Australasian Journal of

Educational Technology
2007, 23(4), 435-454

Bullying in the new playground: Research into

cyberbullying and cyber victimisation
Qing Li
University of Calgary
This study examines the nature and extent of adolescents’ cyberbullying
experiences, and explores the extent to which various factors, including
bullying, culture, and gender, contribute to cyberbullying and cyber
victimisation in junior high schools. In this study, one in three adolescents
was a cyber victim, one in five was a cyberbully, and over half of the
students had either experienced or heard about cyberbullying incidents.
Close to half of the cyber victims had no idea who the predators were.
Culture and engagement in traditional bullying were strong predictors not
only for cyberbullying, but also for cyber victimisation. Gender also played a
significant role, as males, compared to their female counterparts, were more
likely to be cyberbullies.

Cyberbullying and cyber victimisation

School bullying has been widely recognised as a serious problem and it is
particularly persistent and acute during junior high and middle school
periods (National Center for Educational-Statistics, 1995). In the USA, “up
to 15% of students … are frequently or severely harassed by their peers. …
Only a slim majority of 4th through 12th graders … (55.2%) reported
neither having been picked on nor picking on others” (Hoover & Olsen,
2001). Universally, bullying is reported as a significant problem in many
countries of the world including European countries, North America, and
Japan (Smith et al., 1999), suggesting that bullying may play a important
role in adolescents’ life in many societies. More importantly, it is reported
that in many cases of school shootings, the bully played a major role
(Dedman, 2001; Markward, Cline & Markward, 2002).

The fast development of new technology in recent years has also increased
school and student access to new technologies such as computers and cell
(or mobile) phones. Substantial research studies have shown that
computers in classrooms can have positive effects on learning of all
subjects. The introduction of electronic communication into classrooms,
however, also brings problems that deserve our attention. One such issue
436 Australasian Journal of Educational Technology, 2007, 23(4)

concerns the growing cyberbullying problem in schools, i.e. the use of

electronic communication devices to bully others.

The problem of school bullying has been examined in the last two decades
and as a result, many schools have developed effective bully-prevention
programs. Many teachers and administrators now recognise this problem.
Few, however, are aware that students are being harassed through
electronic communication. Parallel to this lack of awareness by school
professionals, researchers have yet to examine the nature of cyberbullying
(Beran & Li, 2005). Although cyberbullying incidences have been reported
in different countries around the world, few studies, if any, have explored
the issue taking different variables such as culture, gender, bullying into
consideration. The growing number and the level of severity of
cyberbullying incidents reported worldwide calls for educators,
researchers, and policy makers to take action (Patchin & Hinduja, 2006).

Before we can tackle this problem, however, a thorough understanding of

this issue is necessary. This new phenomenon challenges us, as researchers,
to examine the issue from different perspectives. This research, therefore,
studies cyberbullying in relation to factors such as gender, culture, and
traditional bullying. In this paper, “bullying” refers to bullying in the
traditional sense, and “cyberbullying” refers to bullying via electronic
communication tools.

Definitions
Although bullying traditionally has been equated to physical harassment,
researchers now define bullying as a form of aggression including verbal or
physical harassment. Specifically, “a student is being bullied when he or
she is exposed, repeatedly and over time, to negative actions on the part of
one or more students. The person who intentionally inflicts or attempts to
inflict, injury or discomfort upon someone else is engaging in negative
actions” (Olweus, 2003, p. 15). Bullying may take a wide range of forms
including hitting, pushing, holding, hostile gesturing, threatening,
humiliating, degrading, teasing, name calling, put downs, sarcasm,
taunting, staring, sticking out the tongue, eye rolling, silent treatment,
manipulating friendship and ostracising (Ma, 2001).

Bill Belsey, developer of the award winning webpage www.bullying.org,

defined cyberbullying in his website www.cyberbullying.ca:
Cyberbullying involves the use of information and communication
technologies such as e-mail, cell phone and pager text messages, instant
messaging, defamatory personal Web sites, and defamatory online personal
polling Web sites, to support deliberate, repeated, and hostile behaviour by
an individual or group, that is intended to harm others.
Li 437

Different types of cyberbullying were reported ranging from flaming to

cyberstalking. In her book, Nancy Willard (2004) identified seven different
categories of common cyberbullying actions:
Flaming: Sending angry, rude, vulgar messages about a person to an online
group or to that person via email or other text messaging.
Online harassment: Repeatedly sending offensive messages via email or other
text messaging to a person.
Cyberstalking: Online harassment that includes threats of harm or is
excessively intimidating.
Denigration (put-downs): Sending harmful, untrue, or cruel statements about
a person to other people or posting such material online.
Masquerade: Pretending to be someone else and sending or posting material
that makes that person look bad.
Outing: Sending or posting material about a person that contains sensitive,
private, or embarrassing information, including forwarding private
messages or images.
Exclusion: Cruelly excluding someone from an online group. (Cyberbullying
Survey Section 8)

Objectives
Giving the fact that the cyberbullying is a new phenomenon, the first
objective for this study is to examine nature and extent of adolescents’
experience of cyberbullying and cyber victimisation in Canadian and
Chinese schools. The period of early adolescence for the study of
cyberbullying and cyber victimisation is chosen because adolescence is a
time when bullying is reported to be at its peak and “this period …
witnesses a series of abrupt changes in the social lives of youngsters”
(Pellegrini & Bartini, 2000, p.700). Most importantly, this paper explores the
extent to which certain factors contribute to cyberbullying and cyber
victimisation. These factors are: bullying, culture, gender, knowledge of cyber
safety, frequency of computer use, and academic achievement.

