







































 

 

 

 

 

 

Family Capitals and Education Acquisitions: 

Analysis on the Mediating Effect of 

Shadow Education 

Haiping Xue

Capital Normal University, Beijing 100048, China

Abstract. The purpose of this study is to explore whether family capitals 

and shadow education affect students’ education acquisitions. The study 

constructs the social reproduction theory mode of family capitals influ-

ence on children’s education acquisitions. Using 2014 Chinese educa-

tion tracking data, this study examined shadow education’s mediation 

effects on the process of family capitals affecting on education acquisi-

tions for middle school students. Family capitals and shadow education 

jointly affect the education acquisitions of students, in which, shadow 

education plays an intermediary role between family capitals and educa-

tion acquisitions for students. In sum, shadow education is increasingly 

becoming a supplement to school education and may become a new in-

termediary for family capitals influence on students' education acquisi-

tions. The study adds to the limited literature in this particular field con-

cerning the mediating effects of shadow education on family capitals and 

education acquisitions. 

Best Evid Chin Edu 2019; 1(2):87-107. 

Doi: 10.15354/bece.19.ar1019 

Keywords: Family capitals; Education acquisitions; Shadow education; Social re-

production; Mediation effect 

 

 
 

Correspondence to: Haiping Xue, Professor and Director of the Institute of Educational Economics and Manage-

ment, College of Education, Capital Normal University, China, Email: xuehaiping_416@163.com. 



Xue. Family Capitals and Education Acquisitions. 

Vol.1, No. 2, 2019 88 

 

Research Background and Problem 

AMILY socioeconomic background influence on education acquisitions is a 

major field of study, in China. When it comes to social equity research, many 

scholars in the country have examined the effects of family social, cultural, 

economic and political capital has on academic education opportunity acquisitions of 

children. It turns out that most students with higher family capitals have more chanc-

es to attain better academic education opportunities. Nowadays, there are plenty of 

Chinese compulsory education students participating in extracurricular tutoring, 

called “shadow education”. Shadow education and school education complement 

each other and jointly influence students’ education acquisitions. To date, the vast 

majority of empirical research on supplementary tutoring has only investigated the 

effects of family capitals/shadow education on academic education opportunity ac-

quisitions. Also, empirical research combining school and shadow education out-

comes concerning family capital and its related effects on student academic perfor-

mance and higher-level education acquisitions is rare. Even fewer studies have ex-

plored the mediating effects of shadow education on students’ education acquisitions 

under the affects of family capitals. This paper will establish a theoretical model us-

ing joint social reproduction to investigate how family capitals influence the mecha-

nisms of students’ education acquisitions through shadow education. At the same 

time, an examination of the mediating effects of shadow education on students’ edu-

cation acquisitions performed using family capitals will be conducted. From the per-

spective of shadow education, this paper plans to expand the research about family 

capitals and its effects on student education acquisitions. The results from this study 

will provide governments and education policy-makers another scientific theoretical 

reference for developing educational social equity initiatives.  

This paper is divided into five parts: first, a review of relevant empirical studies; 

second, an introduction to data sources and variables to be used in the study; third, an 

explanation of the constructs and theoretical model for family capitals on education 

acquisitions; fourth, an introduction to the structural equation modeling (SEM) meth-

od to be used; and, fifth, a summary of results, discussion, and implications of this 

study. 

Review of Existing Research 

Connotations and Classification of Family Capitals 

The concept of family capital was derived from the larger theories on social capital. 

 
 

Funding: National Natural Science Foundation of China - “Family Capital, Shadow Education and Social 

Reproduction” Project (Approval Number: 71774112); National Natural Science Foundation of China -- “Fac-

ing the Shadow Education System: Research on Supplementary Tutoring in Compulsory Education in China” 

Project (Approval Number: 71373165). 

Conflict of Interests: None. 

F 



Xue. Family Capitals and Education Acquisitions. 

Vol.1, No. 2, 2019 89 

 

Social capital in the modern era was first proposed in the late 1970s originating from 

the fields of social relations and social networks. French sociologist, Pierre Bourdieu, 

was one of the first researchers that devoted to systematically study the concept of 

social capital. In his view, “family” is the “cell” and basic unit of society. A family 

capital was seen as a means to providing various useful resources for educational 

activities for individual students. 

Coleman (1988), an American scholar, divided family capitals into material, 

human and social capitals. Material capital refers to the material resources that can 

promote the development of production. Human capital applies to the knowledge 

level and ability that people possess to change the society. Lastly, social capital is a 

social resource structure that exists in the interpersonal network and can be used as 

an asset to help individuals achieve their goals. Social capital is believed to be found 

inside and outside of the family construct. More specifically, it is present in parent-

child relationships, parents’ concerns, expectations, supports, investments and partic-

ipation in education of their children. Outside the family, concerns the interpersonal 

relationships within the community where the family is located. 

According to Li Chunling (2003), family background mainly deals with the 

family’s economic capital (family’s income level), cultural (father’s education level), 

social (father’s position level) and political capitals (family’s composition). Jiang 

Guohe and Yan Guangfen (2006) believed that family capitals mainly include family 

economic, cultural and social capitals. Liu Zhimin and Gao Yao (2011) understood 

family capitals as family socio-economic background, which included family eco-

nomic, social, cultural and political capitals. 

Influence of Family Capitals on School Education Acquisition 

Coleman’s (1988) work indicated that family social capital assists students to achieve 

higher education achievement. Teacherman (1987)’s study showed that social capital 

of American families measured by parents’ occupation had an important influence on 

children’s education prospects and academic performance in high school. Wong 

(1998) found that family social capital had a significant positive impact on children’s 

academic achievement in his research from the Czech Republic. James (2000), using 

Australian survey data from 1991 to 1997, indicated that children from families with 

low social capital usually attended lower level colleges and universities. 

