prevention. Peers, who were the people most teenagers would likely turn to when experiencing cyberbullying, usually took no action because of their tendency to avoid conflicts and to maintain group harmony. In its interpretation of the findings, this study emphasizes Tai-wan’s context, including Confucian philosophy.
Ó 2010 Elsevier Ltd. All rights reserved.
1. Introduction
Bullying, as a serious kind of school violence, has long been recognized as a common aggressive behavior among peers and has negative effects on mental development and learning ( Flana-gan, Erath, & Bierman, 2008). The problem of bullying in middle school is serious and cross-cultural, and it has attracted consider-able attention in Europe, North America, and Japan (e.g., Akiba, 2004; Olweus, 2003; Pepler, Jiang, Craig, & Connolly, 2008). Regardless of whether the participants are victims, bullies, or wit-nesses, experiencing bullying can increase the possibility of other victimization, including child maltreatment, conventional crime, and psychological problems (Holt, Finkelhor, & Kantor, 2007a, 2007b). Bullying behavior is now happening in cyberspace and in an even more powerful way than has been the case in conven-tional contexts, because cyberspace is quicker, more comprehen-sive, and almost unstoppable and unavoidable. The physical scars of a beating can heal, and it is often possible for the would-be victim of such a beating to run away; stalked by someone online,
even the strongest mind can break and there is no place to hide.
Young people are socially connected with others through the Internet and other communication technologies, and these tools have become the new medium of bullying behaviors. Cyberbullying hurts teenagers emotionally, rather than result directly in physical damage, and operates by means of cell-phone text messages, photos posted online, mean words on personal blogs, and rumors that spread faster than ever through e-mail, instant messengers (IMs), or any other such communication devices. With the growing popu-larity of social-networking sites, instant messengers, and mobile technology among adolescents, the risk and extent of cyberbullying cannot be underestimated (Juvonen & Gross, 2008).
Owing to the limited number of cyberbullying studies in Taiwan (e.g.,Hokoda, Lu, & Angeles, 2006; Wei, Jonson-Reid, & Tsao, 2007), to the studies’ omission of pertinent issues, and to the studies’ inconsistent results, this study explores the cyberbullying prob-lems among Taiwanese teenagers and examines the frequency, types of tools, gender differences, and other factors relevant to both the issue itself and the East Asian context. It is hoped that the results illustrate how the new form of bullying happens in the context of Taiwan. In addition to examining the prevalence of cyberbullying, the current study emphasizes the cultural differ-ences between Taiwan and Western countries.
0747-5632/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.chb.2010.06.005
* Corresponding author. Tel.: +886 3 5712121x31641; fax: +886 3 5738083. E-mail address:paranoidrocker@gmail.com(Y.-y. Huang).
Contents lists available atScienceDirect
Computers in Human Behavior
2. Literature review 2.1. Definition of cyberbullying
As a prominent researcher in bullying studies,Olweus (1993) explained that bullying occurs when children (1) say mean things about or make fun of another person, (2) ignore or exclude him or her from their group, (3) hit, kick, push, or physically restrain him or her, or (4) tell lies or spread false rumors or send mean notes and try to make other students dislike him or her.
Later,Olweus and Limber (1999)summarized notable features of bullying: ‘‘it is aggressive behavior or intentional ‘harm-doing’, which is carried out repeatedly and over time in an interpersonal relationship characterized by an imbalance of power” (p. 31). The definition covers four features of bullying behaviors: they are intentional, harmful, repeated, and imbalanced in a power rela-tionship. The definition helps to distinguish accidental and just-one-time events from bullying, a willful aggressive behavior.
Cyberbullying, the bullying behavior in cyberspace, is a new kind of bullying happening through Internet applications, cell phones, or any other information technology. According to Bullying Beyond the Schoolyard (Hinduja & Patchin, 2008), cyberbullying is ‘‘the intentional and repeated harm of others through the use of computers, cell phones, and other electronic devices” (p. 5). The aforementioned text points out that the difference between tradi-tional bullying and cyberbullying concerns basically the tools and methods used; the intentional, repeated, and harmful nature re-mains the same.
Indeed, the different tools and methods have been changing the face of bullying; the communication gadgets make cyberbullying generally indirect. Without physical presence, cyberbullying hap-pens in chiefly verbal and relational ways. Relevant behaviors in-clude harassment, denigration, impersonation, outing, exclusion, and cyber-stalking, which are all classified as cyberbullying ( Wil-lard, 2006). These new forms of bullying facilitate attacks on teen-agers not only during school time, but also after school and at home, outside the awareness of parents. We should never underes-timate the harmful effects of cyberbullying, especially on children and teenagers.
The tools that could be used in cyberbullying, meaning all the communication technologies available, bring bullying to a new space and transform bullying into a new form. The tools include cell phones and Internet tools (instant messengers, social-net-working sites, chat rooms, e-mails) and result in the two unique characteristics of cyberbullying, which are rapid dissemination and anonymity; the latter feature remains an issue for further discussion.
2.2. The cyberbullying characteristics of rapid dissemination and anonymity
Owing to the features of information technology, cruel words can spread very quickly through simple copy-and-paste or forward actions in e-mails and instant messages, and the harmful content could be text, photos, drawings, videos, audio, and any combina-tion of these multi-media forms. Once the materials are posted online, school teachers or parents can take notice and ask the authors to delete the original file, but the numerous copies else-where in cyberspace remain; and the victim’s fear or embarrass-ment persists. The maelstrom of cyberbullying can spread quickly and is almost unstoppable. This first unique characteristic of cyberbullying, rapid dissemination, has been confirmed by the vast majority of studies so far (Kowalski, Limber, & Agatston, 2008; Li, 2005).
The second characteristic, anonymity, remains relatively incon-sistent. Different from face-to-face traditional bullies, cyberbullies
usually can easily remain anonymous in cyberspace (Li, 2007b, 2005). The bullies theoretically can hide behind computer screens and keep themselves safe from being recognized and punished. Teenagers have access to free e-mail accounts and other web-based applications and thus have the power to threaten, tease, and spread rumors, without implicating themselves. Physically stronger teenagers, usually male students, no longer hold advanta-ges in this form of bullying. Traditional victims of bullying, who probably are weak and shy, can fight back, or even take revenge in cyberspace. The anonymous nature of cyberbullying may re-verse the bully-victim relationship, or reinforce the role-turning cycle.
Yet, the results of an anonymous Internet-based survey of 1454 teenagers (Juvonen & Gross, 2008) are inconsistent with the com-mon assumption that cyberbullies are usually anonymous, sug-gesting that teenagers are aware of cyberbullying behavior and know exactly who did what. Anonymity is one good cover in cyber-space, but it cannot be inferred that every cyberbully would use it at all or would use it successfully. It is also possible that teenagers simply transfer the bullying field from the school bathroom or the school storage room to blogs and IMs and that teenagers do not tend to hide their identity, just as they typically do not wear masks when beating up a peer in a face-to-face confrontation. While cyberbullying accompanies traditional bullying, the victims might be able to guess the cyberbully’s identity. Interestingly, online materials quite obviously are more open to the public than are the bathrooms or the storage rooms in schools; but teen cyberbul-lies might not be aware of this fact. They do not expect that teach-ers, parents, or other adults would read their blog online or check their IM records, and most adults are indeed neither aware of such goings-on nor able to do anything about the matter if they are.
