Children and Youth Services Review 161 (2024) 107693

Contents lists available at ScienceDirect

Children and Youth Services Review

journal homepage: www.elsevier.com/locate/childyouth

A comparison of definitions of school bullying among students, parents, and

teachers: An experimental study from China
Zepeng Gong a, Zhiwei Tang a, b, Jing Zhou a, Ziqiang Han c, Jingran Zhang d, *
a
School of Public Affairs and Administration, University of Electronic Science and Technology of China, Chengdu 611731, China
b
Shenzhen Institute for Advanced Study, University of Electronic Science and Technology of China, Shenzhen 518110, China
c
School of Political Science and Public Administration, Shandong University, Qingdao 266237, China
d
Tian Co-innovation Center, University of Electronic Science and Technology of China, Chengdu 611731, China

A R T I C L E I N F O A B S T R A C T

Keywords: This study aims to determine whether the four types of bullying (verbal, physical, relational, and cyberbullying)
School bullying are equally likely to be regarded as school bullying by students, parents, and teachers, and whether all three
Definition characteristics of bullying (repetitiveness, intentionality, and power imbalance) are part of the definition of
Repetitiveness
school bullying in China. Students (N = 1,943), parents (N = 1,957), and teachers (N = 409) in pre-college
Intentionality
Power imbalance
schools in China participated in survey experiments. Multi-level models and ordinary linear models were used
Survey experiment to perform data analysis. Results indicated that compared with verbal, physical, and cyberbullying, relational
bullying was less likely to be considered as school bullying. Neither students and parents, nor teachers used all
three characteristics as criteria for identifying school bullying, although parents and teachers considered more
characteristics than students when defining school bullying. Therefore, the three characteristics of bullying may
not be the gold standard for defining school bullying. Practical investigation should give more attention to
relational bullying, and parents and teachers should view school bullying through the students’ broader lens
when responding to bullying.

1. Introduction et al., 2019) (e.g., skipping classes, self-harm, suicidal behavior, violent
behavior, and murder plans). Therefore, it is necessary to implement
School bullying is a global public health issue, which significantly effective prevention and intervention strategies to build a safe school
affects child health and development and increases the risk of poor environment and ensure the healthy growth of children and adolescents.
health and social and educational outcomes in childhood and adoles­ In general, an accurate and in-depth understanding of a phenomenon
cence. According to the United Nations Educational, Scientific, and is crucial to attain the ability to control and change it (Merton, 1987),
Cultural Organization report on school violence and bullying, almost a and the first step in understanding the phenomenon is to provide its
third of school-aged children and adolescents have had the experience of clear definition (Olweus & Limber, 2018). Accordingly, before inter­
being bullied (UNESCO, 2019). Bullying brings forward detrimental vening in school bullying, it is first required to know what school
consequences for the physical and mental well-being of children. bullying actually is. Currently, common definitions of bullying used in
Moreover, bullying experiences have a serious, long-term, negative the research and social policy fields are based on those proposed by the
impact on the physical and psychological health and behavioral devel­ Norwegian scholar Dan Olweus (Olweus, 2013). However, some studies
opment of children and adolescents (Zarate-garza et al., 2017). in recent years have found that this definition is not fully applicable in
Compared to those who were uninvolved, students with bullying expe­ different social contexts such as different institutions, cultures, or lan­
rience (whether being bullied or bullying others) were more likely to guages (Smorti et al., 2003; Liu & Graves, 2011; Carrera et al., 2011;
develop internalization symptoms (Nielsen et al., 2015; Zhou et al., Langos, 2012). Thus, Olweus’s definition should not be taken as
2015; Evans et al., 2018; Gong et al., 2022) (e.g., post-traumatic stress axiomatic, instead, it should be re-examined in the light of different
disorder, low self-esteem, anxiety, depression, loneliness, and insomnia) social contexts (Volk et al., 2014). This study tests the applicability of
and externalizing problems (Ttofi et al., 2012; Takizawa et al., 2014; Su Olweus’ definition in a Chinese context.

* Corresponding author at: No. 715, North Section of Hupan Road, Tianfu New Area, Chengdu 611731, China.
E-mail address: [email protected] (J. Zhang).

https://doi.org/10.1016/j.childyouth.2024.107693
Received 2 January 2023; Received in revised form 19 May 2024; Accepted 19 May 2024
Available online 22 May 2024
0190-7409/© 2024 Elsevier Ltd. All rights are reserved, including those for text and data mining, AI training, and similar technologies.
Z. Gong et al. Children and Youth Services Review 161 (2024) 107693