In the 60 years from the 1940s to the 1990s, family social capital measured us-

ing the occupational status of fathers was viewed as an important impact on the stu-

dent’s education acquisition (Li, 2003). This theory began to change, in the 1980s, 

when research began revealing that the higher the occupational levels of fathers re-

sulted in the higher their children’s academic performance. Li Yu (2006) found that 

from 1966 to 2003, family social capital measuring fathers’ occupational status had 

different degrees of influence on the probability of their children going to high 

schools and colleges. During China’s Great Cultural Revolution (1966-1976), family 

social capital had little influence on the probability of children attending school. In 



Xue. Family Capitals and Education Acquisitions. 

Vol.1, No. 2, 2019 90 

 

the earlier years of educational reform (1977-1991), the family’s social capital on the 

probability of children attending school began playing as a factor. In later years 

(1992-2003), the influence of family social capital rose steadily, to the point, which it 

performed as a significant indicator for university entrance. For instance, Liu and 

Gao (2011) study, using survey data of 14 university students from Jiangsu province, 

learned that the relationship between family capitals and higher education acquisition 

was a significant factor.  

Family Cultural Capital on School Education Acquisition 

Bourdieu (1989) indicated that children with higher cultural capital, in French fami-

lies, received more and higher quality education. As well, Teacherman (1987) found 

that American parents with education degrees had a significant influence on their 

children’s education and academic achievements in high school. Based on TIMSS 

data from Hong, South Korea, Japan, Singapore, and Thailand, Woessmann (2003) 

saw results indicating similar significant positive impact on students’ performance. 

However, two studies, concluded otherwise, Robinson and Garnier (1985) and 

Katsillis and Rubinson (1990) found cultural capital to have little or no impact on 

student academic achievement.  

Family Economic Capital on School Education Acquisition 

American parents’ incomes play an important factor on their children’s education 

opportunities and academic performance in high school (Carneiro & Heckman, 2002; 

Teacherman, 1987). Wong (1998), in a study done in the Czech Republic, found sim-

ilar results indicating that family economic capital had a significant positive impact 

on students education achievement.  

Li (2003) discovered that from the 1940-1970, family economic capital had a 

weak impact on individual education acquisition; but, had a significant impact on 

some special groups (such as rural and female populations). In the years, 1970-1990, 

the relationship between the two concepts would have a greater impact on all popula-

tions. An urban family survey indicated if the family’s economic capital was at the 

top-tier in social class, the education level of their children was generally higher Guo 

& Min, 2006). Finally, Liu & Gao (2011) found that family economic income was a 

significant predictor of acquisition; but, had no significant impact on the quality of 

education acquisition. 

Family Political Capital on School Education Acquisition 

Bian, Wu and Li (2008), believed that political capital refers to the identity, power 

and resources provided by political parties and regimes, as well, as the deterrent and 

influence derived there from. Li (2003) uncovered that during the 1940s up until the 

1990s, the influence of family political capital on personal education acquisition was 

divided with the first 30 years showing a positive relationship between concepts; 



Xue. Family Capitals and Education Acquisitions. 

Vol.1, No. 2, 2019 91 

 

whereas, in the following 20 years, the influence of family political capital on chil-

dren’s education acquisition was just the opposite. Beginning in the 1990s, family 

political capital had no significant influence on individual education acquisition. 

Family political capital mainly played a role in the opportunity distribution of middle 

and higher education. Liu & Gao (2011), observed that, if the father’s political status 

was a member of the Communist Party of China, had a significant negative impact on 

the quantity of higher education acquisition of his children. Meanwhile, the father’s 

national administrative level had also an insignificant impact on the quantity of high-

er education acquisition of his children. Moreover, the father’s political status and 

national administrative level had a significant positive impact on the quality of his 

children’s acceptance to higher education institutions.   

Family Capitals on Shadow Education Acquisition 

Research by Smyth (2009) on Irish families saw that with more social capital a fami-

ly could afford more shadow education acquisition resulting in higher quality educa-

tion for their children. Hong & Zhao (2014) found that parental occupation hierarchy 

had a significant positive impact on whether urban compulsory education students 

participated in shadow education or not. Similar results were discovered in Zhou & 

Zou (2016) review using PISA 2012 data that analyzed the influence of family social 

capital in China and the United States and how children’s shadow education oppor-

tunity acquisition was impacted. Using a multi-layer linear model, results indicated 

that Chinese parents ranked at the highest occupational status index had significant 

positive opportunities for their children to getting supplementary tutoring in mathe-

matics, reading and science; meanwhile, American parents in the highest occupation-

al status index, would have no significant impact on their children receiving similar 

supplementary tutoring. 

Family Cultural Capital on Shadow Education Acquisition 

Tansel and Bircan-Bodur (2008) indicated that parents’ education levels have a sig-

nificant positive impact on supplementary tutoring expenditure for Turkish primary 

and secondary school students. A Chinese survey pointed out that the degree of edu-

cation affects the head of a household significantly and on whether urban students 

participated in supplementary tutoring (Xue & Ding, 2009). Parents’ education de-

grees and expectations had significant positive impacts on whether Chinese primary 

and secondary school students participated in supplementary tutoring and compulso-

ry education (Chu, 2009; Hong & Zhao, 2014; Mark Bray et al., 2014; Xue, 2015). 

However, other studies found differing results, education degrees and education ex-

pectations on the part of parents had no significant influence on whether students 

participated in shadow education (Buchmann et al., 2010; Zeng et al., 2010).    

Family Economic Capital on Shadow Education Acquisition 



Xue. Family Capitals and Education Acquisitions. 

Vol.1, No. 2, 2019 92 

 

Stevenson and Baker (1992) conducted a pioneering research on shadow education 

and discovered that family economic capital had a significant impact on whether Jap-

anese high school students participated in supplementary tutoring, which eventually 

led to a more likely chance of attending university for these particular students. Other 

studies discovered parallel findings, such as, family income significantly impacting 

supplementary tutoring (Bircan, 2008; Buchmann et al., 2010; Tansel and Bircan, 

2008; Xue & Ding, 2009). On the other hand, a survey completed in Gansu, Hunan, 

and Jiangsu provinces, Zeng et al. (2010) found that the monthly income of fathers 

has no significant impact on whether junior high school students chose supplemen-

tary tutoring. Xue’s (2015) study, also done in China, found similar results showing 

that net household income per capita of a family had no significant influence on 

whether students participated in supplementary tutoring.  