While it is assumed that the bullies would make the best use of a given technology’s features to remain anonymous (e.g.Li, 2005), there are other findings suggesting that cyberbullies are no differ-ent from the traditional face-to-face bullies who would not try to hide their identities (e.g. Juvonen & Gross, 2008). Owing to the inconsistent findings of the existing literature, the present study aims to explore whether anonymity is a unique feature of bullying in cyberspace.
2.3. Factors influencing cyberbullying
Previous research on school bullying has identified several fac-tors that likely contribute to cyberbullying. Among them are the significant factors of gender, academic achievement, and culture. In addition, research on cyberbullying has reported that com-puter-use frequency is a key factor (Li, 2005).
Gender has long been a significant factor influencing aggressive behavior, and it may result in different types of bullying among teenagers. Nabuzoka (2003) found that males usually were in-volved in physical and direct bullying (i.e., hitting someone) and that females were more involved in psychological bullying (e.g., ru-mor-spreading and relational aggression). Another study con-ducted in Taiwan found that boys and girls were equally likely to be victimized by indirect bullying (Hokoda, Lu, & Angeles, 2006). In cyberspace, where bullying has no physical form and no face-to-face contact, the gender difference might be changed. Li (2006) reported that there is no significant difference between genders, although males were still slightly more likely to cyberbul-ly than females in the Canadian case. There has been debate on which gender is more likely to engage in or to experience cyberbul-lying, and the findings so far are inconsistent.
Academic achievement is another key factor involved in bully-ing. Especially in East Asian countries (Lai, Ye, & Chang, 2008), which tend to be highly test-oriented comparatively, teenagers generally have suffered under intense academic pressure from
par-ents, teachers, and peers.Ma (2001)identified bullying connected to academic achievement: students with high academic achieve-ment, the so-called ‘‘good students,” usually were the targets of bullying behaviors. Focusing on Canada,Li (2007a)reported that half of cyber victims had above-average school grades and sug-gested that academic achievement might be a universal factor of cyberbullying. In a similar vein, being bullied could result in a drop in school grades and academic performance (Holt, Finkelhor, & Kantor, 2007a, 2007b). Academic achievers are likely targets of bul-lies, and the bullying can negatively affect the achievers’ academic achievement, suggesting that academic achievement is a possible factor acting diversely in different cultures.
Clear cultural difference was also found between English and Zambian pupils (Nabuzoka, 2003). While boys reported more bul-lying experience than girls in both of the cultural settings, it was indicated that English pupils exhibited proportionally more bully-ing-related behaviors than Zambian pupils. In addition to socio-economic diversity, the difference might result from various cul-tural definitions and understandings of bullying behaviors. Li (2005) pointed out that a Canadian student is four times more likely to cyberbully than a Chinese student and concluded that cul-ture was a significant predictor. Different patterns in China and Canada were found in another study byLi (2008), which highlights the importance of culture to the design of effective prevention pro-grams. While athletics often functions as a ticket to elevated social status in American schools (Bishop et al., 2004), the opposite might be true in Taiwan, where the culture is greatly influenced by Con-fucianism and where academic achievement garners the greatest value. A study showed that bullying of Taiwanese adolescents may be similar to Japan-based bullying (Ijime), owing to the shared norms of a collective cultural (Hokoda, Lu, & Angeles, 2006). In col-lective societies, witnesses of bullying may refrain from interven-ing for fear of upsettinterven-ing a sense of security and harmony, and victims might endure the sufferings to avoid conflicts with others. Another study of Ijime (Maeda, 1999) suggested that Japanese stu-dents were more likely than Western children to use indirect rela-tional tactics, such as spreading nasty rumors, and that the tendency was strongly associated with general social conformity. Social context and norms could account for significant differences in bullying behaviors.
In terms of cyberbullying, electronic devices constitute the medium of choice. Frequency of and knowledge of technology use can create circumstances where bullying or being bullied takes place. It was noticed that the widespread availability of cell phones may result in a high incidence of text-message bullying ( Raskaus-kas & Stoltz, 2007). It is reasonable to assume that the same pat-terns would surface in the use of other electronic devices. Ybarra and Mitchell’s study (2004) confirmed that both online aggressors and targets are intense Internet users. In addition to intense use, it is no surprise that networking sites and chat rooms, among various web tools providing frequent interaction among users, may be-come fertile ground for cyberbullying (Mesch, 2009). However, how these tools are used for cyberbullying and the reasons behind it remain unclear, along with the differences between social-net-working sites, chat rooms, and other communication tools (e-mail, blogs, instant messengers, etc.).
The types of tools, access to the tools, and the usage patterns (frequency and location) of communication technology are impor-tant indicators of cyberbullying behaviors and typically differ from the tools attributable to traditional school bullying. The present study tests and discusses these possible indicators.
2.4. The cyberbullying issues of interrelationship and peer support To date, few studies have examined the interrelationships among different roles of cyberbullying. As Olweus and Limber
(1999)noted, in traditional bullying, the victim can hardly defend himself or herself, owing to the physical, mental, or social imbal-ance of power. In contrast, the cyberbullying victim can compara-tively easily fight back with a computer or a cell phone and just a bit of ill intent. The imbalance of power structures in cyberbullying seems not as strong as it has been in traditional bullying. In cyber-space, everyone with Internet access and basic operation skills can easily spread rumors and send harassing e-mails and so can the victims. Even with less physical strength or relational resources, victims can more easily fight back or justify themselves through communication technology. The difficult part of revenge might be to identify the bully who hides behind the computer screen and who spreads rumors by anonymous e-mail accounts and cell phone numbers. Whether the power differential still exists in the cyberbullying realm is an issue requiring more empirical data.
The interwoven relationship among cyberbullies, victims, and the huge group of online bystanders has not yet been comprehen-sively discussed. In fact, a majority of teens are aware of or have witnessed bullying, and studies have pointed out that these teens, both individually and as a group, wield effective power to stop or to minimize bullying (Gini, Pozzo, Borghi,& Franzoni, 2008). Re-search has proposed a peer-support model to enhance bystanders’ responsibility to take action against bullying (Menesini, Codecasa, Benelli, & Cowie, 2003), and the model rests on the active interven-tion of groups of outsiders. The power of peer support might be re-garded as an effective factor in combating cyberbullying, as well as traditional bullying. Especially because a usually sizable popula-tion witnesses online bullying, these witnesses’ attitudes toward bullying and the power of their collective voice should not be underestimated.