2. Literature review one study found that both students and parents were more likely to
consider it to be school bullying than the other three types of bullying
2.1. Olweus’s definition of school bullying (verbal, physical, and relational) (Thomas et al., 2017). Overall, rela­
tional bullying is less likely to be deemed school bullying compared to
According to Olweus’s definition, a student is considered to suffer the other three types of bullying.
from school bullying if he is repeatedly exposed to one or more negative Students’ perceptions of the three characteristics of school bullying
behaviors intentionally committed by other student or students, with the (intentionality, repetitiveness, and power imbalance) and their rele­
existence of an unequal power relationship between the perpetrator and vance to the definition of bullying vary by country/region. For instance,
the victim (Olweus, 2013). Olweus’ definition emphasizes on three a study conducted in Italy shows that students in Italy considered all
characteristics that comprise school bullying: (1) Intentionality, indi­ three characteristics when defining school bullying (Menesini et al.,
cating that the perpetrator deliberately causes harm or makes others feel 2002). By contrast, findings from the studies conducted in the United
uncomfortable, which reflects the active nature of the behavior (Volk Kingdom (Madsen, 1996; Monks & Smith, 2006; Naylor et al., 2006) and
et al., 2014), i.e., the behaviors are not accidental. (2) Repetitiveness, Canada (Vaillancourt et al., 2008) suggest that students pay little
indicating that the perpetrator repeatedly bullies a particular student attention to the three characteristics when identifying school bullying.
over time; to a certain extent this also reflects the intentional nature of Moreover, some studies have found that only some of the three char­
the behavior (Olweus, 2013). Although other researchers believe that acteristics are used by students as the criteria to define bullying. For
some one-time serious bullying behaviors can also be counted as example, students in the Taiwan region were found to judge school
bullying (Dehue et al., 2008a), in general, repetitiveness represents a bullying mainly based on the intentionality of bullying behaviors and
greater likelihood of causing harm, and thus it can be viewed as a proxy the power imbalance between the perpetrator and the victim (Cheng
indicator for intentional harm (Volk et al., 2014). (3) Power imbalance et al., 2011), while students from the United States (Allen, 2015) and
between the perpetrator and the victim, reflecting the victim’s difficulty Ireland (Guerin & Hennessy, 2002) were found to mainly consider the
in resisting or defending against bullying (Olweus, 2013). The power feature of power imbalance. Swedish students were found to define
imbalance can be manifested in terms of age/grade, physical strength, or school bullying in terms of repetition and power imbalance rather than
popularity (Thomas et al., 2015). Among these, intentionality and intentionality (Hellström et al., 2015), while an experimental study from
repetitiveness can be used to distinguish bullying behavior from general Australia emphasized that intentionality was an important indicator of
joking behaviors or unintentional accidental injury behaviors (Langos, school bullying in Australia (Thomas et al., 2017). Moreover, a cross-
2012), and power imbalance can be used to distinguish bullying national study from Europe examined the definition of cyberbullying
behavior from other proactive provocative/aggressive behaviors (Volk and found that students in six European countries (Italy, Germany,
et al., 2014). Spain, Sweden, Estonia, and France) agreed that power imbalance was
an important criterion in defining cyberbullying, with less consideration
2.2. Types of school bullying given to repetitiveness (Menesini et al., 2012). Moreover, there also
existed national differences in the consideration of intentionality, for
School bullying can be divided into four types: verbal, physical, example, compared to students in Germany and Italy, students in the
relational, and cyberbullying (Liu & Graves, 2011; Olweus, 2013). other four countries were more likely to consider whether the bullying
Verbal bullying or bullying with cruel words refers to actively saying was committed intentionally (Menesini et al., 2012).
unfriendly words to others to harm, hurt, or scare them, it involves Compared to students, parents and teachers consider more bullying
insulting nicknames, ridicule, name-calling, or teasing. In contrast, characteristics when identifying school bullying. For instance, Madsen
physical bullying is the act of bullying others by threatening or harming (1996) found that teachers and parents in the United Kingdom perceived
others’ body or property, such as pushing/kicking/spitting on others, or bullying as consisting of repeated and intentional acts that caused harm,
snatching or destroying others’ property. Conversely, relational bullying while students did not consider repetitiveness and intentionality (Mad­
refers to the manipulation of social relationships to destroy or damage sen, 1996). A similar study, also conducted in the UK, compared
others’ social network or social reputation, such as social exclusion, teachers’ and students’ definition of bullying, and found that teachers
isolating someone, or spreading rumors. Cyberbullying refers to an in­ considered all three characteristics, while students considered almost
dividual or group using information technology to deliberately conduct none of them (Naylor et al., 2006). Similarly, it has been found that
aggressive behaviors against others, mainly manifested as relationship teachers in the United States (Allen, 2015) and the Taiwan region
manipulation and verbal insults. It involves abusing, ridiculing, sati­ (Cheng et al., 2011) tended to consider all three characteristics when
rizing, threatening, spreading rumors, ostracizing, or isolating a class­ identifying bullying incidents. However, students in the Taiwan region
mate online or by texting. Compared to cyberbullying, verbal, physical, mainly focused on intentionality and power imbalance (Cheng et al.,
and relational bullying are considered to be traditional forms of bullying 2011), while American students only emphasized on power imbalance
(Smith et al., 2008). (Allen, 2015). Only one study from Italy showed consistent perceptions
of bullying between teachers and students, and both of them agreed that
2.3. The definition of school bullying according to students, parents, and it was required to consider all three characteristics when defining
teachers bullying (Menesini et al., 2002).
In summary, research to date on the definition of school bullying
Different types of bullying behaviors differ in their likelihood of presents four general findings:
being defined as school bullying by students, parents, and teachers. In First, the criteria for defining school bullying proposed by Olweus
terms of traditional bullying, students are more likely to consider overt may not be the gold standard. Olweus’s definition originated from the
and direct harmful behaviors (i.e., physical and verbal bullying behav­ studies conducted in Norway; however, the above-mentioned studies
iors) to be bullying, and less likely to view relational bullying as school indicate that this definition is not fully applicable in other countries or
bullying (Maunder et al., 2010; Naylor et al., 2006). However, adults (i. regions. Accordingly, it is reasonable to doubt the applicability of this
e., parents and teachers) feel that physical injuries can cause more definition in China, with a different cultural, linguistic, and institutional
serious harm (Fry et al., 2020; Sawyer et al., 2011), and thus they are environment than Norway. Notably, although one study explored the
more likely to consider physical bullying as school bullying than verbal definition of school bullying in the Taiwan region (Cheng et al., 2011),
or relational bullying (Hazler et al., 2001; Monks & Smith, 2006). Thus, Chinese mainland and Taiwan region have different social systems.
adults pay less attention to and are less likely to intervene in verbal and Therefore, research conducted in the Taiwan region may not be repre­
relational bullying (Ellis & Shute, 2007). With regard to cyberbullying, sentative of the situation in the mainland of China.

2
Z. Gong et al. Children and Youth Services Review 161 (2024) 107693

Second, students’ definition of bullying is often not the same as that relational/cyberbullying behaviors as school bullying if the behavior is
used by researchers. When defining bullying, researchers are more intentional.
accustomed to use Olweus’ three characteristics of repetition, inten­ H3: Students/parents/teachers are more likely to define verbal/physical/
tionality, and power imbalance as the criteria. However, students do not relational/cyberbullying behavior as school bullying if the behavior is
often consider all or even any of these three characteristics (Slattery repeated.
et al., 2019). For this reason, if researchers still use the three charac­ H4: Students/parents/teachers are more likely to define verbal/physical/
teristics to determine whether students are involved in bullying, they are relational/cyberbullying behavior as school bullying if there is an imbalance
actually underestimating its prevalence (Hellström et al., 2015). It is, of power between the perpetrator and the victim.
therefore, necessary to understand how students define school bullying.
Third, there exist differences between students and adults (parents 4. Methods
and teachers) in the criteria they use for defining school bullying. Peo­
ple’s definition of bullying can affect their intervention behavior 4.1. Participants and procedures
(Sawyer et al., 2011). If adults use more characteristics than students to
identify school bullying, leading to more stringent definition of bullying, The participants in this study were teachers, students, and their
they may end up ignoring some practical bullying incidents that none­ parents from pre-college schools (primary, middle, high, and vocational
theless cause harm to students, and thus they will fail to make timely schools) in a county in Gansu Province, China. This region was selected
interventions (Monks & Smith, 2006). Therefore, it is required to un­ to conduct the survey because of the following two reasons: first, Gansu
derstand how students, parents, and teachers differently perceive school is located in the transition zone between nomadic culture and farming
bullying. Comparative analysis of these differences can help in pro­ culture, and has long been affected by the combination of the two cul­
moting agreement on the definition, thus contributing to the effective tures. Thus, it can roughly reflect the entire cultural background of
prevention of school bullying. China. Second, we conducted a school bullying survey in the same
Fourth, the four types of bullying may not be treated equally as county two years before this study, and popularized the definition of
school bullying. If a certain type of bullying is ignored, it is likely that school bullying among local students, parents, and teachers. Therefore,
the relevant groups (i.e., students, parents, and teachers) will not they have certain degree of cognitive ability with regard to the definition
intervene in bullying incidents in a timely manner. Therefore, to better of school bullying, which laid the foundation for the current investiga­
comprehend the perceptions of the three relevant groups on the four tion into the definition of school bullying. All teachers in the pre-college
types of bullying is of great practical significance for designing and schools of this county were valid participants. Students in grades 4
implementing comprehensive preventive measures and ensuring the through 6 of primary school and all grades of middle, high, and voca­
healthy growth of children and adolescents. tional schools were eligible. Students in grades 1 to 3 of primary school
were not included in the survey on the grounds that younger students
2.4. Necessity of studying the definition of school bullying in Chinese may not fully comprehend the questionnaire (Ba et al., 2019). However,
mainland parents of these students were also eligible for this study.
Herein, the questionnaires were designed for students, parents, and
To the best of our knowledge, empirical studies on the criteria for teachers, respectively, and the core content of these questionnaires was
defining school bullying in the mainland of China have never been re­ basically the same for all three groups. The questionnaires first intro­
ported to date, although the definition lays a strong foundation for the duced the purpose and significance of the study to respondents in order
research on practicing school bullying. Many studies have investigated to improve their cooperation, and then surveyed respondents’ socio-
the prevalence of school bullying in China, but their results vary widely demographic information and their perceptions of the definition of
(Chan & Wong, 2015). For example, Xiao et al. used repetition as a school bullying. These questionnaires were distributed to respondents
criterion to determine whether students were involved in bullying, and through an online questionnaire survey platform known as Question­
found that 1.78 % of students were bullies and 12.41 % were victims naire Star (https://www.wjx.cn/), and the questionnaires could be filled
(Xiao et al., 2021). However, another study that did not consider out using a computer system, WeChat, or QQ. To ensure the quality of
repetitiveness found that the prevalence of bullies and victims was the survey data, we worked closely with local education departments
30.00 % and 69.89 %, respectively (Fei et al., 2022). Furthermore, some and schools to ensure the eligibility of participants. Prior to the survey,
studies have considered both repeatability and intentionality (Gong respondents were informed that their participation in this study was
et al., 2022; Liu et al., 2023). Thus, the prevalence rates assessed by voluntary and anonymous and that they could withdraw from the survey
these inconsistent criteria are not conducive to comparative analysis and at any stage. Students were asked to fill out the questionnaires in the
hinder the development of practical application. Thus, it is extremely computer classroom of their respective schools. However, the school
necessary to conduct a study on the definition of school bullying in sent a link to the questionnaire to the parents and teachers and asked
Chinese mainland. them to complete it on their own. A trained research assistant provided
assistance to respondents in completing the survey. This study was
3. The current study approved by institutional review board (Grant number: K2019067) and
supported by the local education departments and schools of the survey
The main objective of this study is to explore how the relevant groups sites. The survey was conducted in December 2019. Finally, 1,943 stu­
(students, parents, and teachers) define school bullying in the context of dents, 1,957 parents, and 409 teachers participated in this study.
Chinese society. Specifically, this study aims to analyze whether the
relevant groups consider the three characteristics of intentionality, 4.2. Survey experimental design
repetitiveness, and power imbalance when defining school bullying, and
whether there exist differences in the likelihood that the four types of This study adopted a survey experimental design. The experimental
bullying (verbal, physical, relational, and cyberbullying) are considered scenarios were designed according to the types and characteristics of
as school bullying. According to Olweus’s definition, we posit the school bullying. This study considered four types of school bullying:
following hypotheses: verbal, physical, relational, and cyberbullying, and each of the charac­
H1: There are no significant differences in the likelihood that verbal, teristics of bullying (intentionality, repetitiveness, and power imbal­
physical, relational, and cyberbullying behaviors will be defined by students/ ance) was assigned one of the two values (no = 0, yes = 1). Therefore,
parents/teachers as school bullying. this study included 32 (4 × 2 × 2 × 2) scenarios. Based on previous
H2: Students/parents/teachers are more likely to define verbal/physical/ studies (Olweus, 2013; Thomas et al., 2017; Bell & Willis, 2016), the