Influence of Shadow Education on School Education Acquisition 

Family capitals directly influence the students’ school and shadow education acquisi-

tions and, at the same time, students’ shadow education acquisitions influence their 

school education acquisitions. Since students’ shadow education will be helpful for 

them to improve their academic performance, also will position them favorably for 

higher education success.  

Studies on the influence of shadow education acquisition on school education 

acquisition have basically centered on the influence of shadow education acquisition 

on students’ academic performance at school. Due to the influence of individual, 

family, school, and other factors on students’ academic performance, in order to con-

trol the endogenous problems, some scholars, in China, and abroad have adopted 

advanced statistical methods to estimate the influence of supplementary tutoring on 

students’ academic performance. These studies, so far, have presented mixed results.  

Sunderman (2006) attempted to estimate the influence of supplementary tutor-

ing on Indonesian fourth-graders by using the instrumental variable method and 

found that the influence was not significant. As well, Dang (2007), made use of the 

instrumental variable method and using such data as the hourly after-class tuition 

fees paid by parents predicted the positive impact on student learning. ZHAO (2015), 

making use of the instrumental variable method, also, found that after-school tutoring 

expenditures, though small, showed an impact on the math performance. Yet, studies 

from Guill and Bonsen (2010), using a multilayer linear model to observe the influ-

ence of supplementary tutoring on grade 5 and 6 students, in Hamburg, Germany, 

saw no significance in student learning. Zhang (2013) and Hu et al. (2015) studies 

had similar findings.  

Summary of Existing Research 

Scholars have paid extensive attention to the influence of family capitals on educa-

tion opportunity acquisition in schools and most of the relevant research concludes 



Xue. Family Capitals and Education Acquisitions. 

Vol.1, No. 2, 2019 93 

 

that family capitals is an important influence on student learning. With the rise of 

shadow education, the influence of family capitals on shadow education acquisition 

has only begun to attract the attention of scholars, in recent years. Related studies 

tended to focus on the influence of family cultural capital and economic capital on 

the opportunity acquisition of shadow education, and rarely discussed the influence 

of family social capital and political capital. At present, in the compulsory education-

al system now present in China, shadow and school education complement each oth-

er.  

The vast majority of existing studies focused mainly on the family capitals im-

pact on school education acquisition or shadow education acquisition. Few studies 

combined school education and shadow education to investigate the influence of 

family capitals on the students’ academic performance and higher-level education 

acquisition. As well, few studies investigated the mediating effects of shadow educa-

tion on the influence of family capitals, students’ performance, and the acquisitions 

of higher-level education.  

Data Source and Variable Description 

The data used in this study came from the 2014 China Education Panel Survey 

(CEPS) conducted by the China Survey and Data Center of Renmin University of 

China. The variables used in the statistical analysis are described in Table 1. The 

variables of extracurricular tutoring in the dataset refer to the education activities of 

academic course supplements. The sample number of junior high school students in 

this data set was 19,487. 

According to the literature research results from above, this study, divides fami-

ly capitals into four capital categories: social, cultural, economic and political. Fami-

ly social capital measurements come from Li Chunling’s research on occupational 

stratification, which divides the occupation of students’ parents into upper, middle 

and lower levels. The parent with the higher occupational stratification was used to 

measure the index of student’s social capital outside the family. The higher the occu-

pational stratification of the parent, the more social capital outside the family the stu-

dent attained. 
1
Cultural capital is measured by the number of books in the family, 

outside of the parent’s highest education and textbooks and magazines. Economic 

capital is measured by family financial status, which can be divided into three levels: 

difficult, medium and rich. Political capital is measured by parents’ political status, 

which can be divided into three categories
2
: party members, democratic parties and 

the masses. 

From previous literature, education acquisitions of students mainly referred to 

the level and quality of education acquired by students. Generally, the difference of 

education acquisitions of students can be measured from two dimensions. One is the 

level of education received by students or the number of years receiving education, 

which represents the difference of education acquired by students. Second, to meas-

ure the education obtained by students is the difference in education quality under the 



Xue. Family Capitals and Education Acquisitions. 

Vol.1, No. 2, 2019 94 

 

same education level or years, usually measured by students’ grades or school quality. 

Although the quantity and quality of education are both important measures of edu-

cation acquisition, in studies of family capitals and student’s education acquisitions, 

affected by data availability, the researchers always paid much more attention to the 

number than to the quality of education acquisition. Most studies were not involved 

in aspects of the difference of quality education, which may be because the quality of 

education information data is difficult to collect. Different from existing studies, this 

paper will mainly focus on the influence of family capitals on quality of students’ 

education acquisitions, which is measured by students’ academic performance at 

school. The variables involved in the article are described in Table 1 below. 

Theoretical Construction Model of the Influence of Family Capitals on Edu-

cation Acquisition 

According to existing literature studies, family capitals has an important impact on 

acquisitions of school education and shadow education; but few studies combine 

them to investigate the impact of family capitals on education acquisitions of chil-

dren. It will underestimate the influence of family capitals on children’s education 

acquisitions if they are investigated separately, and the mechanisms by which family 

capitals affect children’s education acquisitions will also not be fully revealed. In 

order to fully and accurately evaluate the influence of family capitals on children’s 

education acquisitions, this study, combined school education and shadow education 

to construct a theoretical model of the influence of family capitals on children’s edu-

cation acquisitions. Based on Li’s (2006) study on the mechanisms of education ine-

quality and Liu & Gao’s (2011) study on the mechanisms of family capitals influ-

ences on acquisition of higher education quantity and quality, this study, constructs 

the theoretical model of family capitals impact on acquisition of education (see Fig-

ure 1). Firstly, family capitals can directly influence children’s academic achieve-

ment at school. Secondly, family capitals can also influence their children’s academic 

performance at school by influencing the child’s access to education opportunities 

(for example, go to a better school) (see corresponding solid arrows in Figure 1). 

