Influences of Teachers, Students and School

Climate on Bullying Victimization:

Evidence from China

Ying Zhu,1,3 Yiying Teng2,3

1. Shanghai Sanda University, Shanghai, China

2. Shanghai Zhongqiao Vocational and Technical University, Shanghai,
China;
3. Chapman University, California, USA

Abstract: Given the common occurrence of school bullying incidents and

high prevalence rates of victimization in China, this study aims to
explore the association between multiple school-related predictors and
Chinese adolescents’ overall experiences with bullying victimization.
Guided by the social-ecological framework for violence prevention, this
study integrated different factors involving teachers, students, and
school climate into one single research to detect the bullying problem
more systematically and holistically. Further, the study applied the data
from the 2018 Program for International Student Assessment while
utilizing binary logistic regression analyses to perceive anti-bullying
variables and their relationships with bullying victimization. The study
revealed that teachers’ emotional, instrumental, and informational
support other than appraisal support were negatively associated with
students’ experiences of being bullied and victimized. The research also
indicated that students’ sense of belonging, a cooperative school
environment, and classroom disciplinary climate played a buffering role
in bullying victimization while competition in school settings adversely
increased overall bullying exposure. The results have highlighted the
integration of various predictors in influencing multiple social relations
and school environments to battle against in-person bullying. The results
also have implications for stopping cyberbullying to achieve inclusive
school education for all in the post-COVID-19 era.
Best Evidence in Chinese Education 2022; 12(1):1547-1571.
Doi: 10.15354/bece.22.or065

How to Cite: Zhu, Y., & Teng, Y. (2022). Influences of teachers, students and school
climate on bullying victimization: Evidence from China. Best Evidence in Chinese
 Zhu & Teng. (Shanghai & California). School Bullying Victimization in China.

Education, 12(1):1547-1571.

Keywords: Bullying Victimization, Classroom Disciplinary Climate, Cooperation and

Competition, Sense of Belonging, Teacher Support

About the Authors: Ying Zhu, Shanghai Sanda University, Shanghai, China; and Chapman University, California,
USA. E-mail: [email protected]
Correspondence to: Yiying Teng, Shanghai Zhongqiao Vocational and Technical University, Shanghai, China; and
Chapman University, California, USA. E-mail: [email protected]
Author Contribution: Zhu and Teng contributed equally to the design, draft, critical revision, and final agreement
of the present article to be published.
Conflict of Interests: None.

© 2022 Insights Publisher. All rights reserved.

Creative Commons Non Commercial CC BY-NC: This article is distributed under the terms of the
Creative Commons Attribution-Non Commercial 4.0 License
(http://www.creativecommons.org/licenses/by-nc/4.0/) which permits non-commercial use, reproduction and distribution of
the work without further permission provided the original work is attributed by the Insights Publisher.

BECE, Vol.12, No.1, 2022 1548

 Zhu & Teng. (Shanghai & California). School Bullying Victimization in China.

T HE significance of bullying prevention and intervention has been well

addressed in a series of initiatives taken by the Chinese government,
including The Guidance on the Prevention and Treatment of Bullying
and School Violence in 2016, Comprehensive Plan on Strengthening the
Management of Bullying among Primary and Middle School Students in
2017, and the newly revised Minor Protection Law that took effect in 2021.
However, school bullying still topped the agenda of this year’s two plenary
sessions of China’s political bodies (2022 two sessions) and incidents of bul-
lying victimization are often reported in media outlets. In one case, a teenag-
er in Guangdong Province drowned after being slapped and knocked by
peers in a restroom (Zuo, 2021); in another case, an adolescent from Anhui
Province suffered from being bullied by a group of 12 schoolmates (China
National Radio [CNR], 2022). These single incidents are just a tip of the ice-
berg and in fact, more instances of school bullying have surfaced with vic-
timization constituting a major source of injuries (Li et al., 2020). Statistics
from a national survey led by Central China Normal University has revealed
that the proportion of students who had bullying experiences in 2019 and
2020 amounted to 32.4 percent (Zhao, 2021). Worse still, when classes were
moved online due to the outbreak of the COVID-19 pandemic, cyberbullying
(a form of bullying using the Internet) still persisted in China (Han et al.,
2021). That is why legislators and deputies at the 2022 two sessions called
on schools and communities to speed up the process of building a bullying-
free, equitable, and inclusive campus for all. To achieve this goal, it is im-
perative for Chinese researchers and educators to identify intervention pre-
dictors pertaining to school violence and bullying victimization and work out
detailed plans to reduce the incidence of bullying behaviors.
As for the factors predicting bullying victimization, teachers, students
and school climate are generally regarded as three major components for as-
sessing an engaging or a horrifying learning experience (e.g., Ahnert et al.,
2012; Maxwell et al., 2017). Among them, teacher support (Flaspohler et al.,
2009; Huang & Zhao, 2019), a fundamental need to belong (Underwood &
Ehrenreich, 2014), a socially cooperative environment (Van Ryzin & Roseth,
2018a), and benign competition in one’s immediate settings (Volk et al.,
2015) have been respectively evidenced in previous bullying literature. De-
rived from classroom settings, management style or disciplinary climate is
more associated with students’ academic achievement (Sortkær & Reimer,
2018), but seldom mentioned to be directly linked to school bullying. Further,
researchers’ renewed attention to the fundamental relationships involving
teachers, students, classroom settings, and school cultures could help to ex-
plain cyberbullying patterns during the COVID-19 pandemic (Vaillancourt

BECE, Vol.12, No. 1, 2022 1549

 Zhu & Teng. (Shanghai & California). School Bullying Victimization in China.

et al., 2021). But in the Chinese context, little has been known about the
multiple school-related factors that jointly work on anti-bullying endeavors.
Therefore, this empirical study aims to explore the association between the
different facets of school learning and Chinese students’ overall experiences
with bullying victimization while controlling for such demographics as gen-
der, age, student backgrounds, and grade repetition. Three overarching re-
search questions have guided this exploration.
(i) How much variance in overall bullying experience is explained by
school-related factors and demographics?
(ii) Do school-related predictors have associations with high or low fre-
quencies of bullying experiences?
(iii)What intervention factors predict the likelihood of Chinese students’ high
or low exposure to bullying victimization?

Literature Review
Theoretical Framework
Guided by the social-ecological framework for violence prevention (CDC,
2004), this study addresses the complex interplay between individual, rela-
tionship, community, and societal factors in relation to bullying and bullying
victimization. Originating from Bronfenbrenner’s (1979) social-ecological
model (SEM), this updated framework offers an ecological angle to “investi-
gate the combined impact of social contexts and influences on behavioral
development” (Swearer et al., 2010, p. 42).
As Figure 1 shows, a range of factors work to alert people to the
possible risks for experiencing or perpetrating violence while the overlap-
ping circles indicate how one level of factors influences another and how
different levels of factors coexist to function on the entire model. At the in-
dividual level, prevention strategies may involve personal attitudes, beliefs,
or behaviors that prevent or stop violence. At relationship and community
levels, a person’s close connection to family members and active interaction
with schools/workplaces/ neighborhoods may provide a pathway for risk re-
duction and violence prevention, including teacher-student interaction, posi-
tive peer relations, and reliable school/work environments. The final level
(societal factors) features social and cultural norms that encourage or hamper
violence. The following sections briefly introduce the variables that fall
within this framework, which will form the basis of the bullying model to be
tested in our empirical study and help to understand anti-bullying efforts
over time and beyond human-level impact.

BECE, Vol.12, No. 1, 2022 1550

 Zhu & Teng. (Shanghai & California). School Bullying Victimization in China.

Figure 1: Social-Ecological Model - A Framework for Prevention.

Note: Adapted from Centers for Disease Control and Prevention (p.5), 2004, CDC.

Teacher Support
Students’ healthy development cannot be sustained without the involvement
of teachers, who provide support for, build trust in, and form close interper-
sonal relationships with students (e.g., Hughes et al., 2012; Quin et al., 2018;
Ryan & Patrick, 2001; Wentzel et al., 2010). Moreover, such teacher-student
relationships seem more collaborative rather than hierarchical (Schleicher,
2018). Tardy (1985) pointed out that teacher support as a form of social sup-
port was composed of four categories, namely emotional, instrumental, ap-
praisal, and informational support. Emotional support is the type of teacher
support that reflects love, concern, empathy, confidence, and patience; in-
strumental support refers to the substantial help passed on to the students in
real need; appraisal support could be interpreted as the constructive feedback
and appropriate evaluation system offered for students who will then know
where and how to improve their learning behaviors; informational support is
viewed as instruction, guidance, assistance, and counselling provided to help
with students’ problem-solving (Wentzel, 2016).