Related literature
According to Nelson (2003), “cyber bullying is often very serious, including
stalking and death threats… I can say anything I want. It’s impersonal.
Face to face is a little intimidating.” Many news stories have reported
cyberbullying incidents all over the world. Cyberbullying also takes
various forms and electronic communication tools - from email, listserve,
mobile phone, to websites. For example, a 15 year-old boy in Quebec,
Canada became an unwilling “celebrity when a film he made of himself
emulating a Star Wars fight scene was posted on the Internet by some
classmates. Millions downloaded the two-minute clip… He was so
humiliated he sought counseling, and his family has launched a lawsuit
against his tormentors” (Snider & Borel, 2004). In Japan, mobile phone
438 Australasian Journal of Educational Technology, 2007, 23(4)

pictures of an overweight boy, which was taken on the sly in the locker
room, were distributed to many of his peers (Paulson, 2003). Another
incident happened in Calabasas High School in California. “It was a
website – schoolscandals.com – on which vicious gossip and racist and
threatening remarks grew so rampant that most of the school was affected”
(Paulson, 2003). Combating cyberbullying is more difficult for schools than
people initially expected. Many bullies are anonymous. Further, under the
free speech rights, it is difficult to take down a website.

Aside from the many reported news stories, several surveys have been
conducted to explore cyberbullying issues. In a survey conducted in Britain
in 2002, it was found that one out of four youngsters aged 11 to 19 had
been cyberbullied (National Children's Home, 2002). An earlier survey
conducted in New Hampshire in 2000 found that about 6 percent of youths
had the experience of being harassed online (Thorp, 2004). A study
reported at a cyberbullying conference in Westchester indicated that “only
six kids out of 200 said they had not been involved in some form of
cyberbullying” (Dickstein, 2005).

Cyber harassment is a form of bullying happening in cyberspace. Various

studies report different types of cyber harassment. For example, Katz
(2002) found that many developmentally delayed adolescents experienced
sexual harassment over the Internet. Other researchers (Spitzberg &
Hoobler, 2002) found that one third of undergraduate students reported
being stalked over the Internet. For example, a textually enacted “rape”
was conducted on a MOO (an online environment enabling multi-user
interactions) in which a male user controlled two female players’ characters
to force the performance of sexually degrading actions on themselves
(Dibbell, 1996). Another incident occurred in a support MUD (a
multiplayer computer game) sexual abuse survivors in which a male
enacted graphic sexual abuse to all participants (Reid, 1994). Anonymity
inherited in many electronic communication modes “not only fosters
playful disinhibition but reduces social accountability, making it easier for
users to engage in hostile, aggressive acts” (Herring, 2001).

These studies and the stories suggest that cyberbullying indeed occurs
around the world, yet it is unclear how different factors affect adolescents’
cyberbullying behaviors.

Contributing factors
Earlier research has indicated that many dimensions have negative impacts
on youngsters’ school lives (Wigfield, Eccles, MacIver, Reuman, &
Midgley, 1991). Many factors relate to bullying and victimisation, and I
hypothesise that they also relate to cyberbullying and cyber victimisation.
Li 439

Considering that there are limited research studies available that address
cyberbullying and that it may be seen as bullying moving to the new
medium, it makes sense to base hypotheses about the contributing factors
of cyberbullying on the relatively well researched literature in traditional
bullying. This section discusses the elements that are identified as
significant factors effecting bullying in the existing literature, and
consequently hypothesised as contributing factors to cyberbullying. These
are: bullying, gender, culture, use of technology, knowledge of cyber safety and
academic achievement.

The number one factor is, of course, student engagement in bullying or

victimisation in face to face environments. Previous research (Beran & Li,
June 2005), although limited, indicates a relationship between bullying and
cyberbullying. Engagement in bullying activities, therefore, is hypothesised
to predict cyberbullying and cyber victimisation.
The second factor considered is gender. As reported in previous research
and many news stories, gender plays a significant role in school violence
and bullying. Extensive research work has shown that male students,
compared to female students, are more likely to be bullies (Hoover &
Olsen, 2001; Pellegrini & Bartini, 2000) as well as being bullied (Nabuzoka,
2003). One important factor to consider is that using aversive tactics with
peers is an important social behavior for male students (Pellegrini &
Bartini, 2000). In keeping with extant work with bullying (Pellegrini,
Bartini & Brooks, 1999), it is posited that gender should predict
cyberbullying and cyber victimisation.

The third factor relates to culture. Bullying is a universal problem with

evidence from a wide range of countries including Norway and Sweden
(Olweus, 1994), Spain (Ruiz, 1992), Australia (Rigby & Slee, 1991), Canada
(Bentley & Li, 1995), Japan (Crystal, 1994), and USA (Hoover & Juul, 1993;
Hoover, Oliver & Thomson, 1993). Although bullying has been identified
around the world, previous research (e.g. Nabuzoka, 2003) suggests that
students from different countries and cultures behave differently with
involvements in bullying. This may result from a number of factors. For
example, people in different cultures may hold different beliefs or
religions. Since culture is related to bullying and victimisation, it is logical
to argue that culture should be considered as a predictor for cyberbullying
and cyber victimisation.
The fourth factor is the use of technology. Since cyberbullying occurs in
cyberspace, it is reasonable to assume that if students have limited
opportunities to access to technology, they should have fewer
opportunities to be involved in cyber harassments. The frequency of
technology use by students, therefore, should predict cyberbullying and
cyber victimisation. It is worth noting that mobile (cell) phones are not
440 Australasian Journal of Educational Technology, 2007, 23(4)

considered in this study because they are not allowed in many Canadian
schools.

The fifth factor is awareness of cyber safety. According to the past two
decades of research into the prevention of bullying, awareness is identified
as one of the four most important areas that can reduce bullying in schools
(Campbell, 2005), and educating students about safety approaches plays a
major role in any effective anti-bullying programs. Campbell (2005) further
suggests that increasing awareness should also help reduce cyberbullying.