Thirdly, family capitals can also influence their children’s academic performance at 

school by influencing their shadow education opportunity advantage (see correspond-

ing dotted arrows in Figure 1). Finally, advantage opportunity in children’s school 

education acquisitions will also affect their advantage opportunity acquisitions in 

shadow education (see corresponding part of the dotted line arrow in the Figure 1). 

The mechanism faction that family capitals affects is their children’s education 

acquisitions that are essentially a mechanism of resource exclusion, meaning that the 

advantage class families use their family capital advantage to reduce competition for 

entrance, or to exclude some competitors from the competition so to increase oppor-

tunities of accessing high-quality school education and shadow education. This kind 

of exclusion is embodied in four types: economic resource, privileged resource, im-

plicit resource, and cultural resource. Economic resource exclusion, which is the 



Xue. Family Capitals and Education Acquisitions. 

Vol.1, No. 2, 2019 95 

 

most common form in modern society, mainly refers to the rich class’s exchange of 

family economic capital for high-quality education resources. Privileged resource 

exclusion refers to the fact that in the design of education system, special positions 

are always reserved for some privileged classes. The setting of high-quality educa-

tion resources often favors some privileged classes. Implicit exclusion of resources 

means that when making decisions on admission to higher education and shadow 

education, vulnerable class families have to withdraw from competition for shadow 

and better school education opportunities due to their poor abilities to bear the risk of 

low evaluation on the expected return of education. Lastly, the exclusion of cultural 

resources indicates that families with parents, who have low educational backgrounds 

and poor cultural resources, do not attach great importance to their children’s educa-

tion, so therefore they may withdraw from the competition for shadow and better 

school education opportunities. 

Based on the theoretical model of family capitals influencing education, in order 

to test the mediating effect of shadow education on family capitals and education 

acquisition, the following research hypotheses are as follows: 

Hypothesis 1: Family capitals have a significant positive impact on education acquisi-

tions of junior high school students. 

Hypothesis 2: Whether participation in shadow education has a significant positive 

impact on education acquisitions of junior high school students. 

Hypothesis 3: Shadow education plays an intermediary role in the process of family 

capitals influence on students’ education acquisitions, that is, family cap-

itals influences students’ education acquisitions by influencing their 

shadow education participation opportunities. 

Analysis of the Mediating Effect of Shadow Education 

Based on the theoretical model and research hypothesis established, this study will 

use structural equation models to analyze influences of family capitals to students’ 

education acquisitions and test mediating effects of family capitals to students’ edu-

cation acquisitions through shadow education. 

Intermediary Model of Family Cultural Capital on Students’ Education Acqui-

sitions 

Using the theoretical model and Wen et al.’s research to verify the methods and pro-

cedures for mediating effects, this study constructs an intermediary model for family 

cultural capitals impact on students’ education acquisitions (the normalized regres-

sion path coefficient results of the model are shown in Figure 2). Results indicated 

that the path coefficients passed the significance level test except for the path coeffi-

cient of shadow education participation affects to students’ performance. According 

to the joint significance test, the mediating effect of family cultural capital to students’ 

achievements is not significance; shadow education does not play an intermediary 

role between family cultural capitals and students’ performance. 



Xue. Family Capitals and Education Acquisitions. 

Vol.1, No. 2, 2019 96 

 

Table 1. Description of Variables in Statistical Analysis. 

Variable 

Type 
Variable Name Variable Description 

Shadow 

Education 

Whether to participate in extracur-

ricular tutoring 
0=No,1=Yes 

Education 

Acquisition 
Student achievement 

The total score of standardized scores of stu-

dents in the mid-term examination of foreign 

languages and numerals in 2013 

Family 

Capitals 

Social 

capital 

Parents have the 

highest career stratifi-

cation 

1= lower, 2= middle, 3= upper 

Cultural 

capital 

Highest degree of 

parents 

1= illiteracy, 2= primary school,  3= junior high 

school, 4= high school, 5= junior college, 6= 

university undergraduate, 7= graduate student 

and above 

Books, magazines 

outside the family 

library 

1=little, 2= less, 3= average, 4= more, 5= a lot 

Parents' expectations 

of their children's edu-

cation level 

1= don’t read now, 2= junior high school gradua-

tion, 3= high school, 4= junior college, 5= Big 

Ben, 6= graduate student, 7= doctor 

Economic 

capital 
Family financial status 1= difficult, 2= medium, 3= rich 

Political 

capital 

Political status of par-

ents 

1= communist party member 2= Democratic 

Party 3= masses 

Individual 

Factors 

Gender 0 = female, 1 = male 

Grade 0= seventh grade, 1= ninth grade 

 

 

Table 2. Fitting Results of the Intermediary Model of Family Cultural Capital 

Influencing Students’ Education Acquisitions. 

Fitting Index CMIN DF CMIN/DF RMSEA NFI RFI IFI TLI CFI 

Numerical 358.10 12 29.84 0.038 0.967 0.922 0.968 0.925 0.968 

Judgment 

Standard 
  <5.0 <0.10 >0.90 >0.80 >0.90 >0.80 >0.90 

 

 

Table 3. Results of the Intermediary Model – Family Social Capital Influencing 

Students’ Education Acquisitions. 

Fitting Index CMIN DF CMIN/DF RMSEA NFI RFI IFI TLI CFI 

Numerical 140.01 5 28.00 0.034 0.957 0.919 0.959 0.875 0.959 

Judgment 

Standard 
  <5.0 <0.10 >0.90 >0.80 >0.90 >0.80 >0.90 

 



Xue. Family Capitals and Education Acquisitions. 

