Cyberbullying in Canada

3 Cyberbullying in Canada Julia Riddell, Debra Pepler, and Wendy Craig Cyberbullying in Canada Cyberbullying is a significant problem in Canada, with high prevalence rates that have remained stable over the past decade (Boak et al. 2016; Craig et al. 2016). A number of highly publicized Canadian cases have underscored the significant implications of involvement in cyberbullying. For example, Amanda Todd, a 15-year-old from western Canada, was lured by a man who manipulated her to share intimate pictures of herself with him (BBC News 2014). When these pictures were posted to Facebook, Amanda was severely bullied by peers at school, causing her to change schools a number of times. She posted a video on YouTube that detailed her struggles with cyberbullying before taking her own life (BBC News 2014). Another young woman, Rehtaeh Parsons, had a photo of her sexual assault shared with peers via text messages (Gillis 2013). After J. Riddell (*) • D. Pepler York University, Keele Campus, Toronto, ON, Canada W. Craig Queen’s University, Kingston, ON, Canada © The Author(s) 2018 A. C. Baldry et al. (eds.), International Perspectives on Cyberbullying, Palgrave Studies in Cybercrime and Cybersecurity, https://doi.org/10.1007/978-3-319-73263-3_3 39 40 J. Riddell et al. this picture was shared, she was severely bullied both online and in person, leading her to commit suicide at age 17 (Gillis 2013). Canada is currently developing policies at both the provincial and school levels to address cyberbullying and prevent tragedies such as the ones described above. Cyberbullying legislation was first developed in the province of Ontario through an amendment to the Education Act in 2012 called Bill 13 (PREVNet 2015). This bill (entitled the Accepting Schools Act) outlines the rights and responsibilities of principals, teachers, schools, school boards, and the Ministry of Education when preventing or dealing with bullying incidents. This law specifies that the rights and responsibilities apply to all incidents of bullying that affect the school’s learning climate, whether on or off school property, face-to-face or electronically (PREVNet 2015). Similarly, the western Canadian province of Alberta amended its Education Act in 2012 to include bullying behaviors that occur both within the school building and by electronic means (Education Act 2012). Other provinces, such as Manitoba, state in their revised Education Act that parents are responsible for their children’s cyberbullying if they were aware of this problem behavior, could have reasonably predicted the effect, and did nothing (The Public Schools Act 2015). Efforts to prevent cyberbullying, however, are not consistent across Canada, partly due to the fact that there is still much to learn about this important social problem. In order to develop effective prevention and intervention programs for cyberbullying, and to continue developing effective legislation, more information on cyberbullying is needed. In this chapter, we describe the prevalence of cyberbullying and cybervictimization in Canada, as well as known risk and protective factors. We also describe the state of evidence on cyberbullying prevention and intervention programs in Canada, closing with recommendations for advancing this important work. The Prevalence of Cyberbullying and Cybervictimization in Canada The national rate of cyberbullying in Canada ranges from 4.5% of university-aged youth engaging in cyberbullying at any time (Cunningham et al. 2015) to 33.7% of adolescents in the past three months (Mishna Cyberbullying in Canada 41 et al. 2010). Estimates of cybervictimization prevalence in university range from 5.7% of youth being cybervictimized at any time (Cunningham et al. 2015) to 49.5% of adolescents reporting cybervictimization in the past three months (Mishna et al. 2010). These prevalence rates vary greatly due to sample characteristics, including the age, gender, and geographic location of the individuals sampled (e.g., province, language spoken, urban or rural location). Canada is a large country with provincial jurisdiction that has both urban and rural areas within each province. Therefore, cultural differences between urban and rural areas may partly explain the large differences in prevalence rates of cyberbullying and cybervictimization across studies. Further, there are methodological differences in the studies such as the time period of reporting, response rate, method of sampling, and whether or not a definition of cyberbullying was provided. Each of these key features will be explored in detail to explain differences in rates of cyberbullying and cybervictimization in Canada. Table 3.1 presents a summary of the studies in Canada examining the prevalence of cyberbullying and cybervictimization. Differences in Prevalence Due to Sample Characteristics Age Most studies of the prevalence of cyberbullying and cybervictimization in Canada have been conducted with adolescents. The estimated prevalence of cyberbullying in adolescence ranges from 7% to 33.7% over the past few months, and the estimated prevalence of cybervictimization ranges from 5.1% to 49.5% over the past few months. When Boak et al. (2016) sampled 10,426 students aged 12 to 18, they found that the prevalence of cybervictimization ranged from 19.0% to 21.3% over the past year. The highest rate of cybervictimization occurred when youth were about 15 years old (21.3%), although the differences in age were not statistically significant. A recent systematic review of the prevalence of cyberbullying in Canada indicated that prevalence rates tended to increase over childhood, with the highest rates of victimization reported when youth 42 J. Riddell et al. Table 3.1 Prevalence of cyberbullying and cybervictimization in Canada Sample N Boak et al. (2016) using OSDUHS Craig et al. (2016) HBSC 2014 data 10,426 12 to 18 Not reported 25.8% for girls; 14.0% for boys Past year 29,784 11 to 15 7% (as calculated by the authors of this chapter) 4.5% 8% to 12% for boys; 14% to 22% for girls; 14.4% overall Past couple of months 5.7% 10.2% at Time 1; 13% at Time 2 17% of boys; 12% of girls 22.0% at Time 1; 26.8% at Time 2 During university Past 30 days Cunningham 1,004 et al. (2015) Holfeld and 714 Leadbeater (2015) Li and Craig (2015) 800 8,194 Cénat et al. (2014) using QYRRS Cappadocia 1,972 et al. (2013) using HBSC 2006 and 2007 data Ages Prevalence of Prevalence