BECE, Vol.12, No. 1, 2022 1551

 Zhu & Teng. (Shanghai & California). School Bullying Victimization in China.

Previous literature has revealed that teacher support positively corre-

lates with students’ emotional and cognitive development (Quin, 2016; Quin
et al., 2018) as well as with their interest in classroom activities (Lapointe et
al., 2005) and their academic enjoyment (Ahmed et al., 2010; Ma et al.,
2021). Despite much attention devoted to the connection between teacher
support and academic performance, scholars have begun to realize the im-
portance of anti-bullying efforts in the presence of teachers. Furlong and
Chung (1995) early indicated in their empirical research that those non-
victims are almost twice as likely as bullying victims to report the presence
of a teacher to whom they can express their problems and pour out their feel-
ings; other studies (e.g., Baik et al., 2019; Colarossi & Eccles, 2003; Mazzer
& Rickwood, 2015) then emphasized the close relationship between teacher
influences and students’ well-being, which could further determine students’
sense of security, problem behavior and victimization experience (Berkowitz
& Benbenishty, 2012; McNeely & Falci, 2004). Besides, through analyzing
the index of informational support, Huang and Zhao (2019) implied that
teacher support could slightly mitigate bullying exposure, but they did not
discuss the remaining three categories. Hence in this study, all four types of
teacher support will be examined in relation to bullying victimization in
China.

Students’ Sense of Belonging

In educational settings, belonging is identified as the feeling of security, ac-
ceptance and inclusion a student connects to peers, teachers, and the school
(Goodenow, 1993). Research shows that students’ sense of belonging is
closely linked to academic achievement (Wang & Eccles, 2012; Wang &
Holcombe, 2010) and psychological adjustment (Lester et al., 2013; Loukas
et al., 2016), and has sparked renewed concern for educators as a result of a
high prevalence of bullying exposure (Duggins et al., 2016; Ma, 2003). Pre-
vious studies have suggested that the stronger a sense of belonging students
have, the lower the level of bullying perpetration and victimization occurs
(Goldweber et al., 2013; Raskauskas et al., 2010). Accordingly, we hypothe-
size that students’ sense of belonging might play a buffering role in bullying
victimization while influencing multiple relationships and different commu-
nities.

Classroom Disciplinary Climate

A major component of a positive classroom environment, disciplinary cli-
mate is provided when students listen to their teachers and peers attentively
and focus only on their academic tasks without the interference of noise or
BECE, Vol.12, No. 1, 2022 1552
 Zhu & Teng. (Shanghai & California). School Bullying Victimization in China.

disorder (Moos, 1979). Three key elements characterize the variable: stu-
dents’ reverence for classroom rules as always, teachers’ utmost responsibil-
ity for ensuring a favorable learning environment, and peers’ disruptive-free
behavior in the classroom (Cheema & Kitsantas, 2014; Matsumura et al.,
2008). A positive disciplinary classroom secures more teaching and learning
opportunities (Mostafa et al., 2018), better academic performance (Blank &
Shavit, 2016), students’ psychological wellbeing (Kim et al., 2021; Wang et
al., 2020), and the efforts against bullying involvement (e.g., victimization
and witnessing) (Kim et al., 2021). Additionally, an orderly, nondisruptive
classroom can also facilitate students’ sense of belonging (Organization for
Economic Cooperation and Development [OECD], 2017) and benefit ethnic
minorities and students in a disadvantaged position (Cheema & Kitsantas,
2014). In this research, we hypothesize that classroom disciplinary climate
might predict bullying victimization while influencing various social rela-
tions.

Cooperation and Competition

Skills of cooperation is nurtured in an educational setting where students
learn together (usually in small groups) to fulfill well-structured learning
tasks and accomplish a well-defined learning goal (Van Ryzin & Roseth,
2019; Zook, 2018). If properly created, a cooperative learning environment
brings about a high level of peer relatedness or peer relations (Roseth et al.,
2008; Van Ryzin & Roseth, 2018a), which in turn promotes the attainment
of learning goals (Johnson et al., 2014) as well as encourages positive feel-
ings and supportive interactions. Ultimately, benign spiral emerges with the
repetition of collaborative learning atmosphere and positive social interac-
tions within the group (Deutsch, 1949, 1962). After class, students still have
the opportunity to enhance their favorable peer relations by relating each in-
dividual’s contribution to the performance of the entire group. This type of
cooperation with an emphasis on social contact and skills benefits anti-
bullying behaviors and helps to reduce bullying victimization (Van Ryzin &
Roseth, 2019).
The opposite side of school learning is perceived as having a climate
of competition, in which students compare one another or work against each
other to obtain their academic or social benefits (e.g., excellent grades and
admission to prestigious universities) (Johnson & Johnson, 1978; Volk et al.,
2015). Views differ as to whether competition should be fostered in the
learning process. One perspective considers competition as being supported
by in-group distinctions, social hierarchy, and socioeconomic backgrounds
and hence linked to increased bullying (Volk et al., 2015); another opinion
BECE, Vol.12, No. 1, 2022 1553
 Zhu & Teng. (Shanghai & California). School Bullying Victimization in China.

identifies competition as a catalyst for student engagement and students’

competitive instincts (Anderson, 2006); a third perception regards competi-
tion as potentially feasible depending on different disciplines and education-
al levels (Chen & Chen, 2014; Vandercruysse et al., 2013). Whatever views
prevail and whatever environmental attribute is outlined, students need to
learn to create and bolster positive social relations while taking advantage of
every factor in the learning community (e.g., teacher and peers) to achieve
what is to be achieved (Johnson & Johnson, 1978; Gutiérrez-Braojos et al.
2019). In this study, we hypothesize that both cooperation and competition
might work together to influence school violence and bullying victimization
while affecting different relationships and school environments.

Study Design
Previous bullying-related literature has identified that in both Chinese and
non-Chinese contexts, gender, grade repetition (or the practice of having stu-
dents remain in the same grade without promotion), and economic, social
and cultural status (ESCS) all closely correlate with bullying exposure and
bullying victimization. Therefore, this study regards the above literature-
supported predictors as student-level control variables (as they might con-
found the bullying model) and centers the underlying mechanism of the as-
sociation between the fore mentioned school-related predictors and bullying
victimization for Chinese adolescents. In line with the theoretical model and
out of the urge to identify anti-bullying factors in the Chinese context, we
proposed the following three hypotheses for the present study (see Figure 2).
H1: Teacher support (at the relationship level) significantly influences
bullying victimization at school.
H1-1: Emotional support significantly influences bullying victimiza-
tion.
H1-2: Instrumental support significantly influences bullying victimiza-
tion.
H1-3: Appraisal support significantly influences bullying victimization.
H1-4: Informational support significantly influences bullying victimi-
zation.
H2: Students’ sense of belonging (at the community level) strongly
predicts bullying victimization.
H3: School climate (at the community level) greatly affects bullying
victimization while influencing various social relations.
H3-1: Classroom disciplinary climate strongly predicts bullying vic-
timization.

BECE, Vol.12, No. 1, 2022 1554

 Zhu & Teng. (Shanghai & California). School Bullying Victimization in China.

Figure 2: School-Related Predictors Hypothesized for Bulling Victimization.

H3-2: Cooperation greatly influences bullying victimization.

H3-3: Competition greatly influences bullying victimization.

Methods
Sample
This study analyzed the publicly available dataset from the 2018 Program for
International Student Assessment (PISA 2018) conducted by OECD in Chi-
na’s Beijing, Shanghai, Jiangsu, and Zhejiang provinces (B-S-J-Z, China).
PISA, a triennial survey featuring 15-year-olds, measures to what extent they
have obtained the critical knowledge and skills essential for active involve-
ment in society. PISA 2018 adopted a stratified two-stage sampling proce-
dure to select sample students. In the initial phase, schools were chosen
through systematic probability proportional to size (PPS) sampling (OECD,
2019c). In the subsequent phase, 42 students were selected with equal prob-
ability within each sampled school. For schools with a required number of
students fewer than 42, all 15- year-olds were singled out (OECD, 2019c).
Ultimately, 12,058 adolescents were selected from 361 schools, among
which 4306 (35.7%) came from lower-secondary schools, 5594 (46.4%)
from upper secondary schools, and 2158 (17.9%) from vocational schools.
As stipulated by PISA’s international protocol (OECD, 2019b), PISA 2018

BECE, Vol.12, No. 1, 2022 1555

 Zhu & Teng. (Shanghai & California). School Bullying Victimization in China.