Academic achievement is the sixth factor relating to cyberbullying and

cyber victimisation. Previous research on bullying and victimisation found
that academic achievement and academic pressure connect to bullying in
schools (Ma, 2001). Academic pressure refers to “the expectations of peers
and teachers” (Ma, 2001, p. 357). Students with lower academic pressure
are more likely to bully than students in schools with high academic
pressure. Victimisation, on the other hand, relates to various factors
including academic achievement. Regardless of the type of bullying,
victims suffer various negative consequences. Some significant and well
documented such consequences include students having difficulty
concentrating on study, decreased academic achievement levels, decreased
interest in school, and increased absences (National Association of State
Boards of Education, 2003). This link between academic work and bullying
suggests possible connections between academic achievement and issues of
cyberbullying. Consequently, academic achievement is considered to be
connected to cyberbullying and cyber victimisation.

In summary, it is hypothesised that cyberbullying in seventh grade should

be predicated by traditional bullying. Further it is posited that cultural
differences should result in different bullying behaviours. Similarly,
gender, frequency of computer use and culture should drive cyber
victimisation. In addition, I posit that victimisation in real world should
predict cyber victimisation. Academic achievement, on the other hand,
should relate negatively to cyber victimisation. Knowledge of cyber safety
should act as a protective factor against cyberbullying.

Methods
Data
Two sets of data were collected in 2004: Canadian and Chinese data. For
Canadian data, the participants were randomly selected from three middle
schools in Calgary. The three schools were chosen because of the schools’
enthusiasm about technology. The schools were involved in a large
educational technology integration project of the Province. Two schools are
Li 441

located in areas where residences are mainly at middle class range, and the
third is in a low to middle socio-economic status (SES) area. A total of 264
grade seven students (130 males and 134 females) completed the
questionnaire. Among them, only 7.6 percent are English as a second
language (ESL) students. Further, about three quarters of the students are
white, 6.5% Asian, and about 18% are Black, Hispanic Aboriginal, or other.
Close to 60% of the students reported that their school grades were usually
above average; while about forty% of them were average. Only several
students reported their grades to be below average.

The Chinese data were collected from 197 grade seven students (107 males,
90 females) who were randomly selected from two secondary schools in
Changsha, China. The same questionnaire was administrated to the
students. Amongst them, one in three reported that their school grades
were usually above average, while half of them were average. Close to ten
percent of the students reported their grades to be below average. Detailed
demographic information is presented in Table 1.
Table 1: Demographics of the students
Country Gender (%) Academic achievement (%)
M F Above average Average Below average
Canada (n=264) 49.2 50.8 57.2 39.4 1.9
China (n=197) 54.3 45.7 37.6 53.8 8.6
*Total percentages may not add up to 100 due to missing values.

Instrument
The survey instrument used was a questionnaire developed by Li (2007)
(see Appendix). It included a total of 22 questions which collected
information from two major areas: students’ demographic information and
their experience related to cyberbullying. It is important to note that no
student name or identification was collected and therefore, the survey was
anonymous. Because computers are an integral part of class instruction,
and mobile (cell) phones are not allowed in the Canadian schools surveyed
in this study, the emphasis was on computer use rather than mobile
phones. Frequency of mobile phone use was not included in the questions.
To collect the Chinese data, this original questionnaire was translated into
Chinese by a professor who is a native speaker with over ten years
experience in teaching English in Chinese universities. The Chinese version
of the questionnaire was administered in the two Chinese schools.

Analysis
This paper employed quantitative analyses of student surveys using SPSS.
Specifically, descriptive statistics was first used to explore the nature and
442 Australasian Journal of Educational Technology, 2007, 23(4)

extent of adolescents’ experience of cyberbullying and cyber victimisation.

Then, forward stepwise logistic regressions were employed to examine the
extent to which various factors contributed to cyberbullying and cyber
victimisation. “In logistic regression analysis for a dependent dichotomous
variable, one attempts to predict the probability that an observation
belongs to each of two groups” (Wright, 2000, p. 219). The rationale for
choosing stepwise logistic regression over direct or hierarchical logistic
regression is twofold. First, the exploratory nature of this study calls for a
statistical method to identify significant factors. Second, a paucity of
literature determining the hierarchical arrangement of variables
necessitated a statistical model that enables the variables to emerge and
organise with respect to strength of effect.

In a logistic regression analysis, a common practice is to use a more liberal

significance level than .05. Researchers “recommend[ed] a criterion for
inclusion of a variable that is less stringent than .05; they suggest that
something in the range of .15 or .20 is more appropriate to ensure entry of
variables with coefficients different from zero” (Tabachnick & Fidell, 2001,
p. 535). In this study, therefore, an alpha of .15 rather than .05 is used for
the logistic regression analyses.

The selected variables were: country, gender, knowledge of safety strategy,

frequency of computer using, school grade, and traditional bullying. These
variables were chosen because they have been previously reported as being
predictors in the literature or have been found to be useful in our
experience. The variable “frequency of computer using” was collapsed into
two categories: rare (less than 4 times per month) and often (4 or more
times per month). Similarly, the variable “school grade” was collapsed
into two groups: “above average” and “average and below”. This
procedure was conducted to reduce the number of low size cells. The
traditional bullying variable as a categorical predictor was first recoded
into a set of dummy variables before entering into the logistic regression
model. Specifically, the traditional bullying variable was grouped into the
following four subcategories: bully only (hereafter bully), victim only
(hereafter victim), bully and victim (hereafter bully victim), and not
involved.

Results
To date, cyberbullying has not been examined in relation to traditional
bullying issues. Evaluating traditional bullying behavior as a predictor
may add to the understanding of how cyberbullying is influenced by
traditional ways of harassment. Furthermore, factors like gender, culture,
knowledge of safety strategies, and frequency of computer use, may
provide valuable information to assess possible involvement with
Li 443

cyberbullying. The present study has explored these factors and others that
may predict which students were likely to be involved in cyberbullying.

