







































 

 

 

 

 

Empirical Research on the Relationship 

between Family Economic, Social and 

Cultural Status and Students’ Exposure to 

School Bullying: Mediating Effects of 

Parental Support and Teacher Support 

Liang Huang
1,2

, Decheng Zhao
3
 

1. The Chinese University of Hong Kong, Hong Kong, China 
2. Southeast University, Nanjing, China 

3. Beijing Normal University, Beijing, China 

Abstract. Using survey data collected by the Organization for the Eco-

nomic Cooperation and Development (OECD) in its 2015 Program for 

International Student Assessment (PISA2015), this study explores the 

relationship between family economic, social and cultural status (ESCS) 

and students’ exposure to school bullying for students in Beijing-

Shanghai-Jiangsu-Guangdong (China). Additionally, the study examines 

the mediating effects of parental support and teacher support on the re-

lationship between family ESCS and students’ exposure to school bully-

ing. Lower family ESCS led to significantly increased exposure to school 

bullying. Parental support and teacher support mediated the relation-

ship between family ESCS and students’ exposure to school bullying, 

with slightly different magnitudes. The results suggest that China should 

continue to improve its system for school bullying prevention and treat-

ment, and should especially focus on reducing bullying victimization 

among students from disadvantaged backgrounds. Further, parents in 

disadvantaged families should provide support and care to their children 

to enhance their abilities to tackle bullying victimization, and teachers 

should treat students fairly and guide students in socializing with their 

peers appropriately. 

Best Evid Chin Edu 2019; 1(1):15-27. 

Doi: 10.15354/bece.19.ar1006 

Keywords: PISA2015; Family ESCS; School Bullying; Parental Support; Teacher 

Support; Mediating Effect 

 



 

 

Vol. 1, No. 1, 2019 16 

 

RELATIVELY high prevalence of school bullying in China, with events of 

bullying and school violence being reported, has been observed recently 

(Wang, 2016). Bullying is widely recognized as a specific form of aggressive 

behavior that is repetitive, intentional, and based on a power imbalance (Smith et al., 

1999). School bullying experienced by adolescent students, who are in a period of tran-

sition, has negative effects on their physical and mental health, on their academic capa-

bility and social adaptability, and can be irreversibly detrimental to the students’ long-

term wellbeing (Woods et al., 2004). Research reveals that bullying victimization se-

verely hampers students’ educational attainment and personality development. Students 

frequently exposed to school bullying have lower educational performance and are 

more likely to suffer from various psychological symptoms, such as anxiety, depression, 

loneliness, and suicidal tendencies, compared to their counterparts (Delprato et al., 2017; 

Kaltiala-Heino et al., 2000). In fact, the high prevalence of school bullying and its se-

vere impacts has become a major policy concern in China. In 2016, a national anti-

bullying policy, Guidance on the Prevention and Treatment of Bullying and School Vio-

lence, that required schools’ active participation in counteracting school bullying and 

protecting students’ rights, was promulgated by the Ministry of Education and eight 

other central ministries. Consequently, it is imperative for researchers and practitioners 

to identify the factors associated with school bullying and to help design effective poli-

cies that support and protect at-risk students (Olweus, 1994).  

Of the risk factors for bullying victimization, family socioeconomic background is 

recognized as an important but still under-researched predictor. Prior research by schol-

ars in other countries has shown connections between students’ exposure to school bul-

lying and their family socioeconomic backgrounds. In many cases, students with disad-

vantaged family backgrounds tend to suffer peer victimization at a higher rate than stu-

dents from advantaged families. For example, Due et al. (2009), using data from the 

international ‘Health Behavior in School-aged Children’ study involving 35 countries, 

examined the socioeconomic inequality in students’ exposure to school bullying and 

found that students from families of low affluence were more likely to be bullied. Con-

sistent with this, Analitis et al. (2009) explored the factors associated with bullying vic-

timization in children and adolescents aged 8 to 18 and found that students whose par-

ents had a lower educational level reported more peer victimization. Empirical findings 

suggest that the inequality in students’ exposure to bullying that exists in various cultur-

al contexts may prevent some students from equally healthy development. As a result, 

this has become a concern for policy-makers, researchers, and practitioners worldwide. 

