Article Text
PDF
Abstract
Objectives Cyberbullying research in China is in early stage. This study describes the cyberbullying experiences of junior and senior high school students in Guangzhou, China, and to examine the risk factors associated with cyberbullying perpetrators, victims and perpetrator-victims among students. We also investigated the frequency of cyberbullying and coping strategies of student victims.
Methods Participants were 2590 students in grades 7, 8, 9 and 10 from six junior and senior high schools in October 2015 in Guangzhou, in south China, who completed a questionnaire. Data on participants' experiences with cyberbullying perpetration and victimisation during the previous 6 months were collected. Multinomial logistic regression was used to analyse factors associated with being perpetrators, victims and perpetrator-victims.
Results In this sample, 28.0% (725) of participants reported being a perpetrator and 44.5% (1150) reported being a victim in the previous 6 months. Specifically, 2.9% (74) reported being perpetrators only, 19.3% (499) reported being victims only and 25.2% (651) reported being perpetrator-victims (both perpetrator and victim). In addition, flaming was the most common form of cyberbullying in both perpetration and victimisation. Logistic regression analyses indicated that online game addiction in participants was associated with increased odds of being a perpetrator only; no democratic parenting style in the mother and physical discipline by parents were associated with increased odds of being a victim only; male students, students with low academic achievement, those spending over 2 hours a day online, experiencing physical discipline from parents and online game addiction were associated with increased odds of both perpetration and victimisation.
Conclusions Cyberbullying is a common experience among Chinese junior and senior high school students. These findings add to the empirical data on cyberbullying and reinforce the urgent need for cyberbullying prevention in China. Furthermore, from the perspective of practice, it is important to raise our awareness of cyerbullying and reduce the risk factors.
Statistics from Altmetric.com
Introduction
Cyberbullying has become a common occurrence among adolescents worldwide. With the development of information and communications technology (ICT) and the popularisation of social networks, the internet has become a universal way for teenagers to communicate with others. According to the China Internet Network Information Center survey, China's internet users numbered 630 million in 2012, with preadolescents and adolescents (10–19 years old) accounting for 25% of this total. Heavy use of the internet, such as instant messaging, text messaging, micro blogs and online chat rooms, connects people conveniently and creates new problems. The popularity of the internet in China has made cyberbullying a serious public health problem.
Although there is no consensus on the definition of cyberbullying, most researchers agree that cyberbullying is an international phenomenon and is a repeated, intentional act done with the purpose of harming another person through technologies such as email, text messaging, social networking sites and instant messaging and is carried out by a group or an individual.1–3 Personal computers and mobile phones are the most common means for engaging in cyberbullying. Unlike traditional bullying, cyberbullying does not require face-to-face communication and can be completely anonymous.3 Cyberbullying messages and images can also quickly spread on the internet and distributed to a wide audience, which may heighten children's perceptions of vulnerability. Besides, it is difficult to escape or hide from cyberbullying attacks compared with traditional bullying.
Existing studies on the prevalence of cyberbullying show that it is common among adolescents worldwide. Diverse prevalence rates of cyberbullying throughout the world have been reported due to the different definitions of cyberbullying and the different age groups in which it occurs. American researchers4 found that approximately 9% of American adolescents have been cyberbullied and approximately 16% had cyberbullied others in their lifetime. An online, cross-national study5 for European Union (EU) children in 2009 showed that almost 18% of EU teenagers had been the victims of cyberbullying through the internet or a mobile phone. In the UK,6 the National Children's Home, in a study of 770 children aged 11–19 years, discovered that 20% of children have been cyberbullied or threatened and 11% claimed to have sent a bullying message to someone else. In Canada, a survey7 reported that 24.9% of the sample were victims of cyberbullying. Another survey8 of 461 grade-7 students in both China and Canada discovered one in three students has been a victim of cyberbullying. Further, Slonje and Smith9 determined the overall cyberbullying rate among Swedish students aged 12–20 years was 11.7%.