Our review of studies on traditional bullying and on cyberbully-ing reveals that there are similarities and also differences between traditional bullying and cyberbullying. The repeated and inten-tional nature of bullying remains, but the power differential might change. The differences also lie in the tools used, the ways rumors spread, and the interrelationships among the involved roles. A un-ique characteristic of cyberbullying – anonymity – merits more investigation. Online bystanders’ actions could be a key to prevent-ing or minimizprevent-ing the phenomenon. Therefore, this study explores the following issues:
(1) Do gender and academic-achievement differences influence cyberbullying?
(2) Do different technologies result in any differences in cyberbullying?
(3) Is there a correlation among the experiences attributable to different roles (cyberbullies, victims, and bystanders)? (4) Could cyberbullying be characterized as anonymous
behavior?
(5) How do bystanders respond to cyberbullying?
The current study also explores two cyberbullying features, rep-etition and harmfulness, by asking about frequency and by posing open-ended questions. The focus of the survey is mainly the elec-tronic tools used in cyberbullying and their impact on the behavior. Thus, the remaining two features of cyberbullying, intention and power imbalance, do not receive full coverage in the survey insofar as their identification is difficult to attain in a self-report questionnaire.
3. Methodology
Self-report and peer-report surveys are the most frequently used methods of collecting data on school bullying (Wei & Huang, 2005); surveys are comparatively low-cost and are preferred for
large-scale data collection. Because surveys lack flexibility, six open-ended questions were added for more comprehensive feed-back in this study. Students, rather than teachers and other adults, were chosen to be the source because they are present in most bul-lying incidents (O’Connell, Pepler, & Craig, 1999). Whether being active participants or bystanders, peers are typically the best infor-mants on bullying events.
3.1. Instrument
Olweus (1993)in Bullying at School described eight roles in-volved in bullying behaviors: the bully, the follower, the passive supporter, the supporter, the onlooker, the possible defender, the defender, and the victim. The variety of and the interwoven rela-tionships among these roles reveal the complex nature of bullying behaviors. These roles may change places and overlap; that is, a victim might be a bully at the same time or change from one role to another. The current study has condensed the eight roles, for practical application, into the categories of bullies, victims, and bystanders.
This study’s anonymous survey comprises two major parts: one on personal information and the other on cyberbullying experi-ences. The first part asks about gender, grade level, academic achievement, and computer use, as reported above. The second part of the questionnaire was adapted from Kowalski and Limber’s survey (R. Kowalski, personal communication, December 18, 2008); the original questions were revised substantially for both content and language-use in this study to fit the Taiwanese con-text, and some original questions were deleted. A new section on bystander experience was added to investigate the perceptions attributable to the relatively sizable population of stakeholders. The newly added section aims to combine the advantages of self-report measures and peer-nomination measures. In every question, wording and phrases were changed to better fit the Taiwanese con-text; the survey’s language (the wording, including computer-re-lated phrases) was carefully chosen and reviewed by other stakeholders (two junior high school students and two junior high school teachers) so that the language would be familiar to teenagers.
The second part of the questionnaire includes 37 questions in total and was developed from three different roles’ perspectives of cyberbullying events: respectively, the questionnaire explores the experiences and attitudes of bystanders, victims, and bullies. The second part starts with the topic of bystanders’ experiences to avoid directly asking for bullying confessions, and proceeds to the topic of victims and then the topic of bullies.
The three sections in the second part of the survey are (1) know-ing/being aware of cyberbullying experiences (Cronbach’s
a
= .913), (2) victimization experiences (Cronbach’sa
= .904), and (3) bullying experiences (Cronbach’sa
= .958). Also, in the questionnaire’s sec-ond part, each section includes six open-ended questions asking about both the reasons for not reporting cyberbullying to adults and reactions to cyberbullying; the aim, herein, was to further ex-plore the mechanism underlying the aggression behaviors. 3.2. TerminologyThe translation of the English word ‘bullying’ has been debated, and the concept of bullying differs from culture to culture (Smith, Cowie, Olafsson, & Liefooghe, 2002); both the term ‘bullying’ and the term ‘cyberbullying’ are not so often used in Chinese as they are in English. The direct Chinese translation of the word ‘bullying’ (ba-lin) is considerably negative and is not often used in daily con-texts. Concerning the ambiguities of the concept among students, the term ba-lin was replaced by a longer elaboration of the meaning. In order to minimize the possibility of confusion, the beginning of
each section featured a textual paragraph both explaining the cyber-bullying concept and identifying examples to avoid possible misun-derstanding of bullying-related behaviors; the textual paragraph also serves as a reminder to participants that they should bear in mind the correct cyberbullying concept when answering the questions. The English translation of the paragraph is as follows:
You will be asked about experiences regarding ill-intended behaviors in cyberspace, and these behaviors include threats, harassment, humiliation, insults, and any other emotional put-downs by means of words, fake pictures, peeping-Tom videos, or any combination of digital content. In the following questions, the ‘‘online” environment suggests all kinds of elec-tronic communication tools, such as e-mails, instant messen-gers, chat rooms, online polls, web forums, weblogs, and cell-phone text messages. Please bear in mind that the following questions concern ill-intended behavior and content performed and transmitted online through any and all means. You will be asked about how often – and in what ways – you have come across the above behaviors or content. The behaviors that you mention are not necessarily bad behaviors. Please answer the questions honestly according to your actual experiences. 3.3. Sample
Data were collected from 16 classes of eight junior high schools in both northern and southern Taiwan, including urban and rural areas. Junior high school students were chosen (1) because they would be the most likely to suffer from both online and traditional bullying, (2) because teachers and parents would likely be unaware of cyberbullying, and (3) because teenagers would likely be reluc-tant to report the cyberbullying (Kowalski, Limber, & Agatston, 2008). Among the total of 545 participants, 228 students com-pleted the questionnaire online and the remaining 317 students completed the same questionnaire in print form. The data collected online and in print were assumed equal, because no significant dif-ference was found between the mean scores of the print survey and the mean scores of the online survey relative to each of the three groups (bystanders t(545)= . 1.334, victims t(466)= .810,
bul-lies t(540)= .686, respectively, p < .001).
3.4. Procedure
The anonymous survey was completed at school during school hours and was administered by the students’ teacher. The survey required 15–20 min for each student. Sixteen classes completed the survey, and there were about 35 students in each class. So that students would understand the survey, they were asked to pay attention to the instructions at the beginning of each section, explaining the concept of cyberbullying. Both the hardcopy survey and the online survey followed the same procedure.
3.5. Analysis
This study accounts for the demographics and cyberbullying experiences of participants. The majority of the outcomes were either single dichotomous (yes–no) or 5-point Likert-style ordinal variables, and before the analysis, some 5-point ordinal data regarding frequency were grouped into a ‘‘no cyberbullying expe-rience at all” category and an ‘‘at least one cyberbullying experi-ence” category (once, twice in a month, once a week, and several times a week). Effects of various factors (gender, culture, academic achievement, type of tool, etc.) and three roles in cyberbullying (bystander, victim, and bully) were also examined. We conducted a t-test, a bivariate-correlations test, and a multivariate repeated one-way ANOVA to examine differences among pertinent media/
tools and to explore discrepancies in cyberbullying events as per-ceived by different types of role-players. The SPSS 15 package (SPSS, 2006) was used for statistical analyses.