3
Z. Gong et al. Children and Youth Services Review 161 (2024) 107693

corresponding description statements for each variable are presented in 4.3. Measurements
Table 1. These statements were added to a scenario template “Statement
of power imbalance. Today, Statement of bullying type. Statement of In addition to measuring dependent and independent variables from
Repetitiveness. Statement of Intentionality” to form the scenarios. An survey experiments, this study also investigated the gender and educa­
example is as follows: tion level of the respondents because previous studies have suggested
Xiao Liu and Xiao Li were students of the same grade. Today, Xiao Liu that these variables can affect the definition of school bullying (Aalsma
called Xiao Li a mean name. This was the first time Xiao Liu has done this to & Brown, 2008; Hellström & Lundberg, 2020; Naylor et al., 2006). In
Xiao Li. Xiao Liu did this thing as a joke, and did not want to make Xiao Li this study, gender (female = 0, male = 1) was a binary variable. The
sad. options for education level of students included “primary school”,
Noteworthy, to avoid the potential confounding effect of gender “middle school”, “high school”, and “vocational school”. The possible
(Holfeld, 2014), in this study, specific names of the perpetrator and responses for education level of parents were “primary or below”,
victim were not used because such a name could have a potential gender “middle school”, “high/vocational school”, “junior college”, and
associated with it. Thus, the scenarios used common Chinese surnames “bachelor or above”. For teachers, the options for education level were
to represent the perpetrator (Xiao Liu) and victim (Xiao Li). “high/vocational school or below”, “junior college”, and “bachelor or
This study chose a random selection strategy that assumed interac­ above”.
tion effects were zero or negligible in the survey experiment (Atzmüller
& Steiner, 2010), because this study focused on the main effects of the 4.4. Sample size calculation
three characteristics of school bullying rather than on their interaction
effects. Accordingly, the experimental scenarios were randomly According to the experimental design, eight scenarios (intentionality
assigned to the respondents by using a computer software (a function (no/yes) × repetitiveness (no/yes) × power imbalance (no/yes)) were
provided by Questionnaire Star). One of the objectives of this study was designed for each type of bullying. Therefore, sample size was calculated
to determine whether the four types of bullying were equally defined as to detect a medium effect of Cohen’s f = 0.25 when comparing the eight
school bullying; therefore, herein it was required to assign a similar scenarios with a power of 90 % and an alpha error of 5 % (Cohen, 1988).
number of scenarios for each type of bullying. To this end, the 32 sce­ The results show that the sample size involved in each scenario was not
narios were divided into four groups according to the four types of less than 22. It indicates that the sample size of this study meets the
bullying. Then, one scenario from each group was randomly selected for requirements.
each respondent. Thus, each respondent received a total of four sce­
narios. Notably, to avoid the order effect, the four scenarios could not be 4.5. Statistical analysis
presented to respondents in a uniform and unchanging order. For
example, the four scenarios could not always be presented to re­ In the current study, each respondent answered four scenario ques­
spondents in the order of verbal, physical, relational, and cyberbullying tions, resulting in a two-level nested structure of data, i.e., the scenario-
scenarios. Therefore, the four scenario groups were randomly arranged level data nested within respondents. Therefore, multi-level (two-level)
to form a total of 24 orders. One of these orders was then randomly models were used to analyze differences in the likelihood of the four
assigned to each respondent with the help of computer software. types of bullying being deemed school bullying, and the impact of the
Overall, neither the experimenter nor the subject knew which scenario three characteristics (intentionality, repetitiveness, and power imbal­
they were assigned to, thus it was a double-blind allocation strategy. The ance) on the definition that the three types of respondents gave to school
flow diagrams of participants are shown in supplementary document. bullying. Based on the objectives of the present research, this study
Before reading the experimental scenarios, the respondents were first focused on fixed effects, and thus the maximum likelihood estimation
told “There are four scenarios below, each of them followed by a ques­ method was selected for parameter evaluation (Willits, 2015). First,
tion. Please read the scenario first, and then answer the question.” Re­ empty multi-level models containing only the intercept were fitted to
spondents could then begin reading the scenarios. After they were done calculate intraclass correlations (ICCs). The ICCs for students, parents,
reading each scenario, they were asked to answer the question “Do you and teachers were 0.55, 0.54, and 0.35, respectively, indicating that the
think the situation described in the scenario is school bullying?” The multilevel models were appropriate for data analysis. Scenario-level
responses to the question included five options 1 = definitely not, 2 = variables (the types and the three characteristics of bullying) and
probably not, 3 = not sure, 4 = probably, and 5 = definitely. individual-level variables (gender and education level) were then
simultaneously incorporated into the multi-level model. Moreover, to
determine whether respondents considered the three characteristics of
Table 1 intentionality, repetitiveness, and power imbalance when defining a
Statements of variables in the scenarios. specific category of school bullying, this study used ordinary linear re­
gressions to analyze the associations between the three characteristics
Variables Values Statements
and the definition of each type of bullying. Statistical significance was
Power No • Xiao Liu and Xiao Li were students of the same
considered to be a P value smaller than 0.05. All analyses were per­
imbalance grade.
Yes • Xiao Liu was a higher-grade student than Xiao Li. formed by using Stata/MP 14.0.
Bullying type Verbal • Xiao Liu called Xiao Li a mean name.
Physical • In the hallway where Xiao Li was on the way to 5. Results
class, Xiao Liu pushed Xiao Li with force.
Relational • During group activities, Xiao Liu left Xiao Li out
of the activities.
5.1. Descriptive results
Cyber • Xiao Liu spread bad information about Xiao Li on
the Internet, and Xiao Li and other students could Of the total students who took the survey, 48.02 % were girls. Pri­
see the information. mary, middle, high, and vocational schools’ students accounted for
Repetitiveness No • This was the first time Xiao Liu has done this to
37.62 %, 36.08 %, 16.37 %, and 9.93 % of the total, respectively. Of the
Xiao Li.
Yes • In the past month, Xiao Liu has done this to Xiao parents who participated in this study, 1,148 were females (58.66 %).
Li many times. Moreover, 24.94 %, 44.15 %, 15.02 %, 9.25 %, and 6.64 % of the parents
Intentionality No • Xiao Liu did this thing as a joke, and did not want had education levels of primary school or below, middle school, high/
to make Xiao Li sad. vocational school, junior college, and bachelor’s degree or above,
Yes • Xiao Liu did this to make Xiao Li feel bad.
respectively. Less than half of the teachers were females (45.48 %). Most