Vol.1, No. 2, 2019 97 

 

Table 4. Results of the Intermediary Model – Family Economic Capital Influ-

encing Students’ Education Acquisitions. 

Fitting Index CMIN DF CMIN/DF RMSEA NFI RFI IFI TLI CFI 

Numerical 59.38 5 11.87 0.02 0.970 0.910 0.972 0.917 0.972 

Judgment 

Standard 
  <5.0 <0.10 >0.90 >0.80 >0.90 >0.80 >0.90 

 

 

Table 5. Results of the Intermediary Model – Family Political Capital Influenc-

ing Students’ Education Acquisitions. 

Fitting Index CMIN DF CMIN/DF RMSEA NFI RFI IFI TLI CFI 

Numerical 59.47 5 11.89 0.02 0.964 0.892 0.967 0.900 0.967 

Judgment 

Standard 
  <5.0 <0.10 >0.90 >0.80 >0.90 >0.80 >0.90 

 

 

Table 6. Results of the Intermediary Model – Family Capitals Influencing Stu-

dents’ Education Acquisitions. 

Fitting Index CMIN DF CMIN/DF RMSEA NFI RFI IFI TLI CFI 

Numerical 1136.66 2 35.52 0.042 0.950 0.914 0.951 0.916 0.951 

Judgment 

Standard 
  <5.0 <0.10 >0.90 >0.80 >0.90 >0.80 >0.90 

 

 

 

The results show that the model is recursive, with a sample size of 19487 (the 

results of model fitting are shown in Table 2). It can be seen from the table that all 

other fitting indexes reach a good level except for CMIN/DF=29.84>5, this indicates 

that the model can fit well with the actual observation data and the established model 

for family cultural capital influencing education acquisition. In all, the theoretical 

hypothesis model is stable. The likeliest reason why CMIN/DF>5 rejects theoretical 

model 1 is the large number of valid samples (N=19,487). When the sample size is 

very large, all the models that fit well with the data will be rejected, so whether to use 

the model cannot be decided through a chi-square test. Fitting index cannot be the 

main basis for comparing models, instead, the rationality of the relationships between 

variables described in models and the appropriateness of parameters’ estimation 

should be considered (for example, the correlation coefficient cannot be greater than 

1, and the error variance cannot be a negative number). The relationship between 

variables described in this model is reasonable and the parameters’ estimation is ap-

propriate. Therefore, this study decided to use this model to test the mediating effect 

of family cultural capitals on students’ education through shadow education. 



Xue. Family Capitals and Education Acquisitions. 

Vol.1, No. 2, 2019 98 

 

Intermediary Model of Family Social Capital Influence on Students’ Education Ac-

quisitions 

For family capitals influence on education acquisition, this study, constructed the 

intermediary model that family social capital influences education acquisitions of 

students. All the path coefficients in the model past the significance level test (the 

results of standardized regression path coefficient of the model are shown in Figure 

3). According to the joint significance test, family social capital has a significant me-

diating effect on students’ performance and shadow education plays a partial mediat-

ing role between family social capital and students’ performance. 

The results display a model that is recursive at a sample size of 19,487 (Table 

3). Except for CMIN/DF = 28.00 > 5, all other fitting indexes reached a good fit level 

indicating that the model can be well fitted with the actual observation data. The rela-

tionships between variables described in this model were reasonable and the parame-

ters’ estimation as appropriate. So, this study decided to use the model to test the me-

diating effect of families’ social capital on students’ education through shadow edu-

cation. 

Intermediary Model of Family Economic Capital Influence on Students’ Education 

Acquisitions 

This study measured family capitals influence on education acquisition. The interme-

diary model shows family economic capital influences education acquisitions of stu-

dents (the results of standardized regression path coefficient of the model are shown 

in Figure 4). All path coefficients in the model past the significance level test except 

family economic capital influence students’ performance. The joint significance test 

revealed that family economic capital had a significant mediating effect on students’ 

performance and shadow education played a complete mediating role between family 

economic capital and students’ performance. 

The results of the model analysis suggest that the model is recursive with a 

sample size of 19,487 (the results of model fitting are shown in Table 4). All, except 

CMIN/DF = 11.87 > 5, reached a good fit level indicating that the model can be well 

fitted with the actual observation data. The relationships between variables described 

in this model were reasonable and the parameters’ estimation as appropriate. The 

model was used to test the mediating effect of families’ economic capital on students’ 

education through shadow education. 

Intermediary Model of Family Political Capitals Influence on Students’ Education 

Acquisitions  

This study constructed the intermediary model to test family political capital influ-

ence on education acquisitions of students. The results of standardized regression of 

path coefficients show that all the path coefficients in the model have past the signif-

icance level test (are shown in Figure 5). Family political capital had a significant 



Xue. Family Capitals and Education Acquisitions. 

Vol.1, No. 2, 2019 99 

 

Figure 1. Theoretical Model of Family Capitals Influencing Education. 

 

 

 

Figure 2. Multiple Intermediary Model of Family Cultural Capital Influencing 

Students’ Education Acquisitions. 

 

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

 



Xue. Family Capitals and Education Acquisitions. 

Vol.1, No. 2, 2019 100 

 

Figure 3. Model of Family Social Capital Influencing Students’ Education Ac-

quisitions. 

 

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

 

 

 

 

 

Figure 4. Model of Family Economic Capital Influencing Students’ Education 

Acquisitions. 

 

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

 

 

 

 

 



Xue. Family Capitals and Education Acquisitions. 

Vol.1, No. 2, 2019 101 

 

Figure 5. Model of Family Political Capital Influencing Students’ Education 

Acquisitions. 

 

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

 

 

Figure 6. Model of Family Capitals Influencing Students' Education Acquisi-

tions. 

 

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

 

 



Xue. Family Capitals and Education Acquisitions. 

Vol.1, No. 2, 2019 102 

 

mediating effect on students’ performance and shadow education played a partial 

mediating role between family political capital and students’ performance. 