4. Results and discussion 4.1. Participants’ demographics
The total of 545 students were asked personal information regarding only their gender, grade level, academic achievement, and computer use; in this way, this study ensured the anonymity of the participants. As shown inTable 1, the participants comprised 256 (47.1%) males and 288 (52.9%) females and were fairly evenly distributed from grade to grade (7th graders 31.9%, 8th graders 37.1%, 9th graders 31.0%). Regarding academic work, 225 (41.6%) students reported theirs as above average, 210 (38.4%) reported theirs as average, and 109 (20.0%) reported theirs as below average. AsTable 1shows, the majority of students got online at home (86.5%), and some less common places were school (5.5%), Internet cafés (3.1%) and friends’ homes (2.8%). The Internet availability at home (91.5%) was almost universal. As for the frequency of using computers, almost one-third reported that they used computers everyday (27.6%) and another one-third several times a week (38.5%). Instant messengers (83.1%) and e-mails (66%) were the two most frequently used tools for online activities. Other popular Internet tools were various, including BBS (28.2%) and chat rooms (16.4%).
Our sample of students indicated that among their various pur-poses for using computers, the purpose of entertainment was the most common one (87%), including playing online games, watch-ing videos, and listenwatch-ing to music. The second most common pur-pose was to use computers for communication (76.7%), through instant messengers particularly; other purposes include academic work (67.9%) and self-expression (53.6%), suggesting the editing and the reviewing of personal profiles and blogs.
Also confirmed in the survey was that many of the teens were learning from the online resources, but that more of them were looking for fun, a sense of belonging, and interaction with others. The data again confirmed that the teens’ lives were now rooted in an online culture; they used the Internet for multiple purposes, in contrast to older computer users, who tended to use computers for specific tasks. Among all sampled students, 78% reported that they had been using the Internet for more than 4 years and 75% re-ported using the Internet at least 1 h a day, in addition to their 8-h or more school-based daytime studying. Teenagers take online communication, both verbal and textual, for granted as a daily communication medium. The phenomenon has contributed to teens’ mental dependence on communication technology. Being disconnected from the Internet signifies isolation, with no direct news from friends and no direct interaction with peers.
4.2. Cyberbullying experiences
As shown inTable 2, 346 (63.4%) of the students reported hav-ing witnessed or havhav-ing been aware of cyberbullyhav-ing, 190 (34.9%) had been cyberbullied, and 111 (20.4%) had cyberbullied others. The overall results present a picture wherein Taiwanese adoles-cents have frequent involvement in cyberbullying. The prevalence of the new bullying behaviors cannot be ignored.
Also shown inTable 2, from bystanders’ perspectives, there was generally an even distribution or frequency of different forms of bullying. Of the different forms of cyberbullying, the form of mak-ing jokes about/ makmak-ing fun of others was the most frequent (64.3%) and spreading rumors was less frequent (60.9%). Both vic-tims and bullies experienced being made fun of or being the butt of jokes quite frequently (32.3% and 18.2%, respectively). A possible explanation for the prevalence of this category in cyberspace is that making jokes and fun of others may not be considered partic-ularly harmful by people in general.
Victims’ most common type of cyberbullying experience was ‘‘being threatened,” and bullies’ most common type of cyberbully-ing experience was ‘‘to threaten or to harass.” This type of cyber-bullying might happen only between the victim and the bully, completely escaping the notice of other people. Compared to mak-ing fun of others, threatenmak-ing might generally be perceived as seri-ous and wrong behavior, and for this reason, the bullies would have issued the threats secretly. This type of cyberbullying might actually be the most frequent but appears to remain unnoticed by most stakeholders, including peers, teachers, and parents.
It is noteworthy that the bystander-related part of the survey generally presented the highest percentage of cyberbullying expe-riences, the victim-related part presented the second-highest percentage, and the bully-related part presented the lowest per-centage. According to the observed pattern, bystanders seem to have been the group most frequently involved in bullying epi-sodes; also shown in the survey results was that this very same group was the least likely to report a bullying incident in which they were not directly involved. Conforming to the nature of self-report surveys, the bullies themselves seemed the least willing to report their own behavior, particularly in comparison with the vic-tims, while the overall experience score of the bully-related part is generally higher than that of the victim-related part. Even though Table 1 Demographics of participants. n Percentage (%) Gender Male 256 47.1 Female 288 52.9 Grade level 7th grader 174 31.9 8th grader 202 37.1 9th grader 169 31.0
Internet availability at home
ADSL or cable connection 458 83.3
Dial-up connection 45 8.2 None 46 8.4 Academic achievement Above average 225 41.6 Average 210 38.4 Below average 109 20.0
Most frequent computer-use location
Home 472 86.5 School 31 5.5 Internet café 18 3.1 Friend’s home 16 2.8 Other 8 1.5 Computer-use frequency Everyday 152 28.6
Several times a week 210 39.5
Once a week (and less) 170 31.9
Purpose of using computer (multiple answers)
For entertainment 474 87.0
For communication 418 76.7
For academic work 370 67.9
For self-expression 292 53.6
Table 2
Bullying method witnessed/suffered/used by different role-players.
Form Bystander Victim Bully
n (%)
To threaten or harass 346 (63.5) 190 (34.9) 111 (20.4) To make jokes about/fun of 350 (64.3) 176 (32.3) 99 (18.2) To spread rumors 332 (60.9) 137 (25.2) 66 (12.2)
the current study has preserved the anonymity of the participants, the participants who were more involved in bullying were less willing to discuss the bullying. Another possible explanation of the least-frequent reporting of cyberbullying is that the bully group was indeed the smallest group, much smaller than the by-stander group and the victim group. A cyberbullying event may, rather than exhibit a one-bully-to-one-victim pattern, rest on a multiple-victim combination.
4.3. Gender difference
As shown inTable 3, the results of this study reveal that male students have a mean score of 1.91 on a 5-point Likert scale for each questionnaire item on bystander experiences, 1.41 on victim experiences, and 1.33 on bullying experiences. Female students, compared to male students, have a lower mean score of 1.82 on by-stander experiences, 1.29 on victim experiences, and 1.11 on bully-ing experiences. Male students generally have higher scores on all three kinds of cyberbullying experiences. The t-test revealed signif-icant gender differences regarding both victim score and bullying score. In spite of the anonymous and indirect nature of cyberbully-ing, male-student bullying outdid female-student bullying in terms of both victimization and bullying experiences.