Research on cyberbullying in China is still in the early stages. To date, there has been limited research on cyberbullying among adolescents from mainland China. A survey10 on cyberbullying among Chinese high school students in central China revealed 34% of perpetration and 57% of victimisation. Among 5726 students in grades 7–12, the overall prevalence of cyberbullying was 46.8%.11 The majority of existing studies were conducted in Hong Kong and Taiwan. The Chinese University of Hong Kong found in a sample of 1800 junior high school students that more than 30.9% of Hong Kong teenagers had been cyberbullied and 17.8% had cyberbullied others.12 A survey of 2992 grade-10 students in Taipei13 showed that approximately 18.4% had been cyberbullied and 5.8% had cyberbullied others. Another study14 indicated that 34.9% of junior high school students in Taiwan had been cyberbullied and 20.4% had cyberbullied others. However, most of these studies have focused on the prevalence of cyberbullying, with few examining victimisation and perpetration related to cyberbullying.
The impact of cyberbullying takes various forms. Typical adverse reactions to being a victim of cyberbullying include frustration, anger and sadness.15 Both perpetration and victimisation are related to mental health issues such as lower self-esteem,16 less-developed social skills and higher anxiety and aggression.17 A survey18 by Canadian researchers on cyberbullying among students in grades 7–9 reported that these victims are more likely to have poor concentration, low academic achievement and absenteeism. In addition, the higher the frequency of cyberbullying, the more negative issues the victims experienced offline.19 Cyberbullying victimisation was also associated with behavioural problems such as consumption of alcohol, smoking and low school commitment.20 Moreover, a population-based study of adolescents in Finland found that cybervictims experienced more clinical symptoms, such as headaches, recumbent abdominal pain and sleep disorders, compared with non-victims.21 Hindujia and Patchin22 found that in the worst cases, both perpetrators and victims have stronger suicidal thoughts or an increasing likelihood of suicidal thoughts compared with a non-victim group.
According to Haynie,23 cyberbullying involves three groups: perpetrators, victims and perpetrator-victims, and the researchers indicated that perpetrator-victims were a distinct group showing more problem behaviours and conflicts with parents compared with perpetrators and victims. Knowledge of these characteristics is important for us to identifying students at a risk of involvement in cyberbullying. Due to the lack of existing research, this study aims to describe the cyberbullying experiences of male and female junior and senior high school students in Guangzhou, China, and to examine the risk factors associated with cyberbullying perpetrators, victims and perpetrator-victims among students. We also investigated the frequency of cyberbullying and coping strategies of student victims. These findings add to the empirical data on cyberbullying in junior and senior high school students. Furthermore, from the perspective of practice, it is important to raise our awareness of cyerbullying and reduce the risk factors.
Methods
Study subjects and definition
The study was conducted in May 2015 in Guangzhou, southern China. Participants comprised 2590 students in grades 7–10 from six different junior and senior high schools. We first randomly selected one suburban (Baiyun district) and one urban district (Tianhe district), and then randomly selected three junior and high school from each of the two selected districts, using computer-generated random numbers. All classrooms for these grades, 55 in total, were investigated in the selected schools. This age group was chosen because Slonje and Smith8 claimed that cyberbullying rates are concentrated in adolescents aged 12–15 years.
Based on the existing literature7 ,8 and by combining with our national conditions, we decided the final seven forms of cyberbullying: flaming, harassment, exclusion, denigration, outing, cyberstalking and internet fraud. Flaming is online ‘fighting’ using electronic messages containing angry, vulgar language. Harassment is repeatedly sending offensive, rude and insulting messages. Exclusion is completely ignoring or excluding a person from a group of friends or deliberately leaving a person out of plans. Denigration involves sending cruel, and possibly untrue, information about a person to others. Outing is the sharing of sensitive or private information about a person with others. Cyberstalking entails issuing threats of harm or intimidation or engaging in other online activities that threaten a person's safety. Internet fraud means deceiving people through fraudulent actions on the internet.