4
Z. Gong et al. Children and Youth Services Review 161 (2024) 107693

of the teachers (79.22 %) had a bachelor’s degree or above, followed by Table 3

junior college (16.87 %) and high school or below (3.91 %). The impacts of the types and characteristics of bullying on school bullying
The mean ± standard deviations (SD) of participants’ ratings of the definition.
scenarios as school bullying are presented in Table 2. Overall, students’ Variables Student Parent model Teacher
ratings (mean = 4.27, SD = 1.11) were higher than those of the parents model model
(mean = 3.85, SD = 1.33) and teachers (mean = 4.18, SD = 1.20), Level 1: scenario-level
indicating that students were more likely to define the scenarios as Repetitiveness 0.26***(0.02) 0.40***(0.02) 0.76***(0.05)
bullying. This generally held true for every type of bullying scenario. Intentionality 0.27***(0.02) 0.30***(0.02) 0.43***(0.05)
Power imbalance 0.06**(0.02) 0.07***(0.02) 0.19***(0.05)
Moreover, respondents in all three groups gave the highest average
Bullying types (Ref:
rating to cyberbullying scenarios (students: mean = 4.36; parents: mean relational)
= 3.99; teachers: mean = 4.30) and the lowest average rating to rela­ Verbal 0.17***(0.02) 0.11***(0.03) 0.26***(0.06)
tional bullying scenarios (students: mean = 4.16; parents: mean = 3.74; Physical 0.06**(0.02) 0.08**(0.03) 0.08(0.06)
teachers: mean = 4.03). Cyber 0.20***(0.02) 0.25***(0.03) 0.27***(0.06)
Level 2: individual-level
Male − 0.04(0.04) − 0.20***(0.05) 0.00(0.08)
5.2. Multi-level analysis results Education − 0.05*(0.02) 0.12***(0.02) 0.22**(0.08)
** ***
Note: Standard errors in parentheses, * p < 0.05, p < 0.01, p < 0.001.
Table 3 presents the impacts of bullying types, repetitiveness,
intentionality, and power imbalance on participants’ definitions of
school bullying. Students, parents, and teachers were more likely to 5.4. Ordinary linear regression results for physical bullying
agree that scenarios with characteristics of repetitiveness (students: β=
0.26, P < 0.001; parents: β = 0.40, P < 0.001; teachers: β = 0.76, P < The impacts of repetitiveness, intentionality, and power imbalance
0.001), intentionality (students: β = 0.27, P < 0.001; parents: β = 0.30, on the definition of physical bullying are shown in Fig. 2. Scenarios that
P < 0.001; teachers: β = 0.43, P < 0.001), or power imbalance (students: both students and parents perceived as having repetitive (students: β =
β = 0.06, P < 0.01; parents: β = 0.07, P < 0.001; teachers: β = 0.19, P < 0.33, P < 0.001; parents: β = 0.56, P < 0.001) or intentional (students: β
0.001) should be defined as school bullying than those without repeti­ = 0.25, P < 0.001; parents: β = 0.31, P < 0.001) characteristics were
tiveness, intentionality, or power imbalance. Regarding the bullying more likely to be considered as school bullying. However, the charac­
types, both students and parents were more likely to consider verbal teristic of power imbalance did not significantly affect their definition of
(students: β = 0.17, P < 0.001; parents: β = 0.11, P < 0.001), physical physical bullying. In comparison, teachers felt that all three of Olweus’
(students: β = 0.06, P < 0.01; parents: β = 0.08, P < 0.01), and cyber­ characteristics should play significant roles in defining physical bullying
bullying (students: β = 0.20, P < 0.001; parents: β = 0.25, P < 0.001) (repetitiveness: β = 0.85, P < 0.001; intentionality: β = 0.28, P < 0.05;
scenarios as school bullying, compared to relational bullying. Teachers power imbalance: β = 0.34, P < 0.01).
were more likely to view verbal (β = 0.26, P < 0.001) and cyberbullying
(β = 0.27, P < 0.001) scenarios as school bullying than relational 5.5. Ordinary linear regression results for relational bullying
bullying scenarios. However, there were non-significant differences
between relational bullying scenarios and physical bullying scenarios in Students, parents, and teachers were all more likely to perceive
terms of their likelihood of being defined as school bullying. scenarios with a repetitive feature (students: β = 0.34, P < 0.001; par­
ents: β = 0.50, P < 0.001; teachers: β = 0.87, P < 0.001) or with an
5.3. Ordinary linear regression results for verbal bullying intentional characteristic (students: β = 0.33, P < 0.001; parents: β =
0.34, P < 0.001; teachers: β = 0.68, P < 0.001) to be school bullying than
From the perspective of students, verbal bullying scenarios with a scenarios without these characteristics (Fig. 3). Moreover, there was no
repetitive feature were more likely to be defined as school bullying than significant difference in the likelihood of being defined as school
those without repetition (β = 0.16, P < 0.001) (Fig. 1). Similarly, those bullying between scenarios with and without the characteristic of power
with an intentional characteristic were more likely to be perceived by imbalance.
students as school bullying (β = 0.21, P < 0.001) than unintentional
scenarios. However, the probability of a scenario being regarded as 5.6. Ordinary linear regression results for cyberbullying
school bullying did not differ significantly between scenarios with and
without the characteristic of power imbalance. Nonetheless, both par­ Among the three characteristics, only intentionality showed a sig­
ents and teachers were more likely to perceive scenarios characterized nificant effect on the definition of cyberbullying given by students,
by repetitiveness (parents: β = 0.31, P < 0.001; teachers: β = 0.74, P < parents, and teachers (students: β = 0.22, P < 0.001; parents: β = 0.22, P
0.001), intentionality (parents: β = 0.29, P < 0.001; teachers: β = 0.26, < 0.001; teachers: β = 0.35, P < 0.01) (Fig. 4). Moreover, parents and
P < 0.05), or power imbalance (parents: β = 0.15P < 0.05; teachers: β = teachers were also more likely to feel that scenarios with the feature of
0.38, P < 0.001) to be school bullying than scenarios lacking those repetitiveness were school bullying than those without this character­
characteristics. istic (parents: β = 0.13, P < 0.05; teachers: β = 0.59, P < 0.001).
However, the characteristic of power imbalance did not affect the
definition of cyberbullying among students, parents, and teachers.

