Cyberbullying Among Primary School Students in Turkey: Self-Reported Prevalence and Associations with Home and School Life

Abstract

The current study examined the self-reported prevalence and nature of cyberbullying and victimization among second, third, and fourth grade students (N=372) and explored associated features of home and school life. Of the children in the current sample, 27 percent had been victims of cyberbullying, 18 percent had been aggressors, and 15 percent had been both cyberbullies and victims. Boys were significantly more likely to carry out cyberbullying than girls. Cyberbullying exposure (as both a bully and a victim) was significantly associated with low levels of self-reported school satisfaction (bullies odds ratio [OR]: 2.45; victims OR: 2.10; p<0.05) and achievement (bullies OR: 3.85; victims OR: 3.47, p<0.05). Paternal unemployment was also associated with a three-fold increase in the likelihood of being a cyberbully. Increased awareness and regulation is now required within schools and within the home to tackle this escalating problem.

Introduction

Online technologies and telecommunication systems have many benefits for students, allowing instant information access, rapid communication, and extensive social networking. Despite the advantages of these technologies when used appropriately, a variety of problems can arise from their misuse, including cybercrime, identity theft, the uploading and distribution of embarrassing or explicit videos, Internet addictions, and harassment.^1–4 The current study focuses on cyberbullying, a phenomena that has prompted escalating concerns in recent years in developed countries across Europe and the United States.^1,5 The current study aimed to extend our understanding of cyberbullying in Turkey, a developing country that has shown recent surges in Internet usage and technological development. In particular, it explored the prevalence of cyberbullying among 8–11-year-old students and the associations between bullying exposure and aspects of home life (e.g., number of parents living at home, parental employment, number of siblings) and schooling (e.g., satisfaction, attendance, and achievement).

Cyberbullying is defined as the use of modern technologies by an individual or group to embarrass or intimidate others in a deliberate, repetitive, and hostile manner.^3,5–7 Specific examples of cyberbullying include anonymous phone calls, abusive e-mails, theft of identity information and harassment via instant messaging devices, social networking media, and text messages.^7,8 In recent years, cyberbullying has begun to include the distribution of explicit videos and occurs via a range of websites, including online gaming sites.⁴ In keeping with traditional bullying, cyberbullying is characterized by an imbalance of power between the bully and victim.^9–11 New technologies can exacerbate this power differential, making it easier for bullies to hide their identities^11,12 and to perform repeated, prolonged harassment with a high level of anonymity.^7,13,14 Two previous studies have reported that the identity of the cyberbully is often unknown to the victim.^9,11 It is also common for children to be involved in cyberbullying as both the perpetrator and the victim, although the causes of this association remain unclear.⁶

According to a report by Turkey's National Statistics Institute, there has been a steady increase in the accessibility of modern technologies for young people in Turkey. Among 16–24-year olds, 95 percent were reported to have access to a computer and 89 percent were active Internet users.¹⁵ In light of these advances, a number of studies have explored the occurrence of cyberbullying in Turkish schools.^3,6,16–23 In the most recent study, Yilmaz²⁴ investigated cyberbullying in a sample of seventh grade students (N=756) in Istanbul. Using a self-reported questionnaire, 18 percent of children reported being victims, while 6 percent reported being cyberbullies themselves. Erdur-Baker²⁵ suggested that the prevalence of victimization was even higher, finding that 32 percent of their sample (N=276; aged 14–18 years) were victims of both cyberbullying and traditional bullying. In contrast, Aricak et al.²⁶ showed higher levels of cyberbullying (36 percent) than victimization (6 percent) in a sample of 269 Turkish secondary school students. Elsewhere in the world, cyberbullying rates have been placed between 9 percent and 34 percent for victims and between 4 percent and 21 percent for bullies.⁵ Notably, these prevalence rates appear to have escalated steeply over time, with a 50 percent increase observed between 2000 and 2005.⁷

Previous studies of cyberbullying have focused largely on adolescents aged 14 years and upward. Few studies have included younger children. In a recent study of 413 Turkish youths (aged 12–14), around 19 percent of children were victims of cyberbullying, while 12 percent had been bullies themselves.²³ Younger children were found to be at a greater risk of being bullied, whereas older children were more frequently the aggressor. In a U.S. sample of 3,767 middle school students (aged 11–14), 11 percent had been cyberbullied, 4 percent had bullied other children, and 7 percent had been both cyberbullies and victims.⁹ Finally, a study by Williams and Guerra²⁷ included children as young as 10 (N=3,339, aged 10–14); showing that physical and cyberbullying peaked in middle school and declined in high school. Further replication of these findings is required, along with investigation of cyberbullying emergence in even younger samples.