Data collection procedures
The study design was approved by the school authorities, and questionnaires were distributed to the students through their teachers or our investigator. Surveys were administered after class by the principal investigator in the classroom. Following a brief introduction of each principal investigator, we invited all the selected students to participate in a voluntary, anonymous, paper-and-pencil survey, which required 15–20 min to complete. The questionnaires were retrieved after checking for completeness. We distributed a total of 2700 surveys out of which 2590 were completed and returned (return rate of 95.9%).
Study instrument and variables
The self-designed questionnaire was used to understand participants' cyberbullying experience. Before the formal investigation, we invited two experts of epidemiology to demonstrate the rationality and logic of the questionnaire and asked them to modify unclear expressions in it. The final version of our survey included three sections: demographic information (eg, sex, age, grade, school achievement, relationship status, personality, economic status and living arrangement); experiences of perpetration and victimisation, including frequency and forms of cyberbullying and coping strategies used by students; factors associated with cyberbullying (influencing factors and internet gaming disorder (IGD)).
We think that a single incident of cyberbullying may lead to very serious consequences, and by referring to the specific definition of cyberbullying and other research,24 we chose at least once as the cut-off point for categorisation. Because students can potentially be a perpetrator or a victim, we divided the participants into three groups: (1) victims, defined as engagement in one or more of the seven victimisation behaviours at least one time within the previous 6 months. To account for whether a perpetrator was cyberbullying victim as well, victimisation was further measured by variables of ‘any victims’ or ‘victims only’. (2) Perpetrators, defined as engagement in one or more of the seven perpetration behaviours at least one time within the previous 6 months. To account for whether a perpetrator was cyberbullying victim as well, perpetration was further measured by variables of ‘any perpetrators’ or ‘perpetrators only’. (3) Perpetrator-victims, defined as those who both had been cyberbullied and also had cyberbullied others at least once within the previous 6 months.
In this survey, participants were asked how often they engaged in the seven forms of cyberbullying as both perpetrators and victims. We defined the seven forms of cyberbullying as flaming, harassment, exclusion, denigration, outing, cyberstalking and internet fraud (eg, ‘Have you ever used electronic messages containing angry, vulgar language to anybody online?’; ‘Have you repeatedly sent offensive, rude or insulting messages to anybody online?’). The questions on the seven types of cyberbullying were answered used a 5-point Likert scale: 1 (‘never’), 2 (‘1–2 times per month’), 3 (‘3–5 times per month’), 4 (‘6–10 times per month’) and 5 (‘10 times or more per month’) within the previous 6 months. The coping strategies of cybervictims were classified as (1) got revenge, (2) seeking help from friends, (3) seeking help from parents, (4) seeking help from teachers, (5) took no action and (6) called police. Participants were allowed to select as many options as applicable to adequately describe their behavioural reactions to their experience.
The influencing factors of this study, respectively, were time spent online, relationship status of students, parents' education level, parenting style and use of physical discipline by parents. The IGDS-SF9,25 which was used to access IGD, comprises nine items and has been proven to be highly suitable for measuring IGD. The nine questions comprising the IGDS-SF9 are answered by using a 5-point Likert scale: 1 (‘never’), 2 (‘rarely’), 3 (‘sometimes’), 4 (‘often’) and 5 (‘very often’). The scores can range from 9 to 45, with higher scores indicating a higher degree of IGD. Thirty-six (out of 45) is a proposed cut-off point for distinguishing disordered (≥36) and non-disordered (<36) gamers. The Cronbach's α in this study was 0.87.
Statistical methods
The difference in the number and proportion of cyberbullying experiences was calculated, including any perpetrators, any victims, perpetrators only, victims only and perpetrator-victims, within the previous 6 months. The difference in the number and proportion of cyberbullying experiences between males and females were calculated using a χ2 test. Different types of cyberbullying in perpetrators or victims were calculated, including flaming, exclusion, harassment, denigration, outing, cyberstalking and internet fraud.