The results from the model analysis signified that the model is recursive with a 

sample size of 19,487. Except for CMIN/DF = 11.89 > 5, all other fitting indexes 

have reached a good fit level indicating that the model can be well fitted with the 

actual observation data (he results of model fitting are shown in Table 5). The estab-

lished model of family political capital influencing education was a good theoretical 

hypothesis model because CMIN/DF>5 rejects theoretical model 1 and was the larg-

est number of valid samples in the model statistically analyzes (N=19,487). The rela-

tionships between variables described in this model are reasonable and the parame-

ters’ estimation was appropriate. So it was decided to use this particular model to test 

the mediating effect of family political capitals on students’ education through shad-

ow education. 

Intermediary Model of Family Capitals Influence on Students’ Education Acquisi-

tions 

The results from the standardized regression path coefficients suggest that the model 

past the significance level test (Figure 6). Family capitals had a significant mediating 

effect on students’ performance and shadow education played a partial mediating role 

between family capitals and students’ performance. 

The results of model analysis showed that the model was recursive with a sam-

ple size of 19,487 (see Table 6). Except for CMIN/DF = 35.52 > 5, all other fitting 

indexes reached a good fit level. 

Conclusion and Policy Recommendations 

Conclusion 

This paper combines school education and shadow education to investigate the influ-

ence of family capitals on junior high school students’ education acquisitions. The 

following are the main research conclusions. 

A family capital has a significant positive impact on education acquisitions on 

junior high school students, which, therefore, supports hypothesis 1. In terms of all 

types of family capitals, family cultural, family social, and family political all have a 

direct and significant positive effect on junior high school students’ education acqui-

sitions. However, family economic capital has no significant direct impact on junior 

high school students’ academic education acquisitions.  

Participating in shadow education has a significant positive impact on education 

acquisitions of junior high school students, as well, thus, supporting hypothesis 2. In 

the mediating effects of family capitals and total family capitals affecting students’ 

education acquisitions, participating in shadow education had a significant positive 

impact on junior high school students’ education acquisitions in all categories. 



Xue. Family Capitals and Education Acquisitions. 

Vol.1, No. 2, 2019 103 

 

Shadow education plays a partial mediation intermediary role in the process of 

family capitals influencing students’ education acquisitions. More specifically, fami-

ly capitals influences students’ education acquisitions by influencing their shadow 

education participation opportunities, and therefore, supporting hypothesis 3 of this 

study. Shadow education has different mediating effects among different types of 

family capitals and education acquisitions of students. It does not perform an inter-

mediary role between family cultural capital and students’ education acquisitions. 

The reason may be that families with high cultural capital can provide more direct 

guidance and motivation for their children’s learning and such, families are less de-

pendent on shadow education to improve children’s academic performance. Shadow 

education also showed an intermediary role between family economic capital and 

students’ education acquisitions. Families with high economic capital cannot provide 

direct support and motivation for their children to study on their own, but have the 

ability to buy more shadow education for their children and this largely relies on the 

shadow education to raise their children’s academic performance. Results indicated 

that shadow education functions as an intermediary between family social capital and 

students’ education acquisitions and political capital and students’ education acquisi-

tions. Families with high social capital and political capital not only provide direct 

support and motivation for their children’s learning, but also rely on shadow educa-

tion to improve their children’s academic performance. 

A family capital gets its status through the influence of both school education 

and shadow education on children’s education acquisitions. Family capitals influence 

on junior high school students’ education acquisitions through school education and 

shadow education systems displays the differences in opportunities. Families with 

higher education levels, ultimately, have their children attain higher social stratifica-

tions. Children growing-up in families with more capitals are more likely to enter the 

primary labor market, whereas, children coming from families with poor capital are 

more likely to enter the secondary labor market. Therefore, family capitals are trans-

mitted from the fathers to the children. Intergenerational inheritance of family capi-

tals is realized, and hence, this process of intergenerational inheritance of family cap-

itals is seen in school education and shadow education. This educational social struc-

ture leads to solidifying of class right-up to the job market. Classic social reproduc-

tion theorists, such as, Coleman and Bourdieu, focus on the influence of family capi-

tals on education acquisitions in schools. They elaborate on the mechanisms of edu-

cation social reproduction in schools, but ignore the mechanisms of shadow educa-

tion social reproduction. Currently, shadow education has become an important sup-

plement to education in schools, globally. Shadow education acquisition is not only 

an important form of student education acquisition, but also, has an important impact 

on student school education acquisitions. Therefore, to only investigate the influence 

of family capitals on education acquisition on school education will eventually un-

derestimate the influence of family capitals on education acquisitions of children and 

fully reveal the current mechanisms of education social reproduction. This study 

finds that the shadow education plays an intermediary role between family capitals 



Xue. Family Capitals and Education Acquisitions. 

Vol.1, No. 2, 2019 104 

 

and education acquisitions. It also, ravels the joint mechanism of social reproduction 

between school education and shadow education, to a certain extent. In all, this study 

aims to assist in the development and improvement of the classic education social 

reproduction theory, for the current implementation of education fairness and the 

social mobility policy, in China, has certain significance. 

The influence of shadow education on students’ education acquisitions can be 

divided into short-term and long-term impacts. The short-term impact of shadow ed-

ucation on students’ learning is mainly reflected in the promoting of students’ aca-

demic performance in school. While, the long-term impact of shadow education on 

students’ learning is attained in helping students to enter better schools.  

Limited by data, this study only discusses the short-term impact of shadow edu-

cation on the acquisition of family capitals and education. Therefore, we will analyze 

whether shadow education can help students to attend better schools. And, in subse-

quent studies, explore the long-term impact of shadow education between family 

capitals and education acquisition. 

Policy Suggestions 

Limiting the influence of family capitals on education acquisition and giving more 

emphasis to the role of education in promoting social mobility should become im-

portant responsibilities of governments around the world to achieve social equity. 