China data met the minimum requirements of the weighted response rate for
selected schools within participating countries/regions (85%) and that for
students from those selected schools (80%). Before data collection, informed
consent was acquired from principals, teachers, parents, and students. The
secondary data analyses of PISA 2018 would be crucial for this study as the
large-scale survey through a careful sampling process reflected the real bul-
lying situation in Chinese schools.

Dependent Variable
The dependent variable in PISA 2018 was measured from the perspective of
the bullying victims (OECD, 2019a) and assessed by asking students to re-
port how often they had experienced physical, verbal and social forms of
school bullying during the past 12 months (see Table 1). Participants re-
sponded on a four-point Likert-type scale ranging from 1 = never or almost
never to 4 = once a week or more. The corresponding items were then added
up for the bullying scale with the internal consistency or reliability tested
(Cronbach’s α = 0.843).

Demographic Variables
Included in the demographic variables were such student background predic-
tors as age, gender, grade repetition, and the ESCS index (see Table 2). Age,
a continuous variable, spanned from 15.33 to 15.77 years old (mean = 15.77,
SD = 0.29). Gender and grade repetition were all used as dummy variables
with “0” for boys/no grade repetition and “1” for girls/grade repetition. The
ESCS index, a weighted score rooted in PISA 2018, contained three family
background variables, namely home possessions, parents’ highest education-
al levels, and their highest occupational status (OECD, 2019c).

Independent Variables
All the independent variables (teacher support, students’ sense of belonging,
cooperation and competition, classroom disciplinary climate) in the study
were directly derived from the PISA 2018 database (see Table 3). To ensure
validity and reliability of the independent scales applied to this research,
scaling procedures and construct validation of context questionnaire data
were well documented in the dataset with the internal consistency
(Cronbach’s alpha) all exceeding 0.800 (Cronbach’s α > 0.800) (OECD,
2019c).
Parallel with Tardy’s (1985) and Wentzel’s (2016) categorization of teacher
support, this study applied four indices from PISA 2018: TEACHINT (rep-

BECE, Vol.12, No. 1, 2022 1556

 Zhu & Teng. (Shanghai & California). School Bullying Victimization in China.

Table 1: Dependent Variable: Bullying Scale.

Bullying Scale: 1. Other students left me out of things on purpose.
1 (never or almost never)
to 2. Other students made fun of me.
4 (once a week or more) 3. I was threatened by other students.
4. Other students took away or destroyed things that belonged to me.
5. I got hit or pushed around by other students.
6. Other students spread nasty rumors about me.

Table 2: Descriptive Statistics for Demographic Variables.

Minimum Maximum Mean SD N
Continuous demographic variable(s)
Age (years) 15.33 16.25 15.77 0.29 12,058
Variables Frequency %
Female 5,775 47.9
Categorical demographic variables Gender
Male 6,283 52.1
No 11,237 93.2
Grade Repetition
Yes 753 6.8

Table 3: Independent Variables and Scale Reliabilities.

Emotional Support
It reflects teachers’ love, concern, enthusiasm, and encouragement
(TEACHINT)
for students.
Cronbach’s α = 0.893
Instrumental Support It represents pedagogically tangible help (e.g., clear goals for learn-
(DIRINS) ing, instructions, a short summary of the previous lesson, and check-
Cronbach’s α = 0.805 ing questions on whether teaching has been understood) in student
Teacher
learning.
Support
Appraisal Support
It refers to the feedback on students’ strengths in a subject, the areas
(PERFEED)
that should be improved, and how the areas can be improved.
Cronbach’s α = 0.892
Informational Support
It indicates guidance, assistance, and counseling provided to help
(TEACHSUP)
students understand the subject.
Cronbach’s α = 0.851
Sense of Belonging It refers to students’ perceptions of community connectedness and
Cronbach’s α = 0.822 school inclusion.
Disciplinary Climate
in the Test
It measures order and discipline to be maintained in classroom set-
Language Classroom
tings.
(DISCLIMA)
Cronbach’s α = 0.892
Perception of
School
Cooperation It shows the environment where cooperation is addressed by stu-
Climate
(PERCOOP) dents in target schools.
Cronbach’s α = 0.924
Perception of
Competition It refers to the environment where competitiveness is reported by
(PERCOMP) students in target schools.
Cronbach’s α = 0.829

BECE, Vol.12, No. 1, 2022 1557

 Zhu & Teng. (Shanghai & California). School Bullying Victimization in China.

resenting emotional support), DIRINS (indicating instrumental support),

PERFEED (reflecting appraisal support), and TEACHSUP (pointing to in-
formational support). As for students’ sense of belonging, the index naturally
derived from the PISA data examined how easy it was for students to make
friends at school, how they were welcomed or accepted by peers, and how
they felt isolated or asocial in their community (Goodenow, 1993).
For school climate predictors in PISA 2018, disciplinary climate in
the test language classroom (DISCLIMA), perception of cooperation
(PERCOOP), and perception of competition (PERCOMP) were singled out
for study. The disciplinary climate index described how orderly or disruptive
the entire classroom environment appeared with higher values pertaining to
positive climate appropriate for student learning and classroom communica-
tion. Meanwhile, the cooperation index asked respondents to assess if they
value cooperation, collaborate with each other, and are encouraged to coop-
erate at school; the competition index required participants to report if they
cherish competition and compete with each other in the school environment.

Analytic Procedure
Because the dependent variable (the bullying scale) was not normally dis-
tributed, we split it into a dummy one (high and low frequencies of being
bullied or victimized) in response to the research questions. Low frequency
means that participants never or almost never experienced the listed forms of
bullying events while high frequency indicates that respondents experienced
at least one of the listed bullying events a few times a year or more (see Ta-
ble 4). Further, inspired by previous logistic regression models used to un-
derstand Chinese students’ perceptions of and experiences with an educa-
tional phenomenon (Allen, 2019), we conducted the research developing a
stepwise binary logit model (Pallant, 2016). Utilizing SPSS software Version
28.0, the whole analytical process of data began with the entrance of demo-
graphic variables to create a predictive model (Model 1), which was fol-
lowed by the addition of other variables of interest, namely the teacher-
support model (Model 2), the sense-of-belonging model (Model 3), and the
school-climate model (Model 4). The final model (Model 5) ended up with
all the above predictors.
However, preliminary assumption tests revealed that for Models 3
1
and 4 (see Appendix I), the Hosmer and Lemeshow Goodness of Fit Tests
were not passed (significance values all below 0.05), suggesting no support
for these three models (Hosmer & Lemeshow, 2000). Hence, this study only
concentrated on the other three models (i.e., Model 1: demographic model;
Model 2: teacher-support model; Model 5: full-predictor model).
BECE, Vol.12, No. 1, 2022 1558
 Zhu & Teng. (Shanghai & California). School Bullying Victimization in China.

Table 4: Descriptive Statistics for Categorical Dependent Variable(s).