Nonetheless, empirical research on the mechanisms underlying the inequality of

 

About the Authors: Liang Huang is a Lecturer at Department of Public Administration, Southeast University, 

China; and Decheng Zhao is a Professor at Faculty of Education, Beijing Normal University, China. 

Correspondence to: Liang Huang, Department of Public Administration, Southeast University, China. E-mail: 

eliot_huang@163.com. 

Conflict of Interests: None. 

A 



 

 

Vol. 1, No. 1, 2019 17 

 

students’ exposure to bullying is limited, specifically the ways in which variances in 

family economic, social and cultural status (ESCS) affect the likelihood of a student 

being bullied. An emerging line of empirical evidence suggests social support is im-

portant in protecting students from being bullied. Students from families with lower 

ESCS may be more likely to be bullied because they receive less social support com-

pared with other students (Jansen et al., 2011). As parents and teachers are the most 

important sources of social support for students facing possible bullying, we hypothe-

size that inequality in family ESCS among students may result in an inequality of pa-

rental support and teacher support and that this may further lead to inequalities in stu-

dents’ exposure to school bullying. Specifically, two questions guided this research: 

1. What are the effects of family ESCS on students’ exposure to school bullying? 

2. Do parental support and teacher support mediate the effects of family ESCS on 

students’ exposure to school bullying? 

Data from students in China’s four provinces of Beijing, Shanghai, Jiangsu, and 

Guangdong were used in this study. The findings of this study can inform the design of 

effective anti-school bullying policies to put into practice. 

Methods 

Sample 

The Organization for Economic Cooperation and Development (OECD) collects data 

on students’ abilities from countries worldwide through its Program for International 

Student Assessment (PISA). In the 2015 assessment round (hereafter, PISA2015), PISA 

for the first time included questions to evaluate bullying in schools of participating 

countries and economies (OECD, 2016). The school bullying variable, together with 

other variables in the PISA2015 database reflecting students’ individual characteristics, 

family backgrounds and school characteristics, provided the data needed for this study. 

Listwise deletion was employed to exclude those cases with missing data; 8,671 15-

year-old students from the PISA2015 Beijing-Shanghai-Jiangsu-Guangdong (China) 

database were identified as the sample of this study. 

Variables 

Students’ Exposure to School Bullying 

PISA2015 measured students’ exposure to school bullying from the perspective of the 

victims (OECD, 2017). Students were asked to report their experiences during the past 

one year with six forms of school bullying actions: 1) “Others spread rumors about me”; 

2) “Being left out of things by others intentionally”; 3) “Being made fun of by others”; 

4) “Being threatened by others”; 5) “Others took away or destroyed my things”; and 6) 

“Being hit or pushed by others” (OECD, 2017). The frequency of students experiencing 

any one of the six forms of school bullying was classified into four categories: 1) “never 

or almost never”; 2) “a few times a year”; 3) “a few times a month”; and 4) “once a 



 

 

Vol. 1, No. 1, 2019 18 

 

week or more” (OECD, 2017). Based on Maynard et al. (2016), students with responses 

of “a few times a month” or “once a week or more” were defined as being exposed to 

that specific form of school bullying action. Students exposed to least one form of 

school bullying were defined as being exposed to school bullying and assigned a value 

of 1 for the bullying variable; other students were assigned a value of 0. In this study, 

21.74% of students reported being exposed to school bullying during the past one year. 

Family ESCS 

PISA computed an index of family ESCS using family background information, such as 

parents’ highest education level, parents’ highest occupation status, and home posses-

sions (OECD, 2017). The family ESCS was transformed with 0 representing the aver-

age family ESCS of OECD students and 1 representing the standard deviation (SD) of 

the OECD students’ family ESCS. In the subset of data used for this study, student fam-

ily ESCS had a mean of -0.84 and a SD of 1.12. 