Based on the conclusions and discussions of this paper, the policy implications of 

this study are as follows. 

Governments should take notice of the importance of shadow education and its 

influence on family capitals and students’ education acquisitions. Governments can 

make policies, in particular for shadow education to reduce its impact and create 

more equal schooling. In recent years, Chinese governments at all levels have paid 

more attention to creating a fairer educational system that can control such dilemmas 

as shadow education. As well, new policies have been adopted to weaken the influ-

ence of family capitals. The hope is to lessen the role that the dominant class current-

ly holds over the educational system and, thus, open new opportunities to disadvan-

taged groups.  

In order to maintain their competitive advantage of entering the best schools, the 

dominant class has increasingly attached importance to use of the shadow education 

system and this has resulted in the exclusion of groups with less power. In other 

words, disadvantaged groups stuck in the inescapable intergenerational inheritance of 

family capitals are left with a school education and shadow education that is com-

bined with social reproduction mechanisms. The result is severely weakened gov-

ernment policies on education and a damaged foundation for social equality.  

At present, governments at all levels are fighting the logic that shadow educa-

tion is an outcome of market behavior. Therefore, most lawmakers have a laissez-

faire attitude towards shadow education and so generating a “if it isn’t broke, don’t 

fix it” approach to the current situation. However, governments should focus more on 

the intermediary role of shadow education and family capitals on students’ education 



Xue. Family Capitals and Education Acquisitions. 

Vol.1, No. 2, 2019 105 

 

acquisitions, forming initiatives to establish an educational system that will weaken 

the impact of shadow education through compulsory education fair policies. 

The intermediary role of shadow education between family capitals and students’ 

education acquisitions initiates the joint social reproduction mechanism of school and 

shadow education. Education’s social reproduction function has also been further 

enhanced, making it easier for social strata to replicate and solidify. In order to break 

the class solidification and promote social equality, Chinese governments can also 

provide shadow education subsidies to families of junior high school students with 

poor academic performance and family capitals to weaken the negative impact of 

shadow education. Developed countries like Australia, England, France, Singapore 

and the United States have dealt with shadow education through measures, such as, 

remedial vouchers and allowances and tax breaks. Chinese governments can learn 

from these practices and bring forth explicit policies relevant to China to combat 

shadow education in the hopes to providing an equal opportunity education for all 

Chinese students.  

 

 

Notes 

1 The upper professional class includes: heads of government agencies, enterprises and public 

institutions, middle management personnel, middle and senior professional and technical per-

sonnel, cadres and civil servants of general government agencies, economic business person-

nel, private business owners and employees or above. The middle class includes: grass-roots 

management personnel, other professional and technical personnel/general technicians, em-

ployees of enterprises and institutions, technical workers, military police fire fighters, self-

employed/small business owners (8 employees or less). The lower occupational class includes: 

commercial and service personnel, non-skilled workers, agricultural, forestry, animal hus-

bandry and fishery personnel, freelancers, unemployed and laid-off household workers. 

2 Since this survey did not specifically ask parents about their political status, it asked respond-

ents about their political status. However, since respondents may be people other than their 

parents, parents rather than parents’ political status are used to measure family political cap-

ital. Statistical analysis shows that 92.5% of respondents are parents, 5.3% are grandparents, 

and the rest are other identities. 

 

 

 

References

Bian, Y., Wu, X. & Li, Lulu. (2008). Social 

stratification and mobility: The overseas 

scholar’s advanced research on China. 

Renmin University of China Press. (in 

Chinese) 

Bourdieu, P. (1989). Cultural capital and so-

cial capital. Zhang Renjie translation. Se-

lected Basic Articles of Education Sociol-

ogy Abroad. Shanghai: East China Nor-

mal University Press. (in Chinese) 

Bray, M. (2012). Confronting the shadow 

education system: What government poli-

cies for what private tutoring? Edu Rev, 

26(1), 132-133. 



Xue. Family Capitals and Education Acquisitions. 

Vol.1, No. 2, 2019 106 

 

Bray, M., Zhan, S., Lykins, C., Wang, D., & 

Kwo, O. (2013). Differentiated demand 

for private supplementary tutoring: Pat-

terns and implications in Hong Kong sec-

ondary education. Econ Edu Rev, 38(1), 

24-37. (in Chinese) 

Buchmann, C., Condron, D., & Roscigno J. 

(2009). Shadow education American style: 

Test preparation, the SAT and college en-

rollment. Soc Force, 89(2), 435-461. 

Carneiro, P. & Heckman (2002). The evi-

dence on credit constraints in post-

secondary schooling. Econ J, 112(482), 

705-734. 

Chu, H. (2009). Background characteristics 

and personal factors of extracurricular tu-

toring families of primary and secondary 

school students in China. Edu Res Month-

ly,26 (12), 22-27. (in Chinese) 

Coleman, J.S. (1988). Social capital in the 

creation of human capital. Am J Sociol, 

(supplement),94, 95-120. 

Dang, H. A. (2007). The determinants and 

impact of private tutoring classes in Vi-

etnam. Econ Edu Rev, 26(6), 683-698. 

Dobrzańsk, L.A., Gołombek, K., & 

Hajduczek, E. (2000). Parental cultural 

capital and educational attainment in the 

Netherlands: A refinement of the cultural 

capital perspective. Sociol Edu, 73(2), 92-

111. 

Fang, C., & Feng, X. (2005). How distinction 

of social stratum affect the attainment of 

education: An analysis on split flows of 

education. Tsinghua J Edu, 26(5), 22-30. 

(in Chinese) 

Guill, K. & Bonsen, M. (2013). 

Leistungsvorteiledurch 

Nachhilfeunterricht in Mathematik am 

beginn der sekundarstufeI? 

Unterrichtswissenschaft, 38(2), 117-133. 