6. Discussion
Table 2
Participants’ rating of the scenarios as school bullying.
The criterion for the definition of bullying is the basis of research and
Scenarios Students Parents Teachers practical study on school bullying. Using a survey experimental design,
Mean (SD) Mean (SD) Mean (SD) this study explored how students, parents, and teachers defined school
Verbal 4.32 (1.06) 3.85 (1.32) 4.28 (1.11) bullying in the mainland of China. Specifically, this study analyzed
Physical 4.22 (1.13) 3.83 (1.35) 4.10 (1.25) whether Olweus’ three characteristics of intentionality, repetitiveness,
Relational 4.16 (1.19) 3.74 (1.38) 4.03 (1.31) and power imbalance comprised the standard for defining school
Cyber 4.36 (1.05) 3.99 (1.27) 4.30 (1.11) bullying and whether the four types of bullying (verbal, physical, rela­
Overall 4.27 (1.11) 3.85 (1.33) 4.18 (1.20)
tional, and cyberbullying) were equally likely to be defined as school

5
Z. Gong et al. Children and Youth Services Review 161 (2024) 107693

Fig. 1. The impacts of three characteristics on verbal bullying definition.

Fig. 2. The impacts of the three characteristics on physical bullying definition.

bullying. To the best of our knowledge, this is the first study on school with regard to school bullying, and provides an important scientific
bullying conducted from a multi-subject (students, parents, and teach­ basis for development of prevention and intervention strategies for
ers) perspective in the context of China. Overall, this study helps to school bullying.
uncover the cognitive differences among students, parents, and teachers The four types of bullying differ in their likelihood of being defined

6
Z. Gong et al. Children and Youth Services Review 161 (2024) 107693

Fig. 3. The impacts of three characteristics on relational bullying definition.

Fig. 4. The impacts of three characteristics on cyberbullying definition.

by students, parents, and teachers as school bullying. Compared with such as physical and verbal bullying as school bullying, but do not re­
relational bullying, verbal, physical, and cyberbullying are all more gard indirect or implicit bullying behavior (i.e., relational bullying) as
likely to be considered as bullying. This finding supports the existing school bullying (Naylor et al., 2006; Thomas et al., 2017). This may be
view that people generally define explicit and direct bullying behaviors attributed to the fact that the external manifestations of relational

7
Z. Gong et al. Children and Youth Services Review 161 (2024) 107693

bullying are more subtle and difficult to identify than those of direct group considered power imbalance. Second, there were differences in
bullying. For example, people tend to dismiss behaviors such as exclu­ the criteria used by teachers and students when defining verbal, phys­
sion or isolation as normal interpersonal phenomena. Therefore, those ical, and cyberbullying. Specifically, teachers used all the three char­
who have not experienced relational bullying may have difficulty in acteristics as defining criteria for verbal and physical bullying, but
intuitively perceiving the harm it causes (Frisén et al., 2008), and as a students used only two characteristics, intentionality and repetitiveness,
result may believe that relational bullying is less serious than other as criteria. Furthermore, teachers required repetitiveness before recog­
forms of bullying (Maunder et al., 2010; Bell & Willis, 2016). nizing a behavior as cyberbullying, while students did not. Overall,
Students, parents, and teachers all have different viewpoints when compared with students, adults used more characteristics as the criteria
determining whether a particular behavior constitutes school bullying. for defining school bullying.
In this study, both students and parents were found to be more likely to
believe that compared to relational bullying, physical bullying consti­ 6.1. Implications
tuted school bullying. However, teachers did not see a significant dif­
ference between physical and relational bullying. Moreover, as teachers The current study provides some important practical and theoretical
were less likely to consider relational bullying as school bullying than implications. First, practical investigation should give more attention to
verbal bullying and cyberbullying, physical bullying accordingly also relational bullying. This study exhibits that relational bullying is less
exhibited a lower probability to be perceived as school bullying than likely than verbal, physical, and cyberbullying to be deemed school
verbal bullying and cyberbullying. This is inconsistent with prior find­ bullying by students and parents. This is despite the fact that studies
ings that showed that teachers believe physical bullying to be more have shown that relational bullying causes more negative trauma to the
serious and are more likely to see physical bullying as school bullying mental health of victims than the other three types of bullying (Yuchang
than other types of bullying behaviors (Monks & Smith, 2006; Cheng et al., 2019). Therefore, inability of the people to recognize relational
et al., 2011; Bell & Willis, 2016; Fry et al., 2020). The reason may be that bullying needs to be corrected if it is required to facilitate timely
teachers perceived the descriptions of physical bullying behavior in the detection and intervention in relational bullying incidents. More edu­
experimental scenario of this study as playfulness among students rather cation and knowledge are needed on this harmful pattern of behaviors so
than physical harm. Moreover, prior studies have shown that teachers that this impression can be changed over time. Moreover, this negli­
have difficulty distinguishing between physical bullying and physical gence of relational bullying also suggests that when conducting a school
conflict (e.g., fighting) (Mishna et al., 2005), and thus teachers may also bullying prevalence survey, a general question (e.g., “Have you ever
have seen the scenarios in this study as constituting physical conflict bullied your peers?”) may not reflect the real situation. A student who
rather than bullying. If this is true, the consequences are serious because has committed relational bullying may fail to report his/her bullying
teachers may respond to physical bullying incidents with strategies to experience because he/she is unaware that he/she has actually
resolve physical conflicts and thereby fail to quickly protect students committed an act of bullying. Accordingly, surveys for prevalence
who are being physically bullied. assessment should use questionnaires designed for each type of bullying
Students do not consider all three characteristics of intentionality, to obtain more accurate evaluation results.
repetitiveness, and power imbalance when defining the four types of Second, adults (parents and teachers) should view and respond to
bullying. This is manifested in the following two ways. First, inconsistent school bullying through the lens of students’ definitions. The findings of
with the results of studies from Europe and the United States (Menesini this study indicate that adults consider more characteristics than stu­
et al., 2002; Madsen, 1996; Vaillancourt et al., 2008; Allen, 2015; dents when defining school bullying. This may make adults fail to
Hellström et al., 2015; Mishna et al., 2006; Ybarra et al., 2019), this identify some incidents as bullying even when the victims consider
study found that Chinese students only considered whether bullying themselves to have been bullied and, as a result, they may not intervene
behaviors were intentional and repetitive when identifying verbal, (Monks & Smith, 2006), leading to continuing harm and ignorance of
physical, and relational bullying, and did not focus on whether there the harm suffered by the victimized students. It is critical that students,
were any power imbalances between the perpetrator and the victim. parents, and teachers have a shared understanding of what school
Second, students’ definition of cyberbullying was even looser, i.e., they bullying is (Brown et al., 2022), and to this end they should strengthen
only considered the characteristic of intentionality. The reason they did communication on the issue of school bullying.
not consider the power imbalance highlighted by previous studies on Third, it may be better to use the students’ definitions as criteria in
cyberbullying (Langos, 2012; Menesini et al., 2012; Pieschl et al., 2013) assessing the prevalence of bullying in schools. Noteworthy, this study
is that, with the popularity of the Internet, people’s computer skills have demonstrates that students do not consider all three characteristics
increased, thus they believe that bullying victims are able to easily resist (repetitiveness, intentionality, and power imbalance) when defining
the perpetrators in cyberspace (Thomas et al., 2015). Therefore, there is school bullying. If these three characteristics are still used as the stan­
no obvious power imbalance online between the perpetrator and the dard to measure the prevalence of school bullying, it can lead to under-
victim. Moreover, the absence of the feature of repetitiveness indicates a estimation of prevalence. Therefore, based on the definition of students,
need to revisit the necessity of repetitiveness in the definition of this study indicates that only intentionality and repetitiveness need to be
cyberbullying. This is attributed to the fact that, in the online environ­ considered when defining verbal, physical, and relational bullying, and
ment, some single-shot bullying behaviors can have a long persistence only the characteristic of intentionality needs to be considered when
online and thus can be seen by many people, causing lasting harm to the defining cyberbullying. However, we admit that we cannot claim the full
victim (Dehue et al., 2008b). Overall, students do not use all three justification of this implication, and undeniably a lot more systematic
characteristics as criteria for defining a specific type of school bullying explorations and additional evidences are required to draw definitive
for any of the four types of bullying. conclusions, even though the students in this study have been educated
In comparison, adults’ (parents’ and teachers’) criteria for defining about school bullying, and they have a certain understanding of
school bullying were found to be different from those of students in two bullying.
ways. First, the differences between parents and students are reflected in
their definitions of physical, relational, and cyberbullying. Unlike stu­ 6.2. Limitations and future directions
dents, who mainly considered intentionality and repetitiveness when
defining physical and relational bullying, parents considered all three This study has some limitations that need to be pointed out. First, the
features. Moreover, while students only considered intentionality when random selection strategy used in the survey experiments made it
defining cyberbullying, parents tended to use the two characteristics of difficult to calculate the interaction effects (Rossi & Nock, 1982). This is
intentionality and repetitiveness to identify cyberbullying; and neither due to the fact that the random selection strategy leads to a complex