The results are consistent with results from previous studies on traditional face-to-face bullying (e.g., direct physical and direct verbal abuse), but not with results from earlier studies on cyber-bullying. Previous studies have not reached agreements on the gender difference in bullying; some found that boys were more likely to bully and to be involved in direct bullying (Hokoda, Lu, & Angeles, 2006; Nabuzoka, 2003) and some found that there was no gender difference (Nabuzoka, 2003). However, it is gener-ally believed that females are less likely to bully than their male peers and it is also suggested that females prefer to apply indirect or relational bullying tactics, such as acts of isolating and acts of ignoring (Crick, Grotpeter, & Bigbee, 2002). Communication medi-ums thus could be a less direct way of bullying and could lessen or erase or even reverse the gender difference in cyberbullying.
In this study, male students reported greater levels of both bul-lying and victimization experiences than females, and the result was aligned with previous bullying studies (Olweus, 2003). Although cyberbullying usually is more indirect and relational than traditional bullying and would, in theory, be preferable to girls, our cyberbullying-survey results show no difference between bullying with communication technology and bullying without communi-cation technology. The trend of less-direct bullying behavior in girls may be obvious in relation to use of communication tools, but there is a possibility that boys use computers more often and are better at operation skills than is the case with girls. As a possi-ble factor influencing the inconsistency of gender difference in cyberbullying, technology-use background receives further discus-sion in a later section of this article.
4.4. Academic-achievement difference
In this study, a one-way ANOVA was conducted with the expe-rience scores of the three cyberbullying-related roles (i.e., of
bystanders, victims, and bullies) serving as dependent variables for examining the differences in the three groups of academic achievement (above average, average, and below average). A test of the homogeneity of variances was conducted before analysis. AsTable 4indicates, all three academic-achievement groups had no significant difference in their cyberbullying experiences.
These results were not consistent with those of Ma’s study (2001), which suggested that students who perform better in aca-demics are, in general, the target of bullying. In contrast, all three academic-achievement groups from the current study revealed no difference among one another regarding cyberbullying experi-ence, irrespective of one’s role as a bystander, victim, or bully. The possible reason is that Taiwan society is more test-oriented than the United States or Canada owing to Taiwan’s deep-rooted Confucianism (Hokoda, Lu, & Angeles, 2006). In Taiwan, 7–9th graders usually study 8 h or more a day in school in order to pass entrance exams for high schools. Studying and academic perfor-mance play a huge part in teenagers’ lives and constitute more of an ultimate goal pertaining to three-year school life than an emo-tional commitment or or a personal marker. The current study’s re-sults indicate that academic performance affects neither daily relationships nor the harmony and balance of the class nor a cer-tain group in regards to their cyberbullying experiences. Therefore, there should be relevant factors influencing cyberbullying in addi-tion to the relevant factor of academic achievement in the Taiwan context. However, in this study, academic-achievement level was self-reported (41.6% above average, 38.4% average, and 20.0% be-low average), and the distribution was not totally even. Therefore, the results of our study should be interpreted conservatively, and definitely more research is needed to explore the relationships be-tween academic achievements and cyberbullying experiences. 4.5. Technology-use difference
Do these different technologies make any difference regarding cyberbullying experience? In order to investigate the differences among technologies used in cyberbullying, we conducted a MANO-VA. The results show that Wilks’ lambda was significant on all by-stander (F = 17.96, p < .001), victim (F = 8.37, p < .001), and bully (F = 3.98, p < .001) experiences. AsTable 5shows, for bystanders, a post hoc test further revealed that the mean score of using IM was greater than the mean scores of the other four tools, and that the mean score of using websites and BBSs was greater than the mean scores of using chat rooms, e-mails, and cell phones. The same type of analysis was conducted on victims and bullying ( Ta-ble 5). IM was the dominant tool in all three kinds of cyberbullying experiences. Websites and BBSs were the second most common environment where one would witness cyberbullying. For victims, the most likely bullying tool was chat rooms, the second most likely tool being IM. Also, the results on bullies reveal that both of the synchronous environments, IM and chat rooms, were the two most frequently used spaces in which bullies cyberbullied others.
While social-networking sites and chat rooms have served as fertile ground for cyberbullying in the United States (Mesch, 2009), the current study found that IM was the space in which Table 3
Mean scores on bystander, victim, and bullying experiences.
Male Female T Mean SD Mean SD Bystander 1.91 8.62 1.82 7.66 1.40 Victim 1.41 7.33 1.29 4.79 2.46** Bully 1.33 7.62 1.11 3.65 4.10** **p < .01. Table 4
Cyberbullying-experience scores of the three academic-achievement groups. Above average Average Below average F
Mean SD Mean SD Mean SD
Bystander 1.85 7.50 1.86 8.31 1.88 7.10 .025 Victim 1.33 5.86 1.36 6.45 1.34 6.18 .121
teenagers in Taiwan were most likely to experience cyberbullying. Among Taiwanese teenagers, social-networking sites, such as Face-book and MySpace, are far less popular than the five technologies investigated in the present study. The cases of social-networking sites cannot be put on par with one another in different contexts (e.g., Taiwan and the United States). It is interesting to note that there are basic differences between IM operations and chat-room operations. Chat rooms, asMesch (2009)states, provide potential contact with motivated cyberbullies who might be total strangers to the victim. In contrast, IM requires log-in identification, and par-ticipants in an IM conversation must be users who have, first, mutually approved of both the conversation and its participants. Our finding that IM in Taiwan is the dominant tool of cyberbullying conforms with another finding of our study: that cyberbullying vic-tims and also bystanders can identify the bully, meaning that the anonymity of cyberbullying does not always exist (for more details on this matter, see Section 3.7).
The current study notes that, among participants, cell phone messages constituted the least common tool of cyberbullying. All three groups of roles in cyberbullying reported significantly lower scores when using cell phones. The present study found that the majority of cyberbullying happened online through personal com-puters, which was a finding whose subject matter differs strikingly from Japan-based Ijime (Akiba, 2004; Rios-Ellis, Bellamy, & Shoji, 2000), whose chief characteristics are the rapid spread of rumors and the extensive use of cell phones. There were also quite a few cell phone users among Taiwanese teenagers participating in the present study but they seemed not to use the tool for bullying. Although Taiwan and Japan have similar collectivistic cultures, and although bullying in Taiwan may be similar to Ijime in Japan (Hokoda, Lu, & Angeles, 2006), the related technology-use habits in Taiwan seem to differ considerably from those in Japan.
4.6. Interrelationships in roles of cyberbullying experiences
The experience scores of each role were calculated, and bivari-ate correlations were examined. The results show that all three groups are correlated with each other, while the correlation be-tween the bully group and the victim group is highly significant (Pearson’s r = .720, p < .01). Bystanders were more likely to be vic-tims (Pearson’s r = .660, p < .01) and to be bullies (Pearson’s r = .456, p < .01).