Relatively few studies to date have investigated the association between exposure to cyberbullying and characteristics of home (e.g., social economic status, single-parent families, or parental employment) and school (e.g., achievement, attendance, and satisfaction).^8,11 Ybarra and Mitchell⁸ found that poor caregiver monitoring and a poor emotional bond with the caregiver is associated with cyberbullying. However, they found no association between cyberbullying frequency (as either a bully or victim) and household factors, such as income, parental education, and caregiver status (single or married parent). There also is some evidence that cyberbullying is associated with reduced school attendance,¹¹ school commitment,⁸ and a negative school climate.²⁷

The current study aimed to replicate and extend previous studies of cyberbullying in Turkey, addressing four main topics. First, it assessed the self-reported prevalence of cyberbullying among second to fourth grade students (aged 8–11 years) in Turkey; a younger sample than previous studies. Secondly, it explored the frequency of cyberbullying incidents and the most common methods of harassment for this age group (e.g., via e-mail, mobile phones, chat rooms, or instant messengers/social networking sites). A third goal was to examine the factors associated with cyberbullying. These factors included a child's grade and gender, along with three specific home-related factors (parental employment, number of siblings, caregiver status) and three school-related factors (school satisfaction, attendance, and achievement). Finally, the study sought to examine the overlap between bullying and victimization (the presence of bully-victims) and to determine whether victims of bullying knew the identity of their aggressors.

Methods

Participants

Self-report questionnaires were sent to 389 children, across three primary schools under the Istanbul Provincial Directorate of National Education. These schools were selected using a random cluster sampling method. Thirty-two districts of Istanbul were listed and three districts were selected at random. One public school from each of the chosen districts was selected at random. In the chosen schools, questionnaires were sent to three classrooms from grades 2 to 4, each containing 20–30 children. In total, questionnaire data were available for 372 children aged 8–11 years. Istanbul is a large, metropolitan city with a high level of diversity in terms of ethnicity and socioeconomic factors.²⁸ Of the schools selected, one admitted children based on an entrance exam and generally comprised children with a higher socioeconomic status (SES). The other two schools included children with a broader SES range.

Instruments

Self-report questionnaire

This study used a 26 item self-report questionnaire, created by the authors. Sixteen items concerned aspects of home and school. Home-related questions focused on parental employment (father employed: yes/no, mother employed: yes/no), number of siblings, and caregiver status (single- vs. two-parent homes). Parental employment was assessed in place of SES, as it was felt that children would be better able to report parental employment than parental income. School characteristics included school satisfaction (do you like school? yes/no), attendance (do you frequently miss school without a good reason? yes/no), and achievement (an appreciation certificate, a thanks certificate, a direct pass, passing with responsibility or repeating the year). These levels of attainment were divided into three categories: low, average, or high achievement. Ten questions addressed cyberbullying experiences (e.g., ‘have you been a cyberbully’, ‘have you ever been cyberbullied by other people?’), including the frequency (once or twice a month; once or twice a week, or nearly every day), method of harassment (by e-mail, by telephone, in chat rooms, or via instant messengers and social networking sites), and outcome (‘did you tell anyone about the cyberbullying?’; ‘who did you tell?’). The questions also addressed the anonymity of the bullying (‘Did you know the name of the person/people you have been bullied by?’). Questions were either multiple choice or answered with a yes/no response. The content validity of this Turkish cyberbullying questionnaire was established by conducting a comprehensive literature search of previous cyberbullying and victimization measures.^8,9,13,29 Item content and language were reviewed and approved by six faculty members (psychologists, psychiatrists, pediatric nurses, and education specialists) from the Istanbul University, the Marmara University, and the Düzce University. The clarity of the questions was then assessed with 15 students, requiring only minor adjustments. Of note, the current questionnaire focused on a subset of cyberbullying methods, including bullying via telephone, chat rooms, e-mail, and social media sites (e.g., Facebook and Twitter). Owing to the young age of the sample, questions regarding explicit videos or Internet gaming were not included.