Multinomial logistic regression models were used to estimate the likelihood of a participant being a perpetrator only, a victim only and a perpetrator-victim as compared with contrast students, which was adjusted for grade, residence, relationship of students and parents' education level. SPSS V.13.0 was used for statistical analyses.
Results
Characteristic of study participants
Of the 2590 students, 59.7% (1543) were males and 40.3% (1042) were females, with an average age of 14.1 years. Of these participants, 26.0% (673) were in grade 7, 34.8% (901) were in grade 8, 13.6% (353) were in grade 9, 25.6% (663) were in grade 10. Note that 32.8% (836) of participating students were only-child and nearly half (49.2%) of students lived on campus (table 1).
Demographic characteristics of male and female students in Guangzhou
Cyberbullying experiences
Of the 2590 students included, 725 (28.0%) of participants reported that they had bullied someone online and 1150 (44.5%) reported being victims within the past 6 months. Specifically, 74 (2.9%) reported being perpetrators only, 499 (19.3%) reported being victims only and 651 (25.2%) reported being perpetrator-victims (table 2). Male students were more likely to report as perpetrators (p<0.001), victims (p<0.001) and perpetrator-victims (p<0.001) than female students (p<0.05).
Cyberbullying experiences of male and female students in Guangzhou
Forms and frequency of cyberbullying experienced
Flaming was the most common type of cyberbullying in both perpetrators (22.1%) and victims (32.0%), followed by exclusion in perpetrators (13.2%) and harassment (20.2%) in victims. Other types like denigration (5.8% in perpetrators, 17.1% in victims), outing (5.0% in perpetrators, 11.3% in victims) and cyberstalking (5.0% in perpetrators, 11.9% in victims) are shown in table 3. Specifically, 3.9% and 8.6% of participants reported they had cheated someone or been cheated by someone online.
Forms of cyberbullying experienced by male and female students in Guangzhou
Frequency of cyberbullying of junior and high school students in Guangzhou
Factors associated with cyberbullying among junior and high school students in Guangzhou
For cyberbullying perpetrators, the most common frequency is one to two times a month, accounting to 59.2%–70.3% of total perpetrators; three to five times a month is the case with 12.1%–25.2% of perpetrators and last more than six times a month for 15.6%–24.2% of perpetrators. For cyberbullying victims, 49.4%–59.6% of them are bullied one to two times a month; 21.5%–27.4% for three to five times a month and 16.1%–23.1% for more than six times a month (table 4).
Factors associated with odds of cyberbullying
Logistic regression analyses indicated that online game addiction (OR 4.478, 95% CI 1.567 to 12.800, p<0.05) in participants was associated with increased odds of being a perpetrator only than participants without online game addiction; no authoritative parenting style in the mother (OR 1.256, 95% CI 1.011 to 1.560, p<0.05) and physical discipline by parents (OR 1.356, 95% CI 1.091 to 1.686, p<0.05) were associated with increased odds of being a victim only than participants with authoritative parenting style in the mother and without physical discipline by parents; male students (OR 2.270, 95% CI: 1.829 to 2.817, p<0.05), students with below-average academic achievement (OR 1.396, 95% CI 1.072 to 1.816, p<0.05), spending over 2 or 4 hours a day online (OR 1.401, 95% CI 1.103 to 1.780, p<0.05 or OR 2.056, 95% CI 1.609 to 2.627, p<0.05), experiencing physical discipline from parents (OR 1.329, 95% CI 1.083 to 1.631, p<0.05) and online game addiction (OR 2.327, 95% CI 1.302 to 4.156, p<0.05) were associated with increased odds of both perpetration and victimisation than female students, students with above-average academic achievement, spending less than 2 hours a day online, students without physical discipline from parents and online game addiction (table 5).