8
Z. Gong et al. Children and Youth Services Review 161 (2024) 107693

random confounding structure, making the interaction effects Writing – original draft, Supervision, Writing – review & editing. Zhiwei
confounded within the main effects and difficult to control (Atzmüller & Tang: Investigation, Supervision. Jing Zhou: Visualization, Investiga­
Steiner, 2010). Notably, this study focused on main effects rather than tion. Ziqiang Han: Investigation, Software, Visualization. Jingran
interaction effects; therefore, the failure to calculate interaction effects Zhang: Conceptualization, Methodology, Writing – review & editing.
does not conflict with the purposes of our study. Second, the experi­
mental design did not involve any manipulation check, such as an Declaration of competing interest
instructional manipulation check (Oppenheimer et al., 2009) or a
factual manipulation check (Kane & Barabas, 2019). Nonetheless, the The authors declare that they have no known competing financial
descriptive statements about the three characteristics of school bullying interests or personal relationships that could have appeared to influence
in the experimental scenarios were easy to understand. Respondents’ the work reported in this paper.
understanding of these statements has been confirmed in prior research
(Thomas et al., 2017). Therefore, herein, it was believed that the lack of Data availability
manipulation check has less impact on the validity of the data than it
might have otherwise. Third, the sample of this study was limited, as it Data will be made available on request.
was drawn from only one county in China. Schools in other areas may
have different school climates, which may affect people’s perception of Acknowledgments
school bullying. Therefore, incorporation of school-level variables into
research in the future may be a useful approach. Moreover, differences We acknowledged all the students, parents, and teachers who
in students’ popularity or body size can also be regarded as forms of participated in our survey. Moreover, the authors would like to thank
power imbalance (Thomas et al., 2015), therefore, the power imbalance the help provided by the education department of the participating
characteristic reflected by grade level in the experimental scenarios may school districts and Professor Zhanlong Ba for data collection.
not fully represent this characteristic.
Future studies can continue to examine the definition of school Funding
bullying from broader perspectives. First, previous studies indicate that
gender and grade play very important roles in investigating bullying This work was supported by National Natural Science Foundation of
behavior (Gong et al., 2020; Yang et al., 2022; Luo et al., 2023), and thus China (grant number 72204040), Humanities and Social Sciences Youth
it is of great practical significance to analyze students’ definition of Foundation of Ministry of Education of China (grant number
bullying according to gender and grade levels. Second, some studies 22YJCZH035).
have pointed out that anonymity and publicity are two important
characteristics of cyberbullying (Menesini et al., 2012; Thomas et al., Appendix A. Supplementary data
2015), therefore, future research should examine the impact of these
two characteristics on the definition of cyberbullying. Third, the par­ Supplementary data to this article can be found online at https://doi.
ticipants in this study are from the west of China, where less attention is org/10.1016/j.childyouth.2024.107693.
paid to school bullying than that in the east of China. Therefore, it is
worth exploring whether students, parents, and teachers’ definitions of References
school bullying are consistent with those in the west of China.
Aalsma, M. C., & Brown, J. R. (2008). What Is Bullying? Journal of Adolescent Health, 43
7. Conclusions (2), 101–102. https://doi.org/10.1016/j.jadohealth.2008.06.001
Allen, K. P. (2015). “We Don’t Have Bullying, But We Have Drama”: Understandings of
Bullying and Related Constructs Within the Social Milieu of a U.S. High School.
In this study, a comparison of definition of school bullying was made Journal of Human Behavior in the Social Environment, 25(3), 159–181. https://doi.
experimentally among students, parents, and teachers in the mainland org/10.1080/10911359.2014.893857
Atzmüller, C., & Steiner, P. M. (2010). Experimental vignette studies in survey research.
of China. This study uncovered some valuable findings. First, verbal,
Methodology: European Journal of Research Methods for the Behavioral and Social.
physical, relational, and cyberbullying have different probabilities of Sciences, 6(3), 128–138. https://doi.org/10.1027/1614-2241/a000014
being considered as school bullying. Among them, relational bullying is Ba, Z., Han, Z., Gong, Z., Li, F., Zhang, H., & Zhang, G. (2019). Ethnic differences in
the least likely to be defined as school bullying. Therefore, practical experiences of school bullying in China. Children and Youth Services Review, 104,
Article 104402. https://doi.org/10.1016/j.childyouth.2019.104402
investigation should give more attention to relational bullying. Second, Bell, K. J. S., & Willis, W. G. (2016). Teachers’ perceptions of bullying among youth.
students, parents, and teachers have different definitions of the four Journal of Educational Research, 109(2), 159–168. https://doi.org/10.1080/
types of bullying. Both students and parents are more likely to believe 00220671.2014.931833
Brown, J., Keesler, J., Karikari, I., Ashrifi, G., & Kausch, M. (2022). School Principals
that physical bullying is school bullying than the case for relational Putting Bullying Policy to Practice. Journal of Interpersonal Violence, 37(1–2),
bullying; however, teachers perceive that there is no significant differ­ NP281–NP305. https://doi.org/10.1177/0886260520914553
ence between physical and relational bullying. Third, students, parents, Carrera, M. V., DePalma, R., & Lameiras, M. (2011). Toward a More Comprehensive
Understanding of Bullying in School Settings. Educational Psychology Review, 23(4),
and teachers do not completely consider the three characteristics of 479–499. https://doi.org/10.1007/s10648-011-9171-x
intentionality, repetitiveness, and power imbalance as criteria for Chan, H. C., & Wong, D. S. W. (2015). Traditional school bullying and cyberbullying in
defining school bullying, and parents and teachers consider more Chinese societies: Prevalence and a review of the whole-school intervention
approach. Aggression and Violent Behavior, 23, 98–108. https://doi.org/10.1016/j.
characteristics than students when defining school bullying. Accord­ avb.2015.05.010
ingly, parents and teachers should view school bullying through the Cheng, Y.-Y., Chen, L.-M., Ho, H.-C., & Cheng, C.-L. (2011). Definitions of school bullying
students’ broader lens when responding to bullying. in Taiwan: A comparison of multiple perspectives. School Psychology International, 32
(3), 227–243. https://doi.org/10.1177/0143034311404130
Cohen, J. (1988). Statistical Power Analysis for the Behavioral Sciences (2nd ed.).
Ethical approval Routledge. DOI: 10.4324/9780203771587.
Dehue, F., Bolman, C., & Völlink, T. (2008). Cyberbullying: Youngsters’ Experiences and
Parental Perception. CyberPsychology & Behavior, 11(2), 217–223. https://doi.org/
This study was approved by institutional review board (grant num­
10.1089/cpb.2007.0008
ber: K2019067). Ellis, A. A., & Shute, R. (2007). Teacher responses to bullying in relation to moral
orientation and seriousness of bullying. British Journal of Educational Psychology, 77
CRediT authorship contribution statement (3), 649–663. https://doi.org/10.1348/000709906X163405
Evans, C. B. R., Smokowski, P. R., Rose, R. A., Mercado, M. C., & Marshall, K. J. (2018).
Cumulative Bullying Experiences, Adolescent Behavioral and Mental Health, and
Zepeng Gong: Conceptualization, Methodology, Data curation, Academic Achievement: An Integrative Model of Perpetration, Victimization, and