Procedure

The current study was reviewed by the Istanbul University and carried out with the permission of the Istanbul Governor, the Provincial Directorate of National Education, and the three participating schools. All students were informed that they were free to decline or to withdraw from the survey at any time and that all answers were strictly anonymous (no identifying information was collected). Before distributing the questionnaire, a researcher informed the children about the purpose of the study. The researchers explained that cyberbullying was when children were intentionally and repeatedly harmed, bullied or teased using electronic devices, including phones and the Internet (chat rooms, e-mail, and instant messengers). The questionnaire was completed during class time, with no interaction among the students. Ten students did not consent to participate in the study. Data were also excluded from seven students who completed less than half of the items on the questionnaire.

Data analysis

Descriptive analyses of the demographic data were carried out using the statistical software package SPSS 19. Frequencies and subtypes of cyberbullying and victimization were assessed for each gender. Exploratory logistic regression analyses were used to assess the factors that were significantly associated with being a cybervictim or a cyberbully. Nine categorical variables were assessed within univariate logistic regressions, including grade (second, third, fourth); gender (male/female); paternal employment (yes/no); maternal employment (yes/no); number of siblings (0, 1–3, 4+); caregiver status (single- vs. two-parent family); school satisfaction (likes/dislikes school); achievement (below average/average/above average); and attendance (unexcused absences: yes/no). Variables, which were significantly associated with cyberbullying or cybervictimization, were then included within a forward multivariate logistic regression model, adjusting for the effects of grade and gender. The association between victim status (victim vs. not a victim) and cyberbully status (bully vs. not a bully) was also assessed using univariate logistic regression.

Results

Demographic data from the current sample are shown in Table 1. With regard to home characteristics, 332 (89.2 percent) lived with both their parents and 332 (89.2 percent) had at least one sibling. The majority of fathers (N=315; 84.9 percent) were in paid employment, compared to only 108 (29 percent) of mothers.

Table 1.

Self-Reported Demographic and Household/School Characteristics for the Full Sample (N=372)

Variable	n	Percent
Gender
Male	193	51.9
Female	179	48.1
School grade
2	111	29.8
3	133	35.8
4	128	34.4
Parents' employment status
Mother employed	108	29.0
Father employed	315	84.7
Who does the child live with?
Both parents	332	89.2
One parent	26	7.0
Other relatives	14	3.8
Number of siblings
No siblings	40	10.8
One sibling	170	45.7
Two to three siblings	108	29.0
Four or more Siblings	54	14.5
Academic achievement
Above average	195	52.4
Average	127	34.1
Below average	50	13.4
Child's school satisfaction
Likes school	334	89.8
Dislikes school	38	10.2
Child's school attendance
Unexcused absences	23	6.2
No unexcused absences	349	93.8
Has the child engaged in cyberbullying (using online technologies or mobile phones?)^a
Yes	65	17.5
No	307	82.5
Has the child been a victim of cyberbullying?^a
Yes	102	27.4
No	270	72.6

Children asked about bullying experiences over the last 6 months.

The majority of children stated that they liked school (N=334; 89.8 percent) and had not had unexcused absences from class (N=349; 93.8 percent). In the current sample, 65 (17.5 percent) students had bullied other children using online technologies or mobile phones and 102 (27.4 percent) had been the victims of cyberbullying.

Table 2 shows the characteristics of students who reported that they had engaged in cyberbullying. The most common tools for bullying were instant messaging systems, such as MSN (N=33; 50.8 percent), followed by mobile phones (N=15; 23.1 percent), chat rooms (N=9; 13.8 percent), and e-mail (N=8; 12.5 percent). Of the bullies in this sample, around half (N=34; 52.3 percent) did not know the identities of their victims. With regard to bullying frequency, the majority of the bullies harassed other children a few times a week (N=42, 73.8 percent).