Parental Support 

PISA2015 asked students to report the extent to which their parents support them in 

their daily living, school learning and other activities (e.g., “My parents are interested in 

my school activities,” and “My parents support me when I am facing difficulties at 

school,” and “My parents encourage me to be confident”) (OECD, 2017). From the re-

sponses, PISA computed an index of parental support, which was transformed with 0 

representing the average parental support of OECD students and 1 representing the SD 

of the OECD students’ parental support. In this study, the sample students’ parental 

support had a mean of -0.16 and a SD of 0.88. 

Teacher Support 

PISA2015 asked students to report the frequency of the help, care and support they got 

from their science teachers in classroom learning (e.g., “The teachers give extra help 

when students need it,” and “The teachers help students with their learning,” and “The 

teachers give students an opportunity to express their opinions”) (OECD, 2017) and 

used this to compute an index of teacher support. Once again, the index of teacher sup-

port was transformed with 0 representing the average teacher support of OECD students 

and 1 representing the SD of the OECD students’ teacher support. For the sample subset 

used here, the students’ teacher support index had a mean of 0.28 and a SD of 0.92. 

Control variables 

Included control variables related to individual students were gender (52% males, 48% 

females) and study program (56% lower secondary school students, 35% upper second-

ary school students, and 8% vocational school students). Also included were control 

variables related to school characteristics: school location (37% urban schools, 63% 



 

 

Vol. 1, No. 1, 2019 19 

 

rural schools), school type (90% public schools, 10% private schools), school size (in 

hundreds of students, mean = 19.70, SD = 19.18), class size (number of students, mean 

= 41.66 , SD = 9.37), student-teacher ratio (mean = 12.62, SD = 8.55), proportion of all 

teachers fully certified (mean = 0.97, SD = 0.07), school educational material shortage 

(mean = 0.07, SD = 1.22), and school educational staff shortage (mean = 0.77, SD = 

1.28).  

Data Analysis 

Data analysis involved three phases. In phase one, descriptive statistics were used to 

preliminarily examine whether exposure to school bullying differed among students 

with different family ESCS. In phase two, a binary logit model was constructed to ana-

lyze the predictive effects of family ESCS on students’ exposure to school bullying. 

The binary logit model took the following form, 

Yi = α0 +   
 
   

 

   
 + γ1ESCSi + ui1 (1) 

where Yi refers to students’ exposure to school bullying, α0 is a constant, xij is a vector 

of control variables, ESCSi is the student’s family ESCS, and ui1 is the error term. 

In phase three of the data analysis, a path analysis model was constructed to exam-

ine the mediating roles of parent support and teacher support on the relationship be-

tween family ESCS and students’ exposure to school bullying. The path analysis model 

has the advantage that it can be used to simultaneously estimate casual relationships 

among multiple variables (Wang, 2014). The path analysis model had the following 

form, 

Mti = α1 +  
 
   

 

   
 + γ2ESCSi + ui2 (2) 

Yi = α2 +  
 
   

 

   
 + γ3ESCSi + γ4Mti + ui3 (3) 

where Yi refers to students’ being exposed to school bullying, α1 and α2 are constants, 

Xij is a vector of control variables, Mti are the mediating variables between family 

ESCS and students’ exposure to school bullying, including the parental support variable 

(M1) and the teacher support variable (M2), and ui2 and ui3 are error terms. As parental 

support and teacher support are correlated, the path coefficient between the two mediat-

ing variables in the model was set to be freely estimated. 

Since PISA2015 employed two-stage sampling methods to select sample students
1
, 

for phases two and three of the data analysis, the standard errors of parameter estimates 

were estimated using both the final student weighting and Fay’s balanced repeated rep-

lication (BRR) method with a 0.5 coefficient (OECD, 2009). Moreover, in phase three 

data analysis, a bias-corrected bootstrap method was employed for analysis of mediat-

ing variables. The bootstrap method as a resampling estimation method produces more 

accurate interval estimates (Hayes, 2009). In this study, resampling was set to 2,000 



 

 

Vol. 1, No. 1, 2019 20 

 

times. The software SPSS 23.0 was used to conduct descriptive statistics (phase one) 

and Mplus 7.0 was used to conduct data analysis (phases two and three). 