of Time period cyberbullying cybervictimization of reporting Source ~18 10 to 12 12 to 18 14 to 20 Not reported 42% overall; 45% of boys; 38% of girls 26.4% for girls; 18.1% for boys 5.1% at Time 1 only; 6.5% at Time 2 only; 1.9% at both times 12% to 19% for 25,000 15 to Perreault ages 15 to 17; 17, (2011) 17% for ages 18 18 to using 2009 to 24; 24, Statistics 5% for ages 25 25 and Canada and up up data Mishna et al. 2,186 11 to 33.7% overall 49.5% overall (2010) 16 Vaillancourt 16,799 9 to 9.7% 12.4% et al. (2010) 18 14 to 18 4.9% at Time 1 only; 4.7% at Time 2 only; 1.9% at both times Not reported Past 4 weeks Past 12 months Past 2 months Ever (lifetime prevalence) Past 3 months Past 3 months (continued) Cyberbullying in Canada 43 Table 3.1 (continued) Sample Ages Prevalence of Prevalence of Time period cyberbullying cybervictimization of reporting Source N Estimated national prevalence 96,879 9 to 4.5% to adult 33.7% 5.1% to 49.5% Past 30 days to ever begin high school around age 14 (Bilsbury 2015). This finding is consistent with the results of the 2014 Health Behavior in School-aged Children (HBSC) survey, which indicated that the highest rate of cybervictimization occurred during the first year of high school (Craig et al. 2016). Averaged across genders, the HBSC data revealed that the rate of cybervictimization increased from 15% at age 13 to 17% at age 14, and then decreased back to 15% by age 15 (Craig et al. 2016). Other Canadian studies, however, have found the opposite trend with respect to age. Li and Craig (2015) conducted a study with a representative sample of 800 youth aged 12 to 18 who completed an online survey. They found that older youth were significantly more likely than younger youth to be cybervictimized in the past four weeks (47% of youth aged 17 to 18, compared to 36% of youth aged 12 to 14). These differences may be due to methodological differences between the studies, which will be discussed in detail in the section below. Holfeld and Leadbeater (2015) conducted a study with a sample of preadolescent children aged 10 to 12. They asked these children to complete measures at two time points, corresponding with the beginning and end of the school year. At the beginning of the school year, 10.2% of children reported engaging in one or more cyberbullying behaviors, compared to 13% of children at the end of the year. About twice this number of children (22.0%) reported being cybervictimized at the beginning of the school year and 26.8% reported being cybervictimized at the end of the year. These prevalence rates of cyberbullying and cybervictimization in children appear to be consistent with the prevalence rates in adolescents. Based on the two Canadian studies measuring cyberbullying and cybervictimization in adults, the prevalence appears to be lower for emerging adults (aged 18 to 24) and adults (over age 25) compared to 44 J. Riddell et al. children and adolescents. Cunningham et al. (2015) conducted a survey with 1,004 students in a first-year university class. In this study, 5.7% of the students reported that they had experienced cybervictimization at some point in university, 4.5% had engaged in cyberbullying, and 4.9% had both cyberbullied others and been cybervictimized. Higher prevalence rates for this age group were reported by Perreault (2011) using a large nationally representative survey with a sample of 25,000 participants. Results indicated that 17% of young adults (aged 18 to 24) had been cybervictimized in their lifetime. This study provided the only prevalence rate for cybervictimization in adults over the age of 25 (5% lifetime prevalence), which is much lower than the prevalence rates reported for adolescents (Perreault 2011). This may be related to increased use of social media over time. In conclusion, the rates of cybervictimization in Canada are the highest in adolescence and drop significantly during young adulthood and adulthood. Gender Most Canadian studies suggest that the rates of cybervictimization are higher for girls than for boys. In her systematic review, Bilsbury (2015) determined that there were small but significant sex differences such that girls were more likely to be cybervictimized than boys. Craig et al. (2016) found that girls reported much higher rates of cybervictimization (18.8% over the past few months) compared to boys (10% over the past few months). Two other studies indicated a significantly higher proportion of girls compared to boys reporting cybervictimization over the past year. Boak et al. (2016) found that 25.8% of girls compared to 14.0% of boys reported being cybervictimized in the past year. Similarly, Cénat et al. (2014) found that 26.4% of girls and 18.1% of boys reported being cybervictimized. Much less is known about gender differences in perpetrating cyberbullying. Li and Craig (2015) found that boys were significantly more likely to report engaging in cyberbullying in the past four weeks (17% of boys compared to 12% of girls). Further, Cunningham et al. (2015) found that young men in their first year of university were more likely than young women to report that they both cyberbullied others and were cybervictimized. Cyberbullying in Canada 45 Urban or Rural Location Holfeld and Leadbeater (2015) conducted a study of cyberbullying and cybervictimization with 714 children from predominately rural areas across Canada. At the beginning and end of the school year, these children were asked how many times they engaged in four cyberbullying behaviors. Averaged across the two time points, 11.6% of students reported engaging in one or more cyberbullying behaviors, and 24.4% reported being cybervictimized in the past 30 days. Mishna et al. (2010) used an urban sample of youth and found that 49.5% of students had experienced behaviors that the researchers identified as cybervictimization in the past three months. Further, 33.7% of participants indicated that they had cyberbullied others in the past three months. The differences in the reporting period make it difficult to compare across these two studies, as discussed in detail below. Differences in Prevalence Due to Study Methodology Time Period of Reporting As shown in Table 3.1, the studies reviewed in this chapter vary greatly in their reporting period, with some youth being asked to report experiences of cyberbullying/cybervictimization over the past 30 days and others reporting a lifetime prevalence. These