Guo, C., & Min, W. (2006). The effect of 

economical and cultural capital on educa-

tional attainment in China. J Higher Edu, 

27 (11), 24-31. (in Chinese) 

Hong, Y., & Zhao, Y. (2014). From capital to 

habitus: The class differentiation of fami-

ly educational pattern in urban China. So-

ciol Stud, 29(4), 73-93. (in Chinese) 

Hou, J. Wen, Z., & Cheng, Z. (2004). Struc-

tural equation model and its application. 

Beijing: Education Science Press. (in 

Chinese) 

Hu, Y., Fan, W., & Ding, W. (2015). Does 

“shadow education” expand education in 

an unequal way? -- empirical study based 

on PISA 2012 Shanghai data. Peking 

Univ Edu Rev, 13(3), 40-59. (in Chinese) 

James, R. (2000). Non-traditional students in 

Australian higher education: persistent in-

equities and the new ideology of student 

choice. Tertiary Edu Manag, 6(2), 105-

118. 

Jiang, G., & Yan, G. (2006). Correlation 

analysis on family capital and educational 

attainment disparity between rural and ur-

ban. Edu Sci, 22(8), 26-34. (in Chinese) 

Katsillis, J., & Rubinson, R. (1990). Cultural 

capital, student achievement and Educa-

tional reproduction: The case of Greece. 

Am Sociol Rev, 55(2), 270-279. 

Li, C. (2003). Social political changes and 

inequality of educational opportunity. Soc 

Sci Chin, 24(3), 86-98. (in Chinese) 

Li, Yu. (2006). The mechanism of institution-

al change and education inequality -- edu-

cation acquisition of urban children in 

China (1966-2003). Chin Soc Sci, 27(4), 

97-109. (in Chinese) 

Liu, Z., & Gao, Y. (2011). Family capital, 

social stratification and the attainment of 

higher education -- An empirical study 

based on Jiangsu province. J Higher Edu, 

32(12) 18-27. (in Chinese) 

Melot, L. (2010). Lemarché du soutiens 

colaire [EB/OL]. Retrieved from 

http://www.xerfi.fr/ etudes/7SME04.pdf. 

Robinson, R.V., & Maurice, A.G. (1985). 

Class reproduction among men and wom-

en in France: reproduction theory on its 

home ground. Am J Sociol, 91(2), 250-

280. 

Siu, L. (2008). Issues and procedures in 

adopting structural equation modeling 

technique. J Appl Quantitat Method, 3(1), 

76-83. 

Smyth, E. (2009). Buying your way into col-

lege? Private tuition and the transition to 

higher education in Ireland. Oxford Re-

view of Education, 35(1), 1–22. 

Stevenson, D.L. & Baker, D.P. (1992). Shad-

ow education and allocation in formal 

schooling: Transition to university in Ja-

pan. Am J Sociol, 97(6), 1639-1657. 

Sunderman, G.L. (2006). Do supplemental 

educational services increase for minority 

students? Phi Delta Kappan, 88(2), 117-

122. 

Tan, J. (2009). Private tutoring in Singapore: 

Bursting out of the shadows. J Youth 

Stud,12(1), 93-103. 

Tansel, A., & Bircan-Bodur, F. (2008). Pri-

vate supplementary tutoring in Turkey: 

Recent evidence on its various aspects. 

Soc Sci Electronic Pub, 9(1), 162-171. 



Xue. Family Capitals and Education Acquisitions. 

Vol.1, No. 2, 2019 107 

 

Teacherman, J. (2000). Parental cultural capi-

tal and educational attainment in the 

Netherlands a refinement of the cultural. 

Am Sociol Rev, 73(2), 92-111. 

Tsang, M., Ding X., & Shen, H. (2010). Ur-

ban-Rural disparities in tutoring of lower-

secondary students. Edu Economy, 26(2), 

7-11. (in Chinese) 

Wen, Z., Zhang L., Hou, J., & Liu H. (2004). 

Testing and application of the mediating 

effect TS. ActaPsychol Sinica, 49 (5), 

614-620. (in Chinese) 

Woessmann, A. (2010). Educational produc-

tion in East Asia: The impact of family 

background and schooling policies on 

student performance. German Econ Re, 

6(3), 331-353. 

Wong, R.S.K. (1998). Multidimensional in-

fluences of family environment in educa-

tion: The case of socialist Czechoslovakia. 

Sociol Edu, (1),1-22. 

Xue, H. (2015). From school education to 

shadow education: Education competition 

and social reproduction. Peking Univ Edu 

Rev,13 (3), 47-69. (in Chinese) 

Xue, H. (2016). Extracurricular tutoring, stu-

dent achievements and social reproduc-

tion. Edu Economy, 32(2), 32-43. (in 

Chinese) 

Xue, H. & Ding, X. (2009). A study on addi-

tional instruction for students in cities and 

towns in China. Edu Res, 31 (1), 39-46. 

(in Chinese) 

Zhang, G. (2015). Can money ‘Buy’ school-

ing achievement? Evidence from 19 Chi-

nese cities. Chin Econ Rev, 35: 83-104. 

(in Chinese) 

Zhang, Y. (2013). Does private tutoring im-

prove students’ national college entrance 

exam performance? A case study from Ji-

nan, China. Econ Edu Rev, 32(1), 1-28. 

(in Chinese) 

Zhou, J., & Zou, X. (2016). Comparing the 

private tutoring options between students 

in China and the United State -- Evidenc-

es from 2012 PISA survey and investiga-

tion. Edu Economy, 32(2), 44-52. (in 

Chinese). 

Received: 27 February 2019 

Revised: 15 March 2019 

Accepted: 21 March 2019 

 

 

The Chinese version of this article has been published in Education & Economics 2018; 

34(4):69-78. The English version has been authorized for being publication in BECE by the 

author(s) and the Chinese journal. 

薛海平. 家庭资本与教育获得: 基于影子教育中介效应分析. 教育与经济 2018; 34(4):69-78. 


	Article-HaipingXue-BECE_30March2019
	Article-HaipingXue-MainText-BECE_30March2019