Results 

Descriptive Statistics 

To initially explore whether students’ exposure to school bullying was affected by fami-

ly ESCS background, we grouped students into three categories based on their family’s 

ECSC: (1) advantaged family backgrounds – student’s family ESCS was in the top 

quarter of the index of the sample students’ family ESCS; (2) average family back-

grounds – student’s family ESCS was between the 25th to 75th percentile of the index 

of the sample students’ family ESCS; and (3) disadvantaged family backgrounds – stu-

dent’s family ESCS was in the bottom quarter of the index of the sample students’ fami-

ly ESCS. We then used descriptive statistics for each group to look for differences 

among groups in students’ exposure to school bullying behavior. Table 1 shows the 

results of the descriptive statistics. In general, students’ exposure to any form of school 

bullying was less for students from advantaged family backgrounds (18.96%) than for 

students from average family ESCS backgrounds (21.55%), which, in turn, was less 

than that for students from disadvantaged backgrounds (24.91%). Students’ exposure to 

school bullying differed significantly among the three types of family backgrounds (χ
2
 

= 22.75, p < 0.001). 

The percentage of students exposed to each specific form of school bullying simi-

larly decreased when going from disadvantaged family backgrounds to average family 

backgrounds to advantaged family backgrounds. The percentage of students who re-

ported “Being left out of things by others intentionally,” or “Being made fun of by oth-

ers,” or “Others took away or destroyed my things,” or “Being threatened by others” 

differed significantly among the three types of family backgrounds (Chi-squared p-

value ≤ 0.01 in all tests). This further verified that differences exist in students’ expo-

sure to school bullying under different family backgrounds.  

Binary Logit Model: Effects of Family ESCS on Students’ Exposure to School Bully-

ing 

Given the differences observed above in students’ exposure to school bullying based on 

different family backgrounds, we next analyzed the predictive effects of family ESCS 

on students’ exposure to school bullying by employing a binary logit model (equation 

(1) above) to the data. Table 2 displays these results. Without controlling for other vari-

ables, family ESCS had a significant negative effect on students’ exposure to school 

bullying (γ = -0.055, p < 0.001). The predictive effects of family ESCS on students’ 

exposure to school bullying remained significant after controlling for variables related 

to individual students (γ = -0.052, p< 0.01) or variables related to school characteristics 

(γ = -0.040, p < 0.05). The results of the binary logit model are consistent with the re-



 

 

Vol. 1, No. 1, 2019 21 

 

 

Table 1. Students’ Exposure to School Bullying under Different Family Back-
grounds (%). 

Exposure To 
School Bullying 
Actions 

All 
Students 

Disadvantaged 
Family Back-
grounds 

Average Fami-
ly Back-
grounds 

Advantaged 
Family Back-
grounds 

x
2
 

Exposure to any 
form of school 
bullying 

21.74 24.91 21.55 18.96 22.75*** 

Others spread 
rumors about me 

6.30 6.50 6.34 6.00 0.51 

Being left out of 
things by others 
intentionally 

7.57 9.87 7.24 5.90 25.68*** 

Being made fun of 
by others 

11.50 13.93 11.35 9.36 22.40*** 

Being threatened 
by others 

3.29 4.11 3.37 2.31 11.21** 

Others took away 
or destroy my 
things 

12.57 14.30 12.57 10.84 11.80** 

Being hit or pushed 
by others 

4.29 4.57 4.41 3.78 1.90 

*P<0.05，**P<0.01，***P<0.001 

 

 

sults of the descriptive statistics and both show that students with higher family ESCS 

are less likely to be bullied. 

Path Analysis: Mechanisms Underlying Effect of Family ESCS on Students’ Expo-

sure to School Bullying 

To further explore the mechanisms underlying effect of family ESCS on students’ expo-

sure to school bullying, we conducted path analysis using path analysis equations (2) and 

(3) (see Methods). The weighted root mean-square residual (WRMR), an indicator of the 

path model fit, was 0.002, indicating a good model fit.
2
 

Table 3 shows the results of the path analysis model. Results reveal that the direct 

predictive effects of family ESCS on students’ exposure to school bullying was -0.019 

(p > 0.05). Family ESCS affected students’ exposure to school bullying indirectly 

through its influence on parental support and teacher support of students. The predictive 

effects of family ESCS on parental support was 0.138 and the predictive effects of paren-

tal support on students’ exposure to school bullying was -0.086; both these path coeffi-

cients reached at least a 0.01 level of significance. The predictive effects of family ESCS 

on teacher support was 0.070 and the predictive effects of teacher support on students’ 

exposure to school bullying was -0.142; both path coefficients had p < 0.001. 