Nature and extent of cyberbullying

To what extend do adolescents experience cyberbullying? In general, close
to one third of the adolescents were cyber victims and about 18% were
cyberbullies. Significantly more students, in fact, over half of them, knew
someone who had been cyberbullied. When gender was considered, higher
proportions of males were involved in cyberbullying either as cyber
victims or cyberbullies, or knowing someone being cyberbullied (Table 2).

Table 2: Percentage of students who experienced cyberbullying (N=461)

Cyber victim Cyberbully Aware of cyberbullying
Male 31.2 21.9 53.6
Female 26.3 13.4 52.2
Total 28.9 17.8 52.9

What tools were involved in cyberbullying? For the 133 cyber victims, one
out of 5 was cyberbullied by emails only. About one third was cyberbullied
in chatrooms only and about 13% by mobile phone only. Another one third
was victimised via other technologies or mixed electronic means. What
tools did cyberbullies use to harass others? About one fifth of the
cyberbullies used email only, close to 28% used chatrooms only, and about
6% used mobile phone only. A large majority of the cyberbullies used other
electronic means or mixed types of technology to harass others (Table 3).

Table 3: Tools used

Email Chatroom Mobile Other or
Tools used by
(%) (%) phone (%) mixed (%)
Cyberbully (N=82) 20.4 27.8 5.6 39.4
Cyber victim (N=133) 21.8 30.8 13.5 30.9

How often did cyberbullying occur? Over 30% of the cyberbullies reported
they harassed others less than 4 times, about 43% did it between 4 to 10
times and another 20% did it more than 10 times. The cyber victims
reported a somewhat different pattern: close to 55% were cyber assaulted
less than four times, about 20% were cyber harassed between 4 and 10
times, and another 21% were cyberbullied frequently, i.e. over 10 times.
Table 4 provides details.
Table 4: Frequency of cyberbullying
< 4 times 4-10 times >10 times
Cyberbully(N=82) 30.5 43.4 20.7
Cyber victim (N=133) 54.9 20.3 21.1
444 Australasian Journal of Educational Technology, 2007, 23(4)

Who were the predators? The results showed that out of the 133 cyber
victims, 25.6% were cyberbullied by school mates, 12.8% by people outside
school, 46.6% did not know who cyberbullied them and 8.3% were
cyberbullied by mixed groups of people (i.e. school mates and people
outside). In addition, only 48.9% of the cyber victims informed adults (e.g.
parents or teachers) about the incidents.

When considering the total student sample, only 34.5% reported that when
they knew someone being cyberbullied, they told adults. What were
students perceptions about adults’ intervening in schools? Only 63.6%
believed that adults in schools tried to stop cyberbullying when informed.
Further, over 65% of the students thought that they knew cyber safety
strategies. Who taught them those safety strategies? The results of the
survey showed that 30.2% of the students learned by themselves, about
13% learned from parents, less than 10% learned from schools, and another
9.1% from other sources.

Predicting cyber victimisation and cyberbullying

Cyber victimisation
The potential predictors were: gender, country, traditional bullying (4
categories), frequency of computer use, knowledge of safety, and self
reported school grades. Details are presented in Table 5.
Table 5. Determinants of cyber victimisation
Exp (beta)
Predictor beta S. E. Wald’s chi2 df P
(odds ratio)
1. traditional bullying 11.79 3 .008
- bully only .67 .53 1.58 1 .209 1.92
- victim only .90 .27 11.02 1 .001 2.46
- bully victim .65 .27 5.85 1 .016 1.91
2. Knowledge of .44 .24 3.43 1 .064 1.54
safety strategy
3. Country -.36 .22 2.71 1 .100 .70
Constant -1.50 .28 29.06 1 <.001 .22

Out of the potential predictors list, the most significant predictors of cyber
victim, in descending importance, were: traditional bullying (chi square (3)
= 11.79, p = .008), knowledge of safety strategies (chi square (1) = 3.43, p =
.064), and country (chi square (1) = 2.71, p = .100). The percentage of correct
classification was 71.4% and the goodness of fit (chi square (8) = 6.90, p =
.55) indicated a good fit model.

Considering the individual logistic regression coefficients, the results

indicated that victim (beta = .90, p = .001), bully victim (beta = .65, p =
.016), knowledge of safety (beta = .44, p = .064) and country (beta = -.36, p =
Li 445

.100) were significantly predicting the probability of cyber victimisation.

According to the model, a student being cyberbullied was positively
related to bullying, knowledge of safety strategy and country. A victim was
almost 2.5 times more likely to be a cyber victim than a not involved
student. A bully victim, compared to a not involved student, was about 2
times more likely to be a cyber victim. Further, a bully was almost twice as
likely to be a cyber victim as a not involved student. However, this was not
a significant predictor.

Given the same scores in other variables, Chinese students were more
likely to be cyber victims than their Canadian counterparts. In addition,
when other variables were controlled, cyber victims tended to report that
they knew safety strategies in cyberspace. In fact, students who reported
that they knew safety strategies were 1.54 times more likely, than not, to be
cyber victims compared with students who reported that they had no
knowledge of safety strategies.

Cyberbullying
Again, logistic regressions were used to analyse the same potential
predictors for cyberbullying. Out of the potential predictors list, the most
significant predictors of cyber victim, in descending importance were:
country (chi square (1) = 17.48, p <.001), bullying (chi square (3) = 16.38, p
= .001), gender (chi square (1) = 6.06, p = .014), and frequency of computer
use (chi square (1) = 2.68, p = .102). The percentage of correct classification
was 82.4% and the goodness of fit (chi square (8) = 6.81, p = .56) indicated a
good fit model (see Table 6).