differences in the reporting period present challenges in comparing the prevalence rates across studies. Recently, a systematic review of the prevalence of cyberbullying in Canada was completed based on 45 studies (Bilsbury 2015). An overall estimate of prevalence could not be calculated due to heterogeneity in the definition of cyberbullying used, sample characteristics, and length of the reporting period. Instead, a summary of each study was provided, with studies divided into two categories based on the length of the reporting period (3 months or less, and 4 to 12 months). For studies that reported cyberbullying over the past 3 months, estimates ranged from 2% to 28% of youth engaging in cyberbullying and between 4% and 38% of students being cybervictimized. For studies reporting cyberbullying over the past 12 months, estimates ranged from 2% to 34% of youth engaging in 46 J. Riddell et al. cyberbullying and between 4% and 39% of students being cybervictimized. Adopting a standardized reporting period for all cyberbullying studies in Canada would be helpful to calculate a national prevalence rate and compare across studies with different sample characteristics. Response Rate and Method of Sampling One study with particularly high prevalence rates was conducted by Mishna et al. (2010). They found that 49.5% of students reported being cybervictimized in the past three months and 33.7% of participants reported having cyberbullied others. The response rate on this survey was low (35% in Grades 6 and 7 and 17% in Grades 10 and 11); therefore, it is possible that students involved in cyberbullying were more likely to participate than those not involved. The prevalence rates reported by Mishna et al. (2010) are notably higher than those reported by Perreault (2011), in which the response rate was 61.6%, and by Craig et al. (2016), in which the response rate was 77%. Therefore, the response rate and method of selection may greatly impact the prevalence rates reported. Providing a Definition of Cyberbullying The studies presented in Table 3.1 vary in terms of whether or not they provided a definition of cyberbullying as part of the questionnaire. Vaillancourt et al. (2010) provided students with a definition of bullying and then asked about their experiences of bullying using two questions from the Olweus Bully/Victim Questionnaire (1996). Li and Craig (2015) defined cyberbullying as being threatened, embarrassed, gossiped about, or made to look bad online. Cunningham et al. (2015) asked young adults to read a definition of cyberbullying before rating their involvement in ten randomly presented electronic formats (Facebook, YouTube, Twitter, text messages, etc.) as a witness, perpetrator, or victim. As part of the HBSC survey used by Cappadocia et al. (2013) and Craig et al. (2016), participants were provided with a standard definition of bullying including three main components: intention to harm, repetition, and power differential (Cappadocia et al. 2013; Craig et al. 2016). Prevalence rates across these studies varied widely, as can be seen in Cyberbullying in Canada 47 Table 3.1. Of note, Mishna et al. (2012) used a questionnaire that contained a number of questions about perpetrating or being victimized by different types of online behaviors, without explicitly defining the behaviors as bullying. In this study, 49.5% of students indicated that they had experienced behaviors that the researchers identified as cybervictimization in the past three months, and 33.7% of participants indicated that they had engaging in cyberbullying. The overall prevalence rates of cyberbullying and cybervictimization reported in this study were higher than most other studies. It is possible that since the behaviors were not labeled as bullying, students were more likely to endorse them. Other Considerations Related to Cyberbullying Prevalence Stability Over Time Cappadocia et al. (2013) examined the stability of prevalence estimates using two waves of data collected one year apart. In terms of engaging in cyberbullying, 4.9% reported cyberbullying others at Time 1 only, 4.7% at Time 2 only, and 1.9% at both time points. In terms of experiencing cybervictimization, 5.1% reported being victimized at Time 1 only, 6.5% at Time 2 only, and 1.9% at both time points. Although the rates of cyberbullying and cybervictimization between the two time points were similar, less than half of the youth involved at one time point were involved at both time points, suggesting that short-term estimates of cyberbullying are not completely stable over the school year. Another consideration is how stable the Canadian rates of cyberbullying have been over the past few years. A report by Boak and colleagues examined the well-being of students in Ontario using The Centre for Addiction and Mental Health’s Ontario Student Drug Use and Health Survey (OSDUHS), which has been conducted every two years since 1977 (Boak et al. 2016). The authors noted that the overall percentage of students who reported being cybervictimized in 2015 (19.8%) was not significantly different than the percentages in 2013 (19%) and in 2011 (21.6%; Boak et al. 2016). One of the most representative sources of information on cyberbullying among youth across Canada is the HBSC 48 J. Riddell et al. survey, which is conducted in collaboration with the World Health Organization every four years. In the 2006 edition of this survey, students aged 11 to 15 were asked how often they cyberbullied another student using a computer, email messages or pictures, or a mobile phone in the past couple of months. In this survey, 4.8% of Canadian youth reported engaging in cyberbullying and 5.8% of youth reported being cybervictimized over the past two months (Cappadocia et al. 2013). These prevalence rates are slightly lower than those found in the 2014 edition of the HBSC survey, which indicated that across grades and gender 14.4% of youth reported having been cybervictimized in the past few months (Craig et al. 2016). More information is needed on changes in cyberbullying and cybervictimization over time. Involvement in Both Cyberbullying and Cybervictimization Most research suggests that youth who engage in cyberbullying have also been involved in cybervictimization. Using a nationally representative sample, Craig et al. (2016) found that 4.8% of youth reported that they had both cyberbullied others and been cybervictimized in the past two months (compared to 7% of students who reported only cyberbullying others). Similarly, Cappadocia et al. (2013) found that some students were both cybervictimized and engaged in cyberbullying over the past two months: 1.4% at Time 1 only, 2.7% at Time 2 only, and 0.5% at both time points (Cappadocia et al. 2013). A nationally representative study with a smaller sample found that youth who reported that they had been cybervictimized at least once were also significantly more likely to report they had cyberbullied others (32%) compared to youth who had not been cybervictimized (2%; Li and Craig 2015). Type of Cyberbullying Studies have shown that some types of cyberbullying are more prevalent than others. Holfeld and Leadbeater (2015) found that the most common types of cybervictimization were “received a text message that made you upset or uncomfortable” and “someone posted something on your online page or wall that made you upset or uncomfortable.” Mishna et al. (2010) Cyberbullying in Canada 49 identified the most common cyberbullying experiences as being called names (27%), having rumors spread about the participant (22%), having someone pretend to be the participant (18%), being threatened (11%), and receiving unwelcome sexual photos or text messages (10%). The main cyberbullying behaviors that students endorsed were calling someone names (22%), pretending to be someone else (14%), and spreading rumors about someone (11%). The majority of this cyberbullying took place through instant messages or email, with less cyberbullying occurring over Internet games (12%) or on social networking sites (10%). Understanding the type of cyberbullying and where it occurs is helpful in designing interventions to support those who engage in cyberbullying and those experiencing cybervictimization. Knowledge about the nature of cyberbullying behaviors over particular platforms will need to be continually updated to keep pace with ongoing changes in technology and user preferences. Risk and Protective Factors Six Canadian studies have examined the risk and protective factors associated with cyberbullying and cybervictimization. These risk and protective factors have been organized using a social ecological framework, with a discussion of factors at the individual, family, school, and neighborhood level. Table 3.2 contains a summary of the risk and protective factors associated with cyberbullying and cybervictimization in Canadian children, organized by participant age. Individual-Level Risk and Protective Factors Internet-Use Characteristics Three studies investigated whether particular Internet-use characteristics were related to cyberbullying and cybervictimization (Cappadocia et al. 2013; Mishna et al. 2012; Perreault 2011). Mishna et al. (2012) found that compared to a reference group of uninvolved children, children involved in cyberbullying were more likely to use the computer for more 50 Table 3.2 Risk and protective factors associated with cyberbullying and cybervictimization in Canada Sample N Ages Dafoe (2016) 193 ~12 to 13 Schumann et al. (2014) using 2010 HBSC data 17,777 11 to 15 Cappadocia et al. (2013) using 2006 HBSC data 1,972 14 to 16 Dittrick et al. (2013) 492 children 10 to 17 397 parents Risk factors for cybervictimization Risk factors for cyberbullying Protective factors for being cybervictimized Having more friends; Poor self-awareness Individual-level social capital; Low SES – – – Being male, South Asian, and older; Higher collective efficacy; Community-based opportunities for recreation – Higher levels of traditional victimization; Higher levels of depression; Being in the transition year for high school (Grade 9) – Involvement in traditional bullying; Alcohol use; Fewer prosocial peers Playing highly violent – and mature video games (continued) J. Riddell et al. Source Table 3.2 (continued) Sample Source N Mishna et al. (2012) 2,186 Perreault (2011) using 2009 Statistics Canada data 25,000 Ages ~11 to 16 Risk factors for cybervictimization Protective factors for being cybervictimized More time on computer per day; Giving passwords to friends; Violence toward peers at school; Greater age – – Being Francophone; Trusting family relationships Cyberbullying in Canada More time on computer per day; Giving passwords to friends; Violence toward peers at school; Having parents who speak English at home 15 to 17, Using chat sites or social 18 to 24, networking sites; Being single, separated, 25 and or divorced; up Younger age (ages 18 to 24); Identifying as bisexual or homosexual; Having an activity limitation Risk factors for cyberbullying 51 52 J. Riddell et al. hours a day and more likely to give their passwords to friends. These findings were consistent across three groups: those who had been cybervictimized, those engaged in cyberbullying, and those who were involved with both cyberbullying and cybervictimization. Perreault (2011) found that adolescents and adults who used chat sites or social networking sites were almost three times more likely than non-users to report being cybervictimized. Conversely, Cappadocia et al. (2013) did not find the total amount of time spent online to be a significant predictor of cyberbullying or cybervictimization. Demographic Variables A number of demographic characteristics were associated with a higher risk of experiencing cybervictimization. In one study, these risk factors included being single, being homosexual or bisexual, and having a longterm physical or mental health condition (Perreault 2011). Specifically, individuals who were single, separated, or divorced were more likely than married or common-law individuals to have been cybervictimized. Among Internet users, 24% of those who identified as bisexual and 18% of those who identified as homosexual were cybervictimized compared with only 7% of individuals who identified as heterosexual. Lastly, people who were limited in the amount or type of activity they could do because of a long-term physical or mental health problem were more likely than those with no condition to report having been cybervictimized (22% of those with a condition compared to 10% of those with no condition). A few Canadian studies have explored risk factors associated with ethnicity, language use, and immigrant status. In one study, being South Asian was identified as a protective factor against cybervictimization (Schumann et al. 2014). Mishna et al. (2012) found that compared to a reference