Variables Frequency %
Bullying Low 6,463 54.5
Scale High 5,390 45.5

Table 5: Logistic Regression Analysis of Demographic, Teacher-Support and

Full-Predictor Models.
Model 1a Model 2b Model 5c
Variables B S.E. Exp(B) B S.E. Exp(B) B S.E. Exp(B)
Gender 0.506‡ 0.037 1.658 0.524‡ 0.039 1.690 0.495‡ 0.040 1.640
Age 0.004 0.064 1.004 -0.104 0.066 0.901 -0.235‡ 0.069 0.791
Grade Repetition 0.245† 0.078 1.278 0.264‡ 0.080 1.302 0.211* 0.084 1.235
ESCS -0.072‡ 0.017 0.930 -0.017 0.018 0.983 0.030 0.019 1.031

TEACHINT -0.264‡ 0.025 0.768 -0.098‡ 0.027 0.907

DIRINS -0.110‡ 0.025 0.896 -0.064* 0.026 0.938
PERFEED -0.022 0.024 0.978 0.30 0.025 1.031
TEACHSUP -0.200‡ 0.028 0.819 -0.142‡ 0.030 0.867
BELONG -0.454‡ 0.027 0.635
DISCLIMA -0.311‡ 0.022 0.733
PERCOOP -0.168‡ 0.023 0.846
PERCOMP 0.248‡ 0.022 1.281
Notes. *<0.05, †p < 0.01, ‡p < 0.001
a
Demographic model, b Teacher-support model, c Full-predictor model

Results
Through binary logistic regression analyses, this study explored the likeli-
hood that Chinese adolescents were exposed to high or low frequencies of
being bullied or victimized. For Model 1 using age, gender, grade repetition,
and ESCS, the likelihood-ratio test produced a chi-square value of 16,101.62
and was statistically significant (p < 0.001), indicating the logistic model
was better than the null model without predictors. The chi-square value for
the Hosmer-Lemeshow test was 11.07 with a significance level of 0.198 (p >
0.05), suggesting the data fit the above model well (Hosmer & Lemeshow,
2000). The statistical significance (χ2(4, N = 11,847) = 224.125, p < 0.001)
of the demographic model implied that it was able to distinguish between
respondents’ high or low exposure to bullying experiences. Moreover, the
BECE, Vol.12, No. 1, 2022 1559
 Zhu & Teng. (Shanghai & California). School Bullying Victimization in China.

model as a whole explained 1.9% (Cox & Snell R square) and 2.5%
(Nagelkerke R square) of the variance in the two bullying groups, and cor-
rectly classified 56% of the cases with a very small improvement of 1.5% in
predictions. As presented in Table 5, three of the four demographic variables
made a unique statistically significant contribution to the model (except for
age, p > 0.05). This meant that male students were more likely to be found in
the high-level bullying groups than girls when controlling for other factors in
the model. The strongest predictor of reporting the high-level bullying
groups was gender, recording an odds ratio of 1.658. Likewise, the odds ratio
of being in the high-level bullying groups was positively associated with
grade repetition (p < 0.01). For ESCS, its negative correlation with bullying
exposure and bullying victimization was identified (p < 0.001).
In Model 2 with all teacher-support predictors added to the de-
mographics, the likelihood-ratio test and the Hosmer-Lemeshow test again
identified the model as being worthwhile with the former yielding a χ2(8) of
15,498.97 (N = 11,081, p < 0.001) and the latter obtaining the chi-square
value of χ2(8) of 9.67 (p > 0.05). Moreover, able to distinguish between par-
ticipants’ high or low exposure to bullying behaviors (χ2(8, N = 11818) =
786.503, p < 0.001), the teacher-support model explained 6.4% (Cox & Snell
R square) and 8.6% (Nagelkerke R square) of the variance in bullying
groups, and correctly classified 61.2% of the cases with a modest improve-
ment of 6.7% in predictions. As observed in Table 5, five of the eight pre-
dictors made a unique statistically significant contribution to the model
(gender, grade repetition, TEACHSUP, DIRINS, TEACHINT). The two
demographic variables (gender and grade repetition) of reporting the high-
level bullying groups echoed those in Model 1, recording a slightly higher
odds ratio of 1.690 and 1.302 respectively when compared with those in
Model 1. Additionally, adolescents with more teacher support (TEACHINT,
DIRINS, and TEACHSUP) were 0.768, 0.896, and 0.819 times less likely to
be reported in the high-level bullying groups (p < 0.001). This pointed to the
fact that emotional, instrumental, and informational support all played a
buffering role in bullying victimization with emotional support being most
conducive to anti-bullying efforts. Thus, Hypothesis 1 was partially support-
ed except for the role of appraisal support.
In the final model (Model 5) with all independent variables added,
the likelihood-ratio test achieved a χ2(12) of 14,525.39 (N = 11,723, p <
0.001), suggesting that the model was better than the null model with all
predictors removed. The Hosmer-Lemeshow test yielded a χ2(8) of 9.68 (p >
0.05), indicating that the data fit the model well. The whole model again sig-
nified its ability to differentiate between respondents’ high or low frequen-

BECE, Vol.12, No. 1, 2022 1560

 Zhu & Teng. (Shanghai & California). School Bullying Victimization in China.

cies of bullying exposure (χ2(8, N = 11,723) = 1,627.086, p < 0.001). More-

over, the model as a whole explained 13% (Cox & Snell R square) and 17.3%
(Nagelkerke R square) of the variance in bullying groups, and correctly clas-
sified 65.8% of the cases with a remarkable improvement of 11.2% in pre-
dictions. Ten of the 12 predictors made a unique statistically significant con-
tribution to the model (gender, age, grade repetition, TEACHSUP, DIRINS,
TEACHINT, DISCLIMA, PERCOMP, PERCOOP, and BELONG) except
for ESCS and PERFEED (p > 0.05). The two demographic variables (gender
and grade repetition) of reporting the high-level bullying groups recorded a
slightly lower odds ratio of 1.640 and 1.235 compared with those in Models
1 and 2. Statistically insignificant in the previous two models, age (p < 0.001)
this time witnessed a 0.791 likelihood increase in the high-level bullying
groups. For the three forms of teacher support (TEACHINT, DIRINS, and
TEACHSUP), they displayed similar patterns as were revealed in Model 2,
reporting a mitigating effect on the high level of bullying exposure with in-
strumental support (DIRINS) replacing emotional support (TEACHINT) in
Model 2 as the most effective form of teacher support for anti-bullying en-
deavors. Here in the full-predictor model, the first hypothesis was again ba-
sically supported. Regarding the newly added factors in the model, students
with a strong sense of belonging, in a better disciplinary climate, and in a
cooperative environment were respectively 0.733, 0.846, and 0.635 times
less likely to be identified in the high-level bullying groups (p < 0.001). On
the contrary, students in the school settings that fostered competition among
peers were 1.281 times more likely to report a high level of bullying victimi-
zation (p < 0.001). Thus, Hypothesis 2 was fully supported while Hypothesis
3 only was partially supported.

Discussion
Applying PISA 2018 China data and utilizing binary logit regression anal-
yses, this study has added to the empirical experience by investigating the
relationship between multiple intervention factors and bullying victimization
involving teachers, students, and school climate in the Chinese context.
From the perspective of student characteristics, male adolescents, grade re-
peaters (regardless of their gender), younger students, and low-income stu-
dents/students from disadvantaged families are more likely to be bullied and
victimized, as has been reported in a variety of Chinese literature (e.g., Ba et
al., 2019; Huang & Zhao, 2019; Li et al., 2020; Lian et al., 2021; Yu & Zhao,
2021). In this sense, top priority needs to be given to the above types of stu-
dents when intervention measures are introduced locally or nationwide.

BECE, Vol.12, No. 1, 2022 1561

 Zhu & Teng. (Shanghai & California). School Bullying Victimization in China.

Second, the research findings have verified that teacher support emotionally,
informationally, and instrumentally moderates school bullying and bullying
victimization. This could link teachers to their practical assistance and sub-
stantial help students can seek in case of bullying occurrence. Such support
at the relationship level could extend beyond daily instructions provided for
adolescents and penetrate further into almost every aspect of classroom
learning, eventually benefiting multiple levels of influencing factors within
the social-ecological framework. For example, student-perceived love, con-
cern, enthusiasm, encouragement, and guidance are beneficial not only to
students themselves, but to teacher-student communication channels, peers’
recognition, positive school climate, and a set of core social values to be pre-
served. To minimize bullying exposure for all students, teachers are encour-
aged to treat students more fairly, evaluate their behavioral and mental pro-
gress more proactively, and offer timely support to those in academic diffi-
culties and with life or socializing problems (Bosworth et al., 2018). Howev-
er, appraisal support, identified as a teacher’s feedback on students’ strengths
or weaknesses, is more directly associated with their behavioral engagement
(Carvalho et al., 2020) and academic performance (e.g., Carvalho et al. 2014;
Fyfe & Rittle-Johnson, 2016; Wei & Xie, 2017), thus requiring more empiri-
cal evidence to identify its possible correlation with bullying victimization.
Third, consistent with previous Chinese studies (Huang, 2020; Li et
al., 2020), this research has illustrated that students’ sense of belonging acts
as an important buffer against the detriment of bullying victimization. This
indicates that students’ identification with and connectedness to a peer group
or a school environment at the community level are viewed as major social
resources to protect student learning against being hampered by school vio-
lence/bullying victimization (Adams & Hannum, 2018).
For school climate, the findings have demonstrated that a disciplinary,
cooperative environment supported anti-bullying endeavors. Specifically,
this study augments prior research on cooperative learning by showing that a
cooperative environment reduces bullying victimization while promoting
peer relatedness across the entire student population (Ryzin & Roseth, 2019).
However, a socially competitive climate in this study tends to catalyze bully-
ing victimization, urging schools and educators to reconsider the balanced
relationship between cooperation and competition in bringing about student
engagement and facilitating anti-bullying behaviors in school settings
(OECD, 2019a). Besides, a positive school climate cannot be maintained
without the participation of teachers, who help students cultivate such class-
room habits as being friendly to and respectful for peers as well as becoming
social, feeling inclusive, and fostering values that will form certain school