Table 6: Determinants of cyberbullying

Exp (beta)
Predictor beta S. E. Wald’s chi2 df P
(odds ratio)
1. Country 1.47 .35 17.48 1 <.001 4.33
2. Traditional bully 16.38 3 .001
- bully only 1.03 .54 3.69 1 .055 2.81
- victim only -.16 .40 .17 1 .683 .85
- bully victim 1.02 .31 10.60 1 .001 2.76
3. Gender .67 .27 6.06 1 .014 1.96
4. Frequency use computer .64 .39 2.68 1 .102 1.90
Constant -3.84 .48 64.52 1 <.001 .02

The individual logistic regression coefficients indicated that country (beta =

1.47, p <.001), bully (beta = 1.03, p = .055), bully victim (beta = 1.02, p =
.001), gender (beta = .67, p = .014), and frequency of computer use (beta =
.64, p = .102) significantly predicted the probability of cyberbullying. The
influence of culture was strong. When other variables were controlled,
Canadian students were 4.33 times more likely to be cyberbullies than
446 Australasian Journal of Educational Technology, 2007, 23(4)

Chinese students. Another strong indicator was the involvement in

bullying activities. In fact, both ‘bully’ and ‘bully victim’ were significant
indicators. A bully was 2.81 times more likely to be a cyberbully compared
to a not involved student. Similarly, a bully victim being a cyberbully was
2.76 times greater than a not involved student. A student who was a victim
had a slightly lower risk of being a cyberbully than a not involved student.
However, this was not a significant predictor.

Gender was another significant predictor for cyberbullying. Given the

same score for other variables, a male student was almost twice as likely to
be a cyberbully as his female counterpart. In addition, frequency of
computer use was a significant predictor. When other variables were
controlled, frequent users (more than 3 times a month) were 1.9 times more
likely to be cyberbullies than infrequent computer users (less than 4 times
per month).

Discussion
Nature and extent
As evidenced in this study, cyberbullying has become an increasingly
significant problem in schools and it deserves our serious consideration.
One in three adolescents participating in this study were cyber victims and
about one out of five students had cyber harassed others. Over half of the
sample reported that they knew someone being cyberbullied. It is
important to also note that over forty percent of the cyber victims had been
cyber harassed more than three times and 2 in 3 cyberbullies harassed
others four or more times. These results suggest that cyberspace has
become another “playground” for bullying. Bullying does occur in
cyberspace and it occurs frequently.

The tools used for cyberbullying includes emails, chatrooms, mobile (cell)
phones and other type of technologies. Cyberbullies and cyber victims
show similar patterns in terms of the type of tools used for cyber
harassment. A majority of the cyberbullying incidents involve the use of
computers or a mix of technologies, which suggests that a range of
technological means have been used negatively.

One important and unique characteristic of cyberbullying is anonymity. In

this study, close to half of the cyber victims did not even know who
cyberbullied them. This brings a great challenge for us in combating
cyberbullying. Cyberbullies can hide their identities and therefore avoid
consequences. This indirectly encourages them to cyber harass others
again. Not needing to face the victims may also promote this behavior
because it is impersonal. In addition, victims of bullying may try to
Li 447

retaliate indirectly via cyberbullying because it probably constitutes a

lesser consequence.

Another important aspect of this study is the exploration of adult

involvement. When variables related to adult involvements are considered,
a typical trend is identified in the results: over half of the cyber victims did
not inform any adults about the incidents. Less than 35% of all students
report that they would inform adults when they knew about someone
being cyberbullied. This replicates the pattern identified in research of
traditional bullying: the majority of the bystanders and the victims keep
quiet (Hoover & Olsen, 2001). One possible explanation for this comes from
the result of this study: only six out of ten students believe adults in
schools would try to stop cyberbullying when notified. Because many
students think that adults would not even try to stop cyberbullying when
informed, they would, therefore, keep quiet. Another possible explanation
is that the victims and bystanders may have feared getting into trouble if
they told adults.

The fact that a majority of the bystanders or the cyber victims chose to keep
quiet underscores the importance of systematic education in safety
strategies from early age. A preliminary analysis of student narratives
suggests that many students held shallow understandings of safety in
cyberspace, even though many reported that they had the knowledge. For
example, students had some naïve understanding of cyber safety. Some
students believed that as long as they avoided online chat, they would be
safe. Just like approaches proven to be effective in fighting bullying,
educating cyber victims and bystanders may provide some key strategies
in combating cyberbullying. Focusing attention on empowering cyber
victims and bystanders has the potential to prevent a significant amount of
cyberbullying.

Predicting cyberbullying and cyber victimisation

The model predicting seventh grade cyberbullying indicates that culture is
the strongest predictor in the variable examined. A Canadian student is
more than four times likely to be a cyberbully than a Chinese student.
Similarly, culture is also a significant predictor for cyber victimisation. This
is similar to the results of previous bullying research which indicate that
culture plays an important role in bullying and cyberbullying. This study
indicates that Chinese students are more likely to be cyber victims whereas
Canadian students tend to be cyberbullies. This has implications for how
problems of cyberbullying are addressed in the two countries, but perhaps
more importantly, in multicultural societies such as Canada, USA and
Australia. It is worth noting that the students in the two Chinese schools
generally had access to networked computers and about one third had
448 Australasian Journal of Educational Technology, 2007, 23(4)

access to mobile phones (though not allowed in classes). Similarly, the

Canadian students have good access to the technologies, but use of mobile
phones was not allowed in schools.