group of uninvolved children, children involved in cyberbullying were more likely to have parents who speak English at home. Further, English-speaking adolescents and adults were more likely to report being cybervictimized than French-speaking individuals (Perreault 2011). The reasons why English-speaking individuals are at greater risk for involvement in cyberbullying and cybervictimization is unclear. Cyberbullying in Canada 53 Consistent with the gender-based prevalence rates discussed above, young women are more likely to be involved in cyberbullying and cybervictimization compared to young men. Schumann et al. (2014) identified being male as a protective factor against cybervictimization, and Mishna et al. (2012) identified being female as a risk factor for cyberbullying others and being cybervictimized. As reported in the prevalence section above, being an adolescent is linked to the highest levels of cybervictimization. Two studies illustrated that youth aged 14 to 16 were more likely to cyberbully others and be cybervictimized compared to youth aged 11 and 12 (Mishna et al. 2012; Cappadocia et al. 2013). Further, Perreault (2011) found that young adults between 18 and 24 years of age were about three times more likely than those aged 25 and over to report having been cybervictimized. Within adolescence, however, it is unclear when the rates of cybervictimization are the highest. Cappadocia et al. (2013) found that students in a lower grade (i.e., Grade 9) at Time 1 were more likely to be involved in cybervictimization at Time 2, suggesting that the rates of cybervictimization increase over the first year of high school. Another study using the 2010 HBSC dataset found the opposite—being older (ages 14 to 15) was a protective factor against cybervictimization (Schumann et al. 2014). More information on the role of age in cybervictimization is needed. Socio-emotional and Behavioral Variables Cappadocia et al. (2013) analyzed one-year longitudinal data from the 2006 HBSC survey to examine risk factors associated with cyberbullying among Canadian children. The main individual-level risk factor they identified for engaging in cyberbullying was alcohol use, and the main individual-level risk factor for cybervictimization was depression. In a recent doctoral dissertation, Dafoe (2016) asked 193 students aged 12 and 13 from a large urban area to complete questionnaires related to social-emotional learning (i.e., self-awareness, self-management, social awareness, relationship skills, and responsible decision-making) and a range of bullying behaviors (i.e., physical, verbal, social, sexual, and cyber). Results of this study suggested that having poor self-awareness was associated with a higher level of cybervictimization (Dafoe 2016). 54 J. Riddell et al. Other Individual-Level Factors Dittrick and colleagues examined the association between cyberbullying and violent video games using a stratified random sample of 492 Canadian children aged 10 to 17 years, as well as 397 of their parents (Dittrick et al. 2013). Parents and children completed an online survey of children’s bullying behaviors and listed their three favorite video games. Dittrick and colleagues determined that playing highly violent and mature video games was associated with cyberbullying according to both parent and child reports. Schumann and colleagues used the 2010 edition of the HBSC to examine individual-level variables associated with both traditional bullying and cyberbullying (Schumann et al. 2014). The two individual-level variables that predicted cybervictimization were individual-level social capital (involvement in clubs, groups, sports or volunteering) and low socioeconomic status (i.e. poverty). This study also replicated the high overlap between children involved in cyberbullying and cybervictimization (e.g., Cappadocia et al. 2013; Craig et al. 2016; Li and Craig 2015). Risk and Protective Factors Associated with the Family A few Canadian studies have investigated family-level factors associated with a reduced risk of cyberbullying and cybervictimization. In particular, trusting family relationships have been identified as a protective factor against cybervictimization (Perreault 2011). Internet users who indicated that they can trust people in their family “a lot” were less likely to be cybervictimized than those who indicated that they could “more or less” trust them (6% compared to 13%; Perreault 2011). Further, childreported frequency of parent involvement in school activities was identified as an important protective factor for cybervictimization (Leadbeater et al. 2015). Conversely, Cappadocia et al. (2013) tested whether parental trust and communication, parental support, and parental involvement with school were related to cyberbullying and cybervictimization, and did not find significant results. Mishna et al. (2012) found that compared to a reference group of uninvolved children, children involved in cyber- Cyberbullying in Canada 55 bullying were more likely to have parents who supervised their Internet use and had blocking programs for the Internet. The direction of effect for parent behavior is not clear: parents of youth engaged in cyberbullying may have been stricter with monitoring Internet activity prior to involvement in cyberbullying, or parents may have used this software as a result of youths’ involvement in cyberbullying. Overall, there are few studies examining family factors associated with cybervictimization, and those that have been conducted have demonstrated inconsistent results with respect to the association between cybervictimization and the quality of relationships in the family. Risk and Protective Factors Associated with Peers and School The role of relationships with teachers has been investigated as a potential risk and protective factor for involvement in cyberbullying and cybervictimization. Positive student–teacher relationships were identified as a protective factor against cybervictimization (Leadbeater et al. 2015). Conversely, poor relationships with teachers were found to be both a risk factor and a consequence of peer cybervictimization. That is, parent ratings of student–teacher relationships were reciprocally related to their children’s reports of peer victimization from the beginning to the end of the school year (Leadbeater et al. 2015). These findings illustrate the importance of relationships with teachers in protecting