The high correlation between victims and bullies in our results indicates that there may exist a bully-victim phenomenon in
cyberspace. Bully-victims, who bully others and also are bullied, represent a highly intricate relationship in cyberbullying. In tradi-tional bullying, the victims hardly can fight back because of an imbalanced physical-power relationship between victims and bul-lies. Cyberspace changes the rules, so that everyone with an e-mail account or other such tools can spread rumors quickly and harass others by text messages, provocative images, and the like. The imbalanced power relationship is no longer a matter of course; therefore, there could be more bully-victims in cyberspace, and the role-turning cycle gains strength.
The roles involved in cyberbullying seem more complicated than the roles involved in traditional bullying, which are not lim-ited to bystanders, victims, and bullies (see alsoOlweus & Limber, 1999). Owing to the nature of communication technology, bullies can easily hide behind the computer screen and there could be many accessories or followers who forward the cruel words and embarrassing pictures that typically characterize cyberbullying. In this case, bystanders who forward the ill-intentioned messages to others can be considered members of the bully group. Similarly, bystanders who receive the ill-intentioned messages about friends may consider themselves members of the victim group. That is pos-sibly why these three roles are highly related to each other in this study. It is also very possible that the original message provider does not intend to harm, but that anyone in cyberspace might viciously alter or widely disseminate the out-in-the-open message, thereby creating a vicious cycle that does not rest on an initial act of ill will. In contrast, it is also very possible that anyone in cyberspace might put either a permanent stop or a temporary stop to vicious out-in-the-open messages that do, indeed, rest on an initial act of ill will. Therefore, the ability of the many bystanders viewing ill-inten-tioned materials to take positive actions to stop the spread of the materials could be a powerful key to cyberbullying prevention.
4.7. Anonymity of cyberbullying
The findings of this study suggest that most of the teenage par-ticipants knew the given bully’s identity and that bullies in cyber-space did not try to disguise themselves by using the Internet as a cover. As Table 6 indicates, among those who were bullied (n = 281), only 25.1% of participants had no idea who had bullied them. Among bystanders of a cyberbullying event (stating that they had witnessed cyberbullying), 43.1% were unaware of the gi-ven bully’s identity. More than a half of each of the two groups claimed that they were aware of the given bully’s identity. Table 5
Results of the MANOVA and the post hoc test of bystander, victim, and bully experience scores.
Role tool Mean SD F value Sorting of different paired samples in the post hoc test
Bystander
IM 1.90 1.128
Chat room 1.79 1.124 17.96** IM > chat room > website & BBS > e-mail > cell phone
Website & BBS 1.87 1.067
E-mail 1.80 1.083
Cell phone 1.60 .920
Victim
IM 1.42 .858 8.37** IM > e-mail > chat room > cell phone > website & BBS
Chat room 1.31 .792
Website & BBS 1.25 .673
E-mail 1.32 .808
Cell phone 1.25 .699
Bully
IM 1.23 .701 3.98** IM > chat room > website & BBS > cell phone > e-mail
Chat room 1.22 .816
Website & BBS 1.19 .637
E-mail 1.16 .615
Cell phone 1.17 .667
The current study’s findings are inconsistent with (Li, 2007a, 2007b) identifying anonymity as a characteristic of cyberbullying. On the other hand, the current study’s findings confirmed the find-ings ofJuvonen and Gross (2008), showing that more than two-thirds of victims knew or at least could suspect who had bullied them in cyberspace. A possible explanation of this last finding con-cerns the prominent IM usage, where account names identify for users the correspondent with whom they are communicating; thus, less impersonation happens. Another possible reason is that cyberbullying was not independent of face-to-face bullying and, more generally, of daily school life. Cyberbullying might happen alongside traditional bullying, and the victim would consequently know the identity of the bullies. Victimized students could clearly make connections between people who would perpetrate an act of bullying to the victims’ face and on-line bullies who would perpe-trate an act of similar content but of different form.
Therefore, the assertion that anonymity is a universal character-istic of cyberbullying should remain in doubt. The topic may reflect influences from technology used (e.g., with or without log-in iden-tification, synchronous or asynchronous) and social factors (e.g., the classroom climate as either actively anti-cyberbullying or not, the classroom climate as exhibiting more individualism than collectivism or vice versa, the teacher’s attitude toward and re-sponse to cyberbullying).
4.8. Response to cyberbullying
This study confirms the findings of Li’s study (2007b) that teen-agers are reluctant to report cyberbullying. In each section of the current study’s survey, students were asked to report their re-sponse to cyberbullying in multiple-answer questions and open-ended questions. If they had had no previous experience as a by-stander, as a victim, or as a bully in such an event, then they might have skipped these questions. When asked about the person to whom they would talk about or report the event, 545 respondents yielded 200 answer counts in the as-a-bystander section, 342 in the as-a-victim section, and 321 in the as-a-bully section. Bystand-ers, as the largest group involved in cyberbullying, actually were the least likely to take action. A prevalent idea was that people considered the act of reporting neither their business nor their responsibility and that the cyberbullying itself was ‘‘no big deal.” In short, bystanders did not feel responsible for reporting to adults or taking action against cyberbullying events. Moreover, bystand-ers radically suggested that these events constitute othbystand-ers’ privacy in which the bystanders should not get involved. A few argued that they did not report the events because the people involved were not their friends. AsTable 7shows, among those 200 who did re-port as bystanders of cyberbullying, only 11.2% told parents and 3.7% told teachers. Reasons for not reporting either at all or to adults include (1) being afraid of getting into trouble and (2)
feel-ing a sense of uselessness in lookfeel-ing to adults for assistance. Fear of getting into trouble might refer to being threatened by the bully, being regarded as an informer or a gossip, or being excluded from an ‘‘in” group. The clear and common attitude of apathy and indif-ference should be a worry for prevention efforts.
In this study’s survey, many more victims than bystanders re-ported cyberbullying events. According to the multiple-choice and open-ended questions (Table 7), peers, especially classmates (33.4%) who knew the involved people and the corresponding school’s context, and siblings (16.1%) were the participants to whom victims would most likely turn. This finding is consistent with previous research (Gini, Pozzoli, Borghi, & Franzoni, 2008; Unnever & Cornell, 2004). Also asTable 7 indicates, the partici-pants in the current study were clearly less likely to turn to adults, including parents (11.6%) and teachers (5.9%). According to the rea-sons provided in the open-ended part, many victims told no one because they wanted to avoid both triggering parental concern and appearing to be the ‘‘the weak one.” The lowest percentage of reporting to teachers, combined with the open-ended questions’ results regarding reasons, shows that students possibly did not be-lieve that teachers would handle the matter effectively and that reporting an event might be not only useless but indeed counter-productive. As for bullies, the pattern was the same but all percent-ages were the lowest of the bunch, which is reasonable insofar as the bullies were typically unwilling to admit their own behavior to people.