One of the most important results of this study is that engagement in the
traditional form of bullying is a very strong predictor for both
cyberbullying and cyber victimisation. Victims of traditional bullying,
regardless of whether they are victims or bully victims, are more likely to
be cyberbullied than the students who do not engage in bullying activities.
Likewise, bullies, including bullies or bully victims, have higher chance to
cyber harass others than those who are not involved in bullying. This
finding, though not surprising, has significant implications. It suggests that
bullying and cyberbullying are closely related. It is possible that bullying
started in the real world but extended to the cyberspace. This means that
cyberspace provides bullies another venue to harass others. It is also
possible that the harassment began in the virtual space, but the
perpetrators took it to real world which leads to face to face bullying. As
suggested by Beran and Li (2005),
… as a result of not receiving consequences for engaging in cyber
harassment, students may then continue the harassment when in close
contact with a student at school. In addition, cyber harassment may alter
bullying at school. For example, if “electronic bullies” remain undetected,
their bullying behaviors at school may become more severe and directly,
rather than indirectly, exerted against a victim. (p. 271)

This close link between bullying and cyberbullying highlights a significant

issue to be considered when dealing with cyberbullying issues and
working on prevention programs. This calls for a holistic design of
prevention programs that consider bullies, cyberbullies, as well as their
victims as an integrated whole. Another implication of this result is that the
effective strategies for combating bullying may also be effective in fighting
cyberbullying.

One surprising result is that knowledge of cyber safety predicts cyber

victimisation. It was originally hypothesised that knowing safety strategies
in cyberspace would help prevent students from being cyberbullied. The
analysis of the data indicated that students reported to be knowledgeable
about cyber safety, compared to those who do not know the strategies, are
more likely to be victimised in cyberspace. It is probable that those
students who have no knowledge of cyber safety strategies are not as active
in cyberspace; hence they have less chance to be cyberbullied. Another
possible explanation is that many of the students in this sample have
superficial understanding of cyber safety. As discussed above, the
preliminary examination of students' open ended comments indicates that
many students held very naïve beliefs about cyber safety.
Li 449

Gender is a significant predictor for cyberbullying. Consistent with the

results from previous research into traditional forms of bullying (Hoover &
Olsen, 2001; Kumpulainen, Rasanen, Henttonen & Almqvist, 1998), males
are more likely to cyberbully others than females. The fact that gender does
not predict cyber victimisation in this study merits discussion. This result
contradicts research about traditional bullying which suggests that males
are more likely to be bullied than females (Eslea & Mukhtar, 2000;
Kumpulainen, Rasanen, Henttonen & Almqvist, 1998). This contradictory
result may relate to the nature of cyberspace. In cyberspace, users can hide
their identities. In fact, females tend to disguise their gender when using
computer mediated communication because it does not allow judgment on
the basis of gender (Gopal, Mirana, Robichaux & Bostrom, 1997; Li, 2006).
Male and female students (because their identities could be masked) are
equally likely to be targeted in cyberspace.

Student academic achievement, as measured by students' self reported

grades, accounts for minimal variance in cyberbullying and cyber
victimisation. This indicates that students’ academic learning has minimal
effect on cyberbullying and cyber victimisation. This is somewhat different
from previous research into bullying indicating that academic pressure can
help reduce bullying (Ma, 2001). Maybe anonymity and feeling impersonal
creates courage from a distance and make those less likely to cause trouble
to do so. For example, with masked identities and not facing the victims
directly, people feel less intimidated about harassing others and therefore,
even the typical “good” students (e.g. those who are high academic
achievers) can engage in this type of action. As suggested by a teacher
posting to an internal listserver at a Canadian university in 2005, “the
thought of being able to get away [with it] lures even those who may not
really wish to hurt some, to lash out at anyone who has annoyed them in
some fashion. To humiliate someone just because they are an ‘easy’ target
and because you can get away with it is part of the evil cyberbullying has
brought to the table of student life”.

Conclusion
This study contributes to the extant literature on bullying in several
conceptual areas. Firstly, cyberbullying is a bullying problem occurring in
a new territory. To date, few research studies have examined bullying issue
in this new context. The astonishing high proportion of adolescents who
had experiences of cyberbullying observed in both Canada and China
suggests that cyberbullying is becoming an increasingly significant
problem for schools and society generally.

According to Beran and Li (2005), the severity of cyberbullying varies, with

incidents ranging from annoying to dangerous with the occurrence of
450 Australasian Journal of Educational Technology, 2007, 23(4)

death threats. In addition, cyberbullying may alter traditional bullying at

school. For example, if cyber bullies remain undetected, their bullying
behaviors at school may become more severe and directly, rather than
indirectly, exerted against a victim. On the other hand, it is also possible
that if bullies have difficulty in attacking victims face to face, they may
choose to use cyberspace to harass them. The consequences of such
harassment can be as serious as, if not more serious, than face to face
bullying. As evidenced in previous research (Beran & Li, 2005), cyber
victims can be affected in many ways, including psychological, emotional,
and academic problems. They could suffer from emotional depression and
decreased academic achievement. Future research must explore how
cyberbullying decreases, maintains or exacerbates other forms of bullying.

Secondly, cyberbullying is investigated in this study taking culture into

consideration. Canadian and Chinese adolescents’ experiences indicate that
culture plays a significant role in cyberbullying. This result provides
empirical groundwork for programs aiming to combat cyberbullying in
different cultural and school contexts. Most importantly, this work
establishes a relationship between bullying, cyberbullying, and
victimisation. This has significant implications for intervention and
prevention programs in schools. It suggests that anti-bullying actions
should also take cyberbullying into consideration and vice versa. Remedies
and treatments for bullying in schools may help to reduce cyberbullying.

Thirdly, in this paper, bullying and cyberbullying are examined at “a point

where it had seldom been studied. Extant studies, for the most part,
studied primary school children. The early adolescent period merits
attention because it is a period, labeled a ‘brutalising period’, where
disruptions in social networks afford opportunities for peer victimisation
and aggression to establish peer status” (Pellegrini & Bartini, 2000).
Consistent with this, the results show that high percentages of the students
are involved in cyberbullying.