 

 

Vol. 1, No. 1, 2019 22 

 

 

 

 

Analysis of mediating variables was conducted using the bias-corrected bootstrap 

method. Table 4 shows the results. The 95% bootstrap confidence interval (CI) of both 

the indirect effects of family ESCS on students’ exposure to school bullying through 

parental support and the indirect effects of family ESCS on students’ exposure to school 

bullying through teacher support did not include zero, indicating significant mediating 

effects. The magnitude of the mediating effects of parental support and teacher support 

on the relationship between family ESCS and students’ exposure to school bullying 

were slightly different. 

Thus, the mechanism underlying family ESCS effects on students’ exposure to 

school bullying has been elucidated: students with a higher family ESCS get more pa-

rental support and teacher support, and students with more parental support and teacher 

support have a lower risk of being bullied. Conversely, students with lower family 

ESCS get less parental support and teacher support, and students with less parental sup-

port and teacher support have a higher risk of being bullied. 

Table 2. The Predictive Effects of Family ESCS on Students’ Being Exposed 
to School Bullying. 

 Students’ Exposure to School Bullying 

Coefficient SE Coefficient SE Coefficient SE 

Family ESCS -0.055*** 0.016 -0.052** 0.019 -0.040* 0.020 

Male (Reference group: Female)   0.423*** 0.046 0.422*** 0.046 

Lower secondary school students 
(Reference group: Vocational 
school students) 

  -0.030 0.079 -0.114 0.122 

Upper secondary school students 
(Reference group: Vocational 
school students) 

  -0.105 0.087 -0.166 0.108 

Urban school (Reference group: 
Rural school) 

    -0.025 0.067 

Public school (Reference group: 
Private school) 

    -0.108 0.113 

School size     -0.002 0.002 

Class size     0.007 0.005 

Student-teacher ratio     0.003 0.003 

Proportion of all teachers fully 
certified 

    0.027 0.317 

School educational material 
shortage 

    0.008 0.029 

School educational staff shortage     0.012 0.036 

Threshold 0.833*** 0.033 1.018*** 0.076 1.148** 0.375 

*P<0.05, **P<0.01, ***P<0.001 



 

 

Vol. 1, No. 1, 2019 23 

 

 

 

Discussion and Implications 

Discussion 

In this study, we found that students’ exposure to school bullying differed depending on 

family ESCS background. Students from advantaged family backgrounds had decreased 

exposure to bullying (18.96%) compared to those from disadvantaged family back-

grounds (24.91%). We also verified that the inequality of family ESCS is a key factor 

affecting the likelihood of students being bullied in the four China provinces of Beijing, 

Shanghai, Jiangsu, and Guangdong. Bullying victimization can hinder students’ physi-

cal and mental development, their educational attainment, and undermine their social 

Table 3. Results of the Path Analysis Model.
 3
 

 

Parental support 
(M1) 

Teacher support 
(M2) 

Students’ Exposure to 
School Bullying 

Coefficient SE Coefficient SE Coefficient SE 

Family ESCS 0.138*** 0.013 0.070*** 0.015 -0.019 0.020 

Parental support     -0.086** 0.028 

Teacher support     -0.142*** 0.022 

Male (Reference group: Fe-
male) 

-0.078*** 0.023 -0.090*** 0.026 0.402*** 0.047 

Lower secondary school stu-
dents (Reference group: Vo-
cational school students) 

0.026 0.094 0.113 0.071 -0.095 0.118 

Upper secondary school stu-
dents (Reference group: Vo-
cational school students) 

0.109 0.094 -0.024 0.064 -0.161 0.103 

Urban school (Reference 
group: Rural school) 