BECE, Vol.12, No. 1, 2022 1562

 Zhu & Teng. (Shanghai & California). School Bullying Victimization in China.

norms. The connectedness within the entire school system will in turn bene-
fit teachers, students, and schools, and eventually bullying prevention and
control efforts.
Therefore, a single bullying incident is interpreted not as an isolated
phenomenon that probably causes victimization, but as the social behavior
dynamically incorporated into an intricate network of interpersonal relation-
ships, institutional factors, community contexts, and policy components. As
is reflected in the socio-ecological framework, its interactive and reciprocal
nature implies that individual behavior influences and is influenced by mul-
tiple levels of factors while shaping and being shaped by the socio-cultural
environment (Salihu et al., 2015; Townsend & Foster, 2011). In this sense, to
better understand students’ bullying behaviors is not to split them from the
larger social context—schools in which students acquire knowledge, com-
municate with teachers, get along with peers, develop egos, and gradually
form the way a school culture is fostered or ruined. This may point to the
necessity of creating improved school conditions for increasing anti-bullying
awareness on campus and encouraging sustained behavioral change on bul-
lies. The close contact between schools and individuals also promotes the
building of a nonviolent campus and effectively supports the implementation
of anti-bullying programs/policies.

Limitations
Despite our focus on school-based factors in bullying victimization and the
strength in integrating multiple predictors to address the issue holistically,
there exist limitations in this study. First, at the individual and relationship
levels, only student-perceived teacher support was taken into consideration,
but in fact in the entire school community, the involvement of school admin-
istrators, principals, and other school leaders would never be ignored (Minis-
try of Education [MOE], 2021). Their active participation in building an ap-
propriate school climate and use of process-oriented strategies (Bosworth et
al., 2018) could exert a far-reaching impact on anti-bullying endeavors and
reduction of victimization. Besides, this study only featured 15-year-olds in
general in the test language classroom, but seldom explored the subcatego-
ries of this student body including those marginalized or at-risk groups who
are prone to be bullied or victimized (e.g., students with disabilities and un-
der-achieving students) (MOE, 2017, 2021). Hence, future studies can be
realized in showing concern for special student groups as well as considering
professionals (e.g., school counselors) and school leadership in the battle
against bullying.

BECE, Vol.12, No. 1, 2022 1563

 Zhu & Teng. (Shanghai & California). School Bullying Victimization in China.

Implications
The present study has provided implications for minimizing the impacts of
bullying victimization and the development of different intervention
measures from the perspective of school settings, especially in the context of
Chinese schools. Theoretically, the research allows for the room to perceive
school bullying and school violence as something complicated yet intercon-
nected that springs from a combination of multiple influences upon individu-
al behaviors. Thus, the interaction of various levels of factors for violence
prevention could offer an effective mechanism for the understanding of how
school bullying occurs and what to do with bullying before it begins. Empir-
ically, the findings adds to the Chinese-based evidence of the association be-
tween school-based predictors and anti-bullying awareness. Since the enact-
ment of the Comprehensive Plan on Strengthening the Management of Bul-
lying among Primary and Middle School Students in 2017, great importance
has been attached to anti-bullying education for all and bullying prevention
before it develops (MOE, 2017). Based on the experience of anti-bullying
legislation in other countries, China adheres to the principle of prevention as
a primary concern and intervention as a top priority in an effort to push for-
ward the establishment of an anti-bullying campus. Under such circumstanc-
es, this research regards a harmonious classroom climate and a positive
school environment as the starting point to develop a sense of security, form
social values (e.g., justice, unity, collaboration, and responsibility) amid cul-
tivation of self-awareness, and eventually curb bullying behaviors on campus.
At the micro level, teachers play an indispensable part in teaching
students what to learn and how to shape values as they spend a large propor-
tion of time listening to adolescents’ voices and building rapport (Schleicher,
2018). In this process, teachers’ role in assisting the young to identify bully-
ing behaviors and timely stop victimization cannot be taken no notice of.
Noteworthily, imperative to bullying intervention is teacher training, which
means teachers’ skilled coping strategies relieve students’ anxiety in con-
fronting bullying and help to form appropriate management styles of class-
room learning (Di Stasio et al., 2016). As suggested in the study, teacher
training is feasible in promoting a sense of school belonging, improving
classroom disciplinary climate, and balancing the skills of cooperation and
competition. For school climate, our study points to all stakeholders who
join the entire school community in creating a caring, inclusive atmosphere,
not only teachers and student victims in the forefront of this anti-bullying
battle. Also in climate building, our findings have recognized social compar-
ison and intense competition as crucial characteristics that deserve educa-

BECE, Vol.12, No. 1, 2022 1564

 Zhu & Teng. (Shanghai & California). School Bullying Victimization in China.

tional practitioners’ scrutiny and deliberation over their own teaching prac-
tices (Di Stasio et al., 2016).
However, the lingering impacts of the COVID-19 pandemic have
made bullying patterns somewhat different. While strengthening online
learning, the pandemic still secures cyberbullying as mentioned above. But
the fact lies in that with the decrease in in-person bullying, cyberbullying
follows a downward trend (Bacher-Hicks et al., 2021). This echoes previous
evidence that cyberbullying rarely occurs independently of in-person bully-
ing (Waasdorp & Bradshaw, 2015). In order to work out strategies for cop-
ing with cyberbullying, close attention would be paid to such fundamental
relationships in educational reforms and anti-bullying attempts as teacher-
student interactions, peer-to-peer relations, classroom environments and
school cultures (Vaillancourt et al., 2021). Therefore, this study may provide
a route to the understanding of cyberbullying in China that in a sense repli-
cates the patterns of in-person bullying and bullying victimization. Although
it is still pending to see how a positive classroom climate plays a role in
adapting the bullied to the school life, the emphasis upon multiple relations
within the school system offers an angle to propose intervention measures
aimed at achieving equity and inclusivity for all schoolchildren in the post-
COVID-19 era.

Note:
1. For the failed models (Model 3: the sense-of-belonging model and Model 4: the
school-climate model), please refer to the online appendix for details.

Acknowledgement
We wish to express our sincere gratitude to Dr. Ryan Allen and Dr. Doug Havard, who
have offered professional assistance and constructive suggestions during our research,
upon the completion of this academic paper, and in the process of revision. Their will-
ingness to spare the time and energy so generously has been very much appreciated.
Our grateful thanks are also extended to our colleagues for their firm support and val-
uable insight in the research. Finally, we would like to thank OECD for making the
PISA database publicly available for researchers and their analyses.

BECE, Vol.12, No. 1, 2022 1565

 Zhu & Teng. (Shanghai & California). School Bullying Victimization in China.