One important point which needs to be stressed is that schools should play
a significant role in combating cyberbullying, yet, restricting Internet
usage, emails, or any other technological tools cannot stop bullies from
finding ways around the schools’ suspension of privileges. Establishing
effective anti-bullying programs as well as creating a culture of caring and
kindness in schools has never been more important. Awareness is the first
step and it is an absolute must. As indicated in student narratives, many of
the students do not even know that this is a form of bullying and not just
another disadvantage of this age group. If they do not even know they are
victimised, how can we help them? Similarly, the cyberbullies sometimes
do not understand that what they are doing is considered to be a form of
bullying. Most students do not want the reputation of being a bully. Once
Li 451

this becomes clear to them, some will stop. In addition, an anti-bullying,

anti-cyberbullying program should be a systematic and joint effort by
society generally rather than a one time or one school project.

This study has some limitations, just like any research studies. For
example, the survey question intended to discover frequency of computer
use did not consider that a majority of students in the schools have easy
access to computers in this urban city. Simply categorising it into rare, 1-3
times per month, and over 4 times a month does not appropriately reflect
the current trend of access to computers. Further, as the data were collected
from urban schools in a large city in Canada and in a large city in China,
we need to be cautious about generalising the findings to other regions.

Acknowledgement
I express my appreciation to Ms Shelley Luo for translating the question-
naire and help in collecting data from the Chinese schools.

References
Bentley, K. M. & Li, A. (1995). Bully and victim problems in elementary schools and
students' beliefs about aggression. Canadian Journal of School Psychology, 11, 153-
165.
Beran, T. & Li, Q. (2005). Cyber-harassment: A new method for an old behavior.
Journal of Educational Computing Research, 32(3), 265-277. http://www.ucalgary.ca/
~qinli/publication/cyber-bully-Journal_of_Educational_Computing_Research_final.doc
Beran, T. & Li, Q. (June 2005). Cyber-harassment: A new method for an old
behavior. Paper presented at the Annual meeting of the Canadian Psychological
Association, Montreal.
Campbell, M. (2005). Cyber-bullying - An old problem in a new guise. Australasian
Journal of Guidance and Counselling, 15(1), 68-76.
http://eprints.qut.edu.au/archive/00001925/
Crystal, S. (1994). Concepts of deviance in children and adolescence: The case of
Japan. Deviant Behaviour, 15(3), 241-266.
Dedman, B. (2001). Schools may miss mark on preventing violence. [viewed 16 Oct
2003, not found 29 Sep 2007] http://www.suntimes.com/shoot/shoot16.html
Dibbell, J. (1996). A rape in cyberspace, or how an evil clown, a Haitian trickster
spirit, to wizards, and a cast of dozens turned a database into a society. In R.
Kling (Ed.), Computerization and controversy (2 ed.). NY: Academic Press.
Dickstein, L. (2005). School bullies find a new playground. [viewed 7 Nov 2005, not
found 29 Sep 2007] http://www.nydailynews.com/front/story/298461p-255094c.html
Eslea, M. & Mukhtar, K. (2000). Bullying and racism among Asian schoolchildren in
Britain. Educational Research, 42(2), 207-217.
452 Australasian Journal of Educational Technology, 2007, 23(4)

Gopal, A., Mirana, S. M., Robichaux, B. P. & Bostrom, R. P. (1997). Leveraging

diversity with information technology: Gender, attitude, and intervening
influences in the use of group process. Organizational Behavior and Human
Performance, 25, 97-106.
Herring, S. (2001). Gender and power in online communication. Center for Social
Informatics Working Papers. [viewed 5 Sep 2004, verified 29 Sep 2007]
http://rkcsi.indiana.edu/archive/CSI/WP/WP01-05B.html
Hoover, J. & Juul, K. (1993). Bullying in Europe and the United States. Journal of
Emotional and Behavioral Problems, 2, 25-29.
Hoover, J., Oliver, R. & Thomson, K. (1993). Perceived victimization by school
bullies: New research and future directions. Journal of Humanistic Education and
Development, 32, 76-84.
Hoover, J. & Olsen, G. (2001). Teasing and harassment: The frames and scripts approach
for teachers and parents. Bloomington, IN: National Educational Service.
Kumpulainen, K., Rasanen, E., Henttonen, I. & Almqvist, F. (1998). Bullying and
psychiatric symptoms among elementary school-age children. Child Abuse &
Neglect: The International Journal, 22(7), 705-717.
Li, Q. (2006). Computer-mediated communication: A meta-analysis of male and
female altitudes and behaviors. International Journal on E-Learning, 5(4), 525-570.
http://www.ucalgary.ca/~qinli/publication/meta_analysis_Li_IJEL_final_web.doc
Li, Q. (2007). New bottle but old wine: A research on cyberbullying in schools.
Computers and Human Behavior, 23(4), 1777-1791. [verified 29 Sep 2007]
http://www.ucalgary.ca/~qinli/publication/cyber_chb2005.pdf
Ma, X. (2001). Bullying and being bullied: To what extent are bullies also victims?
American Educational Research Journal, 38(2), 351-370.
Markward, M. J., Cline, S. S. & Markward, N. J. (2002). Group socialization, the
Internet and school shooting. International Journal of Adolescence and Youth, 10(1-
2), 135-146.
Nabuzoka, D. (2003). Experiences of bullying-related behaviours by English and
Zambian Pupils: A comparative study. Educational Research, 45(1), 95-109.
National Center for Educational Statistics (1995). Student victimization in schools.
Washington, DC: Hemisphere.
National Children's Home (2002). NCH 2002 Survey. [viewed 15 Jul 2004, not found 29
Sep 2007] http://www.nch.org.uk/itok/showquestion.asp?faq=9&fldAuto=145
National Association of State Boards of Education (2003). Bullying in schools. Policy
Update, 11(10), 1-2.
Nelson, M. (2003). School bullies going high tech. [viewed 14 July 2004, not found 29
Sep 2007] http://www.canoe.ca/NewsStand/LondonFreePress/News/2003/09/02/
174030.html
Olweus, D. (1994). Bullying at school: Long-term outcomes for the victims and
effective school-based intervention program. In L. Huesmann (Ed.), Aggressive
Behavior: Current Perspectives (pp. 97-130). NY: Wiley.
Li 453

Olweus, D. (2003). A profile of bullying at school. Educational Leadership, 60(6), 12-19.