Table 2.

Descriptive Data of the Students Who Reported Engaging in Cyberbullying in the Last 6 Months (N=65)

	Males (N=43)		Females (N=22)		Total (N=65)
Variable	n	Percent	n	Percent	n	Percent
Method of cyberbullying^a
E-mail	6	14.0	2	9.1	8	12.3
In chat rooms	7	16.3	2	9.1	9	13.8
By telephone	7	16.3	8	36.4	15	23.1
Instant messaging tools (e.g., MSN, Facebook, Twitter)	23	53.5	10	45.5	33	50.8
Target of cyberbullying
Classmates	6	13.3	2	10.0	8	12.3
Children outside of school (known to child)	13	28.9	10	50.0	23	35.4
Children outside of school (unknown to child)	26	57.8	8	40.0	34	52.3
Frequency of cyberbullying
Nearly every day	4	9.3	0	0	4	6.2
Once or twice a week	33	76.7	15	68.2	48	73.8
Once or twice a month	6	14.0	7	31.8	13	20.0

Children reported the most common method (only one answer was given).

In the univariate logistic regression analyses, a significant effect of gender was observed, with boys around twice as likely as girls to carry out cyberbullying (odds ratio [OR]: 2.05; confidence interval [CI]: 1.17–3.58; p<0.01). Being a cyberbully was not significantly associated with a child's grade (p>0.05). In terms of home-related factors, number of siblings and maternal employment were not significant predictors of being a cyberbully (p>0.05). However, cyberbullies were over twice as likely to come from single-parent homes (OR: 2.59; CI: 1.26–5.35; p<0.05) than homes with two parents. Children whose fathers were unemployed were also over three times as likely to be cyberbullies as children whose fathers were in employment (OR: 3.08; 95 percent CI: 1.63–5.83; p<0.01) With regard to school, a child's grade and attendance were not predictors of cyberbullying behaviors (p>0.05). Children with a below average level of achievement were over three times as likely to engage in cyberbullying (OR: 3.85 95 percent CI: 1.86–7.99; p<0.01) compared to children with a higher school attainment.

The significant variables from the univariate analyses were included within a multivariate regression model, adjusting for grade and gender. Table 3 shows the results from the final model, which provided a good fit to the data according the Hosmer–Lemeshow test (χ²=5.27, df=8, p=0.73).When controlling for the other variables, children who were achieving below average grades in school were almost three times as likely to be involved in bullying. Children from families, where the father was unemployed were also twice as likely to be bullies compared to children with employed fathers.

Table 3.

Results from the Two Multivariate Logistic Regression Models Assessing the Predictors of Cyberbullying Exposure

			Cyberbully ^a		Cybervictim ^b
			Adjusted odds ratio ^c (95 percent CI)	p value	Adjusted odds ratio ^c (95 percent CI)	p value
Predictor variable	Paternal employment	Employed	—	—	—	—
		Unemployed	2.11 (1.04–4.30)	0.04^*	1.93 (1.02–3.63)	0.04^*
	Caregiver status	Two-parent family	—	—	—	—
		One-parent family	NS	NS	NS	NS
	School achievement	High achievement	—	—	—	—
		Average achievement	1.88 (1.00–3.55)	0.05^*	1.56 (0.92–2.63)	0.10
		Low achievement	3.41 (1.57–7.39)	<0.01^**	3.10 (1.61–6.33)	<0.01^**

Comparison group: never been a cyberbully.

Comparison group: never been a victim of cyberbullying.

Odds ratios taken from a multivariate logistic regression, with grade and gender entered as covariates.

p<0.05, ^**p<0.01.

CI, confidence interval; NS, nonsignificant.