-0.045 0.034 0.011 0.040 -0.028 0.067 

Public school (Reference 
group: Private school) 

-0.100 0.067 -0.086 0.081 -0.129 0.109 

School size 0.000 0.001 -0.001 0.001 -0.002 0.002 

Class size 0.002 0.002 -0.007* 0.003 0.006 0.005 

Student-teacher ratio -0.004* 0.002 -0.008 0.005 0.001 0.003 

Proportion of all teachers fully 
certified 

0.850*** 0.221 0.698 0.385 0.194 0.284 

School educational material 
shortage 

-0.017 0.016 -0.054* 0.027 -0.001 0.029 

School educational staff 
shortage 

-0.027 0.015 -0.011 0.024 0.008 0.035 

Intercept -0.816*** 0.208 0.101 0.397   

Threshold     1.207*** 0.386 

*P<0.05, **P<0.01, ***P<0.001 



 

 

Vol. 1, No. 1, 2019 24 

 

 

Table 4. Mediation Analysis Using Bias-Corrected Bootstrap Method. 

Influence Path 
Estimate 

95% Bootstrap 
CI 

Coefficient SE Lower Upper 

Family ESCS → parental support → students’ exposure to 
school bullying 

-0.012** 0.004 -0.015 -0.008 

Family ESCS → teacher support → students’ exposure to 
school bullying 

-0.010*** 0.003 -0.012 -0.008 

Total indirect effects -0.022*** 0.005 -0.027 -0.017 

*P<0.05, **P<0.01, **P<0.001 

 

 

and economic welfare in adulthood (Brown et al., 2008). As a result, students from dis-

advantaged families in the four provinces of China may not enjoy equal opportunities to 

flourish due to an increased exposure to school bullying.  

We identify the mechanism by which family ESCS influences students’ exposure 

to school bullying: family ESCS indirectly impacted students’ exposure to school bully-

ing through its effect on parental and teacher support available to students. Specifically, 

family ESCS inequality leads to inequality in levels of parental support and teacher 

support, which further created inequality of students’ being bullied. In terms of the me-

diating role of parental support, parents from families with higher ESCS tend to have 

higher educational levels. This may mean they make more conscious efforts to get in-

volved in their children’s education and development and are more willing to com-

municate with their children to help solve the learning and living problems their chil-

dren confront (Lereya et al., 2013). Caring and supportive ties between parents and 

children are conducive to the development of students’ academic adaptability and social 

skills and thus can protect students from being bullied (Rivara et al., 2016). In contrast, 

parents from families with lower ESCS have limited resources, capabilities, and time 

and thus may be less able to provide their children with timely and effective help when 

their children face difficulties, making these students more likely to be bullied. In terms 

of the mediating role of parental support, parents from families with higher ESCS can 

mobilize more social and cultural resources to become better involved in their child’s 

school educational activities and can actively build cooperative relations with teachers, 

which will help their children get more support from teachers (McNeal et al., 1999). 

Support from teachers helps students gain peer acceptance and enhance their social 

competency, which can reduce students’ exposure to school bullying (Troop-Gordon et 

al., 2011). In contrast, parents from families with lower ESCS lack the relevant re-

sources and capabilities to create a favorable external developmental environment for 

their children, making these students more likely to be bullied. 

Implications 



 

 

Vol. 1, No. 1, 2019 25 

 

In light of the discussion above, the following recommendations are offered. 

First, China should continue to improve its system of school bullying prevention 

and treatment to reduce bullying victimization among students from disadvantaged 

backgrounds. School bullying prevention and treatment has become a major concern for 

the country’s policy-makers. To promote the implementation of effective school bully-

ing prevention and treatment initiatives, China should continue to improve its system of 

school bullying prevention and treatment to comprehensively tackle all forms of school 

bullying and to guarantee that all students have equal and sufficient opportunities for 

healthy physical and mental development. For students in schools within Beijing, 

Shanghai, Jiangsu, and Guangdong provinces, family ESCS inequality resulted in stu-

dents from disadvantaged families experiencing disproportionately more bullying. 

Therefore, while formulating the school bullying prevention and treatment policy, the 

country should give more policy support to students from disadvantaged ESCS families. 