References
Adams, J., & Hannum, E. (2018). School vio- 29590). Available at:
lence in China: a multilevel analysis of stu- https://www.nber.org/papers/w29590
dent victimization in rural middle schools. Baik, C., Larcombe, W., & Brooker, A. (2019).
Research in the Sociology of Education, How universities can enhance student mental
20:41-69. DOI: wellbeing: The student perspective. Higher
https://doi.org/10.1108/S1479- Education Research & Development,
353920180000020003 38(4):674-687. DOI:
Ahmed, W., Minnaert, A., van der Werf, G., & https://doi.org/10.1080/07294360.2019.1576
Kuyper, H. (2010). Perceived social support 596
and early adolescents’ achievement: The Berkowitz, R., & Benbenishty, R. (2012). Per-
mediational roles of motivational beliefs and ceptions of teachers’ support, safety, and
emotions. Journal of Youth and Adolescence, absence from school because of fear among
39:36-46. DOI: victims, bullies, and bully‐victims. Ameri-
https://doi.org/10.1007/s10964-008-9367-7 can Journal of Orthopsychiatry, 82(1):67-74.
Ahnert, L., Harwardt-Heinecke, E., Kappler, G., DOI: https://doi.org/10.1111/j.1939-
Eckstein-Madry, T., & Milatz, A. (2012). 0025.2011.01132.x
Student–teacher relationships and classroom Blank, C., & Shavit, Y. (2016). The association
climate in first grade: How do they relate to between student reports of classmates’ dis-
students’ stress regulation? Attachment & ruptive behavior and student achievement.
Human Development, 14(3):249-263, DOI: AERA Open. DOI:
https://doi.org/10.1080/14616734.2012.6732 https://doi.org/10.1177/2332858416653921
77 Bosworth, K., Garcia, R., Judkins, M., & Saliba,
Allen, R. M. (2019). What Do Students Know M. (2018). The impact of leadership in-
about University Rankings? Testing Famili- volvement in enhancing high school climate
arity and Knowledge of Global and Domes- and reducing bullying: An exploratory study.
tic University League Tables in China. Fron- Journal of School Violence, 17(3):354-366
tiers of Education in China, 14(1):59-89. DOI:
DOI: https://doi.org/10.1080/15388220.2017.1376
https://doi.org/10.1177/1028315317697539 208
Anderson, J. R. (2006). On cooperative and Bronfenbrenner, U. (1979). The ecology of hu-
competitive learning in the management man development: Experiments by nature
classroom. Mountain Plains Journal of and design. Harvard University Press.
Business and Technology, 7(1): 35-47. Carvalho, C., Santos, J., Conboy, J., & Martins,
Available at: D. (2014). Teachers’ feedback: Exploring
https://openspaces.unk.edu/mpjbt/vol7/iss1/4 differences in students’ perceptions. Social
/ and Behavioral Sciences, 159:169-173. DOI:
Ba, Z., Han, Z., Gong, Z., Li, F., Zhang, H., & https://doi.org/10.1016/j.sbspro.2014.12.351
Zhang, G. (2019). Ethnic differences in ex- Carvalho, C., Santos, N. N., António, R., &
periences of school bullying in China. Chil- Mendonça Martins, D. S. (2021). Supporting
dren and Youth Services Review, students’ engagement with teachers’ feed-
104:104402. DOI: back: The role of students’ school identifica-
https://doi.org/10.1016/j.childyouth.2019.10 tion. Educational Psychology, 41(7):863-882,
4402 DOI:
Bacher-Hicks, A., Goodman, J., Green, J. G., & https://doi.org/10.1080/01443410.2020.1849
Holt, M. (2022). The COVID-19 Pandemic 564
Disrupted Both School Bullying and Centers for Disease Control and Prevention
Cyberbullying. National Bureau of Econom- [CDC]. (2004). Sexual violence prevention:
ic Research (NBER Working Paper No. Beginning the dialogue. Available at:

BECE, Vol.12, No. 1, 2022 1566

 Zhu & Teng. (Shanghai & California). School Bullying Victimization in China.
https://www.cdc.gov/violenceprevention/pdf https://psycnet.apa.org/doi/10.1037/a003943
/SVPrevention-a.pdf 9
Cheema, J., & Kitsantas, A. (2014). Influences Flaspohler, P. D., Elfstrom, J. L., Vanderzee, K.
of disciplinary classroom climate on high L., Sink, H. E., & Birchmeier, Z. (2009).
school student self-efficacy and mathematics Stand by me: The effects of peer and teacher
achievement: A look at gender and racial- support in mitigating the impact of bullying
ethnic differences. International Journal of on quality of life. Psychology in the Schools,
Science and Mathematics Education, 46(7):636-649. DOI:
12(5):1261-1279. DOI: https://doi.org/10.1002/pits.20404
https://doi.org/10.1007/s10763-013-9454-4 Furlong, M. J., & Chung, A. (1995). Who are the
Chen, Z. H., & Chen, S. Y. (2014). When educa- victims of school violence? A comparison of
tional agents meet surrogate competition: student non-victims and multi-victims. Edu-
Impacts of competitive educational agents cation & Treatment of Children, 18(3):282-
on students’ motivation and performance. 299.
Computers & Education, 75:274-281. DOI: http://www.jstor.org/stable/42899414?origin
https://doi.org/10.1016/j.compedu.2014.02.0 =JSTOR-pdf
14 Fyfe, E. R., & Rittle-Johnson, B. (2016). Feed-
China National Radio [CNR]. (2022, April 4). back both helps and hinders learning: The
Comments on punishing 12 bullies: School causal role of prior knowledge. Journal of
violence should be contained before it esca- Educational Psychology, 108(1):82-97. DOI:
lates. China National Radio. Available at: https://doi.org/10.1037/EDU0000053
https://www.thepaper.cn/newsDetail_forwar Goldweber, A., Waasdorp, T. E., & Bradshaw, C.
d_17462363 P. (2013). Examining associations between
Colarossi, L. G., & Eccles, J. S. (2003). Differ- race, urbanicity, and patterns of bullying in-
ential effects of support providers on adoles- volvement. Journal of Youth & Adolescence,
cents’ mental health. Social Work Research, 42(2):206-219. DOI:
27(1):19-30. DOI: https://doi.org/10.1007/s10964-012-9843-y
https://doi.org/10.1093/swr/27.1.19 Goodenow, C., & Grady, K. E. (1993). The rela-
Deutsch, M. (1949). A theory of cooperation and tionship of school belonging and friends’
competition. Human Relations, 2:129-152. values to academic motivation among urban
DOI: adolescent students. The Journal of Experi-
https://doi.org/10.1177%2F00187267490020 mental Education, 62(1):60-71. DOI:
0204 https://doi.org/10.1080/00220973.1993.9943
Deutsch, M. (1962). Cooperation and trust: 831
Some theoretical notes. In M. Jones (Ed.), Gutiérrez-Braojos, C., Montejo-Gamez, J., Mar-
Nebraska symposium on motivation (pp. in-Jimenez, A., & Campaña, J. (2019). Hy-
275-319). University of Nebraska Press. brid learning environment: Collaborative or
Di Stasio, M. R., Savage, R., & Burgos, G. competitive learning? Virtual Reality,
(2016). Social comparison, competition and 23:411-423. DOI:
teacher-student relationships in junior high https://doi.org/10.1007/s10055-018-0358-z
school classrooms predicts bullying and vic- Han, Z., Wang, Z., & Li, Y. (2021).
timization. Journal of Adolescence, 53:207- Cyberbullying involvement, resilient coping,
216. DOI: and loneliness of adolescents during Covid-
https://doi.org/10.1016/j.adolescence.2016.1 19 in rural China. Frontiers in Psychology,
0.002 12:664612. DOI:
Duggins, S. D., Kuperminc, G. P., Henrich, C. https://doi.org/10.3389/fpsyg.2021.664612
C., Smalls-Glover, C., & Perilla, J. L. (2016). Hosmer, D. W., & Lemeshow, S. (2000). Ap-
Aggression among adolescent victims of plied logistic regression (2nd Edition). John
school bullying: Protective roles of family Wiley & Sons.
and school connectedness. Psychology of Huang, L. (2020). Exploring the relationship
Violence, 6(2):205-212. between school bullying and academic per-