9
Z. Gong et al. Children and Youth Services Review 161 (2024) 107693

Bystander Behavior. Journal of Child and Family Studies. https://doi.org/10.1007/ Olweus, D. (2013). School Bullying: Development and Some Important Challenges.
s10826-018-1078-4 Annual Review of Clinical Psychology, 9(1), 751–780. https://doi.org/10.1146/
Fei, L., Liao, M., Ke, L., Zou, Y., Li, X., Chen, Y., & Zhang, R. (2022). School bullying annurev-clinpsy-050212-185516
among Chinese third to fifth grade primary school students in a cross-sectional study: Olweus, D., & Limber, S. P. (2018). Some problems with cyberbullying research. Current
The protective effect of psychological resilience. PLOS ONE, 17(12), e0278698. Opinion in Psychology, 19, 139–143. https://doi.org/10.1016/j.copsyc.2017.04.012
Frisén, A., Holmqvist, K., & Oscarsson, D. (2008). 13-year-olds’ perception of bullying: Oppenheimer, D. M., Meyvis, T., & Davidenko, N. (2009). Instructional manipulation
Definitions, reasons for victimisation and experience of adults’ response. Educational checks: Detecting satisficing to increase statistical power. Journal of Experimental
Studies, 34(2), 105–117. https://doi.org/10.1080/03055690701811149 Social Psychology, 45(4), 867–872. https://doi.org/10.1016/j.jesp.2009.03.009
Fry, D., Mackay, K., Childers-Buschle, K., Wazny, K., & Bahou, L. (2020). “They are Pieschl, S., Porsch, T., Kahl, T., & Klockenbusch, R. (2013). Relevant dimensions of
teaching us to deliver lessons and that is not all that teaching is”: Exploring teacher cyberbullying—Results from two experimental studies. Journal of Applied
trainees’ language for peer victimisation in schools. Teaching and Teacher Education, Developmental Psychology, 34(5), 241–252. https://doi.org/10.1016/j.
89, Article 102988. https://doi.org/10.1016/j.tate.2019.102988 appdev.2013.04.002
Gong, Z., Han, Z., Zhang, H., & Zhang, G. (2020). Weight Status and School Bullying Rossi, P. H., & Nock, S. L. (1982). Measuring social judgments: The factorial survey
Experiences in Urban China: The Difference Between Boys and Girls. Journal of approach. Sage Publications.
Interpersonal Violence, 35(15–16), 2663–2686. https://doi.org/10.1177/ Sawyer, J.-L., Mishna, F., Pepler, D., & Wiener, J. (2011). The missing voice: Parents’
0886260519880170 perspectives of bullying. Children and Youth Services Review, 33(10), 1795–1803.
Gong, Z., Reinhardt, J. D., Han, Z., Ba, Z., & Lei, S. (2022). Associations between school https://doi.org/10.1016/j.childyouth.2011.05.010
bullying and anxiety in children and adolescents from an ethnic autonomous county Slattery, L. C., George, H. P., & Kern, L. (2019). Defining the word bullying:
in China. Psychiatry Research, 314, Article 114649. https://doi.org/10.1016/j. Inconsistencies and lack of clarity among current definitions. Preventing School
psychres.2022.114649 Failure: Alternative Education for Children and Youth, 63(3), 227–235. https://doi.org/
Guerin, S., & Hennessy, E. (2002). Pupils’ definitions of bullying. European Journal of 10.1080/1045988X.2019.1579164
Psychology of Education, 17(3), 249–261. https://doi.org/10.1007/BF03173535 Smith, P. K., Mahdavi, J., Carvalho, M., Fisher, S., Russell, S., & Tippett, N. (2008).
Hazler, R. J., Miller, D. L., Carney, J. V., & Green, S. (2001). Adult recognition of school Cyberbullying: Its nature and impact in secondary school pupils. Journal of Child
bullying situations. Educational Research, 43(2), 133–146. https://doi.org/10.1080/ Psychology and Psychiatry, and Allied Disciplines, 49(4), 376–385. https://doi.org/
00131880110051137 10.1111/j.1469-7610.2007.01846.x
Hellström, L., & Lundberg, A. (2020). Understanding bullying from young people’s Smorti, A., Menesini, E., & Smith, P. K. (2003). Parents’ Definitions of Children’s
perspectives: An exploratory study. Educational Research, 62(4), 414–433. https:// Bullying in a Five-Country Comparison. Journal of Cross-Cultural Psychology, 34(4),
doi.org/10.1080/00131881.2020.1821388 417–432. https://doi.org/10.1177/0022022103034004003
Hellström, L., Persson, L., & Hagquist, C. (2015). Understanding and defining bullying – Su, P.-Y., Wang, G.-F., He, H., Han, A.-Z., Zhang, G.-B., & Xu, N. (2019). Is involvement in
adolescents’ own views. Archives of Public Health, 73(1). https://doi.org/10.1186/ school bullying associated with increased risk of murderous ideation and behaviours
2049-3258-73-4 among adolescent students in China? Bmc Psychiatry, 19, 121. https://doi.org/
Holfeld, B. (2014). Perceptions and attributions of bystanders to cyber bullying. 10.1186/s12888-019-2108-5
Computers in Human Behavior, 38, 1–7. https://doi.org/10.1016/j.chb.2014.05.012 Takizawa, R., Maughan, B., & Arseneault, L. (2014). Adult Health Outcomes of
Kane, J. V., & Barabas, J. (2019). No Harm in Checking: Using Factual Manipulation Childhood Bullying Victimization: Evidence From a Five-Decade Longitudinal British
Checks to Assess Attentiveness in Experiments. American Journal of Political Science, Birth Cohort. American Journal of Psychiatry, 171(7), 777–784. https://doi.org/
63(1), 234–249. https://doi.org/10.1111/ajps.12396 10.1176/appi.ajp.2014.13101401
Langos, C. (2012). Cyberbullying: The Challenge to Define. Cyberpsychology Behavior and Thomas, H. J., Connor, J. P., Baguley, C. M., & Scott, J. G. (2017). Two sides to the story:
Social Networking, 15(6), 285–289. https://doi.org/10.1089/cyber.2011.0588 Adolescent and parent views on harmful intention in defining school bullying.
Liu, J., & Graves, N. (2011). Childhood Bullying: A Review of Constructs, Concepts, and Aggressive Behavior, 43(4), 352–363. https://doi.org/10.1002/ab.21694
Nursing Implications. Public Health Nursing, 28(6), 556–568. https://doi.org/ Thomas, H. J., Connor, J. P., & Scott, J. G. (2015). Integrating Traditional Bullying and
10.1111/j.1525-1446.2011.00972.x Cyberbullying: Challenges of Definition and Measurement in Adolescents – a Review.