Likewise, educational authorities should invest more resources and professional support 

in rural schools, disadvantaged schools, and migrant schools that have higher propor-

tions of students from disadvantaged family backgrounds and should help these schools 

build up their bullying prevention and treatment system to reduce the risk of school bul-

lying in disadvantaged students. 

Second, parents in disadvantaged families should give more support and care to 

their children to enhance their abilities to resist bullying victimization. The role of par-

ents in helping students counteract school bullying is critically important (Liao et al., 

2017). However, parents from families with lower ESCS often cannot offer such help, 

care, and support to their children, leaving their children exposed to bullying victimiza-

tion. In this study, we propose that even though parents of disadvantaged families may 

lack the corresponding resources and capabilities to provide their children with a better 

developmental environment, they can still be actively involved in their children’s edu-

cation, pay more attention to their children’s mental and behavioral development, and 

improve the parent-child relationship through increased close communication and inter-

action to help their offspring cope with difficulties in learning, living and peer interac-

tion and to protect them from the harmful effects of being bullied. We suggest that par-

ents make a conscious effort to learn how to prevent and treat school bullying and to 

help their children recognize school bullying and make appropriate responses. In addi-

tion, parents who find that their children have been exposed to school bullying should 

seek the support of communities and schools to help these students overcome bullying 

victimization. 

Third, teachers should treat students fairly and guide students to socialize with 

their peers appropriately. Teachers are a valuable resource that students can rely on for 

help with school bullying. The help and care teachers give to students is conducive to 

broadening student-teacher communication channels and building positive student-

teacher relations. This enables students to actively seek their teachers’ support when in 

need and helps student get the acceptance and recognition of peers. In this study, we 

found that the teacher support that students receive varied for students from different 



 

 

Vol. 1, No. 1, 2019 26 

 

family backgrounds, such that students with higher family ESCS have a lower risk of 

being bullied and students with lower family ESCS have a higher risk of being bullied. 

To reduce exposure to school bullying for all students, we suggest that teachers treat 

students fairly, analyze the behavioral and mental progress of their students, and pro-

vide timely help and support to students facing learning, living and socializing difficul-

ties. Additionally, teachers can help students cultivate good conduct, such as being 

friendly and helpful, and respecting others, and can guide students to socialize with 

their peers appropriately so that students from disadvantaged family backgrounds better 

integrate themselves into their peer group. 

 

Notes 

1. In the first stage of sampling, PISA employed probability proportional to size (PPS) to select sam-

ple schools. In the second stage of sampling, PISA randomly selected students from sample 

schools. 

2. For a path analysis model with continuous and categorical variables, model fit is acceptable if the 

weighted root mean-square residual (WRMR) is less than 0.9. For more detail, please refer to 

Schreiber, J. B., Nora, A., Stage, F. K., Barlow, E. A., & King, J. (2006). Reporting structural 

equation modeling and confirmatory factor analysis results: a review. The Journal of Educational 

Research, 99(6), 323-338. 

3. Results of the path analysis model estimate the correlation coefficient between parental support 

and teacher support as 0.150 (P < 0.001). This result was not included in Table 3. 

References

Analitis, F., Velderman, M. K., Ravenssieberer, 

U., Detmar, S., Erhart, M., & Herdman, M., 

et al. (2009) Being bullied: associated fac-

tors in children and adolescents 8 to 18 years 

old in 11 european countries. Pediatrics; 

123(2):569-577. 

Brown, S., & Taylor, K. (2008) Bullying, educa-

tion and earnings: evidence from the nation-

al child development study. Econ Edu Rev; 

27(4):387-401. 

Delprato, M., Akyeampong, K., & Dunne, M. 

(2017) The impact of bullying on students’ 

learning in Latin America: a matching ap-

proach for 15 countries. Int J Edu Dev; 

52:37-57. 

Due, P., Merlo, J., Harelfisch, Y., Damsgaard, M. 

T., Soc, M.S. , & Holstein B.E., et al. (2009) 

Socioeconomic inequality in exposure to 

bullying during adolescence: a comparative, 

cross-sectional, multilevel study in 35 coun-

tries. Am J Pub Health; 99(5):907-914. 