Frequent behavioural responses to victimisation
The most frequent behavioural responses (table 6) to cyberbully victimisation is ‘got revenge’ (33.7%) followed by ‘seeking help from friends’ (23.6%) and ‘take no action’ (18.6%). Male students' top three behavioural responses is ‘got revenge’ (24.5%), ‘seeking help from friends’ (20.6%) and ‘seeking help from parents’ (16.2%), while female students' top three behavioural responses is ‘take no action’ (52.5%), ‘seeking help from teachers’ (29.6%) and ‘seeking help from parents’ (19.0%). Male students are significantly more likely to ‘got revenge’ and ‘called police’ than female students (p all<0.05), while female students are significantly more likely to ‘seeking help from friends’, ‘seeking help from parents’ and ‘take no action’ than male students (p all<0.05).
Students' reaction to cyberbullying victimisation by gender
Discussion
With the development and popularisation of the internet in recent years, traditional bullying has moved into the technological realm with cyberbullying becoming a growing problem and a common public health issue among adolescents worldwide. However, it has not received enough attention in China, especially in Mainland China. Few studies have explored the incidence, frequency, type and risk factors for cyberbullying victims in Mainland China. These findings add to the empirical data on cyberbullying and reinforce the urgent need for cyberbullying prevention in China.
Our data indicated that cyberbullying is not an uncommon occurrence in China. Previous Chinese studies reported a total prevalence of cyberbullying ranging from 8.9% to 70%.11 ,26 ,27 Differences in the definition of cyberbullying, measurement tools, area and research approaches are potential reasons for the diversity in findings. In this study, of the 2590 participants surveyed, nearly half (47.3%) had experienced at least one form of cyberbullying, either as a perpetrator or a victim. Specifically, 28.0% of participants reported they had cyberbullied others and 44.5% reported they had been cyberbullied within the previous 6 months. Nearly 25% of students had experienced both victimisation and perpetration. Our study found that the number of students who identified as being a victim only (n=499) was almost seven times higher than of those who identified as being a perpetrator only (n=74). These differences may due to understatement of perpetration. Although we guaranteed anonymity to participants, some may still not be willing to report because cyberbullying perpetrators may worry about how the teacher and classmates would think of them. However, there are probably more victims than perpetrators because one victim tends to be cyberbullied by more than one perpetrator in Guangzhou.
In this study, the most frequent form of cyberbullying is ‘flaming’. It is worth noting that the incidences of ‘denigration’ and ‘outing’, respectively, are 17.1% and 11.3%. These two actions are likely to cause significant damage to students, such as denigrating a person or outing a person by revealing embarrassing pictures of him or her online and allowing public comments to be made. Repetitive harm can occur due to the anonymity and virality of cyberbullying, which can lead to serious psychological problems. This suggests that schools and parents should pay attention to students' mental health and, in particular, look for ways of preventing serious cyberbullying.
Boys were significantly more likely to be cyberbullying perpetrator-victims than girls, but there were no statistically significant differences in perpetrators only or victims only. The findings on which gender is more likely to be involved in cyberbullying are inconsistent. Some studies14 ,28 have reported that males are more likely to be cyberbullies than females. But other studies8 ,29 have reported opposite results, finding that female students have a higher risk of cyberbullying experiences than male students. Gender has long been a significant factor related to cyberbullying behaviour, and this may due to males being more impulsive and aggressive compared with females. Another possible reason is that male students are more accepting of in-class use of technology than female students.30
A high level of internet use was positively connected with both victimisation and perpetration, and the longer the period of internet use, the greater the danger of a student being involved in cyberbullying. Students who spend more time online had a higher risk of being involved in cyberbullying. Therefore, reducing internet use may lower the possibility of being a part of cyberbullying. We also examined family variables as risk factors for cyberbullying because parents are a major influence on children's experiences and personality. We discovered that parenting style is closely related to cyberbullying. Inappropriate parenting styles (authoritarian, indulgent or doting) may cause more danger for victims only of cyberbullying, compared with an authoritative style. Parenting styles have been found to be connected to children's development, and authoritative parenting in particular has beneficial effects on adolescent lifestyles. Researchers have demonstrated that children whose parents are authoritative are more psychosocially mature and engage less in delinquent behaviour31 than their peers, which may reduce their tendency towards violence, including cyberbullying. Due to the one-child policy in China, most Chinese families have just one child. Children who grow up with indulgent parents may not be taught to control their emotion or to obey and may, therefore, be more likely to commit a crime.32 This finding is supported by previous research on the relation between parenting styles and violence,33 which suggests that parents should take a responsive and demanding approach to cultivating children's good character.