Table 4 shows the characteristics of students who were the victims of cyberbullying. Children were most frequently victimized using instant messaging tools (54.9 percent), mobile telephones (26.5 percent), e-mail (14.7 percent), and chat rooms (3.9 percent). Of the victims in this sample, 78.4 percent did not know the identity of the person bullying them and most (82.4 percent) had never told anyone about the cyberbullying incidents. Of the 18 children that had disclosed the bullying, the majority had spoken to friends (61.1 percent) rather than parents (22.2 percent) or teachers (16.7 percent).

Table 4.

Descriptive Data of the Students Who Reported Being the Victims of Cyberbullying in the Last 6 Months (N=102)

	Males (N=59)		Females (N=43)		Total (N=102)
Variable	n	Percent ^a	n	Percent ^a	n	Percent ^a
Method of cybervictimization
E-mail	9	13.6	10	16.3	19	17.9
Chat rooms	3	5.1	1	2.3	4	3.8
Telephone	12	20.3	13	30.2	25	23.6
Instant message tools (e.g., MSN, Facebook, Twitter)	37	61.0	21	46.5	58	54.7
Frequency of cybervictimization
Nearly every day	6	10.2	9	18.6	15	13.7
Once or twice a week	12	20.3	12	27.9	24	22.5
Once or twice a month	41	69.5	22	51.2	63	61.8
Do you know the identity of the person/people that you have been bullied by?
Yes	12	20.7	11	25.0	23	22.5
No	46	79.3	33	75.0	79	77.5
Have you told someone (e.g., a parent, teacher, friends) that you have been bullied?
Yes	9	23.7	9	18.0	18	20.5
No	41	76.3	29	82.0	70	79.5
Whom have you told about being bullied?
Parent	1	11.1	3	33.3	4	22.2
Teacher	0	0	3	33.3	3	16.7
Friend	8	88.9	3	33.3	11	61.1

Owing to missing data, percentages do not add up to 100 percent, data were missing for some participants (some children did not fill out the full cyberbullying section).

Univariate logistic regression analyses revealed that the grade and gender of a child were not significantly associated with cyberbullying victimization (p>0.05). The number of siblings, caregiver status, and maternal employment were not significant predictors of being a cybervictim (p>0.05). However, children whose fathers were unemployed were twice as likely to be victims compared to children whose fathers were in employment (OR: 2.49; 95 percent CI: 1.39–4.48; p<0.01). Children with below average levels of school achievement were over three times as likely to be victims of cyberbullying (OR: 3.47; CI: 1.81–6.66; p<0.01) as those with above average attainment. Low school satisfaction was also associated with a doubled risk of victimization (OR: 2.10; CI: 1.05–4.18; p<0.05). Significant variables from the univariate regressions were included in a multivariate logistic regression model, adjusting for grade and gender. Table 3 shows the final model, which provided a reasonable fit to the data according to the Hosmer–Lemeshow test (χ²=6.67, df=8, p=0.57). When controlling for all other variables in the model, school achievement and paternal employment remained significant predictors of being a cybervictim (for ORs see Table 3).

Of the 372 children sampled, 54 (14.5 percent) reported that they had been the victim of cyberbullying and had been bullies themselves—so-called bully-victims. Of the bully-victims, a significant proportion were boys (61 percent) than girls (42 percent) (χ²=5.53; p<0.05). Logistic regression confirmed that being a victim was associated with a greater than 20-fold increase in the chance of also being a bully (OR: 26.49; CI: 12.92–54.30; p<0.01).

Discussion

The current study examined cyberbullying and victimization in the youngest Turkish sample to date. Of the 8–11-year-olds sampled, 27 percent had been victims of bullying via the Internet or mobile phones, 18 percent had carried out cyberbullying themselves, and 15 percent of children were both the victims and perpetrators of cyberbullying (bully-victims). Previous findings in older children (aged 14 years and upward) in Turkey have been mixed, placing estimates of cyberbullying at 4–36 percent and victimization at 6–32 percent.^24–26 Our findings in younger children, fall within a similar range. These estimates are also in keeping with reports from the United States of America and other more developed countries, which suggest that 9–34 percent of school aged children are the victims of bullying and 4–21 percent are bullies.^2,9,30–34 The variation across estimates is likely attributable to differences in sample characteristics (e.g., ascertainment and age range) and the measure of cyberbullying employed.