Hayes, & Andrew, F.. (2009) Beyond baron and 

kenny: statistical mediation analysis in the 

new millennium. Commun Monograph; 

76(4):408-420. 

Jansen, D.E., Veenstra R., Ormel, J., Verhulst, 

F.C., & Reijneveld, S.A.. (2011) Early risk 

factors for being a bully, victim, or bul-

ly/victim in late elementary and early sec-

ondary education. the longitudinal trails 

study. BMC Pub Health; 11(1):440. 

Kaltiala-Heino, R., Rimpel, M., Rantanen, P., & 

Rimpel, A. (2000) Bullying at school—an 

indicator of adolescents at risk for mental 

disorders. J Adolesc; 23(6):661-674.  

Lereya, S.T., Samara, M., & Wolke, D. (2013) 

Parenting behavior and the risk of becoming 

a victim and a bully/victim: a meta-analysis 

study. Child Abus Negl; 37(12):1091-1108. 

Liao, J.X., Jin, Y.L. (2017) The prevention of 

the American school bullying: history, expe-

rience and enlightenment. Edu Sci; 33(5):89-

96. 

Maynard, B.R., Vaughn, M.G., Salaswright, C.P., 

& Vaughn, S.R. (2016) Bullying victimiza-

tion among school-aged immigrant youth in 



 

 

Vol. 1, No. 1, 2019 27 

 

the united states. J Adol Health Offic Pub 

Soc Adol Med; 58(3):337-344. 

McNeal Jr, R.B., & Ralph, B. (1999) Parental 

involvement as social capital: differential ef-

fectiveness on science achievement, truancy, 

and dropping out. Soc Forc; 78(1):117-144. 

OECD (2016) PISA 2015 Assessment and Ana-

lytical Framework: Science, Reading, Math-

ematics and Financial Literacy, PISA, 

OECD Publishing, Paris. 

OECD (2017) PISA 2015 Results (Volume III): 

Students’ Well-Being, PISA, OECD Pub-

lishing, Paris. 

OECD (2009) PISA data analysis manual: SPSS, 

second edition. PISA, OECD Publishing: 72-

74. 

Olweus, D. (1994) Bullying at school: basic 

facts and effects of a school based interven-

tion program. J Child Psychol Psych, 

35(7):1171-1190. 

Rivara, F., & Le Menestrel, S. (2016) Preventing 

Bullying Through Science, Policy, and Prac-

tice. Washington, DC: The National Acade-

mies Press.  

Smith, P. & Morita, Y., Junger-Tas, J., Olweus, 

D., Catalano R.F., & Slee, P. (1999) The Na-

ture of School Bullying: A Cross National 

Perspective. London, United Kingdom: 

Routledge. 

Troop-Gordon, W., & Kopp, J. (2011) Teacher–

child relationship quality and children’s peer 

victimization and aggressive behavior in late 

childhood. Soc Dev; 20(3):536-561. 

Wang, M.C. (2014) Latent variable modeling 

using Mplus. Chongqing, P.R.China: 

Chongqing University Press, 77. 

Wang, W.D. (2016) How much school bullying 

that should not have occurred? Guangming 

Daily. Retrieved September 10, 2017 from 

http://news.gmw.cn/2016-

12/15/content_23265749.htm. 

Woods, S., & Wolke, D. (2004). Direct and rela-

tional bullying among primary school chil-

dren and academic achievement. J Sch 

Psychol; 42(2):135-155. 

Received: 27 October 2018 

Revised: 11 November 2018 

Accepted: 25 November 2018 

 

 

 

 

The Chinese version of this article has been published in Education Science 2018; 34(1):7-13. The 

English version has been authorized for being publication in BECE by the author(s) and the Chi-

nese journal. 

黄亮, 赵德成. 家庭社会经济文化地位与学生遭受校园欺凌关系的实证研究-家长支持和教师

支持的中介作用. 教育科学 2018; 34(1):7-13. 


	Article-Liang Huang_Title_20Jan2019
	Article-Liang Huang_Main text_20Jan2019