In December 2013, internet usage among Chinese youth had reached 256 million, accounting for 71.8% of the total. Specifically, mobile internet users now number 221 million. The rapid development of online games for young people has also triggered a series of problems. The youth network game addiction problem has attracted great attention from people of all ages. Compared with other studies34–36 in China from 2.5% to 5.8%, we found a lower rate of online gaming disorder (2.0%), which may be the result of a different sample, questionnaire or definition of online gaming disorder. This study explores the relationship between cyberbullying and IGD. A multifactor analysis showed that IGD is associated with greater odds of being a perpetrator only and being a perpetrator-victim. This finding may provide evidence for further research.
To conclude, this study reported a high level of overall prevalence rates of cyberullying in Guangzhou, China. This is one of the few studies to examine the risk factors associated with perpetrators, victims and perpetrator-victims. Our finds may help researchers to understand the differences among three cyberbullying participant groups and provide an empirical data on bidirectional cyberbullying in China. These findings also have implications for parents and educators. School administrators should work to educate students, teachers and staff about the impacts and coping strategies of cyberbullying. Besides, those students with high risk of cybebullying (eg, man, students with low academic performance) need to focus on cyberbullying education. Appropriate punishment could be considered for cyerbullying perpetrators, but only if sufficient evidence has been present. Parents need to make an appropriate regulation for students' use of internet and other cybertechnologies. They should also take an authoritative parenting style and never use corporal punishment on their children to reduce the possibility of cyberbullying. Of course, there are certain limitations to this study. First, such a cross-sectional study can generate inevitable recall bias and cannot differentiate the direction of effects. Second, there is a shared method bias in this study because information about cyberbullying was based on self-reports. Third, participants may not have been willing to answer sensitive questions, even though we guaranteed anonymity, so the investigation rate of cyberbullying may differ from the actual rate. Fourth, we have not reached consensus on the items used to assess the perpetrators and victims of cyberbullying. Finally, we chose at least once as the cut-off point for categorisation may lose information about variability. Thus, future research should consider developing a reliable and valid scale of cyberbullying in Chinese junior and high school students. Besides, more longitudinal studies are needed to analysis the long-term effects of cyberbullying victims.
Conclusion
Cyberbullying is a common experience among Chinese junior and senior high school students. These findings add to the empirical data on cyberbullying and reinforce the urgent need for cyberbullying prevention in China. Furthermore, from the perspective of practice, it is important to raise our awareness of cyerbullying and reduce the risk factors.
What is already known on the subject?
Cyberbullying has become a common occurrence among adolescents worldwide.
Cyberbullying research in China is in its early stage.
What this study adds?
Over 28.0% (725) reported that they had bullied someone online and 44.5% (1150) reported being victims within the previous 6 months in Guangzhou, China.
2.9% (74) reported being perpetrators only, 19.3% (499) reported being victims only and 25.2% (651) reported being perpetrator-victims (both perpetrator and victim).
Flaming was the most common form of cyberbullying in both perpetration and victimisation.
Online game addiction was associated with increased odds of cyberbullying.
Acknowledgments
The authors thank the medicine college students of Jinan university for their help on the field survey. The authors have no conflicts of interest to declare.
References
Footnotes
Contributors JR designed and implemented the study, conducted data analysis and write the manuscript; HW designed the study and revising the manuscript; MP and JY designed the statistical analysis and help with data analysis; YY helped with analysis plan and result interpretation; XC designed and helped with data analysis; SW contributes for study design and revising the manuscript; XD designed the study and developed the manuscript.
Competing interests None declared.
Patient consent Obtained.
Ethics approval This study was conducted with the approval of the Medical College of Jinan University.
Provenance and peer review Not commissioned; externally peer reviewed.