against cybervictimization. Relationships with peers have also been linked with involvement in cyberbullying and cybervictimization. Mishna et al. (2012) found that compared to a reference group, children involved in cyberbullying were more likely to act violently toward peers at school. Similarly, Cappadocia et al. (2013) found that involvement in traditional bullying was a risk factor for involvement in both cyberbullying and cybervictimization. In terms of cybervictimization, the authors determined that students who reported higher levels of traditional victimization at Time 1 were more likely to be involved in cybervictimization at Time 2. Further, students 56 J. Riddell et al. who reported traditional victimization at Time 1 were almost four times more likely than peers who had not been victimized to report simultaneous cyberbullying and cybervictimization at Time 2 (Cappadocia et al. 2013). In this study, one of the main risk factors linked to higher levels of cyberbullying was having fewer prosocial peers (Cappadocia et al. 2013). Conversely, Dafoe (2016) found that having more close friends was associated with a higher level of cybervictimization. This discrepancy in findings may be due to the influence of other peer relationship factors, such as the quality of relationships, peer culture, and status within that peer group. This suggestion is consistent with the Canadian research on in-person bullying which found that low friendship quality was a significant predictor of subsequent in-person victimization, and having highquality friendships was a protective factor against victimization (Goldbaum et al. 2003). Therefore, models of cybervictimization need to consider multiple peer relationship factors, including the number and quality of friendships, social status, peer culture, and wider school culture, as well as involvement in in-person or traditional forms of bullying. Neighborhood and Societal-Level Risk and Protective Factors Using the 2010 edition of the HBSC, Schumann and colleagues examined a number of community-level variables associated with both traditional bullying and cyberbullying (Schumann et al. 2014). They examined the following community-level variables: community socioeconomic status (SES), built social capital, built recreational opportunity, community stability, and population density. Only one community-level factor was significantly associated with cybervictimization: community recreation. In this study, high levels of community recreation were a protective factor associated with lower levels of cybervictimization. Further, Schumann et al. (2014) measured how students felt about the safety, cooperation, and trust that existed in their neighborhoods, which they termed collective efficacy. Results indicated that higher collective efficacy was associated with Cyberbullying in Canada 57 lower rates of cybervictimization. As there is only one known Canadian study exploring the impact of neighborhood and societal-level variables, more research is needed in order to effectively inform public policy. Cyberbullying and Cybervictimization Intervention Programs Many cyberbullying and digital citizenship resources and programs have been developed in Canada including The Canadian Red Cross Respect Education resources, MediaSmarts resources, Canadian government resources, and some provincial Ministries of Education programs. There are, however, no evaluations of these resources and programs available in the literature. The authors located two international systematic reviews on cyberbullying interventions that contained at least one Canadian intervention (Della Cioppa et al. 2015; Mishna et al. 2009a). Both of these systematic reviews contained one Canadian study, which is an evaluation of the Missing cyber safety program (Crombie and Trinneer 2003). Since this evaluation was summarized in a report submitted to a national funding agency and was not publicly available, the results discussed below are from the systematic review by Mishna and colleague (2009). The Missing cyber safety program included an interactive computer game designed to encourage youth to develop guidelines for safe Internet use. In this game, youth assumed the role of a police officer and solved a series of puzzles to find a missing teenager who was lured away from home by an Internet predator (Crombie and Trinneer 2003). The goal of the game was to highlight that revealing personal information about oneself on the Internet may lead to cybervictimization. The outcomes included the frequency of disclosing personal information online, attitudes regarding the safety of disclosing personal information online, and attitudes about trusting people met online. Crombie and Trinneer (2003) measured these behaviors and attitudes before and three weeks after the intervention. As part of their systematic review, Mishna and colleagues (2009) calculated effect size measures for all 64 individual behaviors and attitudes 58 J. Riddell et al. measures; they then conducted z-tests to determine whether the effect sizes were significantly different between the intervention and control group. Only three variables were significant: disclosing one’s gender to a stranger in a chat room or by email, disclosing one’s age to a stranger in a chat room or by email, and posting one’s school name on a web page. That is, youth who participated in the Missing program showed a greater reduction in these three specific behaviors compared to the control group. There was no significant change in the other 61 behaviors and attitudes measured in this study, leading the authors of the systematic review to conclude that the program did not result in significant change overall (Mishna et al. 2009a, b). In summary, there are few documented cyber intervention programs in Canada, and those that do exist are old and/or have not been evaluated. This lack of evidence-based programming specifically for cyberbullying may be due to the fact that there is a strong link between involvement in traditional bullying and involvement in cyberbullying, as described in the prevalence and risk sections above. Therefore, it may not be necessary to develop interventions specifically targeting cyberbullying. Instead, the most effective strategy to address cyberbullying may be to focus on creating healthy relationships in general (Mishna et al. 2012; Pepler 2006; Craig and Pepler 2007). This approach includes promoting healthy