The most common channels of cyberbullying were instant messaging tools and social networking devices (including MSN, Facebook, and Twitter), reported by half of the current sample. These findings are in line with previous reports.^9,11,35 Although these technologies allow extensive socializing and freedom of expression, they are particularly susceptible to misuse. Bullies can harass their victims over a prolonged period (both inside and outside of school) and can maintain a high level of anonymity.^9,11,34 In the current sample, the majority of victims (nearly 80 percent) were unaware of the identity of the person bullying them, in keeping with previous work by Kowalski and Limber.⁹ By hiding their identity bullies are potentially able to escalate their level of intimidation beyond that seen in traditional bullying. Intervention is also made much more difficult. A further concern was that around three quarters of the victims in this sample had not told anyone about their bullying experiences, in line with some previous evidence.³⁶ According to Ayas and Horzum,²³ some victims are concerned that disclosing the bullying will lead to a loss of their Internet or cell phone access. This finding emphasizes the need to educate children about cyberbullying and to provide appropriate support and prevention strategies at home and school.

Boys in the current sample were more frequently cyberbullies than girls, in keeping with previous reports in the United States of America³¹ and Turkey.^6,26 This finding suggests that boys have a greater tendency towards aggressive behavior, even in a potentially anonymous online environment. In Turkey, this gender divide may be exacerbated as aggressive behavior may be more tolerated in boys.²⁵ Of note, the current finding stands in contrast to claims that cyberbullying is a ‘relational’ or verbal form of aggression and thus would be more common in girls.³⁷ In terms of victims, both sexes were comparably likely to have experienced cyberbullying, challenging some previous findings.^27,30 The anonymity of cyberbullying may make boys and girls equal targets for harassment.

There was a high degree of overlap between bullying and victimization in the current sample, with 15 percent of children reporting being both a bully and a victim. Being a bully was associated with a twenty fold increased risk of also being a victim. Importantly, within a cross-sectional design, it is not possible to determine the direction of this association. Being a victim of cyberbullying may lead to retaliation as a protective strategy. In contrast, cyberbullies may place themselves at increased risk of being bullied in return. Further investigation is required to the interplay between cyberbullying and cybervictimization over time.

Cyberbullying exposure in the current sample, as both a victim and perpetrator, was found to be associated with lower school attainment; in keeping with previous findings.³⁸ This relationship is likely to be complex. One interpretation is that cyberbullies achieve lower grades as a result of their time spent engaged in harassment. Alternatively, children who are already low achievers may be less motivated at school and may be more likely to start targeting their peers. In terms of victimization, a bidirectional relationship can also be considered. First, children achieving low grades may targeted by bullies owing to their academic difficulties. In contrast, children who are bullied may achieve poorer grades owing to the distress and distraction arising from their experiences. This finding is in keeping with studies of traditional bullying.^32,39–40

In the current sample, paternal (although not maternal) employment was a significant predictor of both cyberbullying and victimization. Children whose fathers were unemployed were over three times as likely to be cyberbullies and twice as likely to be cybervictims compared to children whose fathers were in employment. Paternal employment is often an indicator of increased stress in the home, and in some cases, may be associated with lower SES. Although speculative, this stress may negatively impact the child–parent relationship or lead to reduced child supervision; factors that have been linked previously to increased cyberbullying involvement.^8,13,41 Of note, children in the current sample from single-parent homes were around twice as likely to engage in cyberbullying. Although this finding was no longer significant when accounting for all other variables, it supports the potential role for home-related stressors and the parent–child bond in cyberbullying exposure.

A number of limitations of the current study must be considered. First, the self-report questionnaire was developed specifically for this study and did not constitute a standardized cyberbullying measure. Although the representation validity and content validity of the questionnaire was ensured, reliability was not assessed. It relied largely on yes/no answers and children may not be able to report reliably about certain aspects of home and school life (e.g., SES). Further investigation is warranted using standardized tools from multiple informants. New dimensional tools will also be vital to assess the impact of recent social networking advances (such as Facebook and Twitter). The developmental course of cyberbullying behaviors warrants further investigation as the current study uses solely cross-sectional data. The current study would also have benefited from quantifying Internet and mobile phone usage. If some children had reduced access to these resources, this would have affected their risk of cyberbullying exposure. Finally, the current study included multiple comparisons carried out with the same dataset. The results should be interpreted with a degree of caution as there is a risk of Type 1 error.