Liu, J., Han, Z., Ma, X., & Xin, R. (2023). Moral disengagement and cyberbullying Educational Psychology Review, 27(1), 135–152. https://doi.org/10.1007/s10648-
perpetration among adolescents: The moderating role of empathy. Children and 014-9261-7
Youth Services Review, 155, Article 107178. https://doi.org/10.1016/j. Ttofi, M. M., Farrington, D. P., & Loesel, F. (2012). School bullying as a predictor of
childyouth.2023.107178 violence later in life: A systematic review and meta-analysis of prospective
Luo, S., Ban, Y., Qiu, T., & Liu, C. (2023). Effects of stress on school bullying behavior longitudinal studies. Aggression and Violent Behavior, 17(5), 405–418. https://doi.
among secondary school students: Moderating effects of gender and grade level. org/10.1016/j.avb.2012.05.002
Frontiers in Psychology, 14, 1074476. https://doi.org/10.3389/fpsyg.2023.1074476 UNESCO. (2019). Behind the numbers: Ending school violence and bullying. https://
Madsen, K. C. (1996). Differing perceptions of bullying and their practical implications. hivhealthclearinghouse.unesco.org/library/documents/behind-numbers-ending-
Educational and Child Psychology, 13(2), 14–22. school-violence-and-bullying.
Maunder, R. E., Harrop, A., & Tattersall, A. J. (2010). Pupil and staff perceptions of Vaillancourt, T., McDougall, P., Hymel, S., Krygsman, A., Miller, J., Stiver, K., & Davis, C.
bullying in secondary schools: Comparing behavioural definitions and their (2008). Bullying: Are researchers and children/youth talking about the same thing?
perceived seriousness. Educational Research, 52(3), 263–282. https://doi.org/ International Journal of Behavioral Development, 32(6), 486–495. https://doi.org/
10.1080/00131881.2010.504062 10.1177/0165025408095553
Menesini, E., Fonzi, A., & Smith, P. K. (2002). Attribution of meanings to terms related to Volk, A. A., Dane, A. V., & Marini, Z. A. (2014). What is bullying? A theoretical
bullying: A comparison between teacher’s and pupil’s perspectives in Italy. European redefinition. Developmental Review, 34(4), 327–343. https://doi.org/10.1016/j.
Journal of Psychology of Education, 17(4), 393. https://doi.org/10.1007/BF03173593 dr.2014.09.001
Menesini, E., Nocentini, A., Palladino, B. E., Frisén, A., Berne, S., Ortega-Ruiz, R., Willits, D. (2015). Situational predictors of violent intentions: Results from a factorial
Calmaestra, J., Scheithauer, H., Schultze-Krumbholz, A., Luik, P., Naruskov, K., survey. The Social Science Journal, 52(2), 176–187. https://doi.org/10.1016/j.
Blaya, C., Berthaud, J., & Smith, P. K. (2012). Cyberbullying Definition Among soscij.2014.10.004
Adolescents: A Comparison Across Six European Countries. Cyberpsychology, Xiao, Y., Jiang, L., Yang, R., Ran, H., Wang, T., He, X., Xu, X., & Lu, J. (2021). Childhood
Behavior, and Social Networking, 15(9), 455–463. https://doi.org/10.1089/ maltreatment with school bullying behaviors in Chinese adolescents: A cross-
cyber.2012.0040 sectional study. Journal of Affective Disorders, 281, 941–948. https://doi.org/
Merton, R. K. (1987). Three Fragments From a Sociologist’s Notebooks: Establishing the 10.1016/j.jad.2020.11.022
Phenomenon, Specified Ignorance, and Strategic Research Materials. Annual Review Yang, M., Guo, H., Chu, M., Leng, C., Qu, C., Tian, K., Jing, Y., Xu, M., Guo, X., Yang, L.,
of Sociology, 13(1), 1–29. https://doi.org/10.1146/annurev.so.13.080187.000245 & Li, X. (2022). Sex Differences in Traditional School Bullying Perpetration and
Mishna, F., Pepler, D., & Wiener, J. (2006). Factors Associated With Perceptions and Victimization among Adolescents: A Chain-Mediating Effect. International Journal of
Responses to Bullying Situations by Children, Parents, Teachers, and Principals. Environmental Research and Public Health, 19(15), Article 15. https://doi.org/
Victims & Offenders, 1(3), 255–288. https://doi.org/10.1080/15564880600626163 10.3390/ijerph19159525
Mishna, F., Scarcello, I., Pepler, D., & Wiener, J. (2005). Teachers’ Understanding of Ybarra, M. L., Espelage, D. L., Valido, A., Hong, J. S., & Prescott, T. L. (2019). Perceptions
Bullying. Canadian Journal of Education/Revue Canadienne de l’éducation, 28(4), of middle school youth about school bullying. Journal of Adolescence, 75, 175–187.
718–738. https://doi.org/10.1016/j.adolescence.2018.10.008
Monks, C. P., & Smith, P. K. (2006). Definitions of bullying: Age differences in Yuchang, J., Junyi, L., Junxiu, A., Jing, W., & Mingcheng, H. (2019). The Differential
understanding of the term, and the role of experience. British Journal of Victimization Associated With Depression and Anxiety in Cross-Cultural Perspective:
Developmental Psychology, 24(4), 801–821. https://doi.org/10.1348/ A Meta-Analysis. TRAUMA VIOLENCE & ABUSE, 20(4), 560–573. https://doi.org/
026151005X82352 10.1177/1524838017726426
Naylor, P., Cowie, H., Cossin, F., de Bettencourt, R., & Lemme, F. (2006). Teachers’ and Zarate-garza, P. P., Biggs, B. K., Croarkin, P., Morath, B., Leffler, J., Cuellar-barboza, A.,
pupils’ definitions of bullying. British Journal of Educational Psychology, 76(3), & Tye, S. J. (2017). How Well Do We Understand the Long-term Health Implications
553–576. https://doi.org/10.1348/000709905X52229 of Childhood Bullying? Harvard Review of Psychiatry, 25(2), 89–95. https://doi.org/
Nielsen, M. B., Tangen, T., Idsoe, T., Matthiesen, S. B., & Magerøy, N. (2015). Post- 10.1097/HRP.0000000000000137
traumatic stress disorder as a consequence of bullying at work and at school. A Zhou, Y., Guo, L., Lu, C., Deng, J., He, Y., Huang, J., Huang, G., Deng, X., & Gao, X.
literature review and meta-analysis. Aggression & Violent Behavior, 21, 17–24. (2015). Bullying as a Risk for Poor Sleep Quality among High School Students in
https://doi.org/10.1016/j.avb.2015.01.001 China. Plos One, 10(3), e0121602.

10