The current study has noted that, in comparison with the indi-vidualism in many Western countries, Taiwan and Japan exhibit collectivism, where people would put the group’s well-being ahead of individuals’ well-being (Bond & Hwang, 1986). The social con-formity might, in the context of these two societies, contribute to the indifference among bystanders of cyberbullying. Teenagers are taught to obey orders and to follow rules, and social norms can contribute to the teens’ passive responses. Partly in line with Japan’s Ijime, Taiwanese teenagers responding to this study’s open-ended questions did not report that they were afraid of becoming the next victim if they took some action against the bul-lying. On the whole, they neither justified cyberbullying nor tried to stamp it out.
Classmates were the students to whom teenagers would like to talk about cyberbullying experiences, regardless of whether they (the respondents) were involved in the event as bystanders, vic-tims, or bullies. The results herein provide a direction for cyberbul-lying prevention, which should start with and focus on schoolmates. Education on responsible reactions by teenagers is imperative, for they would always be the first to know about and to suffer from cyberbullying before teachers, parents, or any other stakeholder enters the picture.
5. Implications
Cyberbullying, a mixture of traditional bullying behaviors and modern technology, is an offshoot of advances in technology. As discussed, gender, academic achievement, and the use of technol-ogy might contribute to cyberbullying, but these factors are not as significant in this study as they are in other contexts. For exam-ple, academic achievement in Taiwan did not reflect a significant Table 6
Awareness of a bully’s identity.
Number of answers Unaware (%) Aware (%)
Bystander 459 198 (43.1) 261 (57.9)
Victim 281 70 (25.1) 211 (74.9)
Table 7
People who reported cyberbullying (by role).
Number of answers Participants to report, as determined from the total number of answers (%)
None Sibling Classmate Net-pal Parent Teacher
Bystander 200 22.9 15.8 36.7 9.7 11.2 3.7
Victim 342 21.6 16.1 33.4 11.4 11.6 5.9
difference in cyberbullying experiences as it did in a Canadian case (Ma, 2001). Technology with the greatest effect on cyberbullying is instant messaging programs (e.g., MSN), rather than the popular social-networking sites (e.g., Facebook) in the United States and Canada (Kowalski et al., 2008). Insofar as Taiwan has an enormous population of Internet users, and insofar as many Taiwanese teen-agers (76.7%) go online for communicative purposes, it is reason-able to conclude that bullying behavior characterizes cyberspace experiences in Taiwan. Furthermore, it appears that the methods and the tools accompanying this new form of bullying are varied. Cyberbullying behavior might be highly context-dependent and influenced by educational systems, school climates, and cultural norms, which differ from country to country. Although the defini-tion of cyberbullying refers to certain standard themes such as intention, repetition, and harm, the factors that influence the cyberbullying seem different across cultures.
The results of this study indicate that the attitude of Taiwanese teenagers toward cyberbullying is generally one of indifference. Bystanders’ responses could be a key to preventing the spread of cyberbullying. As the major group involved in bullying behavior in cyberspace, everyone with Internet access could become a mem-ber of the bystander group. In this regard, the questions of critical importance are how to transform bystanders’ general indifference to active and positive reactions, and how to equip bystanders with power and tools that facilitate such reactions. In the United States, there are specific laws addressing bullying through electronic com-munication, but they vary from state to state (Kowalski et al., 2008); however, because Chinese culture values peace and har-mony and the avoidance of conflict in daily life, seldom would Chi-nese display their personal disagreement and caring in public. Students’ answers to our open-ended questions about why they did not report cases of cyberbullying point to a strong tendency to avoid personal and group conflicts, a tendency that might be a formidable obstacle to cyberbullying prevention in Taiwan.
Being in a highly IT-rich environment, most Taiwanese teenag-ers nowadays are equipped with cell phones and get online every-day. Compared with many other countries, Taiwan is unique in both the electronic communication tools that its teenagers use and the communication-related attitudes that its teenagers hold. It seems that Taiwan’s most popular electronic communication tool is neither the cell phone (which dominates in Japan), nor so-cial-networking sites (which are increasingly a ‘‘cultural require-ment” among U.S. high school students) (Kowalski et al., 2008). In Taiwan, the most popular tool is instant messengers (IMs), such as MSN, Yahoo Messenger, and Skype. The trend might serve as a unique fertile ground for cyberbullying.
Cyberbullying is serious but has gone relatively unnoticed so far in Taiwan, partly owing to the novel nature of the phenomenon and to most teachers’ and parents’ lack of familiarity with the tech-nology and the media (Dehue, Bolman, & Vollink, 2008). Adults’ unawareness of cyberbullying reflects the relative absence of adult supervision in cyberspace. And yet, the majority of teenagers use the Internet at home, where parental mediation has proven to be effective for cyberbullying prevention (Mesch, 2009). Middle school teachers, parents, and policymakers should all be concerned with the issue and facilitate prevention in the near future. While technology use differs from country to country, more empirical studies and detailed examinations of factors influencing cyberbul-lying in East Asian contexts are needed for development of a rigor-ous and more effective prevention program. The findings of this study constitute an important step in addressing these issues. 5.1. Limitations and recommendations for future studies
We acknowledge limitations of this study and hope to provide avenues for further exploration and research. First, this study
adopted the self-report method, that is, participants were asked to answer the survey concerning their perception (knowledge) of cyberbullying. Though relevant terms were defined and relevant behaviors were explained throughout the survey, participants’ judgments were, in all likelihood, unavoidably subjective to a cer-tain degree. Future research should use other methods, such as those involving focus groups, in-depth interviews, and observa-tions, to triangulate and verify the findings. Second, owing to its limited resources, the current study collected research data only from students. Future studies can collect data not only from stu-dents but also parents, teachers, school staff, and any other stake-holders. Various types and sources of data would be helpful in gaining a more comprehensive view on the phenomenon.
Third, only correlations of possible factors about cyberbullying were identified in the present study. However, causal relationships might also exist among these factors. Other statistical analyses (e.g., multiple regression), if applied to this topic, could shed more light on the bullying behaviors taking place in cyberspace. Fourth, though perhaps not yet included in existing literature, other possi-ble factors, such as school climate and the psychological conditions of the three different cyberbullying roles (bystander, victim, and bully), should receive serious attention in future studies.
Appendix A. The first section of the cyberbullying survey
1. Have you ever been aware of classmates being harassed, hurt emotionally, or threatened online?
2. Have you ever been aware of classmates being made fun of or humiliated online?
3. Have you ever been aware of classmates being slandered or defamed online?
4. Have you ever been aware of others’ malicious Internet-based behaviors or others’ fake e-mail/IM accounts? 5. Have you ever been aware of classmates being hurt
emo-tionally on websites (e.g. web forums)?
6. Have you ever been aware of classmates being hurt emo-tionally in chat rooms?
7. Have you ever been aware of classmates being hurt emo-tionally through e-mails?
8. Have you ever been aware of classmates being hurt emo-tionally through instant messengers (e.g., Skype, Yahoo Mes-senger, or MSN Messenger)?