Patchin, J. W. & Hinduja, S. (2006). Bullies move beyond the schoolyard: A
preliminary look at cyberbullying. Youth Violence and Juvinile Justice, 4(2), 148-169.
Paulson, A. (2003). Internet bullying. Christian Science Monitor, 30 Dec. [viewed 15
July 2004, verified 29 Sep 2007 http://www.csmonitor.com/2003/1230/p11s01-legn.htm

Pellegrini, A., & Bartini, M. (2000). A longitudinal study of bullying, victimization,

and peer affiliation during the transition from primary school to middle school.
American Educational Research Journal, 37(3), 699-725.
Pellegrini, A., Bartini, M. & Brooks, F. (1999). School bullies, victims, and aggressive
victims: Factors relating top group affiliation and victimization in early
adolescence. Journal of Educational Psychology, 91, 216-224.
Reid, E. (1994). Cultural formations in text-based virtual realities. Unpublished Master's
thesis, University of Melbourne, Melbourne. [verified 29 Sep 2007]
http://www.zacha.net/articles/reid.html
Rigby, K., & Slee, P. (1991). Bullying among Australian school-children: Reported
behaviour and attitudes towards victims. Journal of Social Psychology, 131(5), 615-
627.
Ruiz, O. (1992). Violence in schools: Problems of bullying and victimization in Spain.
Paper presented at the European Conference on Developmental Psychology,
Seville, Spain.
Smith, P. K., Morita, Y., Junger-Tas, J., Olweus, D., Catalano, R. & Slee, P. (1999). The
nature of school bullying: A cross-national perspective. London & NY: Routledge.
Snider, M. & Borel, K. (2004). Stalked by a cyberbully. Macleans, 117(19 May), 76-77.
Spitzberg, B. & Hoobler, G. (2002). Cyberstalking and the technologies of
interpersonal terrorism. New Media & Society, 4, 71-92.
Tabachnick, B. & Fidell, L. (2001). Using multivariate statistics (4 ed.). Needham
Heights, MA: Allyn & Bacon.
Thorp, D. (2004). Cyberbullies on the prowl in schoolyard. The Australian, 15 July.
http://www.australianit.news.com.au/
Wigfield, A., Eccles, J., MacIver, D., Reuman, A. & Midgley, C. (1991). Transitions
during early adolescence: Changes in children's domain specific self-perceptions
and general self-esteem across the transition to junior high school. Developmental
Psychology, 27, 552-566.
Willard, N. (2004). An educator's guide to cyberbullying and cyberthreats. [viewed
6 Oct 2005, not found 29 Sep 2007] See http://cyberbully.org/
Wright, R. (2000). Logistic regression. In L. Grimm & P. Yarnold (Eds.), Reading and
understanding multivariate statistics (pp. 217-244). Washington, DC: American
Psychological Association.
454 Australasian Journal of Educational Technology, 2007, 23(4)

Appendix: Survey
Section one: About you (circle one):
1. Sex: Male Female Grade level______ Year of birth____________
2. How do you describe yourself:
Asian Hispanic Black White Aboriginal Other
3. English is my second language: Yes No
4. My school grades are usually: above average average below average
5. I use computers: rarely 1-3 times/month at least 4 times/month

Section two: Cyberbully is defined as harassing using technology such as email,

computer, cell phone, video cameras, etc. Bullying occurs when people say mean
and hurtful things or make fun of another person or calls him/her mean and
hurtful names, completely ignore or exclude him/her from their group of friends or
leaves him/her out of things on purpose, tells lies or spreads false rumors about
him/her, sends mean notes and tries to make other students dislike him/her, and
other hurtful things like that. When we talk about bullying, it is difficult for the
person being bullied to defend himself or herself. We also call it bully, when a
person is teased repeatedly in a mean and hurtful way. But we don’t call it bullying
when the teasing is done in a friendly and playful way. Also, it is not harassment
when two people of about equal strength or power argue or fight.

1. I have been bullied during school: yes no

2. I have bullied others: yes no
3. I have been cyberbullied (e.g. via email, chat room, cell phone): yes no
4. If yes, I was cyberbullied via (circle all that apply):
email chat room cell phone other, specify______________
5. If yes, I was cyberbullied by:
school mates people outside school I don’t know who
6. If yes, I have been cyber-bullied:
less than 4 times 4-10 times over 10 times
7. I have cyberbullied others: yes no
8. If yes, I cyberbullied others via (circle all that apply):
email chat room cell phone other, specify______________
9. If yes, I have cyberbullied others:
less than 4 times 4-10 times over 10 times
10. I know someone who has been cyberbullied: yes no
11. When adults in school know cyberbullying, they try to stop it: yes no
12. When I was cyberbullied, I told adults (e.g. parents, teachers): yes no
13. When I knew someone being cyberbullied, I told adults: yes no
14. I know safety strategies in cyberspace: yes no
15. If yes, I learned safety strategies:
by myself taught by parents taught in schools other, specify: _________

Qing Li PhD (Toronto), Associate Professor

Faculty of Education, University of Calgary
2500 University Drive NW, Calgary, AB T2N 1N4, Canada
Email: [email protected] Web: http://www.ucalgary.ca/~qinli/