It is important to consider the current findings in the context of the social and cultural characteristics of Turkey. As a country on the brink of being considered developed, Turkey has seen a surge in technological advancements and the accessibility of telecommunications for young people. However, the country is still characterized by strong family and cultural traditions, often with high levels of discipline both at home and school. It is therefore interesting to see that the cyberbullying trends in Turkey mirror, at least in part, the patterns seen in more developed counties. Increased awareness among families, educators, and health professionals is crucial to detect instances of cyberbullying early and to prevent exposure to online harassment.

Footnotes

Acknowledgment

The author(s) received no financial support for the research and/or authorship of this article.

Disclaimer

The material has not been published elsewhere and it is not under consideration at any other journal.

Author Disclosure Statement

The author(s) declared no potential conflict of interests with respect to the authorship and/or publication of this article.

References

Kappos

. The impact of electronic media on mental and somatic children's health. International Journal and Hygiene Environmental Health, 2007; 210:555–562.

Price

, Dalgleish

. Cyber-bullying experiences, impacts and coping strategies as described by Australian young people. Youth Studies Australia, 2010; 29:51–59.

Topçu

, Erdur-Baker

, Çapa Aydin

. Examination of cyber-bullying experiences among Turkish students from different school types. Cyberpsychology and Behavior, 2008; 11:644–648.

Rideout

, Roberts

, Foehr

. 2005. Generation M: media in the lives of 8–18 year-olds: executive summary. Menlo Park, CA: Henry J. Kaiser Family Foundation.

Ferdon

, Hertz

. Electronic media, violence, and adolescents. An emerging public health problem. Journal of Adolescent Health, 2007; 41:1–5.

Erdur-Baker

, Kavsut

. Cyber bullying: a new face of peer bullying. Eurasian Journal Education Research, 2007; 27:31–42.

Wolak

, Mitchell

, Finkelhor

. Does online harassment constitute bullying? An exploration of online harassment by known peers and online-only contacts. Journal of Adolescent Health, 2007; 41:51–58.

Ybarra

, Mitchell

. Youth engaging in online harassment: associations with caregiver-child relationships, Internet use, and personal characteristics. Journal of Adolescence, 2004; 27:319–336.

Kowalski

, Limber

. Electronic bullying among middle school students. Journal of Adolescent Health, 2007; 41:22–30.

10.

Smith

, Mahdavi

, Carvalho

et al. Cyber-bullying: its nature and impact in secondary school pupils. Journal of Child Psychology and Psychiatry, 2008; 49:376–385.

11.

Ybarra

, West

, Leaf

. Examining the overlap in internet harassment and school bullying: implications for school intervention. Journal of Adolescent Health, 2007; 1:42–50.

12.

Patchin

, Hinduja

. Cyber-bullying and self-esteem. Journal of School Health, 2010; 80:614–621.

13.

Ybarra

, Mitchell

. Online aggressor/targets, aggressors, and targets: a comparison of associated youth characteristics. Journal Child Psychology and Psychiatry, 2004; 45:1308–1316.

14.

Maher

. Cyber-bullying an ethnographic case study of one Australian upper primary school class. Youth Studies Australia, 2008; 27:50–57.

15.

Turkish Statistic Institute Household ICT Usage Survey. 2011. News bulletin. 179. www.tuik.gov.tr/PreHaberBultenleri.do?id=1&tb_id=1. 2012 Mar. 24.

16.

Erdur-Baker

, Tanrıkulu

. Cyber bullying in Turkey: its correlates and links to depressive symptoms. Journal of eHealth Technology and Application, 2009; 7:16–23.

17.

Akbulut

, Sahin

, Eristi

. Cyberbullying victimization among Turkish online social utility members. Educational Technology and Society, 2010; 13:192–201.