BECE, Vol.12, No. 1, 2022 1567

 Zhu & Teng. (Shanghai & California). School Bullying Victimization in China.
formance: The mediating role of students’ an Journal of Guidance and Counselling,
sense of belonging at school. Educational 23(2):157-171. DOI:
Studies, 48(2):216-232. DOI: https://doi.org/10.1017/jgc.2013.20
https://doi.org/10.1080/03055698.2020.1749 Li, L., Chen, X., & Li, H. (2020). Bullying vic-
032 timization, school belonging, academic en-
Huang, L., & Zhao, D. (2019). Empirical re- gagement and achievement in adolescents in
search on relationship between family, eco- rural China: A serial mediation model. Chil-
nomic, social and cultural status and students’ dren and Youth Services Review,
exposure to school bullying—Mediating ef- 113:104946. DOI:
fects of parental support and teacher support. https://doi.org/10.1016/j.childyouth.2020.10
Best Evidence in Chinese Education, 4946
1(1):15-27. DOI: Lian, Q., Yu, C., Tu, X., Deng, M., Wang, T., Su,
https://doi.org/10.15354/bece.19.ar1006 Q., & Zuo, X. (2021). Grade repetition and
Hughes, J. N., Wu, J. Y., Kwok, O. M., Villar- bullying victimization in adolescents: A
real, V., & Johnson, A. Y. (2012). Indirect global cross-sectional study of the Program
effects of child reports of teacher-student re- for International Student Assessment (PISA)
lationship on achievement. Journal of Edu- data from 2018. PLoS Medicine,
cational Psychology, 104(2):350-365. DOI: 18(11):e1003846. DOI:
https://doi.org/10.1037/a0026339 https://doi.org/10.1371/journal.pmed.100384
Johnson, D. W., & Johnson, R. T. (1978). Coop- 6
erative, competitive, and individualistic Loukas, A., Cance, J. D., & Batanova, M. (2016).
learning. Journal of Research & Develop- Trajectories of school connectedness across
ment in Education, 12(1):3-15. the middle school years examining the roles
https://psycnet.apa.org/record/1980-26458- of adolescents’ internalizing and externaliz-
001 ing problems. Youth & Society, 48(4):557-
Johnson, D. W., Johnson, R. T., Roseth, C., & 576. DOI:
Shin, T. S. (2014). The relationship between https://doi.org/10.1177/0044118X13504419
motivation and achievement in interdepend- Ma, L., Luo, H., & Xiao, L. (2021). Perceived
ent situations. Journal of Applied Social teacher support, self-concept, enjoyment and
Psychology, 44(9):622-633. DOI: achievement in reading: A multilevel media-
https://doi.org/10.1111/jasp.12280 tion model based on PISA 2018. Learning
Kim, S., Spadafora, N., Craig, W., Volk, A. A., and Individual Differences, 85:101947. DOI:
& Zhang, L. (2021). Disciplinary structure https://doi.org/10.1016/j.lindif.2020.101947
and teacher support in Chinese and Canadian Ma, X. (2003). Sense of belonging to school:
schools: Examining how authoritative disci- Can schools make a difference? The Journal
plinary practices protect youth involved in of Educational Research, 96(6):340-349.
bullying at school. School Mental Health, DOI:
13:501-517. DOI: https://doi.org/10.1080/00220670309596617
https://doi.org/10.1007/s12310-021-09431-z Matsumura, L., Slater, S., & Crosson, A. (2008).
Lapointe, J. M., Legault, F., & Batiste, S. J. Classroom climate, rigorous instruction and
(2005). Teacher interpersonal behavior and curriculum, and students’ interactions in ur-
adolescents’ motivation in mathematics: A ban middle schools. The Elementary School
comparison of learning disabled, average, Journal, 108(4):293-312. DOI:
and talented students. International Journal https://doi.org/10.1086/528973
of Educational Research, 43(1-2):39-54. Maxwell, S., Reynolds, K. J., Lee, E., Subasic,
DOI: E., & Bromhead, D. (2017). The impact of
https://doi.org/10.1016/j.ijer.2006.03.005 school climate and school identification on
Lester, L., Waters, S., & Cross, D. (2013). The academic achievement: Multilevel modeling
relationship between school connectedness with student and teacher data. Frontier in
and mental health during the transition to Psychology, 8:2069. DOI:
secondary school: A path analysis. Australi- https://doi.org/10.3389/fpsyg.2017.02069

BECE, Vol.12, No. 1, 2022 1568

 Zhu & Teng. (Shanghai & California). School Bullying Victimization in China.
Mazzer, K. R., & Rickwood, D. J. (2015). results (Volume III): What school life means
Teachers’ role breadth and perceived effica- for students’ lives. OECD Publishing. DOI:
cy in supporting student mental health. Ad- https://doi.org/10.1787/acd78851-en
vances in School Mental Health Promotion, Organization of Economic Cooperation and
8(1):29-41. DOI: Development [OECD]. (2019c). PISA 2018
https://doi.org/10.1080/1754730X.2014.978 technical report. OECD Publishing.
119 http://www.oecd.org/pisa/data/pisa2018tech
McNeely, C., & Falci, C. (2004). School con- nicalreport
nectedness and the transition into and out of Pallant, J. (2016). SPSS Survival Manual (6th
health-risk behavior among adolescents: A Edition). McGraw-Hill Education.
comparison of social belonging and teacher Quin, D. (2016). Longitudinal and contextual
support. Journal of School Health, associations between teacher-student rela-
74(7):284-292. DOI: tionships and student engagement: A sys-
https://doi.org/10.1111/j.1746- tematic review. Review of Educational Re-
1561.2004.tb08285.x search, 87(2):345-387. DOI:
Ministry of Education [MOE]. (2017). MOE and https://doi.org/10.3102%2F00346543166694
other 11 relevant departments print and dis- 34
tribute “Comprehensive Plan on Strengthen- Quin, D., Heerde, J. A., & Toumbourou, J. W.
ing the Management of Bullying among (2018). Teacher support within an ecological
Primary and Middle School Students”. MOE model of adolescent development: Predictors
of the People’s Republic of China. Available of school engagement. Journal of School
at: Psychology, 69:1-15. DOI:
http://www.moe.gov.cn/srcsite/A11/moe_17 https://doi.org/10.1016/j.jsp.2018.04.003
89/201712/t20171226_322701.html Raskauskas, J. L., Gregory, J., Harvey, S. T.,
Ministry of Education [MOE]. (2021). MOE Rifshana, F., & Evans, I. M. (2010). Bully-
carries out supervision and appraisal work ing among primary school children in New
on kindergarten headmasters and school Zealand: Relationships with prosocial be-
principals after their terms of office expire. haviour and classroom climate. Educational
MOE of the People’s Republic of China. Research, 52(1):1-13. DOI:
Available at: https://doi.org/10.1080/00131881003588097
http://www.moe.gov.cn/srcsite/A11/s7057/2 Roseth, C. J., Johnson, D. W., & Johnson, R. T.
02201/t20220124_596061.html (2008). Promoting early adolescents’
Moos, R. (1979). Evaluating educational envi- achievement and peer relationships: The ef-
ronments. Jossey-Bass. fects of cooperative, competitive, and indi-
Mostafa, T., Echazarra, A., & Guillou, H. (2018), vidualistic goal structures. Psychological
The science of teaching science: An explora- Bulletin, 134(2):223-246. DOI:
tion of science teaching practices in PISA https://doi.org/10.1037/0033‐
2015. OECD Education Working Papers. 2909.134.2.223
DOI: https://doi.org/10.1787/f5bd9e57-en Ryan, A. M., & Patrick, H. (2001). The class-
Organization of Economic Cooperation and room social environment and changes in ad-
Development [OECD]. (2017). PISA 2015 olescents’ motivation and engagement dur-
results (Volume III): Students’ well-being. ing middle school. American Educational
OECD Publishing. DOI: Research Journal, 38:437–460. DOI:
https://doi.org/10.1787/9789264273856-en https://doi.org/10.3102%2F00028312038002
Organization of Economic Cooperation and 437
Development [OECD]. (2019a). PISA 2018 Ryzin, M. J., & Roseth, C. J. (2019). Effects of
assessment and analytical framework. cooperative learning on peer relations, em-
OECD Publishing. DOI: pathy, and bullying in middle school. Ag-
https://doi.org/10.1787/b25efab8-en gressive Behavior, 45(6):643-651. DOI:
Organization of Economic Cooperation and https://doi.org/10.1002/ab.21858
Development [OECD]. (2019b). PISA 2018