9. Have you ever been aware of classmates being hurt emo-tionally through cell-phone text messages?
10. Have you ever been aware of classmates being hurt emo-tionally through any other tools online? If you have, please specify: _____________.
11. If you have ever seen or been aware of somebody being threatened, harassed, or humiliated online, what type of per-son was this victim? Siblings, classmates, netfriends, strang-ers, cannot identify, other ________ (please specify) 12. To whom have you ever reported online bullying incidents?
Siblings, classmates, netfriends, strangers, school teachers or staff, parents, other (please specify)
13. If you have told no one, why? And did you take any other action?
References
Akiba, M. (2004). Nature and correlates of Ijime: Bullying in Japanese middle school. International Journal of Educational Research, 41(3), 216–236.
Bishop, J. H., Bishop, M., Bishop, M., Gelbwasser, L., Green, S., Peterson, E., et al. (2004). Why we harass nerds and freaks: A formal theory of student culture and norms. Journal of School Health, 74(7), 235–251.
Bond, M. H., & Hwang, K. K. (1986). The social psychology of Chinese people. In M. H. Bond (Ed.), The psychology of Chinese people. New York: Oxford University Press.
Crick, N. R., Grotpeter, J. K., & Bigbee, M. A. (2002). Relationally and physically aggressive children’s intent attributions and feelings of distress for relational and instrumental peer provocations. Child Development, 73, 1134–1142. Dehue, F., Bolman, C., & Vollink, T. (2008). Cyberbullying: Youngsters’ experiences
and parental perception. CyberPsychology & Behavior, 11(2), 217–223. Flanagan, K. S., Erath, S. A., & Bierman, K. L. (2008). Unique associations between
peer relations and social anxiety in early adolescence. Journal of Clinical Child and Adolescent Psychology, 37(4), 759–769.
Gini, G., Pozzoli, T., Borghi, F., & Franzoni, L. (2008). The role of bystanders in students’ perception of bullying and sense of safety. Journal of School Psychology, 46(6), 617–638.
Hinduja, S., & Patchin, J. W. (2008). Bullying beyond the schoolyard: Preventing and responding to cyberbullying. Thousand Oaks CA:Corwin Press.
Hokoda, A., Lu, H.-H. A., & Angeles, M. (2006). School bullying in Taiwanese adolescents. Journal of Emotional Abuse, 6(4), 69–90.
Holt, M. K., Finkelhor, D., & Kantor, G. D. (2007a). Multiple victimization experiences of urban elementary school students: Associations with psychosocial functioning and academic performance. Child Abuse and Neglect, 31, 503–515.
Holt, M. K., Finkelhor, D., & Kantor, G. K. (2007b). Hidden forms of victimization in elementary students involved in bullying. School Psychology Review, 36(3), 345–360.
Juvonen, J., & Gross, E. F. (2008). Extending the school grounds? Bullying experiences in cyberspace. Journal of School Health, 78(9), 46–54.
Kowalski, R. M., Limber, S. P., & Agatston, P. W. (2008). Cyber Bullying. Malden: Blackwell Publishing.
Lai, S.-L., Ye, R., & Chang, K.-P. (2008). Bullying in middle school: An Asian-Pacific regional study. Asia Pacific Education Review, 9(4), 393–405.
Li, Q (2005). Cyber-bullying in schools: Nature and extent of adolescents’ experience. In Paper presented at the annual American Educational Research Association (AERA) conference, Montreal.
Li, Q. (2006). Cyberbullying in schools: A research of gender differences. School Psychology International, 27(2), 157–170.
Li, Q. (2007a). Bullying in the new playground: Research into cyberbullying and cyber victimisation. [Electronic version]. Australasian Journal of Educational Technology, 23(4), 435–454.
Li, Q. (2007b). New bottle but old wine: A research of cyberbullying in schools. Computers in Human Behavior, 23, 1777–1800.
Li, Q. (2008). A cross-cultural comparison of adolescents’ experience related to cyberbullying. Educational Research, 50(3), 223–234.
Ma, X. (2001). Bullying and being bullied: To what extent are bullies also victims? American Educational Research Journal, 38(2), 351–370.
Maeda, R. (1999). Ijime: An exploratory study of a collective form of bullying among Japanese students. (ERIC Document Reproduction Service No. ED438015) Retrieved August 19, 2009, from ERIC database.
Menesini, E., Codecasa, E., Benelli, B., & Cowie, H. (2003). Enhancing children’s responsibility to take action against bullying: Evaluation of a befriending intervention in Italian middle schools. Aggressive Behavior, 29(1), 1–14. Mesch, G. S. (2009). Parental mediation, online activities, and cyberbullying.
CyberPsychology & Behavior, 12(4), 387–393.
Nabuzoka, D. (2003). Experiences of bullying-related behaviours by English and Zambian pupils: A comparative study. Educational Research, 45(1), 95–109. O’Connell, P., Pepler, D., & Craig, W. (1999). Peer involvement in bullying: Insights
and challenges for intervention. Journal of Adolescence, 22(4), 437–452. Olweus, D. (1993). Bullying at school: What we know and what we can do. Oxford,
U.K.: Blackwell.
Olweus, D. (2003). A Profile of bullying at school. Educational Leadership, 60(6), 12–17.
Olweus, D., & Limber, S. (1999). Bullying prevention program. In D. Elliott (Ed.), Blueprints for Violence Prevention. Boulder. Institute of Behavioral Science, Regents of the University of Colorado: Colorado.
Pepler, D., Jiang, D., Craig, W., & Connolly, J. (2008). Developmental trajectories of bullying and associated factors. Child Development, 79(2), 325–338.
Raskauskas, J., & Stoltz, A. D. (2007). Involvement in traditional and electronic bullying among adolescents. Developmental Psychology, 43(3), 564–576. Rios-Ellis, B., Bellamy, L., & Shoji, J. (2000). An examination of specific types of
Ijime within Japanese schools. School Psychology International, 21(3), 227–241.
Smith, P. K., Cowie, H., Olafsson, R. F., & Liefooghe, A. P. D. (2002). Definitions of bullying: A comparison of terms used, and age and gender differences, in a fourteen-country international comparison. Child Development, 73(4), 1119–1133.
SS, S. P. (2006). Version 15. Chicago: SPSS.
Unnever, J. D., & Cornell, D. G. (2004). Middle school victims of bullying: Who reports being bullied? Aggressive Behavior, 30(5), 373–388.
Wei, H.-S., & Huang, C.-K. (2005). Reviewing school bullying research: Empirical findings and methodical considerations. NTTU Educational Research Journal, 16(1), 69–112.
Wei, H.-S., Jonson-Reid, M., & Tsao, H.-L. (2007). Bullying and victimization among Taiwanese 7th graders: A multi-method assessment. School Psychology International, 28(4), 479–500.
Willard, N. (2006). Flame retardant: Cyberbullies torment their victims 24/7: Here’s how to stop the abuse. School Library Journal, 52(4), 54–56.