18.

Çetin

, Peker

, Eroglu

et al. Interpersonal cognitive distortions as a predictor of cyber victimization and bullying: a preliminary report in adolescents. International Online Journal of Educational Sciences, 2011; 3:1064–1080.

19.

Özdemir

, Akar

. Examination of high school students' opinions on cyberbullying in terms of various variables. Educational Administration: Theory and Practice, 2011; 17:605–626.

20.

Muraa

, Topcu

, Erdur-Baker

, Diamantinia

. An international study of cyber bullying perception and diffusion among adolescents. Procedia Social and Behavioral Sciences, 2011; 15:3805–3809.

21.

Topcu

, Erdur-Baker

. The revised cyber bullying inventory (RCBI): validity and reliability studies. Procedia Social and Behavioral Sciences, 2010; 5:660–664.

22.

Erdur

Baker Ö

, Tanrikulu

. Psychological consequences of cyber bullying experiences among Turkish secondary school children. Procedia Social and Behavioral Sciences, 2010; 2:2771–2776.

23.

Ayas

, Horzum

. On being cyber bully and victim among primary school students. Elementary Education Online, 2012; 11:369–380.

24.

Yilmaz

. Cyber-bullying in Turkish middle schools: an exploratory study. School Psychology International, 2011; 32:645–654.

25.

Erdur-Baker

. Cyberbullying and its correlation to traditional bullying, gender and frequent and risky usage of internet-mediated communication tools. New Media and Society, 2010; 12:109–125.

26.

Aricak

, Yahhan

, Uzunhasanoglu

et al. Cyber-bullying among Turkish Adolescents. Cyberpsychology and Behavior, 2008; 11:253–261.

27.

Williams

, Guerra

. Prevalence and predictors of internet bullying. Journal of Adolescent Health, 2007; 41:14–21.

28.

Seker

. 2011. Quality of life research in Istanbul, İstanbul Ticaret Odası (İTO) Yayınları170, publishing, Istanbul Chamber of Commerce Publications [Publication no.]: 2010–10.

29.

Beran

, Violato

. A model of childhood perceived peer harassment: analyses of the Canadian national longitudinal survey of children and youth data. Journal of Psychology, 2004; 138:129–147.

30.

Beran

, Li

. Cyber-harassment: a study of a new method for an old behavior. Journal of Education Computer Research, 2005; 32:265–277.

31.

Wang

, Lannotti

, Nansel

. School bullying among adolescents in the United States: physical, verbal, relational, and cyber. Journal of Adolescent Health, 2009; 45:368–375.

32.

Woods

, Wolke

. Direct and relational bullying among primary school children and academic achievement. Journal of School Psychology, 2004; 42:135–155.

33.

Nocentini

, Calmaestra

, Schultze-Krumbholz

et al. Cyberbullying: labels, behaviors and definition in three European countries. Australian Journal of Guidance and Counseling, 2010; 20:129–142.

34.

. Cyber-bullying in schools: a research of gender differences. School Psychology International, 2006; 27:157–170.

35.

Grigg

. Cyber-aggression: definition and concept of cyberbullying. Australian Journal of Guidance and Counselling, 2010; 20:2.

36.

Slonje

, Smith

. Cyberbullying: another main type of bullying? Scandinavian Journal of Psychology, 2008; 49:147–154.

37.

Keith

, Martin

. Cyberbullying: creating a culture of respect in a cyber world. Reclaiming Children and Youth, 2005; 13:224–228.

38.

Beran

, Li

. The relationship between cyber-bullying and school bullying. Journal of Student Wellbeing, 2007; 1:15–33.

39.

Glew

, Fan

, Katon

et al. Bullying, psychosocial adjustment, and academic performance in elementary school. Archive Pediatric Adolescent Medicine, 2005; 159:1026–1031.

40.

Slee

. Peer victimization and its relationship to depression among Australian primary-school students. Personality Individual Difference, 1995; 18:57–62.

41.

, Wu

. Adolescents' paternal attachment and internet use. Cyberpsychology and Behavior, 2007; 10:633–639.