relationships with peers, teachers, family members, and neighbors. To design effective cyberbullying interventions, it is important to consider youths’ voices in terms of the kinds of interventions they would find most helpful. A few recent studies have examined youths’ preferences for cyberbullying programs using discrete choice conjoint experiments: one with university students (Cunningham et al. 2015) and another with younger children (Cunningham et al. 2011). In the study by Cunningham et al. (2015), the participants were 1,004 university students in an introductory psychology class. While the response rate for this study was excellent (95.7%), the sample is not representative of all young adults in Canada. More than 90% of the sample expressed a preference for a comprehensive approach that included teaching strategies to prevent cyberbullying, encouraging anonymous reporting, and imposing consequences when cyberbullying is detected. In particular, students preferred a policy that encourages but does not require students to report cyberbullying. Cyberbullying in Canada 59 Students also expressed a preference for the university suspending Internet privileges of students who cyberbully. The preference for comprehensive programs combining prevention and consequences is consistent with the recommendations of younger students (Cunningham et al. 2011). Conclusion Estimates of the prevalence of cyberbullying and cybervictimization in Canada vary widely depending on a multitude of sample characteristics and methodological considerations. In particular, the differences in the reporting period used across studies make it impossible to calculate an overall estimate of cyberbullying prevalence in Canada. Adopting a standardized reporting period for all studies in Canada would make it possible to calculate a national prevalence rate and to compare across studies with different sample characteristics. Other methodological differences between studies are currently a barrier to developing a fulsome understanding of cyberbullying in Canada. Standardized assessment and regular monitoring of cyberbullying are needed to understand the extent of the problem, identify risk and protective factors, and guide the development of prevention and intervention strategies. Regardless of the reporting period assessed, cyberbullying is a significant problem in Canada. There are common risk factors for cyberbullying and cybervictimization, most notably, involvement in traditional bullying. There have been limited studies on protective factors in Canada, and ones that have been conducted are not consistent in their findings. Therefore, more research on this important topic is needed. The few studies on protective factors that have been conducted in Canada have highlighted the importance of close relationships with teachers and parents, highlighting the need for education for adults on how to cultivate healthy relationships with youth. High levels of community recreation were shown to be a protective factor against cybervictimization. It may therefore be worthwhile to invest in community recreation programs for youth as a cyberbullying prevention strategy, particularly if these programs are run by supportive adults who can create safer spaces for youth. Collective efficacy (how students feel about the safety, cooperation, and trust in 60 J. Riddell et al. their neighborhoods) is a protective factor against cybervictimization. Each of us has a role in increasing neighborhood cooperation, and we can all play a part in reducing cyberbullying. Finally, lower individual social capital and low socioeconomic status predict cybervictimization. There is a need to address inequities and structural barriers that may contribute to cyberbullying and cybervictimization, such as poverty and classism. Moving forward, there is a need to identify youth involved in cyberbullying and cybervictimization to provide support, as well as a need to teach strategies for online bystander intervention. Canada must develop and implement evidenced-based interventions to address the problem; however, given the dearth of research, we may need to rely on the research conducted in other countries to guide the development of these interventions. If we are to improve the lives of youth and reduce the risk of other tragedies due to cybervictimization, we need to start this work immediately. Currently, there is a lack of effective programming to identify vulnerable youth and ensure that they are stabilized and supported. Without these interventions, the rates of cyberbullying and cybervictimization are likely to remain stable over time. We cannot truly address cyberbullying until we address racism, sexism, homophobia, ableism, and fat phobia. When we look carefully at the content of cyberbullying messages, they commonly fall along these societal lines of discrimination and marginalization. Students learn to use the word “gay” as an insult because they have heard adults do this. Children cyberbully each other by making comments about each other’s physical appearance and weight because our culture has modeled this for them. Children learn to exclude and vilify those who are different because, as adults, we do this, usually in more subtle and insidious ways. As is the case with other forms of bullying, cyberbullying involves a power imbalance between individuals. To truly address cyberbullying, we need to think about the ways in which we use power over others as opposed to sharing power with others. These different ways of negotiating power between individuals shape the structures in our society and send powerful messages to children. To eradicate cyberbullying, we need more public education on how to have respectful relationships with each other—in the workplace, at the grocery store, at the dinner table, in our places of worship, and Cyberbullying in Canada 61 through digital media. If we truly want to end cyberbullying, we need a cultural shift in how we interact with each other each and every day. References BBC News. (2014). Man charged in Netherlands in Amanda Todd suicide case. Retrieved from http://www.bbc.com/news/world-europe-27076991 Bilsbury, T. (2015). A systematic review of the prevalence of cyberbullying in Canada (Master’s thesis). Dalhousie University. Retrieved from http://hdl. handle.net/10222/58197 Boak, A., Hamilton, H. A., Adlaf, E. M., Henderson, J. L., & Mann, R. E. (2016). The mental health and