BECE, Vol.12, No. 1, 2022 1569

 Zhu & Teng. (Shanghai & California). School Bullying Victimization in China.
Schleicher, A. (2018). Educating learners for students’ perceptions in a game-based lan-
their future, not our past. ECNU Review of guage learning environment. Educational
Education, 1(1):58-75. DOI: Technology Research and Development,
https://doi.org/10.30926/ecnuroe201801010 61:927-950. DOI:
4 https://doi.org/10.1007/s11423-013-9314-5
Sortkær, B., & Reimer, D. (2018). Classroom Van Ryzin, M. J., & Roseth, C. J. (2018a). Co-
disciplinary climate of schools and gender– operative learning in middle school: A
Evidence from the Nordic countries. School means to improve peer relations and reduce
Effectiveness and School Improvement, victimization, bullying, and related out-
29(4):511-528. DOI: comes. Journal of Educational Psychology,
https://doi.org/10.1080/09243453.2018.1460 110(8):1192-1201. DOI:
382 https://doi.org/10.1037/edu0000265
Salihu, H. M., Wilson, R. E., King, L. M., Marty, Van Ryzin, M. J., & Roseth, C. J. (2019). Ef-
P. J., & Whiteman, V. E. (2015). Socio- fects of cooperative learning on peer rela-
ecological model as a framework for over- tions, empathy, and bullying in middle
coming barriers and challenges in random- school. Aggressive Behavior, 45(6):643-651.
ized control trials in minority and under- DOI: https://doi.org/10.1002/ab.21858
served communities. International Journal Volk, A. A., Cioppa, V. D., Earle, M., & Farrell,
of MCH and AIDS, 3(1):85-95. Available at: A. H. (2015). Social competition and bully-
https://www.ncbi.nlm.nih.gov/pmc/articles/P ing: An adaptive socioecological perspective.
MC4948176/ In V. Zeigler-Hill, L. Welling, & T. Shackel-
Swearer, S. M., Espelage, D. L., Vaillancourt, T., ford (Eds.), Evolutionary perspectives on so-
& Hymel, S. (2010). What can be done cial psychology (pp. 387-399). Springer.
about school bullying? Linking research to DOI: https://doi.org/10.1007/978-3-319-
educational practice. Educational Research- 12697-5_30
er, 39(1):38-47. DOI: Waasdorp, T. E., & Bradshaw, C. P. (2015). The
https://doi.org/10.3102/0013189X09357622 overlap between cyberbullying and tradi-
Tardy, C. H. (1985). Social support measure- tional bullying. Journal of Adolescent
ment. American Journal of Community Psy- Health, 56(5):483-488. DOI:
chology, 13:187-202. DOI: https://doi.org/10.1016/j.jadohealth.2014.12.
https://doi.org/10.1007/BF00905728 002
Townsend, N., & Foster, C. (2011). Developing Wang, M. T, Degol, J. L., Amemiya, J., Parr, A.,
and applying a socio-ecological model to the & Guo J. (2020). Classroom climate and
promotion of healthy eating in the school. children’s academic and psychological well-
Public Health Nutrition, 16(6):1101–1108. being: A systematic review and meta-
http://doi.org/10.1017/S1368980011002655 analysis. Developmental Review, 57:100912.
Underwood, M. K., & Ehrenreich, S. E. (2014). DOI:
Bullying may be fueled by the desperate https://doi.org/10.1016/j.dr.2020.100912
need to belong. Theory into Practice, Wang, M. T., & Eccles, J. S. (2012). Social sup-
53(4):265-270. DOI: port matters: Longitudinal effects of social
https://doi.org/10.1080/00405841.2014.9472 support on three dimensions of school en-
17 gagement from the middle to high school.
Vaillancourt, T., Brittain, H., Krygsman, A., Child Development, 83(3):877-895. DOI:
Farrell, A. H., Landon, S., & Pepler, D. https://doi.org/10.1111/j.1467-
(2021). School bullying before and during 8624.2012.01745.x
COVID‐19: Results from a population‐ Wang, M. T., & Holcombe, R. (2010). Adoles-
based randomized design. Aggressive Behav- cents’ perceptions of the school environment,
ior, 47:557-569. DOI: engagement, and academic achievement in
https://doi.org/10.1002/ab.21986 middle school. American Educational Re-
Vandercruysse, S., Vandewaetere, M., Cornillie, search Journal, 47(3):633-662. DOI:
F., & Clarebout, G. (2013). Competition and https://doi.org/10.3102/0002831209361209

BECE, Vol.12, No. 1, 2022 1570

 Zhu & Teng. (Shanghai & California). School Bullying Victimization in China.
Wei, W., & Xie, Y. (2018). University teachers’ countries in PISA. Social Sciences & Hu-
reflections on the reasons behind their manities Open, 4(1):100151. DOI:
changing feedback practice. Assessment & https://doi.org/10.1016/j.ssaho.2021.100151
Evaluation in Higher Education, 43(6):867- Zhao, X. (2021, November 19). Strict punish-
879, DOI: ment urged to prevent bullying. China Daily
https://doi.org/10.1080/02602938.2017.1414 (Global Edition). Available at:
146 http://global.chinadaily.com.cn/a/202111/19/
Wentzel, K. R. (2016). Teacher-student relation- WS6196fe63a310cdd39bc763f2.html
ships. In K. R. Wentzel, & D. B. Miele Zook, C. (2018, October 18). What is coopera-
(Eds.), Handbook of motivation at school tive learning and how does it work? Applied
(pp. 211-230). Routledge. Educational Systems [AES]. Available at:
Wentzel, K. R., Battle, A., Russell, S. L., & https://www.aeseducation.com/blog/what-is-
Looney, L. B. (2010). Social supports from cooperative-learning-and-how-does-it-work
teachers and peers as predictors of academic Zuo, M. (2021, November 11). China’s bullying
and social motivation. Contemporary Educa- problem in spotlight after teenager’s death.
tional Psychology, 35(3):193-202. DOI: South China Morning Post. Available at:
https://doi.org/10.1016/j.cedpsych.2010.03.0 https://www.scmp.com/news/people-
02 culture/social-
Yu, S., & Zhao, X. (2021). The negative impact welfare/article/3155602/teenage-boys-
of bullying victimization on academic litera- drowning-spotlights-chinas-bullying
cy and social integration: evidence from 51

Received: 21 July 2022

Revised: 10 August 2022
Accepted: 18 August 2022

BECE, Vol.12, No. 1, 2022 1571

 Zhu & Teng. (Shanghai & California). School Bullying Victimization in China.

Appendix I

Figure S1. Normal P-P Plot of Regression Standardized Residual.

Note. The points did not lie in a reasonably straight diagonal line from bottom left to
top right. This would suggest major deviations from normality and multiple regression
analyses did not fit for statistics.

BECE, Vol.12, No. 1, 2022

 Zhu & Teng. (Shanghai & California). School Bullying Victimization in China.

Figure S2. Scatterplot of Standardized Residuals.

Note. In the Scatterplot of the standardized residuals, the residuals appeared as a

clear or systematic pattern, higher on one side than the other. This suggested that
multiple regression analyses did not fit for statistics.

BECE, Vol.12, No. 1, 2022

 Zhu & Teng. (Shanghai & California). School Bullying Victimization in China.

Table S1. Multiple Regression Analysis of Demographic, Teacher-support,

Sense-of-belonging, School-climate, and Full-predictor Models.

a, b c d e
Model 1 Model 2 Model 3 Model 4 Model 5
Variables β β β β β
Gender .169‡ .165‡ .178‡ .146‡ .149‡
Age -.008 -.019‡ -.030‡ -.030‡ -.039‡
Grade Repetition .048‡ .047‡ .043‡ .045‡ .039‡
ESCS -.050‡ -.025† -.012 -.016 .001
TEACHINT -.113‡ -.016
DIRINS -.019 .009
PERFEED .017 .045‡
TEACHSUP -.108‡ -.076‡
BELONG -.283‡ -.215‡
DISCLIMA -.208‡ -.169‡
PERCOOP -.153‡ -.071‡
PERCOMP .119‡ .112‡
Notes. *<.05, †p < .01, ‡ p < .001
a b c d e
Demographic model, Teacher-support model, Sense-of-belonging model, School-climate model,
Full-predictor model

BECE, Vol.12, No. 1, 2022

 Zhu & Teng. (Shanghai & California). School Bullying Victimization in China.

Table S2. Logistic Regression Analysis of Sense-of-belonging and School-climate

Models.

a b
Model 3 Model 4
Variables B S.E. Exp(B) B S.E. Exp(B)
Gender .568‡ .039 1.765 .457‡ .039 1.580
Age -.136* .066 .873 -.158* .067 .854
Grade Repetition .215† .081 1.239 .238† .082 1.269
ESCS -.008 .018 .651 -.007 .019 .993
BELONG -.640‡ .026 .527

DISCLIMA -.409‡ .020 .664

PERCOOP -.344‡ .021 .709
PERCOMP .246‡ .022 1.279
Notes. *<.05, † p < .01, ‡ p < .001,
a b
Sense of Belonging, School Climate
Preliminary assumption tests revealed that for Models 3 and 4, the Hosmer and Lemeshow Goodness of
Fit Tests were not passed (significance values all below .05), suggesting no support for these two
models.

BECE, Vol.12, No. 1, 2022