well-being of Ontario students, 1991– 2015: Detailed OSDUHS findings (CAMH Research Document Series No. 43). Toronto: Centre for Addiction and Mental Health. Retrieved from https://www.camh. ca/en/research/news_and_publications/ontario-student-drug-use-and healthsurvey/Documents/2015%20OSDUHS%20Documents/2015OSDUHS_ Detailed%20MentalHealthReport.pdf Cappadocia, M. C., Craig, W. M., & Pepler, D. (2013). Cyberbullying prevalence, stability, and risk factors during adolescence. Canadian Journal of School Psychology, 28(2), 171–192. https://doi.org/10.1177/0829573513491212 Cénat, J. M., Hébert, M., Blais, M., Lavoie, F., Guerrier, M., & Derivois, D. (2014). Cyberbullying, psychological distress and self-esteem among youth in Quebec schools. Journal of Affective Disorders, 169, 7–9. https://doi. org/10.1016/j.jad.2014.07.019 Craig, W. M., & Pepler, D. J. (2007). Understanding bullying: From research to practice. Canadian Psychology, 48(2), 86–93. https://doi.org/10.1037/ cp2007010 Craig, W., Lambe, L., & McIver, T. (2016). Bullying and fighting. In J.G. Freeman, M. A. King, & W. Pickett (Eds.), Health Behaviour in Schoolaged Children (HBSC) in Canada: Focus on relationships (pp. 167–182). Ottawa: Public Health Agency of Canada. Retrieved from http://healthycanadians.gc.ca/publications/science-research-sciences-recherches/health behaviour-children-canada-2015-comportements-sante-jeunes/index-eng. php Crombie, G., & Trinneer, A. (2003). Children and internet safety: An evaluation of the missing program. A report to the Research and Evaluation Section of the 62 J. Riddell et al. National Crime Prevention Centre of Justice Canada. Ottawa: University of Ottawa. Cunningham, C. E., Vaillancourt, T., Cunningham, L. J., Chen, Y., & Ratcliffe, J. (2011). Modeling the bullying prevention program design recommendations of students from grades 5 to 8: A discrete choice conjoint experiment. Aggressive Behavior, 37(6), 521–537. https://doi.org/10.1002/ab.20408 Cunningham, C. E., Chen, Y., Vaillancourt, T., Rimas, H., Deal, K., Cunningham, L. J., & Ratcliffe, J. (2015). Modeling the anti-cyberbullying preferences of university students: Adaptive choice-based conjoint analysis. Aggressive Behavior, 41(4), 369–385. https://doi.org/10.1002/ab.21560 Dafoe, T. L. (2016). The role of social-emotional learning skills in bullying behaviour (Doctoral dissertation). University of Toronto. Della Cioppa, V., O’Neil, A., & Craig, W. (2015). Learning from traditional bullying interventions: A review of research on cyberbullying and best practice. Aggression and Violent Behavior, 23, 61–68. https://doi.org/10.1016/j. avb.2015.05.009 Dittrick, C. J., Beran, T. N., Mishna, F., Hetherington, R., & Shariff, S. (2013). Do children who bully their peers also play violent video games? A Canadian national study. Journal of School Violence, 12(4), 297–318. https://doi.org/10 .1080/15388220.2013.803244 Education Act, Revised Statutes of Alberta. (2012, Chapter E-0.3). Retrieved from the Province of Alberta website: http://www.qp.alberta.ca/documents/ Acts/e00p3.pdf Gillis, W. (2013, April 12). Rehtaeh Parsons: A family’s tragedy and a town’s shame. The Star. Retrieved from https://www.thestar.com/news/canada/2013/04/12/rehtaeh_parsons_a_ family s_ tragedy_ and_a_towns_ shame.html Goldbaum, S., Craig, W. M., Pepler, D., & Connolly, J. (2003). Developmental trajectories of victimization: Identifying risk and protective factors. Journal of Applied School Psychology, 19(2), 139–156. https://doi.org/10.1080/1538822 0.2013.803244 Holfeld, B., & Leadbeater, B. J. (2015). The nature and frequency of cyber bullying behaviors and victimization experiences in young Canadian children. Canadian Journal of School Psychology, 30(2), 116–135. https://doi. org/10.1177/0829573514556853 Leadbeater, B., Sukhawathanakul, P., Smith, D., & Bowen, F. (2015). Reciprocal associations between interpersonal and values dimensions of school climate and peer victimization in elementary school children. Journal of Clinical Cyberbullying in Canada 63 Child and Adolescent Psychology, 44(3), 480–493. https://doi.org/10.1080/15 374416.2013.873985 Li, J., & Craig, W. (2015). Young Canadians’ experiences with electronic bullying. Retrieved from http://mediasmarts.ca/sites/mediasmarts/files/publicationreport/full/young-canadians-electronic- bullying.pdf Mishna, F., Cook, C., Saini, M., Wu, M., & MacFadden, R. (2009a). Interventions for children, youth, and parents to prevent and reduce cyber abuse. Campbell Systematic Reviews, 2009, 2. Mishna, F., Saini, M., & Solomon, S. (2009b). Ongoing and online: Children and youth’s perceptions of cyberbullying. Children and Youth Services Review, 31, 1222–1228. https://doi.org/10.1016/j.childyouth.2009.05.004 Mishna, F., Cook, C., Gadalla, T., Daciuk, J., & Solomon, S. (2010). Cyber bullying behaviors among middle and high school students. American Journal of Orthopsychiatry, 80(3), 362–374. https://doi.org/10.1111/j.1939-0025. 2010.01040.x Mishna, F., Khoury-Kassabri, M., Gadalla, T., & Daciuk, J. (2012). Risk factors for involvement in cyberbullying: Victims, bullies and bully-victims. Children &Youth Services Review, 34(1), 63–70. https://doi.org/10.1016/j.childyouth. 2011.08.032 Olweus, D. (1996). The revised bully/victim questionnaire for students. Bergen: University of Bergen. Pepler, D. J. (2006). Bullying interventions: A binocular perspective. Journal of the Canadian Academy of Child and Adolescent Psychiatry, 15(1), 16–20. Perreault, S. (2011). Self-reported Internet victimization in Canada, 2009 (Statistics Canada catalogue No. 85-002-X). Retrieved from http://www.statcan.gc.ca/pub/85-002 x/2011001 /article/11530-eng.htm PREVNet. (2015). For Parents. Retrieved from http://www.prevnet.ca/resources/ policy-and-legislation/ontario/for-parents Schumann, L., Craig, W., & Rosu, A. (2014). Power differentials in bullying: Individuals in a community context. Journal of Interpersonal Violence, 29(5), 846–865. https://doi.org/10.1177/0886260513505708 The Public Schools Act, Revised Statutes of Canada. (2015, C.C.S.M c.P250). Retrieved from the Manitoba Law website: http://web2.gov.mb.ca/laws/statutes/ccsm/_pdf.php?cap=p250 Vaillancourt, T., Trinh, V., McDougall, P., Duku, E., Cunningham, L., Cunningham, C., Hymel, S., & Short, K. (2010). Optimizing population screening of bullying in school-aged children. Journal of School Violence, 9(3), 233–250. https://doi.org/10.1080/15388220.2010.483182